Efficacy and experiences of telephone counselling for informal carers of people with dementia

Abstract

Background

Informal carers of people with dementia can suffer from depressive symptoms, emotional distress and other physiological, social and financial consequences.

Objectives

This review focuses on three main objectives:

To:

1) produce a quantitative review of the efficacy of telephone counselling for informal carers of people with dementia;

2) synthesize qualitative studies to explore carers’ experiences of receiving telephone counselling and counsellors’ experiences of conducting telephone counselling; and

3) integrate 1) and 2) to identify aspects of the intervention that are valued and work well, and those interventional components that should be improved or redesigned.

Search methods

The Cochrane Dementia and Cognitive Improvement Group's Specialized Register, The Cochrane Library, MEDLINE, MEDLINE in Process, EMBASE, CINAHL, PSYNDEX, PsycINFO, Web of Science, DIMDI databases, Springer database, Science direct and trial registers were searched on 3 May 2011 and updated on 25 February 2013. A Forward Citation search was conducted for included studies in Web of Science and Google Scholar. We used the Related Articles service of PubMed for included studies, contacted experts and hand‐searched abstracts of five congresses.

Selection criteria

Randomised controlled trials (RCTs) or cross‐over trials that compared telephone counselling for informal carers of people with dementia against no treatment, usual care or friendly calls for chatting were included evaluation of efficacy. Qualitative studies with qualitative methods of data collection and analysis were also included to address experiences with telephone counselling.

Data collection and analysis

Two authors independently screened articles for inclusion criteria, extracted data and assessed the quantitative trials with the Cochrane 'Risk of bias' tool and the qualitative studies with the Critical Appraisal Skills Program (CASP) tool. The authors conducted meta‐analyses, but reported some results in narrative form due to clinical heterogeneity. The authors synthesised the qualitative data and integrated quantitative RCT data with the qualitative data.

Main results

Nine RCTs and two qualitative studies were included. Six studies investigated telephone counselling without additional intervention, one study combined telephone counselling with video sessions, and two studies combined it with video sessions and a workbook. All quantitative studies had a high risk of bias in terms of blinding of participants and outcome assessment. Most studies provided no information about random sequence generation and allocation concealment. The quality of the qualitative studies ('thin descriptions') was assessed as moderate. Meta‐analyses indicated a reduction of depressive symptoms for telephone counselling without additional intervention (three trials, 163 participants: standardised mean different (SMD) 0.32, 95% confidence interval (CI) 0.01 to 0.63, P value 0.04; moderate quality evidence). The estimated effects on other outcomes (burden, distress, anxiety, quality of life, self‐efficacy, satisfaction and social support) were uncertain and differences could not be excluded (burden: four trials, 165 participants: SMD 0.45, 95% CI ‐0.01 to 0.90, P value 0.05; moderate quality evidence; support: two trials, 67 participants: SMD 0.25, 95% CI ‐0.24 to 0.73, P value 0.32; low quality evidence). None of the quantitative studies included reported adverse effects or harm due to telephone counselling. Three analytical themes (barriers and facilitators for successful implementation of telephone counselling, counsellor's emotional attitude and content of telephone counselling) and 16 descriptive themes that present the carers’ needs for telephone counselling were identified in the thematic synthesis. Integration of quantitative and qualitative data shows potential for improvement. For example, no RCT reported that the counsellor provided 24‐hour availability or that there was debriefing of the counsellor. Also, the qualitative studies covered a limited range of ways of performing telephone counselling.

Authors' conclusions

There is evidence that telephone counselling can reduce depressive symptoms for carers of people with dementia and that telephone counselling meets important needs of the carer. This result needs to be confirmed in future studies that evaluate efficacy through robust RCTs and the experience aspect through qualitative studies with rich data.

Plain language summary

Efficacy and experiences of telephone counselling for informal carers of people with dementia

What is the problem?

Caring for a person with dementia often has a negative impact on an informal carer’s mental and physical health and social life. Therefore, these informal carers should be offered support. They are usually family members and care for the person with dementia at home.

What is telephone counselling?

Eliciting a person’s concerns, listening, and providing support, information, or teaching in response to a persons’s stated concerns, over the telephone.

The purpose of this review

The aim of this review was to investigate whether telephone counselling is an effective way of reducing symptoms of depression and other stresses in the carers of people with dementia. We also investigated which aspects of telephone counselling the people who received it thought could be improved.

Results of the review

We searched for scientific studies that compared telephone counselling with no treatment, or usual care or friendly calls. We found nine studies that investigated how well telephone counselling worked (efficacy) and two studies that examined the quality of the experience.

The efficacy studies investigated three types of telephone counselling: telephone counselling only (six studies); telephone counselling plus video sessions (one study); and telephone counselling plus video sessions and a workbook (two studies). These provided some evidence that telephone counselling is effective for reducing depressive symptoms in carers of people with dementia (three studies), but no clear positive effects could be shown for any other outcome such as stress or anxiety.

The studies that investigated the experience aspect of telephone counselling revealed a range of carers’ needs (16 themes) that covered three main areas: barriers to ‐ and things that enabled ‐ successful implementation of telephone counselling; the counsellor’s emotional attitude; and the content of the telephone counselling.

All studies were of moderate quality.

Conclusions

Analysis of both sets of results, i.e. efficacy compared with information about carers' experiences of telephone counselling, revealed needs that so far have not been met by telephone counselling. The studies that examined the experience aspect covered a very limited range of telephone counselling. The results of this review should be interpreted with caution due to the small number of included studies and their moderate quality.

Summary of findings

Summary of findings for the main comparison. Telephone counselling for informal carers of people with dementia.

Telephone counselling for informal carers of people with dementia
Patient or population: informal carers of people with dementia Settings: Intervention: telephone counselling
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Telephone counselling
Depressive symptoms		The mean depressive symptoms in the intervention groups was 0.32 standard deviations lower (0.63 to 0.01 lower)		163 (3 studies)	⊕⊕⊕⊝ moderate1,2	SMD 0.32 (0.01 to 0.63)
Burden		The mean burden in the intervention groups was 0.45 standard deviations lower (0.90 lower to 0.01 higher)		165 (4 studies)	⊕⊕⊕⊝ moderate3,4,5	SMD 0.45 (‐0.01 to 0.90)
Support		The mean support in the intervention groups was 0.25 standard deviations higher (0.24 lower to 0.73 higher)		67 (2 studies)	⊕⊕⊝⊝ low6,7,8	SMD 0.25 (‐0.24 to 0.73)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; SMD: standardised mean difference
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: we are very uncertain about the estimate

¹ All three studies: no ITT analysis
 ² All three studies: blinding was not possible due to the intervention and the self‐reported outcome measures
 ³ All four studies: no ITT analysis
 ⁴ All four studies: blinding was not possible due to the intervention and the self‐reported outcome measures
 ⁵ Only two out of four studies reported the significant outcomes
 ⁶ Both studies: no ITT analysis
 ⁷ Both studies: blinding was not possible due to the intervention and the self‐reported outcome measures
 ⁸ Too few participants

Background

Description of the condition

Caring for people with dementia is a challenge for the people involved and for society as a whole. In 2013, it was estimated that 44.35 million people worldwide suffered from dementia; 17 million of these were living in high‐income countries and 27 million were living in low and middle‐income countries (Prince 2013). It can be assumed that every person receiving medical/social care for dementia who is living at home has at least one informal carer (Brodaty 2003). Informal carers of people with dementia are on average 55 years old and 65% are female. More than 50% are a child of the person with dementia and 28% are the spouse (Ferrario 2003).

Caring often leads to negative psychological, physiological and financial consequences for these informal carers (Baumgarten 1992; Max 1995; Pinquart 2003). Many of them show high, or moderate to severe, levels of burden (Brodaty 2013). A long caring period results in physical discomfort; though if people with dementia have functional restrictions and show behavioural syndromes, informal carers suffer more often from depression and physical discomfort (Baumgarten 1992). It has been estimated that 30% of informal carers suffer from depressive symptoms (Taylor 2008), with 62% of women and 52% of men stating they suffer emotional stress (Alzheimer's Association 2014). Half of the informal carers are likely to give up their holidays, hobbies or social contacts; and 18% remarked that caring harmed their health (Alzheimer's Association 2004).

Informal male carers deal with care‐related burdens differently from women (Almberg 1998; Larranaga 2008). Men perceive a lack of positive prospects and need support. Women, on the other hand, report an increased strain in their relationships with other family members and an increase in their own health problems. However, caregiver burden can lead to early transfer of the person with dementia to a nursing home (Yaffe 2002), which should usually be avoided. Therefore, informal carers of people with dementia should be offered support opportunities. Throughout this review the term 'carer' always refers to informal carers providing home care for people with dementia.

Description of the intervention

The use of telecommunication for healthcare purposes is widespread. Telephone counselling is one way of using this technology, and is an attractive option for people who do not have the opportunity to participate in traditional face‐to‐face counselling (Reese 2006). Occupational groups such as nurses or psychotherapists use telephone counselling for a wide variety of diseases and situations. In the 5^th Nursing Interventions Classification telephone consultation is defined as: "Eliciting a patient’s concerns, listening, and providing support, information, or teaching in response to a patient’s stated concerns, over the telephone" (College of Nursing 2005).

How the intervention might work

Reese, and colleagues, investigated the use of telephone counselling for a variety of issues such as depression, anxiety, relationship problems, grief or work‐related difficulties (Reese 2002). Participants described telephone counselling as helpful for global and specific improvements, and were satisfied with the counselling they received. The efficacy of telephone counselling was also confirmed by the results of a study on smoking cessation (Stead 2013). Since the carers of people with dementia frequently suffer from depressive symptoms, in theory, they could also benefit from telephone counselling.

Why it is important to do this review

Informal carers are an important care resource for people with dementia, so interventions that support them and reduce their burden should be available. Telephone support practised in clinical settings is one option that might reduce the burden.

Only one narrative review of limited quality has been published on telephone counselling for carers of people with dementia (Mason 2008), so there is a need for a high quality and up‐to‐date systematic review.

The results of this review could show that telephone counselling might be a valid alternative to face‐to‐face interventions in rural areas where healthcare professionals have to travel long distances to reach the carers.

Alongside a summary of randomised controlled trials (RCTs) addressing the question of efficacy, we use qualitative evidence to address peoples' experiences with the intervention. Carers’ perspectives and the range of carers’ needs, for example, may help in understanding whether the carers find telephone counselling helpful, but miss the face‐to‐face contact because of their limited social contacts. Study findings indicate that telephone support is not helpful for some people (Chang 2004). A synthesis of qualitative evidence may help explain why certain people benefit more than others and provide more detailed information about the context, characteristics and circumstances of this group.

An earlier Cochrane review on the efficacy of cognitive reframing for carers of people with dementia did not consider the experiences of the carers of people with dementia (Vernooij‐Dassen 2011). It included interventions aimed at cognitive reframing but excluded interventions aimed at practical support. By comparison, our review includes telephone counselling that gives practical advice, but excludes interventions delivered by ongoing face‐to‐face counselling. In contrast to the earlier review (Mason 2008), we conducted a comprehensive and systematic search strategy, assessed the methodological quality of the included studies, and synthesised the evidence for both aspects ‐ efficacy and experience ‐ separately.

In order to provide clear differentiation of the two aspects being investigated, each section in this review consists of two subsections: 'To evaluate efficacy' and 'To evaluate experience', or, where applicable,'To evaluate efficacy and experience'.

Objectives

This review focuses on three main objectives:

To:

1) produce a quantitative review of the efficacy of telephone counselling for informal carers of people with dementia;

2) synthesize qualitative studies to explore carers’ experiences of receiving telephone counselling and counsellors’ experiences of conducting telephone counselling; and

3) integrate 1) and 2) to identify aspects of the intervention that are valued and work well, and those interventional components that should be improved or redesigned.

Methods

Criteria for considering studies for this review

Types of studies

Evaluation of efficacy

We included all individually randomised, parallel‐group controlled trials that compared telephone counselling for carers of people with dementia with no treatment, usual care or friendly calls for chatting. We also considered cross‐over trials.

Evaluation of experience

We included studies that used qualitative methods of data collection and analysis, that is, either a stand‐alone study or a discrete part of a larger mixed‐method study. Also, we considered qualitative elements from process evaluations conducted alongside trials.

Types of participants

Evaluation of efficacy and experience

We included informal carers who provide care for a person with dementia at home without being reimbursed for their caring work. The adult carer was a relative, a friend or a neighbour of the person with dementia. There were no restrictions regarding sex, age or ethnic background.

In order to evaluate experience, counsellors’ experiences of telephone counselling were considered and involved in the synthesis as well, if their experiences were investigated in combination with caregivers’ experiences.

Types of interventions

Evaluation of efficacy and experience

Telephone counselling is a complex intervention. It is also used in combination with face‐to‐face consultations or computer‐involved interventions. For this review, we considered all telephone counselling interventions that did not include any face‐to‐face counselling (a face‐to‐face contact at the first meeting was acceptable) that addressed informal carers of people with dementia living in the community. The telephone counselling was not allowed to be part of another complex intervention. It had to be conducted for a minimum of two months, but we made no restrictions about the frequency, length or number of telephone calls. The telephone counselling had to be performed by a health professional, such as a nurse or a psychologist, with expertise in dementia.

Telephone counselling had to comprise at least three components:

1. general information about dementia;

2. educating of carers in coping skills and caring for their own health;

3. psychosocial support where carers could share their feelings and were shown how to build up their social network.

Types of outcome measures

Evaluation of experience

We included all experiences regarding the intervention of interest.

Evaluation of efficacy

We included all studies that had either the following primary or secondary outcome(s).

Primary outcomes

Evaluation of efficacy

The primary outcomes were depressive symptoms, since they are frequent among informal carers of people with dementia (Baumgarten 1992).

Secondary outcomes

Evaluation of efficacy

Carer‐related outcomes:

1. caregiver burden;

2. distress;

3. anxiety;

4. quality of life;

5. care‐giving self‐efficacy;

6. satisfaction.

Outcomes concerning the person with dementia:

1. institutionalisation;

2. mood;

3. quality of life.

Search methods for identification of studies

Evaluation of efficacy and experience

The efficacy of and experience with telephone counselling address the same intervention and population in the same situation, but to answer these questions, both quantitative and qualitative study designs are needed.

We searched databases focusing on the intervention (telephone counselling) and the population (carers of people with dementia) without a limitation regarding study designs. We expected that retrieved references would include both quantitative and qualitative studies. We did not identify too many references, so we did not have to limit the search results by quantitative or qualitative filters.

Our search strategy for efficacy (quantitative trials) followed Cochrane Collaboration guidelines (Lefebvre 2011). For the question of experience (qualitative studies), we followed advice in Chapter 20 (Noyes 2011a), section 20.3.2.1, which provides supplemental guidance on searching for qualitative studies supplied by the Cochrane Qualitative and Implementation Methods Group. Details of the full search, with all search strategies and the number of records identified in each database is included in the Appendices 1 to 22 of the review.

Electronic searches

Evaluation of efficacy and experience

We searched ALOIS (www.medicine.ox.ac.uk/alois), the Cochrane Dementia and Cognitive Improvement Group’s Specialized Register. Our approach was to assess all carer‐focused studies within ALOIS.

ALOIS is maintained by the Trials Search Co‐ordinator of the Cochrane Dementia Group and contains dementia and cognitive improvement studies identified from:

1. monthly searches of a number of major healthcare databases: MEDLINE, EMBASE, CINAHL, PsycINFO and LILACS;

2. monthly searches of a number of trial registers: metaRegister of Controlled Trials (mRCT); UMIN (Japan Trial Register); ICTRP (the World Health Organization (WHO) portal that covers ClinicalTrials.gov; ISRCTN; Chinese Clinical Trials Register; German Clinical Trials Register; Iranian Regsitry of Clinical Trials and the Netherlands National Trials Regsiter, plus others);

3. quarterly search of The Cochrane Library’s Central Register of Controlled Trials (CENTRAL);

4. six‐monthly searches of a number of grey literature sources: ISI Web of Knowledge Conference Proceedings; Index to Theses; Australasian Digital Theses.

To view a list of all sources searched for ALOIS see About ALOIS on the ALOIS website.

We ran additional searches in many of the above plus the following sources to ensure that both qualitative studies and the most up‐to‐date results of quantitative trials were retrieved. An overview of the search results is displayed in Appendix 1. The search strategies used can be seen Appendix 2, Appendix 3, Appendix 4, Appendix 5, Appendix 6, Appendix 7, Appendix 8, Appendix 9, Appendix 10, Appendix 11, Appendix 12, Appendix 13, Appendix 14, Appendix 15, Appendix 16, Appendix 17, Appendix 18, Appendix 19, Appendix 20, Appendix 21, and Appendix 22.

We also searched the following: MEDLINE (Ovid SP); MEDLINE in Process & other non‐indexed citations (Ovid SP); The Cochrane Library (Wiley Online Library); CINAHL (EBSCOhost); PSYNDEX (EBSCOhost); PsycINFO (EBSCOhost); EMBASE (DIMDI); Web of Science (Thomson Reuters) (Science Citation Index Expanded (SCI‐EXPANDED), Social Science Citation Index (SSCI), Arts & Humanities Citation Index (A&HCI), Conference Proceedings Citation Index‐Science (CPCI‐S), Conference Proceedings Citation Index Social Science & Humanities (CPCI‐SSH)); DIMDI databases: Global Health, CCMed, DAHTA‐Database, HECLINET (1969 to 2001), ISTPB + ISTP/ISSHP, SOMED (1978 to 2000), Health Technology Assessment, and Publishers’ databases (up to 2011): Hogrefe, Karger, Kluwer, Krause & Pachernegg, Thieme; Springer (via Springerlink); Science Direct (Elsevier). HECLINET and SOMED were only searched up to 2001/2000 as the databases ceased to be updated then. The Publishers’ databases Hogrefe, Karger, Kluwer, Krause & Pachernegg and Thieme were searched up to 2011 since in 2013 we had no access to these databases.

Searching other resources

Evaluation of efficacy and experience

We handsearched abstracts of the following congresses (see Table 2): International Conference of Alzheimer’s Disease International; Alzheimer Europe Conference; Dementia Services Development Centre International Conference; Dementia Congress; and International Congress on Vascular Dementia. We also searched the trial registries of ClinicalTrials.gov; Controlled Trials and the WHO.

1. Details of handsearches of conference abstracts.

Conference	Years handsearched
International Conference of Alzheimer’s Disease International	1991, 1996, 1998, 2002, 2004, 2005, 2006, 2007, 2009, 2010, 2011, 2012
Alzheimer Europe Conference	2001, 2003, 2004, 2005, 2006, 2007, 2009, 2010, 2011, 2012
Dementia Services Development Centre International Conference	2008, 2008‐2, 2009, 2010, 2011, 2012
UK Dementia Congress	2010, 2012
International Congress on Vascular Dementia	2003, 2005, 2007, 2009, 2011

We checked the references of included studies and conducted a Forward Citation Search of these studies in Web of Science and Google Scholar. We also used the Related Articles service of PubMed for the included studies to find further relevant trials. The investigators of the included RCTs were contacted to ask for relevant quantitative studies and qualitative studies that were associated with, or conducted alongside, the effectiveness studies.

Data collection and analysis

Selection of studies

Evaluation of efficacy and experience

We transferred the references identified into reference management software (Endnote). Two independent authors (SL, GM) reviewed the titles and abstracts. Based on the full texts, two un‐blinded authors independently examined whether the studies fulfilled the inclusion criteria. In the preliminary search, the authors became acquainted with the names and institutions of study authors and the journal of publication of most of the relevant articles, so a blinded assessment of articles’ relevance was no longer possible. We resolved disagreement by discussion with a third review author (GL). We linked multiple articles from the same study together.

We have listed studies that initially seemed to meet the eligibility criteria but on further inspection were seen no to, and studies that did not meet all of the criteria, with the reasons for their exclusion (see Characteristics of excluded studies).

Data extraction and management

We performed data extraction differently for the evaluations of efficacy and experience and considered quantitative and qualitative study designs.

Evaluation of efficacy

We based the data extraction form for the quantitative studies on the recommended items from the Cochrane Handbook, Chapter 7 (Higgins 2011), section 7.3.1. The form covered information on the authors; publication date; country; study design; characteristics of the study population including setting; inclusion and exclusion criteria; credentials of those who provided the treatment; treatment type, duration, intensity, frequency; control intervention and outcomes. We conducted a pre‐test using two randomly‐selected studies and adapted the form afterwards. Two researchers (SL and DHB or AG or CH or JR) extracted data independently. We resolved disagreement by discussion. We requested missing data or further information related to the risk of bias assessment from the original investigators. We contacted eight authors of included studies, but only four responded; three authors provided further information for our risk of bias assessment, and two provided numerical data. We documented all essential steps in data extraction as requested in the PRISMA statement (Liberati 2009).

Evaluation of experience

The data extraction form included the following items: trial authors, publication date, country, methodology, methods, phenomena of interest, characteristics of the study population including setting, inclusion and exclusion criteria, credentials of those who provided the study, intervention outcomes, methods of data collection and analysis, and findings such as data from the participants and comments of study authors. One researcher (SL) created the data extraction form and a manual for its use prior to the data extraction. The data extraction form was adapted slightly during data extraction of the first and second studies and checked by another researcher (DHB). One researcher (SL) extracted the data on participants’ quotes and study authors commentaries. A second researcher (DHB) checked the extracted data. Disagreement was resolved by discussion. The data extraction and other aspects of synthesis were reported according to the ENTREQ statement (Tong 2012).

Assessment of risk of bias in included studies

Evaluation of efficacy

We performed critical appraisal of the risk of bias of included studies using The Cochrane Collaboration’s tool for assessing risk of bias. Two researcher (SL and GL) performed the assessment independently. Blinding to the names of the authors, institutions, journal and results of a study was not feasible because most of the studies were known to the review authors after the first specific literature search. We resolved disagreement by discussion and consulted a third researcher (GM) where necessary. We requested further information from the original investigators.

Evaluation of experience

When assessing the methodological quality of qualitative studies, it is important to evaluate researcher bias as a core criterion (cf Hannes 2011). In qualitative studies researchers have an impact on the data, so they must show their ability and the efforts they undertook to prevent this from influencing the data. Researchers have to state that the process was rigorous and trustworthy so that credibility of the findings can be assumed. The appraisal of qualitative research includes three stages (Hannes 2011):

1. filtering the identified hits according to the inclusion and exclusion criteria;

2. technical appraisal with evaluation of the credibility, transferability, dependability and confirmability of the original research papers included;

3. judging the methodological coherence. For this, the congruity between paradigms that guide the research project and the methodology and methods chosen is evaluated. The Cochrane Qualitative and Implementation Methods Groups recommends the third stage whenever the instrument for critical appraisal does not cover a paradigmatic approach.

We used the Critical Appraisal Skills Program (CASP) for judgment of the qualitative studies included. CASP does not include a question relating to the paradigm that guides the study. Therefore an evaluation of methodological coherence was conducted.

Two researchers assessed study quality. One (SL) performed the critical appraisal and another researcher (DHB), with experience in qualitative research, checked the judgments.

The consequence of the critical appraisal of the qualitative research can be a weighting of the studies found in the search. High quality studies can be given more weight than those of low quality. At present there are no established parameters to determine the weight of qualitative studies.

To enhance the applicability of the findings of qualitative synthesis, we used the CerQual (confidence in the qualitative evidence) approach (Glenton 2013). The CerQual approach is similar to GRADE (Grading of Recommendations Assessment, Development and Evaluation) and considers the methodological quality of evidence and the coherence of synthesised findings. For appraisal of methodological limitations, we used the results of the CASP assessment. For assessment of coherence, we determined the extent to which synthesised qualitative findings were grounded in the data of individual studies. If a synthesised finding was grounded across individual studies from different settings, then coherence was considered to be increased.

Measures of treatment effect

To evaluate efficacy

In meta‐analysis, longer ordinal scales are often analysed as continuous data. Depression scales can have different types of outcome data. For continuous data measured by different psychometric scales, we calculated the standardised mean difference. If only one assessment had been used in the trials, the weighted mean difference would have been applied. The scales were checked with regard to whether or not they had been validated and whether the original, or an adapted questionnaire had been used. Due to the expected heterogeneity, pooling was done using a random‐effects model. Pooling of some studies was not possible, so the results of those are presented narratively. We used the standard software of The Cochrane Collaboration (Review Manager 5.2, RevMan 2012) for analysis and graphical display. We used the open statistical environment R (version 2.15.3) for extracting standard deviations from P values (R 2012).

Dealing with missing data

Evaluation of efficacy

We evaluated the missing data and drop‐out rates for each RCT and displayed the results in Characteristics of included studies. Missing data could concern outcomes, summary data, individuals or, in the case of subgroup analysis or meta‐regression, study‐level characteristics. We requested all missing data from the original investigators and a sensitivity analysis was performed.

Assessment of heterogeneity

Evaluation of efficacy

We assessed heterogeneity by Cochran’s Q test and quantified it by using the I² statistic. We used a random‐effects model for pooling data. Where there was pronounced heterogeneity (i.e. I² > 60%) we planned to explain heterogeneity by subgroup analyses or meta‐regression as described under Subgroup analysis and investigation of heterogeneity.

Assessment of reporting biases

Evaluation of efficacy

We took steps to detect reporting bias. We checked each study to see whether outcomes were mentioned in the protocol, the methods section and the results sections of the report and the conference presentation, if available. We asked the authors whether a published protocol was available. If outcomes mentioned in the protocol or in the methods section of the report were not presented in the results section of the report or the conference presentations, we contacted the study authors.

Data synthesis

First synthesis: evaluation of efficacy

In the absence of clinical and statistical heterogeneity, we conducted a meta‐analysis using the random‐effects model in Review Manager 5.2. The separate control interventions were analysed in separate meta‐analyses.

Second synthesis: evaluation of experience

We conducted thematic synthesis for qualitative data. Such a synthesis can produce new insights and understanding as studies with different important aspects, such as population and settings, can be combined to identify concepts that are present across a range of contexts and settings.

The choice of synthesis methods depends on several aspects (see the supplemental Handbook guidance of the Cochrane Qualitative and Implementation Methods Group; Noyes 2011b), and the review question and purpose should be considered. In our case, this concerns the carers’ and counsellors' experiences of telephone counselling. Another aspect to be considered is the identified evidence: if descriptive qualitative studies that are characterised by 'thin descriptions' are available, an aggregative method is more appropriate; if the identified evidence includes a high proportion of in‐depth qualitative studies characterised by 'thick descriptions', an interpretative method can be applied. Authors should also consider whether or not an appropriate theoretical model for the phenomenon of interest exists.

Since we only identified two descriptive studies that were characterised by 'thin descriptions' with no appropriate theoretical model, we used thematic synthesis as described by Thomas 2008. We transferred the extracted findings of the studies into the qualitative data analysis software MAXQDA Version 11.0.1.

The first step of the analysis was an intensive reading of the included studies during which the initial ideas and memos were recorded in a researcher’s diary. This step was followed by a line‐by‐line coding, conducted in the review authors' mother language of German, to describe the content and meaning of the data with the review author’s own words. During the line‐by‐line coding, ideas for grouping the data and descriptive themes were noted in the researcher’s diary.

The third step included summarising codes into descriptive themes by looking for similarities and differences between the codes. This step was influenced by the first idea on possible analytical themes and a structure which displays how they can be related to each other. Afterwards the descriptive themes were grouped into analytical themes, and the first idea of the relationship between the analytical themes was modified. In this step the analytical themes were checked and refined by looking at the underlying descriptive theme, codes and original data of the studies.

We discussed the preliminary findings with another author with experience in qualitative research. We considered the descriptive and analytical themes and their distinctions from each other, afterwards some thematic themes were grouped together and reworded. We discussed and modified the specifications of the descriptive and analytical themes. We also adapted the structure or the relationship between the analytical themes in terms of a mapping. At the end of this process, we derived three analytical themes that were based on 16 descriptive themes (see Figure 1) and 119 codes. There was, however, little opportunity for synthesis with only five themes shared between the two studies. Additional limiting factors concerning the included qualitative evidence are described under the 'Included studies' section below.

1.

Aspects of telephone counselling for informal carers of people with dementia

Third synthesis: integration of the first synthesis and the product of the second synthesis

The aim of the third synthesis was to merge the question of efficacy (quantitative studies) with the question of experience (qualitative studies) while juxtaposing them in a matrix (Thomas 2004). The integration of qualitative and quantitative data was guided by two questions.

1. Which themes of experience are addressed by the interventions of the RCTs?

2. Which aspects of the RCT interventions were not mentioned by the participants?

The first question was captured in the first part of the integration of qualitative and quantitative data within Matrix 1 (Table 3), where the descriptive themes from the qualitative data were juxtaposed with the RCTs. This first comparison should reveal which descriptive themes were already considered in the telephone counselling intervention of the RCTs and which were missing, and also identify those interventions that contain more items of positive experience than others. Consequently, these findings can be used for subgroup analyses.

2. Matrix 1. Descriptive themes of the thematic synthesis investigated in quantitative trials.

Analytical themes	Descriptive themes/study	Davis 2004	Finkel 2007	Glueckauf 2007	Tremont 2008	Wilz unpublished	Winter 2006	Chang 1999	Gant 2007	Steffen 2000
Barriers and facilitators for successful implementation of telephone counselling	Telephone as communication tool	/	/	/	/	/	/	/	/	/
	To know each other	●	●	/	/	●	/	●	/	/
	Multilingualism	/	/	/	/	/	/	/	/	/
	24‐hour reachability of the counsellor	/	/	/	/	/	/	/	/	/
	Continuation of counselling after institutionalisation	‡	‡	‡	●	○	/	/	‡	/
	Debriefing of counsellor	/	/	/	/	/	/	/	/	/
Counsellor's emotional attitude	Non‐judgmental/appreciative	/	/	/	/	/	/	/	/	/
	To be familiar and trusted	/	/	/	/	/	/	/	/	/
	To be reliable	/	/	/	/	/	/	/	/	/
Content of telephone counselling	The range from providing factual information to mediation	●	●	●	●	●	●	○	/	●
	Individual counselling	●	●	●	●	●	○	●	●	●
	Problematic situations	●	●	●	●	●	●	●	●	‡
	Conversational themes that help forget the daily routine	(●)	/	/	/	/	/	(●)	/	/
	Information and advice	●	●	●	●	●	●	●	●	●
	The carer’s emotions, thoughts, and physical condition	○	○	●	●	●	●	●	●	●
	Empowerment for autonomy	●	○	●	○	●	●	●	●	●

● = was investigated; ‡ = was not investigated; ○ = unclear, maybe investigated; / = not mentioned; () = not part of the intervention, but mentioned by the authors in the discussion

The second question was captured in the second part of the integration of qualitative and quantitative data within Matrix 2 (Table 4). Here the results of the telephone counselling intervention of the RCTs were juxtaposed with the qualitative studies. The second comparison made clear which aspects of telephone counselling were already considered by the qualitative studies and which were not considered. This matrix is interesting for appraising the range of findings of the qualitative studies; in addition, aspects that may be less important for the carers were indicated.

3. Matrix 2. Aspects of telephone counselling for carers of people with dementia covered by qualitative studies.

Aspects of telephone counselling mentioned in the included quantitative trials	Chang 2004	Salfi 2005
Individual telephone counselling	●	●
Telephone support groups	/	/
Video‐sessions	●	/
Workbook	/	/
Loose‐leaf notebook	/	/
Individual telephone counselling via the CTIS sytem	/	/
Support group sessions via CTIS system	/	/
Information about dementia	/	○
Information about effects of caregiving	/	/
Self‐care	/	/
Relaxation training	/	/
Breathing techniques	/	/
Pleasant daily activities to offset emotional distress	/	/
Promote carers’ assertiveness	/	/
Self‐instructional training	/	/
General problem solving skills	●	/
Managing behaviour problems	●	●
Managing difficult caregiving challenges	●	●
Strategies for handling response to difficult caregiving situations	/	/
Organising care routines	/	/
Strategies to enhance safety	/	/
Finding and using local community resources	●	●
Social support network	/	●
Emotional support	/	●
Talking about carers' feelings	●	○
Empathy	/	/
Giving permission	/	/
Normalising	/	/
Validation or venting	/	/
Anger‐management	/	/
Bibliotherapy	/	/
Interpretation	/	/
Role play	/	/
Cognitive appraisal and reappraisal techniques	/	/
Cognitive reframing	●	/
Cognitive restructuring	/	/
Coping strategies	●	●
Behavioural management	/	/
Behavioural activation	/	/
Thought stopping	/	/
Distraction techniques	/	/
Coming to the end of the counselling intervention	/	/

● = performed it; ‡ performed it not; ○ = unclear, maybe included; / = mentioned it not; () = was not part of the intervention, but mentioned by the authors in discussion.

Subgroup analysis and investigation of heterogeneity

Evaluation of efficacy

If five studies for each planned aspect had been available, subgroup analysis or meta‐regression would have been conducted for the following areas.

1. Carer‐related: relation to the person with dementia (e.g. spouse, child); support by other family members; pre‐existing stress of the carer; in employment or not; gender.

2. Intervention‐related: duration; intensity of intervention.

3. Disease‐related: stage of dementia.

4. Experience‐related: aspects of the intervention that were experienced positively.

Sensitivity analysis

Evaluation of efficacy

We applied sensitivity analyses to check the impact of different ways of handling missing data as described in Dealing with missing data.

Results

Description of studies

Nine quantitative studies and two qualitative studies met the inclusion criteria. The characteristics of all included studies are described in Characteristics of included studies.

Results of the search

Evaluation of efficacy and experience

The first search was performed in May 2011 and an update search in February 2013. The check for duplicates of records identified by database searching was first conducted electronically using reference management software and then conducted by hand. The records identified by searching non‐electronic sources could not be transferred into reference management software. An electronic check for duplicates of those records was therefore not possible. The titles and abstracts of the 7877 retrieved records were independently screened by two authors (SL and DHB or academic support staff). The remaining 110 records were independently assessed for inclusion on the basis of full texts by two review authors (SL and GM or DHB). Thirteen publications fulfilled the inclusion criteria (for efficacy: Chang 1999; Davis 2004; Finkel 2007, Gant 2007; Glueckauf 2007; Steffen 2000; Tremont 2008; Wilz unpublished; Winter 2007; for experience: Chang 2004; Salfi 2004; Salfi 2005). Salfi 2004 (Thesis) and Salfi 2005 (article) refer to the same study. Wilz 2011 published the results of one outcome and recorded other outcome results in a manuscript that has been submitted for publication (Wilz unpublished). Two ongoing trials were identified: one through searching trial registries (Nichols 2005), and one as a poster abstract (Tremont 2011). In total, we included nine studies that addressed the efficacy and two studies that addressed the experience of telephone counselling (see Figure 2).

2.

Flow diagram of study selection process

Included studies

Evaluation of efficacy

All the included quantitative trials were RCTs; all except one were conducted in the USA, the exception was performed in Germany (Wilz unpublished). Seven studies included between 32 and 71 carers of people with dementia (Chang 1999; Davis 2004; Finkel 2007; Gant 2007; Glueckauf 2007; Steffen 2000; Tremont 2008), one study included 103 carers (Winter 2007), and another included 229 (Wilz unpublished).

Telephone counselling

The interventions in the included studies addressing efficacy differed. In six studies, telephone counselling was provided without an additional intervention (Davis 2004; Finkel 2007; Glueckauf 2007; Tremont 2008; Wilz unpublished; Winter 2007). One study combined telephone counselling with an educational video‐series (Chang 1999), and two studies combined it with educational video‐series plus a workbook (Gant 2007; Steffen 2000).

Since we had decided to include only trials that conducted counselling, we assessed the included studies for whether they provided counselling or therapy. It is difficult to differentiate between counselling and therapy since the boundary between the concepts is blurred. Usually the target group for counselling does not need therapy, and telephone counselling will be performed for a shorter period than therapy. We therefore tried to demarcate the concepts on the basis of the intervention’s duration, theoretical background and the participants' condition. The theoretical background for telephone counselling gives an impression of the way in which the counselling was performed and whether it should be placed into the area of 'counselling' or fits better into the area of 'therapy'. If the theoretical background was not mentioned in the publication, study investigators were contacted and asked to explain it. Table 5 lists the included studies with their intervention, duration, theoretical background and the review authors’ decisions on whether the intervention was to be allocated to the area of counselling or therapy.

4. Included quantitative studies and their allocation to counselling or therapy.

Study	Quotations of intervention judged by review authors as ‐ Education (as part of counselling): printed in bold ‐ Counselling: printed in italics ‐ Therapy: printed with underscores	Duration of intervention	Theoretical background	Overall judgment by review authors
Telephone counselling
Davis 2004	Telephone‐based skill training plus conversation about finding and using local community resources: Thinking constructively about caregiving problems (primary appraisal), acquiring an increased sense of self‐efficacy around personal caregiving abilities (secondary appraisal), being more resourceful in managing (coping with) caregiving situations Telephone: ‐ General problem solving skills ‐ Caregiver appraisal of behaviour problems (problem antecedents, situational context and consequences) ‐ Written behavioural programs for managing specific problems ‐ Strategies for handling affective response to difficult caregiving situations (e.g. the use of anger management and relaxation techniques) Loose‐leaf notebook of suggestions for managing frequently encountered Alzheimer’s disease situations (e.g.; managing confusion, agitation, wandering, bathing/eating difficulties) Finding and using local community resources	3 months	‐ Lazarus 1999 (stress, appraisal and coping theory)	Counselling
Finkel 2007	Telephone counselling via the CTIS system: There were 8 individual educational/skill building sessions and 6 support group sessions (six caregivers and a facilitator), which were interspersed Intervention: ‐ Provision of information about the disease, community resources, strategies to enhance safety,communication self‐care, social support, management of problem behaviours ‐ Via Computer‐Telephone Integration System and social workers: screen phones (text and voice): place and receive calls/massages, information and service, conference calls	6 months	Not stated by the study investigator in publication or by email request	Counselling
Glueckauf 2007	Intervention: telephone‐based CB intervention ‐ overview of the basic characteristics of dementia ‐ relaxing training ‐ effective thinking about caregiving challenges ‐ building in pleasant daily activities as a guard against emotional distress ‐ assertiveness in caregiving situations and with family members ‐ developing problem‐solving skills through personal goal setting ‐ the importance of having a social support network and further refining of problem‐solving skills after the ARCH project	3 months	Not stated by the study investigator in publication or by email request	Counselling
Tremont 2008	Telephone counselling plus information about local resources and educational materials No case management, serve as a question‐and‐answer hotline, or provide psychotherapy over the telephone 1. Initial call (orientation about the study with its content and procedure and psychoeducation) ‐ providing emotional support ‐ directing caregivers to appropriate resources ‐ encouraging caregivers to attend to their own physical, emotional and social needs ‐ teaching caregivers strategies to cope with ongoing problems Psychoeducation: reviewing information about dementia and common psychological, emotional, psychosocial and medical effects of caregiving 2. Telephone counselling: ‐ new problems were identified ‐ positive and negative changes in caregivers or care recipients were discussed ‐ psychoeducational information was reviewed and applied for particular situations ‐ If needed: 1) supportive approaches (i.e.; empathy, giving permission, normalizing, provision information, validation or venting) or 2) active strategies (i.e.; bibliotherapy, interpretation, positive reframing, problem solving, reference to resource packet, referral and setting task directives) ‐ Final 4 calls: issues of termination by allowing caregivers to anticipate FITT‐D contacts coming to an end and foster reliance on the support network established during the intervention	12 months	‐ Tyhurst 1958 (psychosocial transition) ‐ Lazarus &Folkman 1984 (transactional stress &coping) ‐ McMaster Model of Family Functioning; Eptstein, Bishop & Levin 1978 (systems view of family functioning)	Counselling
Winter 2007	Telephone‐based support groups; 5 caregivers per group, professionally led: ‐ Manage daily stressors by providing emotional support and validation ‐ Caregivers express emotions and share coping strategies including cognitive reframing and practical approaches to organizing care routines. Assist each other in problem solving and share educational resources. The mutual support and validation provided by group members normalize experiences and provide a supportive social network, core to the service model	6 months	Not stated by the study investigator in publication or by email request	Counselling
Wilz unpublished	The therapeutic manual consisted of the following CBT intervention modules, matched to the needs of dementia caregivers: 1. Improving caregivers’ problem‐solving skills with problem‐solving training, role‐play and promoting coping strategies for challenging behaviour of the care recipient 2. Increasing self‐care and positive experiences by applying techniques such as listing pleasant activities, making a weekly and daily pleasant activity schedule, and performing self‐care activities 3. Enhancing caregivers’ coping skills using cognitive appraisal and reappraisal techniques, breathing techniques, and anger‐management 4. Cognitive restructuring through recording dysfunctional thoughts and considering alternative thoughts 5. Psychoeducation about dementia in order to improve acceptance and psychological understanding of the disease. Psychoeducation was also used to explain the rationale of each applied intervention technique; it was therefore implemented in almost every session 6. For the issues such as accepting the disease, coping with role change, loss and grief a multi‐conceptual treatment approach was used which includesrestructuring dysfunctional thoughts related to grief, expressing and processing emotions of role change and loss, dealing with grief, coping with loss and change, and defining new roles	3 months	(D'Zurilla & Goldfried, 1971; Kanfer, Reinecker, & Schmelzer, 2006) 2. (Lewinsohn, 1974) 3.(Lazarus, 1999) 4. (Beck, Rush, Shaw, & Emery, 1979) via the ABC Model (Ellis, 1977) 5. (Noyes, Hill, Hicken, Luptak, Rupper, Dailey, & Bair, 2010; Sanders & Sharp, 2004).	Therapy
Telephone counselling + video
Chang 1999	Intervention: Knowledge and skills to improve eating & dressing abilities: 1. Video: behavioural modelling strategies for dressing & eating 2. Telephone: cognitive portion: increasing knowledge about coping strategies and sources of support reinforce the video information;caregivers uncertainties regarding implementation of strategies and caregivers feelings regarding care recipients behaviours and information to assist the caregivers in cognitive reframing and problem solving. Problem solving guidelines for behavioural disorders (aggressive behaviours, incontinence) and CGs referrals when appropriate	2 months	‐ Lazarus & Folkman 1984 – (stress & coping framework); ‐ Lawton, Moss, Fulcomer and Kleban 1982 ‐ Feil 1989 (Validation)	Counselling
Telephone counselling + video + workbook
Gant 2007	Intervention: telephone coaching condition with video and workbook: Focus was primarily behavioural (e.g.; behavioural management, behavioural activation, controlled breathing for relaxation and stress reduction). Not cognitive or emotional Video and Workbook: didactic and experiential materials that reinforce information presented in each video session Telephone: ‐ first 10 calls: reinforced the video sessions ‐ following a coach manual ‐ assisting them in the application of intervention concepts to their unique problems	3 months	Stated by email request: ‐ Lewinsohn 1992 ‐ Rinkston & Linsk 1984 ‐ Teri 1998	Counselling
Steffen 2000	Psychodeucational video series with workbook: ‐ instruction ‐ homework assignments on awareness training, tension‐reduction strategies (relaxation training, thought stopping, distraction techniques) ‐ cognitive strategies (i.e.; cognitive restructuring and self‐instructional training) ‐ assertion training 1. Video (cognitive‐behavioural strategies for anger management) ‐ 10 min : didactics ‐ 5 min: caregiving‐relevant examples ‐ 10 min: guiding participants through self‐monitoring assignments and worksheets ‐ 5 min: imagery‐based relaxation practice ‐ included workbook sections 2. Workbook: information 3. Telephone calls: ‐ reviewed progress on homework from the following week ‐ problem solved (finding time to do relaxation, understanding of concepts and suggestions last week, practicing the assertion skills and positive self‐statements, no practical caregiver advice)	2 months	‐ Novaco (1975/1977/1985) ‐ Ecton & Feindler (1990)	Counselling
Abbreviations: ARCH = Alzheimer’s Rural Care Healthline CB = Cognitive‐Behavioural CBT = Cognitive behavioural therapy CTIS = Computer‐Telephone Integration System FITT = Family Intervention: Telephone Tracking ‐ Dementia min = minutes

Table 6 gives an overview of the treatment and control group details of the included quantitative trials.

5. Overview of treatment and control group details for quantitative studies.

Study	Group versus individual counselling	Frequency	Duration of one telephone call	Duration of the whole intervention	Face‐to‐face contact	Active or usual care control group
Telephone counselling
Davis 2004	Individual	Weekly (12)	Averaged 37 min (SD: 18 min)	3 months	1 initial home visit (45 min)	Friendly calls, average 16 min (SD =12 min)
Finkel 2007	8 individual and 6 group sessions	14 (biweekly)	34 min (mean: 7.87 hours, (SD = 1.82) in 6 months)	6 months	2 in‐home sessions (initial and last)	Usual care (2 brief telephone contacts at 3 and 5 months (15 min)
Glueckauf 2007	5 individual and 7 group sessions	Weekly (12)	6 times 45 min; 5 times 60 min	3 months	No	Usual care
Tremont 2008	Individual	6 weekly calls; 6 months biweekly; 4 monthly (23)	Initial 60 min, rest 15‐30 min	12 months	No	Usual care
Wilz unpublished	Individual	4 weekly, 5‐6 fortnightly, 7 a month later (7)	60 min	3 months	Initial home visit or telephone	Usual care
Winter 2007	Group	Weekly (26)	60 min	6 months	No	Usual care
Telephone counselling + video
Chang 1999	Individual	Weekly (8)	Mean = 18.3 min	2 months	No	Friendly calls, mean = 11.1 min
Telephone counselling + video + workbook
Gant 2007	Individual	Weekly (12)	NR	3 months	No	Friendly calls
Steffen 2000	Individual	Weekly (8)	20 min	2 months	No	Usual care

Abbreviations

min = minutes
 NR = not reported

Most of the included studies conducted telephone counselling on an individual basis (Chang 1999; Davis 2004; Gant 2007; Steffen 2000; Tremont 2008; Wilz unpublished). Two studies combined group sessions over the telephone with individual sessions (Finkel 2007; Glueckauf 2007). One study conducted all the telephone counselling via group sessions (Winter 2007).

Duration of telephone counselling varied amongst the included studies. Two studies were conducted over two months (Chang 1999; Steffen 2000), four studies over three months (Davis 2004; Gant 2007; Glueckauf 2007; Wilz unpublished), two studies over six months (Finkel 2007; Winter 2007), and one study over 12 months (Tremont 2008). In most studies, telephone counselling was performed weekly (Chang 1999; Davis 2004; Gant 2007; Glueckauf 2007; Steffen 2000; Winter 2007). The frequency of calls in one study was biweekly (Finkel 2007), and two studies started with weekly calls and then reduced stepwise after four or six weeks (Tremont 2008; Wilz unpublished). The duration of a single telephone counselling session also varied between studies. Sessions lasted either about 20 minutes (Chang 1999; Steffen 2000), approximately 30 minutes (Davis 2004; Finkel 2007), or 60 minutes (Wilz unpublished; Winter 2007). In one study, the sessions in the first half of the study lasted for 45 minutes and for 60 minutes in the second half of the study (Glueckauf 2007). Another study conducted a 60‐minute session at the beginning and reduced the duration of subsequent sessions to 15 to 30 minutes (Tremont 2008).

Studies with regular face‐to‐face counselling were excluded. However, single face‐to‐face contacts, for example at the beginning of the telephone counselling, were accepted. In six studies, counselling was only conducted via the telephone (Chang 1999; Gant 2007; Glueckauf 2007; Steffen 2000; Tremont 2008; Winter 2007). Two studies conducted an initial home visit (Davis 2004; Wilz unpublished), and one study conducted an additional at‐home session at the end (Finkel 2007).

Study results can be influenced by the type of control group. It must be assumed that a treatment that is compared to usual care might more easily seem significant than a treatment that is compared to an active control treatment such as friendly calls. The study investigators of two trials stated that the participants in the control groups also judged the friendly calls as helpful (Chang 1999; Davis 2004). The differences in outcomes might therefore decrease, and it will become more difficult to find significance. Five studies compared telephone counselling with usual care (Glueckauf 2007; Steffen 2000; Tremont 2008; Wilz unpublished; Winter 2007), and one study treated the control group with usual care but added two brief telephone contacts at three and five months (Finkel 2007). Three studies used friendly calls in their control group (Chang 1999; Davis 2004; Gant 2007).

Study participants

The mean age of the participants in most trials lay between 60 and 66 years (Chang 1999; Finkel 2007; Glueckauf 2007; Steffen 2000; Tremont 2008; Wilz unpublished; Winter 2007). Only in Gant 2007 was the participants’ mean age 71 years, and the figures stated in Davis 2004 seem to be faulty. Gender ratios differed between trials: two studies included only women (Chang 1999; Winter 2007), and one study included only men (Gant 2007). In three studies, the majority of participants were women (Finkel 2007; Glueckauf 2007; Wilz unpublished), and in one study the majority were men (Steffen 2000). The numbers stated in Davis 2004 seem to be faulty in this respect also.

Three trials did not report the duration for which the carers had been caring (Finkel 2007; Gant 2007; Steffen 2000), and this varied in the other included studies. Davis 2004 reported a mean length of caring of approximately two years; Chang 1999 and Tremont 2008, a mean of about three years; Winter 2007, a minimum of about four years and Glueckauf 2007 and Wilz unpublished, a mean of more than five years.

The relationship between the carer and the person with dementia also differed between studies. In five studies, more than half of the participants were caring for their spouses (Chang 1999; Gant 2007; Steffen 2000; Tremont 2008; Winter 2007), but in three other studies, less than half were caring for their spouses (Davis 2004; Finkel 2007; Glueckauf 2007). In Wilz unpublished, 45% of the carers were of the same generation as the person they were caring for and 49% were of a younger generation. Six other studies also included participants who were caring for a parent (Davis 2004; Finkel 2007; Gant 2007; Glueckauf 2007; Steffen 2000; Tremont 2008); in four studies, participants also included other family members ‐ not only the children of the person with dementia (Davis 2004; Gant 2007; Glueckauf 2007; Steffen 2000). Two studies only stated how many of the carers were spouses, but did not identify the family relationship of the remaining participants (Chang 1999; Winter 2007).

Some studies excluded carers who reported alcohol abuse problems (Davis 2004; Gant 2007; Steffen 2000), suicidal ideation (Davis 2004; Gant 2007; Glueckauf 2007; Steffen 2000), or other psychological difficulties (Glueckauf 2007; Steffen 2000; Tremont 2008).

Measurement of outcomes

All the included quantitative trials reported their outcome measures as mean differences with standard deviations (SDs). However, Glueckauf 2007 reported results as medians with semi‐quartile, and inter‐quartile ranges, and Finkel 2007 reported t‐statistics instead of standard deviations (SDs). The conversion from median into mean differences is possible, but, despite a direct request, the study investigators did not supply the necessary minimum and maximum data. Therefore, it was not possible to include the results of Glueckauf 2007 in the meta‐analysis; instead the results were described in narrative form and the t‐values converted into SDs. Winter 2007 did not state the outcome results in combination with the number of participants in each group, and we therefore allocated the dropouts from this trial according to the distribution at baseline.

Outcomes related to the carer

Primary outcome: depressive symptoms

Eight studies assessed depressive symptoms (Chang 1999; Davis 2004; Finkel 2007; Glueckauf 2007; Steffen 2000; Tremont 2008; Wilz unpublished; Winter 2007). Most used the Centers for Epidemiologic Studies Depression Scale (CES‐D) (Finkel 2007; Glueckauf 2007; Wilz unpublished; Winter 2007), two trials used the Geriatric Depression Scale (GDS) (Davis 2004; Tremont 2008), one study used the Brief Symptom Inventory (BSI) (Chang 1999), and one study used the Beck Depression Inventory (BDI) (Steffen 2000).

Secondary outcomes

Burden

Caregiver burden was assessed by seven studies using different tools. The Zarit Burden Interview was used by two studies (Tremont 2008; Winter 2007), and the Caregiver Appraisal Inventory (CAI) was used by two studies (Chang 1999; Glueckauf 2007). The Screen for Caregiver Burden (SCB) was used by Davis 2004, the Revised Memory and Behavior Problems Checklist (RMBPC) by Finkel 2007, and the Subunits Upset and Annoyance from RMBPC by Gant 2007.

Distress

Carer distress was assessed by two studies (Davis 2004; Gant 2007). Davis 2004 used scale B of the Screen Caregiver Burden to measure it, while Gant 2007 used the Target Complaints Interview Assessment Tool.

Anxiety

Only Chang 1999 assessed carer anxiety, using the Brief Symptom Inventory (BSI).

Quality of life

Tremont 2008 was the only study that assessed quality of life, using the quality of life parameter of the SF36 General Health tool.

Self‐efficacy

Four studies assessed carer self‐efficacy (Gant 2007; Glueckauf 2007; Steffen 2000; Tremont 2008). Three used different versions of the Caregiving Self‐Efficacy Scale by Steffen (Gant 2007; Glueckauf 2007; Steffen 2000), while Tremont 2008 used the Self‐Efficacy Scale by Fortinsky.

Satisfaction

Two studies assessed carer satisfaction using different tools: Chang 1999 used the Caregiver Appraisal tool, and Davis 2004 used the Caregiver life satisfaction (LSI‐Z) measure.

Social support

Carers' social support was assessed by three studies with different measures (Davis 2004; Finkel 2007; Tremont 2008). Davis 2004 assessed the social support with the Caregiver Social Support tool (ISEL); Finkel 2007 used a social support scale from the Inventory of Socially Supportive Behaviors; and Tremont 2008 used the Multidimensional Scale of Perceived Social Support (MSPSS).

Outcomes related to the care recipient

None of the included quantitative trials assessed outcomes such as institutionalisation, mood or quality of life of the care recipients. Five of the quantitative trials reported how many of the care recipients were institutionalised as reasons for loss to follow‐up (Davis 2004; Finkel 2007; Gant 2007; Glueckauf 2007; Winter 2007).

Time of assessment of outcome measurements

The included studies assessed outcomes at different time points. The outcomes of studies that conducted telephone counselling without an additional intervention were assessed after three months (Davis 2004; Glueckauf 2007; Wilz unpublished), six months (Davis 2004; Finkel 2007; Winter 2007), nine months (Wilz unpublished), and after 12 months (Tremont 2008). The outcomes of the study that conducted telephone counselling in combination with video sessions were assessed after one, two, and three months (Chang 1999). The outcomes of the studies that conducted telephone counselling in combination with video sessions and a workbook were assessed after two months (Steffen 2000), and after three months (Gant 2007).

Funding source

All except one of the included quantitative trials reported their funding source (Steffen 2000). Four trials were supported by an Alzheimer's organisation (Chang 1999; Gant 2007; Glueckauf 2007; Winter 2007), two received private funding (Finkel 2007; Glueckauf 2007), and four received governmental funding (Davis 2004; Gant 2007; Tremont 2008; Wilz unpublished).

Evaluation of experience

The thematic synthesis included two descriptive studies. Chang 2004 was a qualitative process evaluation of a larger RCT that could not be identified. The review authors contacted the trial authors but received no response. The larger trial, which was partly described within Chang 2004, used the same methods and the same population as Chang 1999, however, these studies were not the same. The Salfi study was first published as a doctoral thesis in 2004 and in 2005 was published as an article (Salfi 2005). We compared and summarised the findings presented in both documents; the study of Salfi 2005, was standalone qualitative research.

Study respondents

Both qualitative studies explored the perspective of the carers of people with dementia who received telephone counselling (Chang 2004; Salfi 2005). The Salfi 2005 trial aimed to understand the characteristics of telephone counselling comprehensively and to learn more about how both the carers and the counsellors experienced it, so observed not only the carers, but also the counsellors.

The process evaluation in Chang 2004 involved the counsellor recording notes in real time during semi‐structured telephone delivery of the cognitive behavioural therapy (CBT)‐based intervention to carers. These handwritten notes were later subjected to content analysis. The process evaluation involved 83 participants in an RCT of an intervention following the same approach described in Chang 1999. Forty‐one of the participants were assigned to the intervention group and 42 to the control group: 75 of the 83 participants were female, and the mean age of the whole study population was 68 years. Sixty‐seven participants were white, 12 were black, and four were Hispanic. On average, all participants went to school for 14 years, and 47% of them earned between USD 10,000 and USD 20,000. They had been supporting people with dementia for approximately 20 hours a day for an average of 3.2 years.

In the Salfi 2005 study eight carers and four counsellors were interviewed. Six of the carers were between 50 and 59 years old, two were between 60 and 80 years old, and all of them were female. Three of the eight were taking care of their husbands while the other five were in charge of their fathers. One carer had been nursing for less than a year and another one for one to two years; four carers had been nursing for between three and five years, and two of them had been in charge of a person with dementia for more than five years. The telephone counsellors were aged between 30 and 65 years. They had between three and 15 years experience counselling, two had a degree in nursing, and two had a Bachelor of Arts degree in gerontology.

Setting

Both studies were conducted in high‐income countries (World Bank 2013). One study was based in Canada (Salfi 2005), and one in the USA (Chang 2004). The Chang 2004 study included community‐based participants, but the Salfi 2005 study did not state clearly if it took place in an urban or rural setting. The telephone counselling in Chang 2004 was performed within an RCT intervention. The telephone counselling in Salfi 2005 was established support offered by two agencies and not part of any evaluation. One agency was a branch of the Alzheimer Society and the other a multiservice community agency. It was not clear whether both agencies were non‐governmental organisations.

Telephone counselling

The telephone counselling in both studies was poorly described. The available information is stated below.

Participants in Chang 2004 who were assigned to the intervention group were visited at the beginning of the study and provided with a video with information about dressing and feeding people with dementia. After that, semi‐structured telephone counselling was performed every two weeks for a total of 12 weeks. During each counselling session, existing problems and problem‐solving strategies were discussed and adapted to each individual case. In addition to dressing and feeding people with dementia, other topics addressed by the carers were discussed according to the problem‐solving guidelines. The telephone counselling sessions lasted from five to 90 minutes, with an average of 18 minutes, and were performed by two counsellors with Masters degrees and a gerontological nursing specialisation. The hand‐written notes of the counsellors from these counselling sessions constituted the raw data for analysis.

Two different counselling institutions attended to the participants in the Salfi 2005 study. The first institution offered counselling for eight hours daily on work days only. The second institution also offered a daily program for adults involved, and training courses in addition to telephone counselling. Carers had access to all of these services and were not limited to telephone counselling. Both institutions offered information and emotional support during their counselling.

Funding source

Both studies stated their source of funding: both were funded by an Alzheimer's organisation, and the Salfi 2005 study received additional financial support from a community care centre.

Excluded studies

Trials were excluded if they did not fulfil the inclusion criteria. Most of the 98 excluded studies were clearly outside the scope. Four studies are listed, with the reasons for their exclusion, in Characteristics of excluded studies since their exclusion was not clear at first glance.

Risk of bias in included studies

Evaluation of efficacy

The results of the 'Risk of bias' assessment are displayed in the 'Risk of bias' graph and the 'Risk of bias' table (see Figure 3 and Figure 4).

3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies

4.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

Allocation

After inquiry by email, random sequence generation was explained and found to be adequate in three studies (Gant 2007; Steffen 2000; Tremont 2008), and was unclear in the other studies (Chang 1999; Davis 2004; Finkel 2007; Glueckauf 2007; Wilz unpublished; Winter 2007). Similarly, after email inquiry, allocation concealment was described and found to be adequate in two studies (Gant 2007; Steffen 2000), but was unclear in the other studies (Chang 1999; Davis 2004; Finkel 2007; Glueckauf 2007; Tremont 2008; Wilz unpublished; Winter 2007).

Blinding

Due to the nature of the intervention, blinding of participants and personnel was not possible. This lack of blinding can have influenced additional interventions and, in consequence, most of the (subjective) outcomes. Thus, all studies were assessed as being at high risk. Blinding of outcome assessment of all studies was also assessed as high risk because the studies included self‐reported measures of carers and the blinding of carers was not possible.

Incomplete outcome data

Three of the nine studies addressed the issue of incomplete outcome data and reported the drop‐out rates, with reasons, for the intervention group and control group (Finkel 2007; Gant 2007; Steffen 2000).

Selective reporting

Six studies appeared to be free of selective reporting (Finkel 2007; Gant 2007; Glueckauf 2007; Steffen 2000; Wilz unpublished; Winter 2007), and two studies presented only the significant outcomes (Davis 2004; Tremont 2008), with Tremont 2008 providing the requested outcomes via email. Chang 1999 did not state caregiver burden.

Other potential sources of bias

All studies appeared to be free of other potential sources of bias.

Evaluation of experience

We used the Critical Appraisal Skills Program (CASP) for the assessment of one qualitative study (Salfi 2005), and one qualitative process evaluation (Chang 2004), and conducted an evaluation of methodological quality.

A short version of the results of the CASP assessment are presented in Table 7, and in a more detailed version in Appendix 23 and Appendix 24. The quality of the two qualitative 'thin description' studies was assessed as moderate. Chang 2004 had specific methodological issues regarding rigour, as interviews were not recorded and transcribed verbatim.

6. Brief overview of the critical appraisal of included qualitative studies with the CASP instrument.

No.	Screening questions	Salfi 2005	Chang 2004
1	Was there a clear statement of the aims of the research?	●	●
2	Is a qualitative methodology appropriate?	●	●
	Is it worth continuing?	●	●
3	Was the research design appropriate to address the aims of the research?	●	/
4	Was the recruitment strategy appropriate to the aims of the research?	▼	●
5	Were the data collected in a way that addressed the research issue?	▼	/
6	Has the relationship between researcher and participants been adequately considered?	●	▼
7	Have ethical issues been taken into consideration?	●	▼
8	Was the data analysis sufficiently rigorous?	●	▼
9	Is there a clear statement of findings?	▼	▼
10	How valuable is the research?	●	●

● = Yes; ▼ = Yes with restrictions; / = No

Confidence in the qualitative synthesis findings

CerQual findings are displayed in the summary of qualitative findings (see Table 8). All 'synthesised' qualitative findings were assessed as being low confidence because they were based on only one or two studies with moderate quality from similar settings. Only 5/16 themes overlapped both studies, so the majority of findings came from only a single study. Therefore very little synthesis across studies was possible.

7. Summary of qualitative findings.

Analytical theme	Descriptive theme	Certainty in the evidence	Explanation of certainty in the evidence assessment
Barriers and facilitators for successful implementation of telephone counselling	Medium of telephone	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
	To know each other	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
	Multilingualism	Low certainty	The study was of moderate quality. However, the finding is only from one study in Canada.
	24 h attainability of the counsellor	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
	Counselling after institutionalisation	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
	Debriefing of counsellors	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
Counsellors' emotional attitude	Not condemning, but appreciative	Low certainty	The studies were of moderate quality. However, the finding is only from two studies in USA and Canada.
	To be acquainted with	Low certainty	The studies were of moderate quality, however, the finding is only from two studies in USA and Canada
	To be reliable	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
Content of telephone counselling	How is the counselling given?
	Dimension tangible information ‐ intermediation	Low certainty	The study was of moderate quality, however, the finding is only from one study in Canada
	Individual counselling	Low certainty	The studies were of moderate quality, however, the finding is only from two studies in USA and Canada
	What is being discussed?
	Problematic situations	Low certainty	The studies were of moderate quality. However, the finding is only from two studies in USA and Canada.
	Conversational themes to forget daily routine	Low certainty	The study was of moderate quality. However, the finding is only from one study in USA.
	Giving information	Low certainty	The studies were of moderate quality. However, the finding is only from two studies in USA and Canada.
	Talk about emotions, thoughts and carer's condition	Low certainty	The study was of moderate quality. However, the finding is only from one study in USA.
	Empowerment for autonomy	Low certainty	The study was of moderate quality. However, the finding is only from one study in Canada.

Effects of interventions

See: Table 1

Evaluation of efficacy

The results of the nine included studies are displayed in three categories based on the different structures of the telephone counselling. The first category includes studies that conducted telephone counselling without an additional intervention (Davis 2004; Finkel 2007; Glueckauf 2007; Tremont 2008; Wilz unpublished; Winter 2007). The second category conducted telephone counselling in combination with video sessions (Chang 1999). The third category conducted telephone counselling in combination with video sessions and a workbook (Gant 2007; Steffen 2000).

Telephone counselling without additional interventions

Primary outcome: depressive symptoms

A meta‐analysis was performed within the first group of three studies (Finkel 2007; Tremont 2008; Winter 2007). Davis 2004 did not report the results because they were not significant, and they did not respond to email inquiries from the review authors. Glueckauf 2007 reported their results as medians and also did not respond to inquiries from the review authors for additional data, so the data are reported in narrative form. The results of Wilz unpublished were also reported in narrative form, because the intervention differed from the others. The review authors assessed the intervention as a therapy for carers who were not in need of therapy.

Pooled results are described in a forest plot (see Figure 5). Two of the studies assessed depressive symptoms at six months (Finkel 2007; Winter 2007), and one after 12 months (Tremont 2008). The forest plot shows a significant pooled difference in depressive symptoms between the telephone counselling group and the control group (standardised mean difference (SMD) 0.32, 95% confidence interval (CI) 0.01 to 0.63, P value 0.04). The forest plot shows a non‐significant pooled difference in caregiver burden in favour of the telephone counselling group.

5.

Forest plot of comparison: 1 Telephone counselling, outcome: 1.1 Depressive symptoms

Glueckauf 2007 reported a reduction of depressive symptoms (as medians) in the intervention group compared to the control group after three months: ‐5.0 percentage points in the intervention group (from 15.0% to 10.0%) and no reduction in the control group (from 10.0% to 10.0%). However, these reductions were not significant. No detailed data about statistical significance were given by the study authors.

Wilz unpublished presented results after three months (directly after the end of the intervention) and after nine months follow‐up. The trial reported no differences between groups, but a reduction in depressive symptoms in both groups after three months (adjusted mean difference 0.66, 95% CI ‐2.81 to 4.14), and nine months (adjusted mean difference 0.06, 95% CI ‐3.95 to 4.07). The greater reduction of depressive symptoms was observed in the control group.

Secondary outcomes

Burden

The review authors performed a meta‐analysis that included (Davis 2004; Finkel 2007; Tremont 2008; Winter 2007). Finkel 2007 assessed caregiver burden with the Revised Memory and Behavior Problems Checklist (RMBPC) and reported the results using the two dimensions frequency and bother, which were both represented in the meta‐analysis. Glueckauf 2007 reported their results as medians and did not respond to the review authors' request for further data, so the available data are reported in narrative form.

Pooled results are described in a forest plot (see Figure 6). Three studies assessed the caregiver burden at six months (Davis 2004; Finkel 2007; Winter 2007). Davis 2004 also assessed burden after three months (directly after the intervention), but the review authors preferred to include the later results for a better comparison of outcome data. One study assessed caregiver burden after 12 months (Tremont 2008). The forest plot shows a non‐significant pooled difference in caregiver burden between the telephone counselling group and the control group (SMD 0.45, 95% CI ‐0.01 to 0.90, P value 0.05). The forest plot shows no significant difference in caregiver burden between groups.

6.

Forest plot of comparison: 1 Telephone counselling, outcome: 1.2 Burden

Glueckauf 2007 reported a reduction of caregiver burden (as medians) for the intervention and control group after three months: ‐0.21 percentage points in the intervention group (from 1.53% to 1.32%) and ‐0.1 percentage point in the control group (from 0.89% to 0.79%). The study authors only stated that the reductions were not significant, but provided no details on statistical significance.

Distress

Davis 2004 presented the results for carer distress for two assessment points, showing a reduction for the intervention and control group directly after the intervention (after three months): ‐1.4 percentage points in the intervention group (from 14.2% to 12.8%) and ‐2.6 percentage points in the control group (from 18.2% to 15.6%). A reduction of carer distress was only reported for the intervention group after six months: ‐3.0 percentage points in the intervention group (from 14.2% to 11.2%), and an increase of carer distress of 2.3 percentage points was found in the control group (from 18.2% to 20.5%). The study authors only reported that the reductions were not significant, without any further details.

Quality of life

Tremont 2008 did not state the quality of life data in their publication but provided them upon email inquiry. They reported a decrease of quality of life in both groups after 12 months: ‐0.31 percentage points in the intervention group (from 50.00% to 49.69%) and ‐4.71 percentage points in the control group (from 55.59% to 50.88%). Reductions were not significantly different (P value 0.07).

Self‐efficacy

A meta‐analysis was not possible since Glueckauf 2007 reported the results in medians and did not respond to the review authors’ request for further data, so the data are reported in narrative form. Tremont 2008 reported the self‐efficacy data upon email inquiry.

Glueckauf 2007 reported an increase of carers self‐efficacy (as medians) for the intervention group and the control group after three months: 11.66 percentage points in the intervention group (from 66.67% to 78.33%) and 14.33 percentage points in the control group (from 67.00% to 81.33%). The study authors reported that the pretest‐posttest gain was significantly higher in the control group (P value 0.02).

Tremont 2008 stated a decrease of carer self‐efficacy for the intervention group and an increase of self‐efficacy for the control group after 12 months: ‐1.69 percentage points in the intervention group (from 43.00% to 41.31%) and 1.11 percentage points in the control group (from 44.24% to 45.35%). The group difference was not statistically significant (P value 0.42).

Satisfaction

Carer satisfaction was assessed by Davis 2004, but the study did not report the results because they were not significant. They did not respond to email inquiries from the review authors.

Social support

A meta‐analysis for social support was conducted for two studies (Finkel 2007; Tremont 2008). Tremont 2008 stated the data upon email inquiry. Davis 2004 did not report the results because they were not significant and did not respond to email inquiries from the review authors.

Pooled results are described in a forest plot (see Figure 7). One study assessed social support after six months (Finkel 2007), while the other assessed it after 12 months (Tremont 2008). The forest plot shows a non‐significant difference in social support between the telephone counselling group and the control group (SMD 0.25, 95% CI ‐0.24 to 0.73, P value 0.32). The forest plot shows no significant difference in caregiver burden between groups.

7.

Forest plot of comparison: 1 Telephone counselling, outcome: 1.3 Support

Telephone counselling in combination with video sessions

Chang 1999 was the only study that investigated telephone counselling in combination with video sessions. The trial presented the outcome results for three assessment points after baseline.

Primary outcome: depressive symptoms

Chang 1999 reported no differences between groups. Instead, the trial reported a reduction of depressive symptoms only in the intervention group after one month: ‐0.08 percentage points in the intervention group (from 0.73% to 0.65%) and no change in the control group (0.74% to 0.74%). After two months (directly after the end of the intervention), there was a reduction in depressive symptoms in the intervention group and an increase in the control group: ‐ 0.04 percentage points in the intervention group (from 0.73% to 0.69%) and 0.12 percentage points in the control group (from 0.74% to 0.86%). A reduction of depressive symptoms in the intervention group and an increase of depressive symptoms in the control group was also found after three months: ‐0.13 percentage points in the intervention group (from 0.73% to 0.60%) and 0.21 percentage points in the control group (from 0.74% to 0.95%). Study investigators reported a significant difference in depression over time (pre‐intervention‐post‐intervention comparison) for the intervention group (P value 0.02).

Secondary outcomes

Burden

Chang 1999 assessed caregiver burden but reported only baseline data and did not respond to email inquiries from the review authors.

Anxiety

Chang 1999 reported no differences between groups. Instead, the trial reported a reduction of carer anxiety in the intervention group and an increase in the control group after one month: ‐0.03 percentage points in the intervention group (from 0.68% to 0.65%) and 0.02 percentage points in the control group (from 0.91% to 0.93%). After two months, anxiety was reduced in both groups: ‐0.09 percentage points in the intervention group (from 0.68% to 0.59%) and ‐0.24 percentage points in the control group (from 0.91% to 0.67%). A reduction in both groups was also reported after three months: ‐0.11 percentage points in the intervention group (from 0.68% to 0.57%) and ‐0.13 percentage points in the control group (from 0.91% to 0.78%). The study investigators stated that carer anxiety was significantly reduced over time in both groups after three months (pre‐intervention‐post‐intervention comparison), but did not provide any details about statistical significance.

Satisfaction

Chang 1999 reported no differences between groups. Instead, the trial reported a decrease in carer satisfaction in both the intervention group and the control group after one month: ‐1.12 percentage points in the intervention group (from 29.03% to 27.91%) and ‐1.91 percentage points in the control group (from 28.23% to 26.32%). A decrease was still visible in both groups after two months: ‐1.74 percentage points in the intervention group (from 29.03% to 27.29%) and ‐1.73 percentage points in the control group (from 28.23% to 26.50%). A decrease of carer satisfaction in both groups was also reported after three months: ‐1.74 percentage points in the intervention group (from 29.03% to 27.29%) and ‐1.52 percentage points in the control group (from 28.23% to 26.71%). Study investigators identified a significant decrease of carer satisfaction over time (pre‐intervention‐post‐intervention comparison) within both groups (P value 0.015).

Telephone counselling in combination with video sessions and a workbook

A meta‐analysis of telephone counselling in combination with video sessions and a workbook was not conducted due to clinical heterogeneity. Gant 2007 conducted the telephone counselling with a focus on behavioural management instead of cognitive or emotional aspects. Steffen 2000, however, included cognitive strategies. Consequently, the results of both studies are presented in narrative form.

Primary outcome: depressive symptoms

Steffen 2000 reported a reduction of depressive symptoms in the intervention group and an increase in the control group after two months: ‐4.1 percentage points in the intervention group (from 7.20% to 3.10%) and 0.56 percentage points in the control group (from 7.44% to 8.00%). The authors stated that the difference between the groups was statistically significant (P value 0.01).

Secondary outcomes

Burden

Gant 2007 showed a reduction of caregiver burden in both the intervention and control groups for upset and annoyance after three months: upset: ‐0.3 percentage points in the intervention group (from 1.3% to 1.0%) and ‐0.5 percentage points in the control group (from 1.4% to 0.9%); annoyance: ‐0.1 percentage points in the intervention group (from 1.1% to 1.0%) and ‐0.4 percentage points in the control group (from 1.2% to 0.8%). The study investigators reported no differences between groups (P value 0.55).

Distress

Gant 2007 presented the distress results for upset, sadness, frustration, and irritation according to the target complaints subsets after three months. The data showed a reduction in upset in both groups: ‐0.9 percentage points in the intervention group (from 3.2% to 2.3%) and ‐0.9 percentage points in the control group (from 3.3% to 2.4%). Sadness was also reduced in both groups: ‐0.8 percentage points in the intervention group (from 3.0% to 2.2%) and ‐0.9 percentage points in the control group (from 3.0% to 2.1%). Frustration decreased in both groups: ‐0.9 percentage points in the intervention group (from 3.1% to 2.2%) and ‐0.5 percentage points in the control group (from 3.1% to 2.6%). Irritation had decreased equally in both groups: ‐0.6 percentage points in the intervention group (from 2.8% to 2.2%) and ‐0.6 percentage points in the control group (from 2.8% to 2.2%). The study investigators stated no differences between groups (P value 0.38).

Self‐efficacy

Gant 2007 reported self‐efficacy results after three months for three subsets: obtaining respites, handling problematic behaviour, and controlling negative thoughts. The capacity to obtain respite increased in both groups: 1.6 percentage points in the intervention group (from 70.7% to 72.3%) and 8.4% percentage points in the control group (from 61.1% to 69.5%). The capacity for dealing with problematic behaviour also increased in both groups: 8.2 percentage points in the intervention group (from 72.9% to 81.1%) and 7.9 percentage points in the control group (from 70.2% to 78.1%). The ability to control negative thoughts increased in both groups: 3.3 percentage points in the intervention group (from 67.2% to 70.5%) and 6.7 percentage points in the control group (from 71.6% to 78.3%). The study investigators stated no differences between groups (P value 0.80).

Steffen 2000 showed an increase of carer self‐efficacy after two months in both the intervention group and control group: 19.05 percentage points in the intervention group (from 61.43% to 80.48%) and 3.22 percentage points in the control group (from 61.22% to 64.44%). The authors stated that the difference between the groups was statistically significant (P ≤ 0.01).

Sensitivity analyses: handling missing data

Winter 2007 reported results for the intervention group and control group but did not state how many carers were in each group, instead the trial authors only stated the total number of carers participating in the trial. Study investigators did not provide these data although we, the review authors, requested it. We decided to allocate the drop‐outs to the intervention and control groups in the same proportion as at baseline. However, this was just an assumption, so we conducted sensitivity analyses for the outcomes of depressive symptoms and burden. Within the sensitivity analysis, we performed meta‐analysis for both outcomes again, but put all drop‐outs into either the intervention group or the control group.

Depressive symptoms

1. All nine drop‐outs allocated to the intervention group: SMD 0.32, 95% CI 0.01 to 0.63, P value 0.04.

2. All nine drop‐outs allocated to the control group: SMD 0.32, 95% CI 0.01 to 0.63, P value 0.05.

Burden

1. All 22 drop‐outs allocated to the intervention group: SMD 0.41; 95% CI ‐0.02 to 0.84; P value 0.06.

2. All 22 drop‐outs allocated to the control group: SMD 0.42; 95% CI ‐0.01 to 0.84; P value 0.05.

The sensitivity analyses showed that the results of meta‐analysis will not be affected by the different possibilities of drop‐out allocation.

Adverse effects or harm

None of the included quantitative studies reported adverse effects or harm due to telephone counselling.

Subgroup analysis

Fewer than five studies were available per planned aspect (carer‐related, intervention‐related, disease‐related, experience‐related), so subgroup analysis or meta‐regression could not be conducted.

Evaluation of experience

The qualitative evidence helped to provide a better understanding of what it is like to receive and deliver telephone counselling. These factors impact on effective implementation and explain why interventions may 'work' (or not) to reduce depressive symptoms.

We were able to derive three analytical themes for conducting telephone counselling based on two studies that investigated the experiences of carers of persons with dementia with telephone counselling (Chang 2004; Salfi 2005) (see also the main findings of included qualitative studies Appendix 25). The first theme concerns the barriers and facilitators for successfully implementing telephone counselling, the second concerns the content of the counselling, and the third concerns the counsellor's emotional attitude. The findings presented by Salfi 2005 are relevant to all three main themes. The findings presented by Chang 2004 are only relevant to the themes of content and counsellor's emotional attitude (see Table 9).

8. Origins of the descriptive themes.

Analytical theme	Descriptive theme	Chang 2004	Salfi 2005
Barriers and facilitators for successful implementation of telephone counselling	Telephone as communication tool		●
	To know each other		●
	Multilingualism		●
	24‐hour reachability of the counsellor		●
	Continuation of counselling after institutionalisation		●
	Debriefing of counsellors		●
Counsellor's emotional attitude	Non‐judgmental/appreciative	●	●
	To be familiar and trusted	●	●
	To be reliable		●
Content of telephone counselling	How is the counselling given?
	The range from providing factual information to mediation		●
	Individual counselling	●	●
	What is being discussed?
	Problematic situations	●	●
	Conversational themes that help forget the daily routine	●
	Information and advice	●	●
	The carer’s emotions, thoughts, and physical condition	●
	Empowerment for autonomy		●

● The descriptive theme is based on data from this study

These three analytical themes include descriptive themes that are introduced and explained further below.

Barriers and facilitators for successful implementation of telephone counselling

Evidence from carers and counsellors includes six themes on the barriers and facilitators for successful implementation of telephone counselling. These themes relate to the counselling itself, as well as to different organisational and conceptual aspects. The following descriptive themes relate particularly to the overall form and structure of a telephone counselling, which is defined by the counselling institution, and not to its content, which is determined by the counsellor.

Telephone as communication tool

Carers find the telephone a good tool for receiving counselling as it helps avoid the stress involved in co‐ordinating an appointment. It is a very comfortable means of counselling because carers do not have to leave their homes and its flexibility allows sessions to be fitted into their daily nursing routines. Salfi 2005 was the only study that evaluated telephone counselling as an effective way to fulfil the needs of carers.

One of the carers who participated in the Salfi 2005 study said:“Well, because you can pick it (the telephone) up anytime, you don't have to go out, you don't have to leave the house, you don't have to get your mom all dressed . . . “ (Carer #5).

However, critical statements regarding the limits of telephone counselling were also made. Salfi 2005 observed that the spatial distance can be a problem because counsellors are not able to see the reactions of carers and interpret how they assimilate their recommendations.

A counsellor included in Salfi 2005 said: “One of the main limitations that I did run into . . . is that you cannot see the person, so you cannot see the expressions on the client's face, their mood. You sometimes can't get a feeling, you know, about how things are being received.” (Provider #4).

The counsellors felt the disadvantage of being off‐site particularly clearly when carers were angry or distraught They were aware of the limited influence they have via the telephone and knew that the carer remained in the same difficult situation as before the call.

To know each other

Carers and counsellors agree on the importance of getting to know each other before counselling. Carers can appreciate the counsellor as a person and know that the counsellor is familiar with their situation and can understand the situations the carer describes. The following example of a carer from the Salfi 2005 study illustrates this point:

“You put a face to the name and you know them and they know you and they know Dad so they know what you're talking about. If you're talking to someone who doesn't know you, you don't know them, they don't know Dad, how can anybody give you advice? If you're speaking strictly people with dementia, there's so many different forms of it and so many different degrees of it, that how can you possibly give someone advice about something that you don't even know the situation? I don't understand how they can do that. I think you have to know the person involved Otherwise I don't think you can actually . . . I mean blanket information is okay, but I don't think you can actually give good help or information . . . if you don't know the situation or the person.” (Carer #1).

Salfi observed that the counsellors might also feel frustrated and helpless during telephone counselling. A reason for this frustration could be that telephone counselling is relatively impersonal as the following quote from the Salfi 2005 study shows:

“ . . . over the phone, there's only so much I can do to really help someone. I mean, I can't be there if someone is extremely agitated or repeating questions over and over. And I can hear it in their voice that they're just at the end of their rope ‐ the caregiver. I feel somewhat limited in what I can really offer them. You know, I can talk to them and I can listen to them and maybe make them feel a little bit better for the moment, but then they hang up and they're still in that situation. I suppose also a limitation is that it is anonymous. And, in a way, I mean people can . . . I don't necessarily know who I'm talking to, and so it could just be a onetime thing. But I could get a pretty distressing phone call, not have a clue who I'm speaking to, and then I hang up and that's it. I'll never know what happened, unless they choose to call back.” (Provider #4).

A carer also expressed reservations about the idea of getting good advice from an unknown person since the disease has a great variety of manifestations. The descriptive theme of 'knowing each other' is only reported in the Salfi 2005 study.

Multilingualism

Telephone counselling of carers who do not speak the national language is problematic. Salfi observed that the needs of carers cannot be appropriately addressed when language barriers exist.

A counsellor expressed his or her irritation in Salfi 2005:“Limitations. Okay. Language barriers. That is major because we have several clients that have trouble with the English language. I would not attempt . . . I mean it's hard enough to find out what the problem is, let alone to try and have a discussion” (Provider #2).

Only the Salfi 2005 study addressed the carer’s need for multilingual counselling. The Chang 2004 study did not provide any information on this descriptive theme.

24‐hour availability of the counsellor

Carers agreed that 24‐hour availability of telephone counselling is necessary. Carers need assistance most acutely in difficult situations when they feel insecure and see that an escalation is possible. Counselling for carers is most needed, for example, in the case of an accident or during the final phase of the disease. Assistance is needed therefore outside regular working hours, for example during the night. A carer in the Salfi 2005 study described her increased need for assistance as follows:

“. . . The time they're available, 'cause what if you get in trouble from eight (pm) to five (am). Oh well, it'd be nice to get maybe a hold of somebody 24 hours if there was an accident or . . . one time, she (her mother) fell a while back and, she was in bed and I was really tired and I would've just like to have been able to get somebody on the phone and ask them some questions . . . like what I should do next.” (Carer #2).

Counsellors also agreed that increased availability would meet the carers’ needs better. Existing arrangements, such as answering machines for the times when counselling is not available, were not sufficient. A counsellor reported that many telephone calls recorded after work hours would actually need immediate counselling. Although the need for extended availability was generally supported, counsellors also questioned its feasibility. A counsellor from the Salfi 2005 study said:

“Well, you know, I'm sure for the caregiver it would be great if we were available in the evening . . . I don't know how feasible that would be. We do have voice mail, but they'll just be able to leave a message. And I do . . . you know, when I come in the morning, there are usually a couple of messages at least . . . A lot of people do call after four‐thirty . . . Now they're mainly calls looking for information . . . but some also for counseling.” (Provider #3).

The need for an increased reachability of counsellors was only addressed in the Salfi 2005 study. Carers in the study of Chang 2004 did not address this aspect.

Continuation of counselling after institutionalisation

Carers experience the abrupt termination of telephone counselling due to the institutionalisation of the person with dementia as burdensome and a revocation of a reliability they depended on. A carer in the Salfi 2005 study who was upset about the abrupt ending of the counselling described her feelings as follows:

“It was my counselor . . . I, I couldn't believe [it] but that was the way things were set up. I thought, I need her now . . . (they stopped the service when her spouse was placed in nursing home) . . . I was a basket case . . . but [the telephone support provider] picked up the pieces after, you know what I mean?” (Carer #7).

Both carers and counsellors expressed the need to continue with telephone counselling. A counsellor said that the need for counselling is especially high during the times when a person with dementia is transferred and suspects that it is even higher after the person with dementia is institutionalised. The Salfi 2005 study addressed this theme, but Chang 2004 did not cover this aspect of carers’ needs.

Debriefing of counsellors

Counsellors expressed the need for debriefing with colleagues after counselling sessions, especially after they had had to address difficult situations. Several different aspects can be discussed in a debriefing. Salfi 2005 suggested themes such as the discussion and evaluation of the assessment tools used, the actions proposed to a carer, discussion of unanswered questions and addressing feelings of frustration and helplessness.

Salfi described the need for debriefing after counselling with these words: “The providers also identified the importance of debriefing with colleagues after the support sessions to validate their assessments and plans of care for the caregiver, discuss any unanswered questions that remained after the call, and vent any feelings of helplessness and frustration. Having the ability to follow up with caregivers was also important to all of the providers, especially after difficult support sessions.“ (Salfi 2005).

Although the need for debriefing was discussed in Salfi 2005, it was not mentioned in the results. Chang 2004 did not identify this aspect, possibly because the study focused on carers, not on counsellors.

Content of telephone counselling

The analytical theme of the 'content of telephone counselling' addressed the counsellor role and could be divided into two sub‐themes. The first sub‐theme, 'How is the counselling given?', covers aspects of how the counselling content is communicated and includes several descriptive themes that deal with the methods available to counsellors and how they should proceed during counselling. The second sub‐theme addresses what content is communicated.

How is the counselling given?

Range of content: providing factual information to providing mediation

In providing information to a carer, a counsellor can choose to provide more or less concrete information. She or he may either advise very specifically on the next steps, or might refer the carer to other institutions or services that might help. Either approach might be perceived as helpful by the carer depending upon her or his individual situation and preferences.

Two counsellors from the Salfi 2005 study reported: “. . . and then I sort of direct them through the system really. I tell them how to sort of navigate through the system" (Provider #1). "I tell them about the CCAC and what they do. I tell them what our role is and what we do . . . so many of the phone calls I get are about that. They’re just looking for: What do I do now? Where do I go now?” (Provider #4).

This quote describes how the counsellor introduces various institutions from whom the carer can get support. The next quotation describes how the counsellor can provide very specific advice for managing a difficult situation.

A carer described: “My husband had very bad problems when he had constipation. He didn't go for ten days. So I called for advice . . .and she was going to send me a recipe for mixing All Bran and applesauce and prune juice . . .and then he went, so I called her back to thank her for being concerned.” (Salfi 2005; Carer #2).

This differentiation of the range of how concrete the information and advice given to carers might be was not described in either Chang 2004 or in the Salfi 2005 study. This descriptive issue emerged during the thematic synthesis and is supported by Salfi’s findings.

Individual counselling

Shaping messages and information specifically for individual carers is a characteristic of counselling. Salfi formulated that both carers and the counsellors were satisfied with the individualisation of counselling and described:

“Both caregivers and telephone support providers described the intervention of telephone support as a source of information and educational advice, tailored to each individual caregiving situation” (Salfi 2005).

When the counselling is not sufficiently individualised, carers can become dissatisfied. Chang reported one instance of telephone counselling that was perceived as unhelpful, but that took a positive turn after a follow‐up call:

“Only one caregiver from the experimental group indicated the information was not applicable; however, in subsequent conversations, this caregiver indicated she tried some of the strategies and found them to be useful” (Chang 2004).

Individual counselling is not only about providing information but also about listening. The Salfi 2005 study described the aspect of listening using the statement of a counsellor who believed that many carers call because they want to talk to someone who really listens to them.

What is being discussed?

The second sub‐theme includes all the descriptive themes that describe the possible contents of counselling: problematic situations; conversational themes that help forget the daily routine; information and advice; the carer’s emotions, thoughts and physical condition; empowerment for autonomy.

Problematic situations

Being able to describe and discuss problematic situations helps carers in their daily lives when they do not know how what to do next, because, from their perspective, they have tried everything but have not been able to resolve the situation. A carer in the Salfi 2005 study reported:

“. . .they told me . . .there is no point in arguing because he was always right. Always right. You know uh, for instance, he was putting both legs in his pant leg ‐ one pant leg. Now he's very frustrated and he's becoming very abusive. The way to handle that I found was oh, to say . . .please don't be mad at me, and he'd say, I won't be mad at you, and I'd take them away and straighten them out.” (Salfi 2005; Carer #7).

According to the Salfi 2005 study, carers recognise that they have training needs. They want to receive information and learn strategies to cope with and resolve problematic situations better. Both the Salfi 2005 study and the Chang 2004 study identified the discussion of problematic situations as meeting a carer need. Chang 2004 reported that carers perceive it as helpful when they can express their feelings and describe and discuss their problematic situations.

Conversational themes that help forget the daily routine

As shown in 'Problematic situations' above, carers find it helpful if they can talk about stressful situations. On the other hand, carers also find it helpful to have conversations about other things that do not remind them of their nursing duties. Chang described this as follows:

“Some caregivers indicated they were able to forget their situation while talking on the telephone to the nurse” (Chang 2004).

Chang interviewed not only participants who had received telephone counselling but also the participants of a control group (attention‐only group). The participants in the control group had conversations only about general topics such as the weather, television, movies, news, or social activities. Such a conversation about common topics was found to be a helpful alternative to relieve carers.

However, Chang 2004 also reported that those conversations could also be perceived as not helpful and describes this with the following words:

“Two caregivers in the attention‐only group indicated the telephone calls were not useful” (Chang 2004).

The descriptive theme of 'conversational themes that to help forget the daily routine' was only addressed and identified as a category by Chang 2004. The Salfi 2005 study only analysed participants who received telephone counselling, and had no control group.

Information and advice

This descriptive topic covers information about various aspects of nursing care for persons with dementia. The authors of the aforementioned studies did not explain what exactly they meant by the term 'information'. The aspect of offering healthcare system orientation, in which counsellors explain the institutions that are available locally and their respective services, was described only in the Salfi 2005 study where it was also identified as a unique theme. Together with the carers, counsellors identify the most suitable services, which helps the carers to understand that they have access to an extensive help system in addition to telephone counselling. A counsellor from the Salfi 2005 study described the provision of information as follows:

“Well, most of the people who phone in here, I would say, are really looking for information. Many people are information seeking . . .And then, you know, I can send some stuff out. Or if they have specific questions, I can answer them . . .Or they'll say: "My parents are having difficulty. Are there any services out there? How do we get in‐home care?" I then tell them about the CCAC and what they do.,. And then I tell them what our role is and what we do, which I think is more of an educational and support service really” (Salfi 2005; Provider #1).

Carers find the information provided helpful and are grateful for it. The information and advice help them to do certain things better, which leads to them feeling happier with themselves. Providing information and advice also helps the counsellors to feel more satisfied and competent in doing their job. The descriptive theme 'information and advice' was identified both in the Chang 2004 and Salfi 2005 studies where, in both studies, carers expressed a great need for information, and counsellors agreed that carers have an extensive need for information.

The carer’s emotions, thoughts, and carer's physical condition

This descriptive theme covers talking about emotions, thoughts and the physical condition of the carer. It is closely related to the descriptive theme 'Discuss problematic situations', but instead of focusing on a problem, it addresses the many personal aspects that are triggered in carers. These can be, for example, the thoughts and feelings the carer experiences in connection with the changes in the personality of the person with dementia and the decrease of their own quality of life or social contacts. A carer in the Chang 2004 study described his perceived loneliness and how telephone counselling was helping him:

“I am sleeping a little better at night, but feel lonely without my wife at night. Sometimes I worry about being alone at night. I’m feeling a little down right now . . . I miss my friends and social life – my friends are getting few and far between . . . Talking to you helps me with some of this“ (Chang 2004; Carer).

Carers benefit from being able to express their helplessness. They feel helpless because they do not want to move their family member with dementia to a nursing home, but at the same time they realise that they are overextended. Overall, this descriptive theme focuses on discussing thoughts, emotions and the physical condition of carers, which carers experience as helpful. Chang 2004 identified the category of 'sharing thoughts and feelings' but did not include any physical aspects and the general condition of carers, even though they were recorded in the study. The Salfi 2005 study did not evaluate any aspect of this descriptive theme.

Empowerment for autonomy

Counsellors are gratified when they can guide carers back to feeling empowered in difficult situations. The carers are then able to recognise their own situation and needs and can themselves find and pursue the appropriate support services. This empowerment means that they can find their own solutions. Only Salfi described this as an individual descriptive theme.

A counsellor from the Salfi 2005 study described:

“Empower them to make their own referrals, to look for their own solutions. To choose from what's available in the community. For instance, a support group is not for everyone. There's a readiness for respite, to accept help in the home or outside of the home” (Salfi 2005; Provider #4).

The counsellor's emotional attitude

This analytical theme describes the emotional attitude of the counsellor towards the carer. The included descriptive themes affect the quality of counselling since the counsellor indirectly communicates his or her attitude to the carer. These themes cannot be ‘switched on’ by the counsellor but develop gradually. Being aware that these descriptive themes are important for carers, counsellors can contribute to their development through many small gestures, remarks and hints. This analytical theme is closely connected to the descriptive theme 'How is the content communicated? but is distinct from it as it focuses on the emotional attitude of the counsellor rather than on tools or techniques for giving information and advice.

Non‐judgmental/appreciative

This descriptive theme describes how the attitude of the counsellor needs to be non‐judgemental towards the ‘non‐professional’ care provided by the carers and needs to convey appreciation for their great efforts. For example, carers might feel sad and guilty after sending their family member with dementia to a nursing home, or they might be ashamed of their anger. With positive recognition of their nursing care, the confirmation that their decisions were correct and their emotions justified, counsellors can help to minimise carers’ concerns.

Salfi described: “They described the comfort of knowing that someone is out there who will listen to them, acknowledge their efforts, and provide encouragement and reassurance . . .” (Salfi 2005).

Carers found the acknowledgment and appreciation given in telephone counselling to be helpful. Both the Chang 2004 and Salfi 2005 studies provided evidence for this descriptive theme, which Chang identified as “seeking reassurance”.

To be familiar and trusted

This descriptive theme is based on the need of carers for a person they can confide in and who is closer to them than an acquaintance. It differs from the descriptive theme 'To know each other' in that feeling familiar with and trusting another person is a sign of a deeper relationship. It is not only about having met the other person, but more about the counsellor creating an atmosphere that nurtures trust and thus creates a true relationship. This somewhat more intimate relationship was described by Chang as follows:

“Expressions in this category [note of review author: The category of 'Sharing thoughts and feeling'] related to the sharing of thoughts and feelings with someone such as a close friend” (Chang 2004).

Carers describe their wish for a counsellor they can confide in with concepts like 'close friend', 'confide' or 'to be private'. Both the Chang and Salfi studies described the theme of familiarity and trust, but did not identify it as a separate category or theme. This thematic synthesis evaluates the descriptive theme of 'To be familiar and trusted' for the first time.

To be reliable

Carers express their desire to know they have a reliable person by their side. Feeling insecure in certain situations, they want to know that they are not alone. A carer from the Salfi 2005 study described how he perceived his counsellor as reliable with the fitting words:

“There's always somebody there. There's an ear at the other end of that phone” (Salfi 2005; Carer #1).

Carers wish for someone to be there to help them whenever they need help, regardless of the time of the day or the situation. Only Salfi’s study provides evidence for the descriptive theme 'To be reliable' but does not identify it as an independent theme. The thematic synthesis discovers 'To be reliable' as a new descriptive theme.

Correlations between the analytical topics

The three analytical themes that developed in telephone counselling to carers of persons with dementia are closely interconnected. Figure 1 gives an overview and explains how these analytical themes relate each other.

Telephone counselling as intervention consists of three analytical topics that all affect carers.

The barriers and facilitators for successful implementation of telephone counselling influence the emotional attitude of the counsellors: the limitations of the telephone as a tool for counselling can lead to frustrations among counsellors and influence their emotional attitude, which in turn might make the building of the relationship across the physical distance more difficult. The descriptive theme of 'Knowing each other' on the other hand, might have a positive impact on the counselling, partly because when counselling is not anonymous it helps counsellors to manage the frustrations of the carers, and partly because it lays the groundwork for becoming 'Familiar and trusted', which is the relationship desired by the carers. Receiving counselling in the native language was also shown to contribute to building trust, while 24‐hour availability and continued counselling after institutionalisation was shown to promote reliability. A debriefing after counselling may help counsellors to reflect on their behaviour and potential prejudices, and thus help them develop a non‐judgmental and appreciative attitude towards the carers.

The emotional attitude of the counsellor is closely connected to the content of counselling, since the manner in which counselling is given reflects the counsellor’s emotional attitude. For example, if the counsellor has prejudices against the carer or judges his or her care giving as insufficient, this would be reflected in the content of counselling (See Figure 1; arrow from emotional attitude towards content of counselling).

Carers will consciously or unconsciously notice the emotional attitude of the counsellor, which can lead them to trust the counselling less and experience it as unhelpful. At the same time, the level of thoroughness with which counsellors convey the content of the discussion has an impact on how reliable and trustworthy the carer experiences the counsellor. In Figure 1 these aspects are illustrated by the arrows pointing from counselling content towards the emotional attitude of counsellors. The emotional attitude of the counsellor plays a central role during telephone counselling, and thus takes an exposed position.

Adverse effects or harm

One study reported a negative effect of telephone counselling (Chang 2004). At the beginning of the intervention, one carer indicated that counsellor's information was not applicable. However, this negative effect was eliminated by discussion with the counsellor.

Integration of quantitative (efficacy) and qualitative (experience) data (third synthesis)

As the level of integration was 'low' due to the paucity and quality of qualitative evidence, it was not possible to develop a sophisticated interpretive synthesis.

First part of the integration of quantitative and qualitative data

Matrix 1 maps the descriptive themes of the thematic synthesis that were investigated in quantitative trials (see Table 3). The following high‐level observations can be made at a descriptive level:

Most of the descriptive themes of the analytical theme 'Content of telephone counselling' were covered by the quantitative trials. Only the descriptive theme 'Conversational themes that help forget the daily routine' was mentioned by two studies (Chang 1999; Davis 2004). Neither study used this strategy in the telephone counselling but mentioned, with respect to the comparison groups that were receiving friendly calls, that talking about things other than caring for a relative with dementia could also be helpful.

The analytical theme 'Counsellor’s emotional attitude' with its three descriptive themes 'Non‐judgmental/appreciative', 'To be familiar and trusted' and 'To be reliable' was not mentioned by any of the quantitative trials. A reason for the lack of these themes could be that the themes are newly derived from the thematic synthesis.

The analytical theme 'Barriers and facilitators for successful implementation of telephone counselling' was covered partly by the studies. The descriptive theme 'To know each other' was included in four trials (Chang 1999; Davis 2004; Finkel 2007; Wilz unpublished). These four studies had arranged face‐to‐face contacts at the beginning of the telephone counselling. It is unknown whether all these meetings were used to get to know each other and to learn more about the carers, their situation and family. The descriptive theme 'Continuation of counselling after institutionalisation' was mentioned in three studies (Chang 1999; Steffen 2000; Winter 2007). Four studies described the termination of telephone counselling at the time of the institutionalisation of the person with dementia (Davis 2004; Finkel 2007; Gant 2007; Glueckauf 2007); here the counsellors acted contrary to carers' needs. Only one study continued the telephone counselling after the person with dementia moved to a nursing home or other care facility (Tremont 2008). In one study, it was unclear whether the telephone counselling ended with the institutionalisation of the person with dementia (Wilz unpublished), one of the possible reasons being a change in the caregiving situation. The four descriptive themes 'Telephone as communication tool', 'Multilingualism', '24‐hour reachability of the counsellor' and 'Debriefing of counsellors' were not covered by any of the RCTs.

Second part of the integration of qualitative and quantitative data

Matrix 2 maps aspects of telephone counselling for carers of people with dementia covered by the included qualitative studies (see Table 4). The following high level observations can be made at a descriptive level: Matrix 2 shows that approximately a quarter of the possible telephone counselling interventions mentioned in the included quantitative trials were covered by the two qualitative studies (Chang 2004; Salfi 2005). One reason for this are the different proportions of quantitative and qualitative studies that were included. The nine quantitative trials illustrated a range of possible aspects within a telephone counselling, while the two qualitative studies were judged as 'thin descriptions' and were based on one or two different counselling agencies, so did not conduct a conscious contrasting in relation to the intervention. Another reason could be the reporting of the qualitative studies, as the telephone counselling was not described in detail in either study. Maybe some aspects of the counselling cannot not be depicted in Matrix 2.

Both the qualitative studies covered the following aspects of telephone counselling (Chang 2004; Salfi 2005): individual telephone counselling, managing behaviour problems, managing difficult caregiving challenges, finding and using local community resources and coping strategies.

Chang 2004 also covered the telephone counselling aspects of general problem solving skills and cognitive reframing.

Salfi 2005 examined telephone counselling with the additional aspects of social support network, emotional support and maybe information about dementia and talking about carers feelings. The last two aspects were not specified by the study investigators, but there were sufficient hints to permit this conclusion.

Neither of the qualitative studies mentioned the following aspects of telephone counselling in their publications: telephone support groups, using a workbook or a loose‐leaf notebook, individual telephone counselling or support group sessions via computer‐telephone integration system (CTIS), information about effects of caregiving, self care, relaxation training, breathing techniques, pleasant daily activities against emotional distress, promoting carer assertiveness, self‐instructional training, empathy, giving permission, normalising, validation or venting, anger‐management, biblio‐therapy, interpretation, role play, cognitive appraisal and reappraisal techniques, cognitive reconstructuring, behavioural management, behavioural activation, thought stopping, distraction techniques or coming to the end of the counselling intervention.

Discussion

Summary of main results

Evaluation of efficacy

The nine RCTs included in this review were sub‐divided into three groups according to how they performed telephone counselling.

Telephone counselling without any additional intervention

The results for the primary end point depressive symptoms were inconsistent. The pooled results of three studies showed a significant reduction of depressive symptoms for the intervention group (Finkel 2007; Tremont 2008; Winter 2007). Two further studies showed no differences between groups (Glueckauf 2007; Wilz unpublished). The results of the secondary outcomes were largely consistent. The pooled results of four studies of caregiver burden showed a non‐significant decrease in the intervention group (Davis 2004; Finkel 2007; Tremont 2008; Winter 2007). Glueckauf 2007 also reported a decrease of caregiver burden, but the results were not significant. Davis 2004 displayed a non‐significant reduction of caregivers' distress. Quality of life was assessed by the Tremont 2008 study, which showed a non‐statistically significant decrease. One study showed an increase of self‐efficacy (Glueckauf 2007), and another showed a decrease of self‐efficacy (Tremont 2008). However, in both studies there were no differences between groups. The pooled results of two studies showed a non‐statistically significant increase in received social support (Finkel 2007; Tremont 2008).

Telephone counselling with video sessions

In Chang 1999 depressive symptoms could be reduced in the intervention group, but there were no differences between groups. The same applied to the secondary outcomes: anxiety and satisfaction.

Telephone counselling with video sessions and workbook

Studies displayed consistent results for the primary and secondary end points. The primary end point of depressive symptoms showed a significant reduction in one study (Steffen 2000). While the secondary outcomes of caregiver burden and distress decreased in the intervention group, the differences to the control group were not significant (Gant 2007). Self‐efficacy increased in the intervention groups. However, the difference to the control group was not statistically significant (Gant 2007, Steffen 2000).

Evaluation of experience

The thematic synthesis included two studies that clarified carers' needs in terms of telephone counselling (Chang 2004; Salfi 2005). The review authors developed three analytical themes based on 16 descriptive themes and 119 codes. The three analytical themes concerned different areas of telephone counselling: barriers and facilitators for successful implementation of telephone counselling, counsellor's emotional attitude, and content of telephone counselling. The first analytical theme 'Barriers and facilitators for successful implementation of telephone counselling' gives conceptual and organisational hints for general configuration of telephone counselling and included six descriptive themes: telephone as a communication tool, to know each other, multilingualism, 24‐hour reachability of the counsellor, continuation of counselling after institutionalisation, debriefing (of counsellors) after counselling. The second analytical theme 'Counsellor's emotional attitude' describes the emotional attitude of the counsellor toward the carer. The counsellor cannot evoke the descriptive themes but can contribute to their gradual increase. The three descriptive themes are non‐judgmental/appreciative, to be familiar and trusted, and individual counselling. The third analytical theme 'Content of telephone counselling' included two subordinated themes. The first subordinated theme 'How is the counselling given?' concerned how telephone counselling can be conducted and included descriptive themes that ranged from providing factual information to mediation and also individual counselling. The second subordinated theme 'What is being discussed?' concerned what content the telephone counselling can have: problematic situations; conversational themes that help forget the daily routine; information and advice; the carer’s emotions, thoughts, and carer's physical condition; empowerment for autonomy. Three of the 16 descriptive themes arose as a result of the consolidation of the two studies.

Integration of qualitative and quantitative data

As the level of integration was 'low' due to the paucity and quality of qualitative evidence, it was not possible to develop a sophisticated interpretive synthesis. At a descriptive level, some intervention components and intervention characteristics (described above) appeared to match with carer and counsellor perceptions of what worked (or not). These findings, although limited by the quality of qualitative evidence and limited context, may provide the basis for further qualitative exploration to help explain why depressive symptoms were reduced by telephone counselling, although the effect was not consistent across interventions.

Overall completeness and applicability of evidence

The review included a small number of quantitative trials that were grouped according to differences in the way they conducted telephone counselling. Only two qualitative studies were included. Adverse effects or harm that might be caused by telephone counselling and endpoints on the level of people with dementia were hardly addressed in the included studies. The results of the meta‐analysis and the integration of qualitative and quantitative data indicated the need for further research in the field of telephone counselling for informal carers of people with dementia.

Evaluation of efficacy

Some studies provided friendly calls in their control group, and some studies treated the control group with usual care. However, usual care was not described in detail. Thus, the applicability of results is restricted.

Evaluation of experience

The applicability of findings to different types of telephone counselling interventions and settings is limited due to several issues. The findings were based on ‘thin data’ and specific interventions delivered in limited settings. One study focused on telephone counselling without additional interventions (Salfi 2005), and the other study focused on telephone counselling in combination with video sessions (Chang 2004). Applicability is likely to be limited to these types of telephone counselling interventions. Telephone counselling in combination with video sessions and workbooks was not covered by the qualitative studies. Therefore, Figure 1, which describes the analytical and descriptive themes of telephone counselling for carers of people with dementia, is a first basis for discussion and does not claim to be complete. The studies were conducted in Canada and the USA. Hence, the applicability of results is likely to be limited to similar contexts. Nonetheless, as new studies emerge, some common themes may become apparent across interventions, thereby offering validation of findings that are not possible with the currently available evidence. Given that none of the descriptive or analytical themes appear to be specific to caring for a person with dementia, and that telephone counselling is delivered to carers caring for a people with a variety of long‐term cognitive and behavioural conditions, it may also be desirable to explore whether there are common themes with other carer groups who receive telephone counselling (e.g. caring for people with stroke). Additional high quality qualitative studies and process evaluations that explore carer and counsellor experiences concerning the full range of available telephone counselling intervention types in‐depth (see Table 4) across a range of settings are needed.

Integration of qualitative and quantitative data

The ability to integrate the quantitative and qualitative evidence was limited by the paucity, spread and quality of the qualitative evidence when mapped against the greater range of RCTs evaluating different types of telephone counselling. The descriptive themes from the qualitative evidence were juxtaposed with the interventions in included RCTs.

When rich and varied contextual qualitative evidence is available, examining it can potentially help explain intervention success and failure. Furthermore, those interventions and intervention components associated with more positive carer or counsellor experience, or both, can be identified. It was acknowledged, however, that one study was a qualitative process evaluation of a similar intervention conducted by the same authors (Chang 1999), but linked with another RCT that was not included in the review since it could not be identified. It is therefore not surprising that the descriptive themes from Chang 2004 mapped onto components of similar RCT interventions (including Chang 1999). Whilst acknowledging these methodological limitations, the qualitative evidence nonetheless provided some useful insights about what it is like to receive and deliver counselling.

Quality of the evidence

Evaluation of efficacy

The methodological quality of the included quantitative studies was moderate for three reasons: the intervention could not be blinded, the outcomes were self‐reported carer outcomes and overall there was insufficient reporting of study details to allow 'Risk of bias' assessment. Overviews of the study qualities are displayed in Figure 3 and Figure 4.

Evaluation of experience

The methodological quality of the two qualitative studies was judged as moderate (see Appendix 23 and Appendix 24). There were greater concerns regarding the rigor and thin description of Chang 2004, although it was the larger study. The review authors’ concerns regarding the quality and coherence of the studies are reflected in the judgement of low confidence in all qualitative synthesis findings (see Table 8).

Potential biases in the review process

The lag time between search date and publication can be a potential bias in the review process. However, the authors plan to update the review soon.

Evaluation of efficacy

For one study, review authors did not know the end‐of‐study participant numbers for either the intervention group or the control group (Winter 2007), so for meta‐analysis the review authors allocated the drop‐outs to both groups according to the distribution of carers at baseline. The sensitivity analyses show that results of the meta‐analyses were not affected by different allocations of drop‐outs.

The interventions used in the studies that contributed data to the three outcomes assessed by meta‐analysis at different time points were all slightly different, as displayed in Table 6 and Figure 8 and described under 'Effects of interventions'.

Evaluation of experience

The thematic synthesis was conducted by one review author and checked by another review author. An independently conducted line‐by‐line coding by two review authors would have been better.

Chang 2004 was classified as a 'qualitative process evaluation' because it was conducted alongside a larger intervention trial and was not a stand‐alone qualitative study. It used the written notes of the nurses who conducted the interview: "The caregivers were asked whether they thought the telephone calls were helpful and if so, to elaborate on their answer" (Chang 2004). The study authors could not ask in‐depth questions. However, the study used qualitative interview methods and analysis methods.

The aim of the thematic synthesis is to display the range of different aspects of telephone counselling judged by the carers as being helpful. Chang 2004 tried to establish a ranking of themes according to the frequency with which they were mentioned. The current thematic synthesis does not pursue this goal and does not attempt to determine which analytical or descriptive themes are the most important.

Potential issues concerning rigour and the qualitative evidence synthesis were minimised by making transparent the limitations of the evidence when mapped against the included interventions.

Agreements and disagreements with other studies or reviews

The review authors are not aware of other systematic reviews that have investigated the efficacy of and experiences related to telephone counselling for carers of people with dementia. One publication that reviewed the literature on telephone support interventions for carers of people with dementia included six studies in which the focus was on reducing caregiver burden (Mason 2008). The Mason 2008 authors reported the studies in a narrative way and concluded that telephone support interventions are effective for reducing caregiver burden. However, the results of this Cochrane Review show that it is still unclear whether telephone counselling can reduce caregiver burden.

Authors' conclusions

Implications for practice.

The review found moderate quality evidence that telephone counselling without any additional intervention for carers of people with dementia can reduce depressive symptoms. However, the restrictions discussed show that these results should be interpreted with caution. It is still unclear whether telephone counselling, in combination with video or telephone counselling or in combination with video and workbook, is effective in reducing depressive symptoms and other secondary outcomes. Single studies reported uncertain effects. Existing or new telephone counselling providers can use the findings of the thematic synthesis as a provisional basis for establishing or improving their counselling.

Implications for research.

Further high quality studies are needed. Telephone counselling can be performed in different ways, as described in the review. It would be interesting to know which aspects of telephone counselling influence the efficacy. Outcomes on the level of people with dementia would be also interesting. Investigators of quantitative studies can use the findings from the thematic synthesis to guide further development in telephone counselling. It would be helpful if future randomised controlled trials included a high quality process evaluation to enable better understanding of carer and counsellor experiences and barriers and facilitators to optimal implementation. Additional high quality qualitative studies would improve knowledge about what works from the carer's and counsellor's perspectives. A synthesis of carer and counsellor experiences of similar telephone counselling interventions for similar groups of patients with long‐term conditions may provide a clearer understanding of common themes and issues in the absence of condition‐specific qualitative research.

Appendices

Appendix 1. Overview search results

Source	Date searched	Records retrieved with duplications
First searches (up to the 03.05.2011)
MEDLINE (Ovid SP)	03.05.2011	274
MEDLINE(R) In‐Process & Other Non‐Indexed Citations (Ovid SP)	03.05.2011	31
Cochrane Library (Wiley Online Library)	03.05.2011	211
Cinahl (EBSCOhost)	03.05.2011	395
Psyndex (EBSCOhost)	03.05.2011	101
PsycINFO (EBSCOhost)	03.05.2011	693
Web of Science (Thomson Reuters) ‐ Scince Citation Index Expanded (SCI‐EXPANDED) ‐ Social Science Citation Index (SSCI) ‐ Arts & Humanities Citation Index (A&HCI) ‐ Conference Proceedings Citation Index‐Science (CPCI‐S) ‐ Conference Proceedings Citation Index Social Science & Humanities (CPCI‐SSH)	04.05.2011	324
DIMDI databases: ‐ AZ72 GLOBAL HEALTH ‐ CC00 CCMED ‐ DAHTA DAHTA‐Datenbank ‐ HG05 Hogrefe‐Verlagsdatenbank und Volltexte ‐ HN69 HECLINET ‐ II78 ISTPB + ISTP/ISSHP ‐ INAHTA Health Technology Assessment ‐ KP05 Krause & Pachernegg Verlagsdatenbank ‐ KR03 Karger‐Verlagsdatenbank ‐ SM78 SOMED ‐ TV01 Thieme‐Verlagsdatenbank ‐ TVPP Thieme‐Verlagsdatenbank PrePrint	04.05.2011	182
ALOIS	14.05.2011	84
EMBASE (DIMDI)	04.05.2011	1591
Update Searches (2011‐25.02.2013)
MEDLINE (Ovid SP)	25.02.2013	40
MEDLINE(R) In‐Process & Other Non‐Indexed Citations (Ovid SP)	25.02.2013	13
Cochrane Library (Wiley Online Library)	25.02.2013	163
Cinahl (EBSCOhost)	25.02.2013	59
Psyndex (EBSCOhost)	25.02.2013	11
PsycINFO (EBSCOhost)	25.02.2013	191
Web of Science (Thomson Reuters)	25.02.2013	97
DIMDI databases: ‐ AZ72 GLOBAL HEALTH ‐ CC00 CCMed ‐ DAHTA DAHTA‐Datenbank ‐ II01 II78 ISTPB + ISTP/ISSHP ‐ INAHTA Health Technology Assessment	25.02.2013	30
ALOIS	24.02.2013	46
EMBASE (DIMDI)	25.02.2013	212
Springer database ( up to 25.02.2013); http://link.springer.com	25.02.2013	552
Science Direct (up to 25.02.2013); http://www.sciencedirect.com	25.02.2013	622
Searching other resources
ClinicalTrials.gov (up to 05.03.2013)	05.03.2013	37
Controlled Trials (up to 05.03.2013)	05.03.2013	67
WHO register (up to 05.03.2013)	05.03.2013	798
Forward Citation Search in Web of Science for the included studies	21.03.2013	165
Forward Citation Search in Google Scholar for the included studies	18.03.2013	315
PubMed Related articles service for the included studies	27.03.2013	1982
Reference Checking of included studies		37
Conference abstracts		27
TOTAL including duplicates		9350
TOTAL after rejecting duplicates and title and if necessary abstract screening		110

Appendix 2. MEDLINE search strategy (first search 2011)

No	Searches	Results
1	exp Dementia/	97,786
2	Delirium/	4435
3	Wernicke Encephalopathy/	1217
4	Delirium, Dementia, Amnestic, Cognitive Disorders/	8433
5	dement*.mp.	74,500
6	alzheimer*.mp.	78,016
7	(lewy* adj2 bod*).mp.	4979
8	deliri*.mp.	17,185
9	(chronic adj2 (cerebrovascular or cerebro‐vascular)).mp.	385
10	("organic brain disease" or "organic brain syndrome").mp.	716
11	("normal pressure hydrocephalus" and "shunt*").mp.	801
12	"benign senescent forgetfulness".mp.	16
13	(cerebr* adj2 deteriorat*).mp.	155
14	((cerebral* or cerebrovascular or cerebro‐vascular) adj2 insufficien*).mp.	1424
15	(pick* adj2 disease).mp.	2394
16	(creutzfeldt or jcd or cjd).mp.	6571
17	huntington*.mp.	11,059
18	binswanger*.mp.	521
19	korsako*.mp.	1222
20	or/1‐19	156,454
21	(caregiver* or care‐giver*).ti,ab.	23,778
22	carer*.ti,ab.	5826
23	relatives.ti,ab.	33,622
24	family*.ti,ab.	448,635
25	families*.ti,ab.	135,313
26	dependents.ti,ab.	1102
27	kin*.ti,ab.	714,076
28	spouse*.ti,ab.	10,423
29	parent*.ti,ab.	237,909
30	folk*.ti,ab.	5140
31	Caregivers/	15,993
32	or/21‐31	1,431,168
33	20 and 32	20,981
34	telephone*.ti,ab.	32,580
35	phone*.ti,ab.	13,824
36	helpline*.ti,ab.	293
37	hotline*.ti,ab.	708
38	telecar*.ti,ab.	427
39	tele‐car*.ti,ab.	21
40	telemedicine*.ti,ab.	4798
41	tele‐medicine*.ti,ab.	38
42	telehome*.ti,ab.	69
43	tele‐home*.ti,ab.	25
44	*counseling/ or directive counseling/	10,916
45	Telephone/	7911
46	or/34‐45	63,104
47	33 and 46	274
48	remove duplicates from 47	270

Appendix 3. MEDLINE search strategy (update search 2013)

No	Searches	Results
1	exp Dementia/	108,673
2	Delirium/	5106
3	Wernicke Encephalopathy/	1312
4	Delirium, Dementia, Amnestic, Cognitive Disorders/	8472
5	dement*.mp.	79,220
6	alzheimer*.mp.	84,698
7	(lewy* adj2 bod*).mp.	5486
8	deliri*.mp.	17,913
9	(chronic adj2 (cerebrovascular or cerebro‐vascular)).mp.	400
10	("organic brain disease" or "organic brain syndrome").mp.	722
11	("normal pressure hydrocephalus" and "shunt*").mp.	865
12	"benign senescent forgetfulness".mp.	17
13	(cerebr* adj2 deteriorat*).mp.	167
14	((cerebral* or cerebrovascular or cerebro‐vascular) adj2 insufficien*).mp.	1448
15	(pick* adj2 disease).mp.	2559
16	(creutzfeldt or jcd or cjd).mp.	6785
17	huntington*.mp.	11,914
18	binswanger*.mp.	524
19	korsako*.mp.	1323
20	or/1‐19	168,327
21	(caregiver* or care‐giver*).ti,ab.	26,731
22	carer*.ti,ab.	6469
23	relatives.ti,ab.	35,620
24	family*.ti,ab.	474,510
25	families*.ti,ab.	143,459
26	dependents.ti,ab.	1141
27	kin*.ti,ab.	738,735
28	spouse*.ti,ab.	10,956
29	parent*.ti,ab.	249,310
30	folk*.ti,ab.	5609
31	Caregivers/	18,894
32	or/21‐31	1,495,544
33	20 and 32	22,827
34	telephone*.ti,ab.	34,982
35	phone*.ti,ab.	15,409
36	helpline*.ti,ab.	324
37	hotline*.ti,ab.	753
38	telecar*.ti,ab.	470
39	tele‐car*.ti,ab.	26
40	telemedicine*.ti,ab.	5089
41	tele‐medicine*.ti,ab.	44
42	telehome*.ti,ab.	67
43	tele‐home*.ti,ab.	24
44	*counseling/ or directive counseling/	12,191
45	Telephone/	8546
46	or/34‐45	68,582
47	33 and 46	297
48	remove duplicates from 47	296
49	limit 48 to yr="2011 ‐Current"	40

Appendix 4. MEDLINE in Process & other non‐indexed citations search strategy (first search 2011)

No	Searches	Results
1	exp Dementia/	0
2	Delirium.mp.	300
3	Wernicke.mp.	29
4	Delirium, Dementia, Amnestic, Cognitive Disorders/	0
5	dement*.mp.	2439
6	alzheimer*.mp.	3267
7	(lewy* adj2 bod*).mp.	176
8	deliri*.mp.	309
9	(chronic adj2 (cerebrovascular or cerebro‐vascular)).mp.	15
10	("organic brain disease" or "organic brain syndrome").mp.	11
11	("normal pressure hydrocephalus" and "shunt*").mp.	22
12	"benign senescent forgetfulness".mp.	0
13	(cerebr* adj2 deteriorat*).mp.	0
14	((cerebral* or cerebrovascular or cerebro‐vascular) adj2 insufficien*).mp.	15
15	(pick* adj2 disease).mp.	44
16	(creutzfeldt or jcd or cjd).mp.	114
17	huntington*.mp.	353
18	binswanger*.mp.	7
19	korsako*.mp.	28
20	or/1‐19	5612
21	(caregiver* or care‐giver*).ti,ab.	1281
22	carer*.ti,ab.	294
23	relatives.ti,ab.	1179
24	family*.ti,ab.	18,654
25	families*.ti,ab.	5281
26	dependents.ti,ab.	48
27	kin*.ti,ab.	41,171
28	spouse*.ti,ab.	313
29	parent*.ti,ab.	10,374
30	folk*.ti,ab.	339
31	Caregivers/	0
32	or/21‐31	72,074
33	20 and 32	806
34	telephone*.ti,ab.	1441
35	phone*.ti,ab.	811
36	helpline*.ti,ab.	11
37	hotline*.ti,ab.	20
38	telecar*.ti,ab.	36
39	tele‐car*.ti,ab.	3
40	telemedicine*.ti,ab.	261
41	tele‐medicine*.ti,ab.	0
42	telehome*.ti,ab.	8
43	tele‐home*.ti,ab.	3
44	counsel*.mp.	2187
45	Telephone/	0
46	or/34‐45	4544
47	33 and 46	31

Appendix 5. MEDLINE in Process & other non‐indexed citations search strategy (update search 2013)

No	Searches	Results
1	exp Dementia/	1
2	Delirium/	0
3	Wernicke Encephalopathy/	0
4	Delirium, Dementia, Amnestic, Cognitive Disorders/	0
5	dement*.mp.	3734
6	alzheimer*.mp.	5359
7	(lewy* adj2 bod*).mp.	301
8	deliri*.mp.	508
9	(chronic adj2 (cerebrovascular or cerebro‐vascular)).mp.	19
10	("organic brain disease" or "organic brain syndrome").mp.	16
11	("normal pressure hydrocephalus" and "shunt*").mp.	38
12	"benign senescent forgetfulness".mp.	0
13	(cerebr* adj2 deteriorat*).mp.	2
14	((cerebral* or cerebrovascular or cerebro‐vascular) adj2 insufficien*).mp.	14
15	(pick* adj2 disease).mp.	75
16	(creutzfeldt or jcd or cjd).mp.	167
17	huntington*.mp.	576
18	binswanger*.mp.	9
19	korsako*.mp.	30
20	or/1‐19	8827
21	(caregiver* or care‐giver*).ti,ab.	2055
22	carer*.ti,ab.	483
23	relatives.ti,ab.	1923
24	family*.ti,ab.	27,176
25	families*.ti,ab.	7843
26	dependents.ti,ab.	65
27	kin*.ti,ab.	54,922
28	spouse*.ti,ab.	487
29	parent*.ti,ab.	14,814
30	folk*.ti,ab.	660
31	Caregivers/	0
32	or/21‐31	100,243
33	20 and 32	1304
34	telephone*.ti,ab.	1817
35	phone*.ti,ab.	1189
36	helpline*.ti,ab.	22
37	hotline*.ti,ab.	28
38	telecar*.ti,ab.	41
39	tele‐car*.ti,ab.	3
40	telemedicine*.ti,ab.	335
41	tele‐medicine*.ti,ab.	2
42	telehome*.ti,ab.	11
43	tele‐home*.ti,ab.	4
44	*counseling/ or directive counseling/	0
45	Telephone/	0
46	or/34‐45	3292
47	33 and 46	19
48	remove duplicates from 47	19
49	limit 48 to yr="2011 ‐Current"	13

Appendix 6. Cochrane Library search strategy (first search 2011)

ID	Search	Hits
#1	MeSH descriptor Dementia explode all trees	3240
#2	MeSH descriptor Delirium, this term only	151
#3	MeSH descriptor Wernicke Encephalopathy, this term only	3
#4	MeSH descriptor Delirium, Dementia, Amnestic, Cognitive Disorders, this term only	114
#5	dement* OR alzheimer* OR (lewy* NEAR/3 bod*) OR deliri* OR (chronic NEAR/3 cerebrovascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* NEAR/3 det	10,994
#6	(#1 OR #2 OR #3 OR #4 OR #5)	11,000
#7	caregiver* or (care NEXT giver*) OR carer* OR relatives OR family* OR families* OR dependents OR kin* OR spouse* OR parent* OR folk*	103,965
#8	MeSH descriptor Caregivers, this term only	1011
#9	(#7 OR #8)	103,965
#10	(#6 AND #9)	2449
#11	telephone* OR phone* OR helpline* OR hotline* OR telecar* OR (tele NEXT car*) OR telemedicine* OR (tele NEXT medicine*) OR telehome* OR (tele NEXT home*)	8393
#12	MeSH descriptor Counseling, this term only	2237
#13	MeSH descriptor Directive Counseling, this term only	149
#14	MeSH descriptor Telephone, this term only	1067
#15	(#11 OR #12 OR #13 OR #14)	10,279
#16	(#10 AND #15)	216

Appendix 7. Cochrane Library search strategy (update search 2013)

ID	Search	Hits
#1	MeSH descriptor: [Dementia] explode all trees	3282
#2	MeSH descriptor: [Delirium] this term only	175
#3	MeSH descriptor: [Wernicke Encephalopathy] this term only	3
#4	MeSH descriptor: [Delirium, Dementia, Amnestic, Cognitive Disorders] this term only	112
#5	dement* or alzheimer* or (lewy* near/3 bod*) or deliri* or (chronic near/3 cerebrovascular) or ("organic brain disease" or "organic brain syndrome") or ("normal pressure hydrocephalus" and shunt*) or "benign senescent forgetfulness" or (cerebr* near/3 deteriorat*) or (cerebral* near/3 insufficien*) or (pick* near/3 disease) or creutzfeldt or jcd or cjd or huntington* or binswanger* or korsako* or (chronic near/3 (cerebro next vascular)) or (cerebrovascular near/3 insufficien*) or ((cerebro next vascular) near/3 insufficien*)	11,595
#6	#1 or #2 or #3 or #4 or #5	11,603
#7	caregiver* or (care next giver*) or carer* or relatives or family* or families* or dependents or kin* or spouse* or parent* or folk*	110,723
#8	MeSH descriptor: [Caregivers] this term only	943
#9	#7 or #8	110,723
#10	#6 and #9	2842
#11	telephone* or phone* or helpline* or hotline* or telecar* or (tele next car*) or telemedicine* or (tele next medicine*) or telehome* or (tele next home*)	10,303
#12	MeSH descriptor: [Counseling] this term only	2454
#13	MeSH descriptor: [Directive Counseling] this term only	193
#14	MeSH descriptor: [Telephone] this term only	1135
#15	#11 or #12 or #13 or #14	12,360
#16	#10 and #15 (from 2011 to 2013)	165

Appendix 8. CINAHL search strategy (first search 2011)

Search ID#	Search Terms	Actions
S19	S12 and S18	395
S18	S13 or S14 or S15 or S16 or S17	40,723
S17	TX directive counsel*	64
S16	(MH "Counseling")	10,210
S15	(MH "Telephone")	9241
S14	(MH "Telehealth+")	5611
S13	TX telephone* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedicine* OR tele medicine* OR telehome* OR tele home*	28,936
S12	S6 and S11	9666
S11	S7 or S8 or S9 or S10	392,845
S10	(MH "Caregiver Support")	2216
S9	(MH "Caregiver Burden")	4465
S8	(MH "Caregivers")	12,204
S7	TX caregiver* or care giver* OR carer* OR relatives OR family* OR families* OR dependents OR kin* OR spouse* OR parent* OR folk*	392,845
S6	S1 or S2 or S3 or S4 or S5	32470
S5	TX dement* OR alzheimer* OR (lewy* N3 bod*) OR deliri* OR (chroni* N3 cerebrovascular) OR (chroni* N3 cerebro‐vascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* N3 deteriorat*) OR (cerebral* N3 insufficien*) OR (cerebrovascular N3 insufficien*) OR (cerebro‐vascular N3 insufficien*) OR (pick* N3 disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*	32,385
S4	(MH "Delirium, Dementia, Amnestic, Cognitive Disorders")	47
S3	(MH "Wernicke's Encephalopathy")	123
S2	(MH "Delirium")	1790
S1	(MH "Dementia+")	25,223

Appendix 9. CINAHL search results (update search 2013)

Search ID#	Search terms	Search options	Actions
S20	S12 and S18	Limiters ‐ Published Date from: 20110101‐20131231 Search modes ‐ Boolean/Phrase	59
S19	S12 and S18	Search modes ‐ Boolean/Phrase	458
S18	S13 or S14 or S15 or S16 or S17	Search modes ‐ Boolean/Phrase	47,024
S17	TX directive counsel*	Search modes ‐ Boolean/Phrase	75
S16	(MH "Counseling")	Search modes ‐ Boolean/Phrase	12,119
S15	(MH "Telephone")	Search modes ‐ Boolean/Phrase	10,331
S14	(MH "Telehealth+")	Search modes ‐ Boolean/Phrase	6817
S13	TX telephone* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedicine* OR tele medicine* OR telehome* OR tele home*	Search modes ‐ Boolean/Phrase	32,936
S12	S6 and S11	Search modes ‐ Boolean/Phrase	11,519
S11	S7 or S8 or S9 or S10	Search modes ‐ Boolean/Phrase	451,109
S10	(MH "Caregiver Support")	Search modes ‐ Boolean/Phrase	2633
S9	(MH "Caregiver Burden")	Search modes ‐ Boolean/Phrase	5152
S8	(MH "Caregivers")	Search modes ‐ Boolean/Phrase	14,587
S7	TX caregiver* or care giver* OR carer* OR relatives OR family* OR families* OR dependents OR kin* OR spouse* OR parent* OR folk*	Search modes ‐ Boolean/Phrase	451,109
S6	S1 or S2 or S3 or S4 or S5	Search modes ‐ Boolean/Phrase	39,739
S5	TX dement* OR alzheimer* OR (lewy* N3 bod*) OR deliri* OR (chroni* N3 cerebrovascular) OR (chroni* N3 cerebro‐vascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* N3 deteriorat*) OR (cerebral* N3 insufficien*) OR (cerebrovascular N3 insufficien*) OR (cerebro‐vascular N3 insufficien*) OR (pick* N3 disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*	Search modes ‐ Boolean/Phrase	39,637
S4	(MH "Delirium, Dementia, Amnestic, Cognitive Disorders")	Search modes ‐ Boolean/Phrase	85
S3	(MH "Wernicke's Encephalopathy")	Search modes ‐ Boolean/Phrase	148
S2	(MH "Delirium")	Search modes ‐ Boolean/Phrase	2296
S1	(MH "Dementia+")	Search modes ‐ Boolean/Phrase	30,909

Appendix 10. PSYNDEX search strategy (first search 2011)

Search ID#	Search Terms	Actions
S9	S5 and S8	101
S8	S6 or S7	3658
S7	TX dement* or demenz* OR alzheimer* OR (lewy* N3 bod*) OR deliri* OR (chroni* N3 cerebrovascular) OR (chroni* N3 cerebro‐vascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* N3 deteriorat*) OR (cerebral* N3 insufficien*) OR (cerebrovascular N3 insufficien*) OR (cerebro‐vascular N3 insufficien*) OR (pick* N3 disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*	3658
S6	((DE "Dementia" OR DE "AIDS Dementia Complex" OR DE "Dementia with Lewy Bodies" OR DE "Presenile Dementia" OR DE "Semantic Dementia" OR DE "Senile Dementia" OR DE "Vascular Dementia" OR DE "AIDS‐Demenz" OR DE "Lewy‐Körper‐Demenz" OR DE "Präsenile Demenz" OR DE "Semantische Demenz" OR DE "Senile Demenz" OR DE "Vaskuläre Demenz") OR (DE "Alzheimer's Disease" OR DE "Creutzfeldt Jakob Syndrome" OR DE "Picks Disease")) OR (DE "Delirium")	1415
S5	S1 or S2 or S3 or S4	6535
S4	DE "Counseling"	3969
S3	DE "Telemedicine"	37
S2	DE "Hot Line Services"	146
S1	TX telephon* OR telefon* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedici* OR tele medi* OR telehome* OR tele home* OR anruf*	2647

Appendix 11. PSYNDEX search strategy (update search 2013)

Search ID#	Search terms	Search options	Actions
S10	S5 and S8	Limiters ‐ Publication Year: 2011‐2013 Search modes ‐ Boolean/Phrase	11
S9	S5 and S8	Search modes ‐ Boolean/Phrase	121
S8	S6 or S7	Search modes ‐ Boolean/Phrase	4131
S7	TX dement* or demenz* OR alzheimer* OR (lewy* N3 bod*) OR deliri* OR (chroni* N3 cerebrovascular) OR (chroni* N3 cerebro‐vascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* N3 deteriorat*) OR (cerebral* N3 insufficien*) OR (cerebrovascular N3 insufficien*) OR (cerebro‐vascular N3 insufficien*) OR (pick* N3 disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*	Search modes ‐ Boolean/Phrase	4131
S6	((DE "Dementia" OR DE "AIDS Dementia Complex" OR DE "Dementia with Lewy Bodies" OR DE "Presenile Dementia" OR DE "Semantic Dementia" OR DE "Senile Dementia" OR DE "Vascular Dementia" OR DE "AIDS‐Demenz" OR DE "Lewy‐Körper‐Demenz" OR DE "Präsenile Demenz" OR DE "Semantische Demenz" OR DE "Senile Demenz" OR DE "Vaskuläre Demenz") OR (DE "Alzheimer's Disease" OR DE "Creutzfeldt Jakob Syndrome" OR DE "Picks Disease")) OR (DE "Delirium")	Search modes ‐ Boolean/Phrase	2059
S5	S1 or S2 or S3 or S4	Search modes ‐ Boolean/Phrase	7274
S4	DE "Counseling"	Search modes ‐ Boolean/Phrase	4403
S3	DE "Telemedicine"	Search modes ‐ Boolean/Phrase	49
S2	DE "Hot Line Services"	Search modes ‐ Boolean/Phrase	164
S1	TX telephon* OR telefon* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedici* OR tele medi* OR telehome* OR tele home* OR anruf*	Search modes ‐ Boolean/Phrase	2967

Appendix 12. PsycINFO search strategy (first search 2011)

Search ID#	Search terms	Actions
S7	S3 and S6	693
S6	S4 or S5	65,142
S5	((DE "Dementia" OR DE "AIDS Dementia Complex" OR DE "Dementia with Lewy Bodies" OR DE "Presenile Dementia" OR DE "Semantic Dementia" OR DE "Senile Dementia" OR DE "Vascular Dementia") OR (DE "Alzheimer's Disease" OR DE "Creutzfeldt Jakob Syndrome" OR DE "Picks Disease")) OR (DE "Delirium")	41,855
S4	TX dement* OR alzheimer* OR (lewy* N3 bod*) OR deliri* OR (chroni* N3 cerebrovascular) OR (chroni* N3 cerebro‐vascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* N3 deteriorat*) OR (cerebral* N3 insufficien*) OR (cerebrovascular N3 insufficien*) OR (cerebro‐vascular N3 insufficien*) OR (pick* N3 disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*	65,142
S3	S1 or S2	59,943
S2	((DE "Hot Line Services") OR (DE "Telemedicine")) OR (DE "Counseling")	18,448
S1	TX telephon* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedici* OR tele medi* OR telehome* OR tele home*	43,045

Appendix 13. PsycINFO search strategy (update search 2013)

Search ID#	Search terms	Search options	Actions
S8	S6 and S3	Limiters ‐ Publication Year from: 2011‐2013 Search modes ‐ Boolean/Phrase	191
S7	S3 and S6	Search modes ‐ Boolean/Phrase	1140
S6	S4 or S5	Search modes ‐ Boolean/Phrase	76,099
S5	((DE "Dementia" OR DE "AIDS Dementia Complex" OR DE "Dementia with Lewy Bodies" OR DE "Presenile Dementia" OR DE "Semantic Dementia" OR DE "Senile Dementia" OR DE "Vascular Dementia") OR (DE "Alzheimer's Disease" OR DE "Creutzfeldt Jakob Syndrome" OR DE "Picks Disease")) OR (DE "Delirium")	Search modes ‐ Boolean/Phrase	48,822
S4	TX dement* OR alzheimer* OR (lewy* N3 bod*) OR deliri* OR (chroni* N3 cerebrovascular) OR (chroni* N3 cerebro‐vascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* N3 deteriorat*) OR (cerebral* N3 insufficien*) OR (cerebrovascular N3 insufficien*) OR (cerebro‐vascular N3 insufficien*) OR (pick* N3 disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*	Search modes ‐ Boolean/Phrase	76,099
S3	S1 or S2	Search modes ‐ Boolean/Phrase	67,988
S2	((DE "Hot Line Services") OR (DE "Telemedicine")) OR (DE "Counseling")	Search modes ‐ Boolean/Phrase	20,561
S1	TX telephon* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedici* OR tele medi* OR telehome* OR tele home*	Search modes ‐ Boolean/Phrase	50,496

Appendix 14. EMBASE search strategy (first search 2011)

No	Hits	Search expression
1	20,218,975	EM74; EA08
2	158,855	CT D DEMENTIA
3	10,162	CT=DELIRIUM
4	1526	CT=WERNICKE ENCEPHALOPATHY
5	60,776	CT=COGNITIVE DEFECT
6	185,819	DEMENT* OR ALZHEIMER* OR DELIRI* OR ORGANIC BRAIN DISEASE OR ORGANIC BRAIN SYNDROME OR (NORMAL PRESSURE HYDROCEPHALUS AND SHUNT*) OR BENIGN SENESCENT FORGETFULNESS OR CREUTZFELDT OR JCD OR CJD OR HUNTINGTON* OR BINSWANGER* OR KORSAKO*
7	6706	(LEWY* AND BOD*)/SAME SENT
8	1449	(CHRONI* AND CEREBROVASCULAR)/SAME SENT OR (CHRONI* AND CEREBRO‐VASCULAR)/SAME SENT
9	1245	(CEREBR* AND DETERIORAT*)/SAME SENT
10	2717	(CEREBRAL* AND INSUFFICIEN*)/SAME SENT OR (CEREBROVASCULAR AND INSUFFICIEN*)/SAME SENT OR (CEREBRO‐VASCULAR AND INSUFFICIEN*)/SAME SENT
11	3702	(PICK* AND DISEASE)/SAME SENT
12	248,405	2 TO 11
13	2,269,993	CAREGIVER* OR CARE GIVER* OR CARER* OR RELATIVES OR FAMILY* OR FAMILIES* OR DEPENDENTS OR KIN* OR SPOUSE* OR PARENT* OR FOLK*
14	24,135	CT=CAREGIVER
15	2,269,993	13 OR 14
16	40,135	12 AND 15
17	72,036	TELEPHON* OR PHONE* OR HELPLINE* OR HOTLINE* OR TELECAR* OR TELE CAR* OR TELEMEDICI* OR TELE MEDI* OR TELEHOME* OR TELE HOME*
18	77,648	CT D COUNSELING
19	16,219	CT=TELEPHONE
20	147,306	17 TO 19
21	1591	16 AND 20
22	1351	21 NOT SU=MEDLINE

Appendix 15. EMBASE search strategy (update search 2013)

No	Hits	Search expression
1	21,758,623	EM74; EA08
2	178,897	CT D DEMENTIA
3	11,924	CT=DELIRIUM
4	1700	CT=WERNICKE ENCEPHALOPATHY
5	70,972	CT=COGNITIVE DEFECT
6	210,997	DEMENT* OR ALZHEIMER* OR DELIRI* OR ORGANIC BRAIN DISEASE OR ORGANIC BRAIN SYNDROME OR (NORMAL PRESSURE HYDROCEPHALUS AND SHUNT*) OR BENIGN SENESCENT FORGETFULNESS OR CREUTZFELDT OR JCD OR CJD OR HUNTINGTON* OR BINSWANGER* OR KORSAKO*
7	7869	(LEWY* AND BOD*)/SAME SENT
8	1618	(CHRONI* AND CEREBROVASCULAR)/SAME SENT OR (CHRONI* AND CEREBRO‐VASCULAR)/SAME SENT
9	1357	(CEREBR* AND DETERIORAT*)/SAME SENT
10	2828	(CEREBRAL* AND INSUFFICIEN*)/SAME SENT OR (CEREBROVASCULAR AND INSUFFICIEN*)/SAME SENT OR (CEREBRO‐VASCULAR AND INSUFFICIEN*)/SAME SENT
11	4049	(PICK* AND DISEASE)/SAME SENT
12	282,802	2 TO 11
13	2,494,243	CAREGIVER* OR CARE GIVER* OR CARER* OR RELATIVES OR FAMILY* OR FAMILIES* OR DEPENDENTS OR KIN* OR SPOUSE* OR PARENT* OR FOLK*
14	27,400	CT=CAREGIVER
15	2,494,243	13 OR 14
16	46,539	12 AND 15
17	81,047	TELEPHON* OR PHONE* OR HELPLINE* OR HOTLINE* OR TELECAR* OR TELE CAR* OR TELEMEDICI* OR TELE MEDI* OR TELEHOME* OR TELE HOME*
18	85,545	CT D COUNSELING
19	17,255	CT=TELEPHONE
20	163,890	17 TO 19
21	1779	16 AND 20
22	1533	21 NOT SU=MEDLINE
23	212	22 AND PY>=2011

Appendix 16. Web of Science search strategy (first search 2011)

Set	Results	Search
# 5	324	#4 AND #3 Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=All Years
# 4	> 100,000	ts=(caregiver* or care giver* OR carer* OR relatives OR family* OR families* OR dependents OR kin* OR spouse* OR parent* OR folk*) Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=All Years
# 3	949	#2 AND #1 Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=All Years
# 2	84,987	ts=(telephone* OR phone* OR helpline* OR hotline* OR telecar* OR "tele car*" OR telemedicine* OR "tele medicine*" OR telehome* OR "tele home*" ) Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=All Years
# 1	> 100,000	ts=(dement* OR alzheimer* OR (lewy* SAME bod*) OR deliri* OR (chronic SAME cerebrovascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* SAME deteriorat*) OR (cerebral* SAME insufficient*) OR (pick* SAME disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*) Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=All Years

Appendix 17. Web of Science search strategy (update search 2013)

Set	Results	Search
# 5	97	#4 AND #3 Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=2011‐01‐01 ‐ 2013‐02‐25
# 4	598.354	ts=(caregiver* or care giver* OR carer* OR relatives OR family* OR families* OR dependents OR kin* OR spouse* OR parent* OR folk*) Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=2011‐01‐01 ‐ 2013‐02‐25
# 3	246	#2 AND #1 Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=2011‐01‐01 ‐ 2013‐02‐25
# 2	14.309	ts=(telephone* OR phone* OR helpline* OR hotline* OR telecar* OR "tele car*" OR telemedicine* OR "tele medicine*" OR telehome* OR "tele home*" ) Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=2011‐01‐01 ‐ 2013‐02‐25
# 1	34.626	ts=(dement* OR alzheimer* OR (lewy* SAME bod*) OR deliri* OR (chronic SAME cerebrovascular) OR ("organic brain disease" or "organic brain syndrome") OR ("normal pressure hydrocephalus" and shunt*) OR "benign senescent forgetfulness" OR (cerebr* SAME deteriorat*) OR (cerebral* SAME insufficient*) OR (pick* SAME disease) OR creutzfeldt or jcd or cjd OR huntington* OR binswanger* OR korsako*) Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH Timespan=2011‐01‐01 ‐ 2013‐02‐25

Appendix 18. DIMDI search strategy (first search 2011)

No	Search terms
27	25 AND 26	182
26	telephon* OR telefon* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedi* OR tele medi* OR telehome* OR tele home* OR anruf*	36,953
25	15 OR 16 OR 17 OR 18 OR 19 OR 20 OR 21 OR 22 OR 23 OR 24	47,848
24	zerebralinsuffi*	1
23	(zerebr* AND insuffi*)/same sent	63
22	(zerebr* AND deteriorat*)/same sent	0
21	(chroni* AND zerebro*)/same sent	19
20	(pick* AND disease)/same sent	445
19	(CEREBRAL* AND INSUFFICIEN*)/SAME SENT OR (CEREBROVASCULAR AND INSUFFICIEN*)/SAME SENT OR (CEREBRO‐VASCULAR AND INSUFFICIEN*)/SAME SENT	302
18	(cerebr* AND deteriorat*)/same sent	153
17	(CHRONI* AND CEREBROVASCULAR)/SAME SENT OR (CHRONI* AND CEREBRO‐VASCULAR)/SAME SENT	280
16	(lewy* AND bod*)/same sent	1162
15	dement* or demenz* OR alzheimer* OR deliri* OR organic brain disease OR organic brain syndrome OR (normal pressure hydrocephalus AND shunt*) OR benign senescent forgetfulness OR creutzfeldt OR jcd OR cjd OR huntington* OR binswanger* OR korsako*	46,600
1	AZ72 CC00 DAHTA HG05 HN69 II78 INAHTA KP05 KR03 SM78 TV01 TVPP	11,755,636

Appendix 19. DIMDI search strategy (update search 2013)

No	Search terms
15	14 AND PY=2011 to 2013	30
14	12 AND 13	173
13	telephon* OR telefon* OR phone* OR helpline* OR hotline* OR telecar* OR tele car* OR telemedi* OR tele medi* OR telehome* OR tele home* OR anruf*	39,271
12	2 OR 3 OR 4 OR 5 OR 6 OR 7 OR 8 OR 9 OR 10 OR 11	48,123
11	zerebralinsuffi*	0
10	(zerebr* AND insuffi*)/same sent	3
9	(zerebr* AND deteriorat*)/same sent	0
8	(chroni* AND zerebro*)/same sent	4
7	(pick* AND disease)/same sent	450
6	(CEREBRAL* AND INSUFFICIEN*)/SAME SENT OR (CEREBROVASCULAR AND INSUFFICIEN*)/SAME SENT OR (CEREBRO‐VASCULAR AND INSUFFICIEN*)/SAME SENT	275
5	(cerebr* AND deteriorat*)/same sent	140
4	(CHRONI* AND CEREBROVASCULAR)/SAME SENT OR (CHRONI* AND CEREBRO‐VASCULAR)/SAME SENT	288
3	(lewy* AND bod*)/same sent	1128
2	dement* or demenz* OR alzheimer* OR deliri* OR organic brain disease OR organic brain syndrome OR (normal pressure hydrocephalus AND shunt*) OR benign senescent forgetfulness OR creutzfeldt OR jcd OR cjd OR huntington* OR binswanger* OR korsako*	46,890
1	AZ72 CC00 DAHTA II01 II78 INAHTA	12,292,246

Appendix 20. Springer search strategy (update search 2013)

Hits	Search
552	(dement OR alzheimer) AND (Caregiver OR carer) AND (tele* OR helpline OR hotline)

Appendix 21. Science Direct search strategy (update search 2013)

Hits

Search

622

TITLE‐ABSTR‐KEY((dement* OR alzheimer* OR deliri* OR cerebr* OR pick* OR creutzfeldt or huntington* OR binswanger* OR korsako*) AND (Caregiver* OR care‐giver* OR carer* OR relative* OR family* OR dependents OR kin* OR spouse* OR parent* OR folk)) and TITLE‐ABSTR‐KEY((telephone* OR phone* OR helpline* OR hotline* OR telecar* OR tele‐car* OR telemedicine* OR tele‐medicine* OR telehome* OR tele‐home*))

Appendix 22. ALOIS search results (update search 2013)

Source	Search strategy	Hits retrieved
1. ALOIS (www.medicine.ox.ac.uk/alois)	Keyword search: telephone OR telemedicine OR phone OR telecare OR helpline OR hotline OR telehome	95 (all dates)
2. MEDLINE In‐process and other non‐indexed citations and MEDLINE 1950‐present (Ovid SP)	1. exp Dementia/ 2. Delirium/ 3. Wernicke Encephalopathy/ 4. Delirium, Dementia, Amnestic, Cognitive Disorders/ 5. dement*.mp. 6. alzheimer*.mp. 7. (lewy* adj2 bod*).mp. 8. deliri*.mp. 9. (chronic adj2 cerebrovascular).mp. 10. ("organic brain disease" or "organic brain syndrome").mp. 11. ("normal pressure hydrocephalus" and "shunt*").mp. 12. "benign senescent forgetfulness".mp. 13. (cerebr* adj2 deteriorat*).mp. 14. (cerebral* adj2 insufficient*).mp. 15. (pick* adj2 disease).mp. 16. (creutzfeldt or jcd or cjd).mp. 17. huntington*.mp. 18. binswanger*.mp. 19. korsako*.mp. 20. or/1‐19 21. caregiver*.ti,ab. 22. carer*.ti,ab. 23. relatives.ti,ab. 24. family*.ti,ab. 25. families*.ti,ab. 26. dependents.ti,ab. 27. kin*.ti,ab. 28. spouse*.ti,ab. 29. parent*.ti,ab. 30. folk*.ti,ab. 31. Caregivers/ 32. or/21‐31 33. 20 and 32 34. telephone*.ti,ab. 35. phone*.ti,ab. 36. helpline*.ti,ab. 37. hotline*.ti,ab. 38. telecar*.ti,ab. 39. tele‐car*.ti,ab. 40. telemedicine*.ti,ab. 41. tele‐medicine*.ti,ab. 42. telehome*.ti,ab. 43. tele‐home*.ti,ab. 44. *counseling/ or directive counseling/ 45. Telephone/ 46. or/34‐45 47. 33 and 46	315
3. EMBASE 1974‐2013 February 22 (Ovid SP)	1. exp dementia/ 2. Lewy body/ 3. delirium/ 4. Wernicke encephalopathy/ 5. cognitive defect/ 6. dement*.mp. 7. alzheimer*.mp. 8. (lewy* adj2 bod*).mp. 9. deliri*.mp. 10. (chronic adj2 cerebrovascular).mp. 11. ("organic brain disease" or "organic brain syndrome").mp. 12. "supranuclear palsy".mp. 13. ("normal pressure hydrocephalus" and "shunt*").mp. 14. "benign senescent forgetfulness".mp. 15. (cerebr* adj2 deteriorat*).mp. 16. (cerebral* adj2 insufficient*).mp. 17. (pick* adj2 disease).mp. 18. (creutzfeldt or jcd or cjd).mp. 19. huntington*.mp. 20. binswanger*.mp. 21. korsako*.mp. 22. CADASIL.mp. 23. or/1‐22 24. caregiver*.ti,ab. 25. carer*.ti,ab. 26. relatives.ti,ab. 27. family*.ti,ab. 28. families*.ti,ab. 29. dependents.ti,ab. 30. kin*.ti,ab. 31. spouse*.ti,ab. 32. parent*.ti,ab. 33. folk*.ti,ab. 34. caregiver/ 35. or/24‐34 36. telephone*.ti,ab. 37. phone*.ti,ab. 38. helpline*.ti,ab. 39. hotline*.ti,ab. 40. telecar*.ti,ab. 41. tele‐car*.ti,ab. 42. telemedicine*.ti,ab. 43. tele‐medicine*.ti,ab. 44. telehome*.ti,ab. 45. tele‐home*.ti,ab. 46. *counseling/ 47. *directive counseling/ 48. or/36‐47 49. 23 and 35 and 48 50. (2011* or 2012* or 2013*).em. 51. 49 and 50	121
4. PSYCINFO 1806‐February week 3 2013 (Ovid SP)	1. exp Dementia/ 2. exp Delirium/ 3. exp Huntingtons Disease/ 4. exp Kluver Bucy Syndrome/ 5. exp Wernickes Syndrome/ 6. exp Cognitive Impairment/ 7. dement*.mp. 8. alzheimer*.mp. 9. (lewy* adj2 bod*).mp. 10. deliri*.mp. 11. (chronic adj2 cerebrovascular).mp. 12. ("organic brain disease" or "organic brain syndrome").mp. 13. "supranuclear palsy".mp. 14. ("normal pressure hydrocephalus" and "shunt*").mp. 15. "benign senescent forgetfulness".mp. 16. (cerebr* adj2 deteriorat*).mp. 17. (cerebral* adj2 insufficient*).mp. 18. (pick* adj2 disease).mp. 19. (creutzfeldt or jcd or cjd).mp. 20. huntington*.mp. 21. binswanger*.mp. 22. korsako*.mp. 23. ("parkinson* disease dementia" or PDD or "parkinson* dementia").mp. 24. or/1‐23 25. caregiver*.ti,ab. 26. carer*.ti,ab. 27. relatives.ti,ab. 28. family*.ti,ab. 29. families*.ti,ab. 30. dependents.ti,ab. 31. kin*.ti,ab. 32. spouse*.ti,ab. 33. parent*.ti,ab. 34. folk*.ti,ab. 35. exp Caregivers/ 36. or/25‐35 37. telephone*.ti,ab. 38. phone*.ti,ab. 39. helpline*.ti,ab. 40. hotline*.ti,ab. 41. telecar*.ti,ab. 42. tele‐car*.ti,ab. 43. telemedicine*.ti,ab. 44. tele‐medicine*.ti,ab. 45. telehome*.ti,ab. 46. tele‐home*.ti,ab. 47. exp Counseling/ 48. or/37‐47 49. 24 and 36 and 48 50. (2011* or 2012* or 2013*).up. 51. 49 and 50	62
5. CINAHL (EBSCOhost)	S1 (MH "Dementia+") S2 MH "Delirium") or (MH "Delirium, Dementia, Amnestic, Cognitive Disorders") S3 (MH "Wernicke's Encephalopathy") S4 TX dement* S5 TX alzheimer* S6 TX lewy* N2 bod* S7 TX deliri* S8 TX chronic N2 cerebrovascular S9 TX "organic brain disease" or "organic brain syndrome" S10 TX "normal pressure hydrocephalus" and "shunt*" S11 TX "benign senescent forgetfulness" S12 TX cerebr* N2 deteriorat* S13 TX cerebral* N2 insufficient* S14 TX pick* N2 disease S15 TX creutzfeldt or jcd or cjd S16 TX huntington* S17 TX binswanger* S18 TX korsako* S19 S1 or S2 or S3 or S4 or S5 or S6 or S7 or S8 or S9 or S10 or S11 or S12 or S13 or S14 or S15 or S16 or S17 or S18 S20 TX caregiver* S21 TX carer* S22 TX relatives S23 TX family* S24 TX families* S25 TX dependents S26 TX kin* S27 TX spouse* S28 TX parent* S29 TX folk* S30 (MH "Caregivers") S31 S20 OR S21 OR S22 OR S23 OR S24 OR S25 OR S26 OR S27 OR S28 OR S29 OR S30 S32 S19 AND S31 S33 TX telephone* S34 TX phone* S35 TX helpline* S36 TX hotline* S37 TX telecar* S38 TX tele‐car* S39 TX telemedicine* S40 TX tele‐medicine* S41 TX telehome* S42 TX tele‐home* S43 (MH "Counseling") S44 S33 OR S34 OR S35 OR S36 OR S37 OR S38 OR S39 OR S40 OR S41 OR S42 OR S43 S45 S19 AND S31 AND S44 S46 EM 2011 S47 EM 2012 S48 EM 2013 S49 S46 OR S47 OR S48 S50 S45 AND S49	73
6. Web of Science (1945‐present) (via Web of Knowledge)	Topic=(dement* OR alzheimer* OR "lewy bod*" OR DLB OR "vascular cognitive impairment*" OR FTD OF FTLD OR "cerebrovascular insufficienc*") AND Topic=(caregiver* OR carer* OR family OR families OR spouse* OR kin OR parent) AND Topic=(telephone OR phone OR telecar* OR tele‐car* OR telemedicine* OR tele‐medicine* OR telehome* OR tele‐home* OR counseling) AND Year Published=(2011‐2013) Timespan=All Years. Databases=SCI‐EXPANDED, SSCI, A&HCI, CPCI‐S, CPCI‐SSH.	108
7. LILACS (BIREME)	telephone OR telefônico [Words] and carer OR caregiver OR family OR families OR kin OR spouse OR parent [Words] and dementia OR demência OR alzheimer [Words]	4
8. CENTRAL (The Cochrane Library) (Issue 1 of 2013)	#1 MeSH descriptor: [Dementia] explode all trees #2 MeSH descriptor: [Delirium] this term only #3 MeSH descriptor: [Wernicke Encephalopathy] this term only #4 MeSH descriptor: [Delirium, Dementia, Amnestic, Cognitive Disorders] this term only #5 dement* #6 alzheimer* #7 "lewy* bod*" #8 deliri* #9 "chronic cerebrovascular" #10 "organic brain disease" or "organic brain syndrome" #11 "normal pressure hydrocephalus" and "shunt*" #12 "benign senescent forgetfulness" #13 "cerebr* deteriorat*" #14 "cerebral* insufficient*" #15 "pick* disease" #16 creutzfeldt or jcd or cjd #17 huntington* #18 binswanger* #19 korsako* #20 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 #21 caregiver* #22 carer* #23 relatives #24 family* #25 families* #26 dependents #27 kin* #28 spouse* #29 parent* #30 folk* #31 #21 or #22 or #23 or #24 or #25 or #26 or #27 or #28 or #29 or #30 #32 #31 and #20 #33 telephone* #34 phone* #35 helpline* #36 hotline* #37 telecar* #38 tele‐car* #39 telemedicine* #40 tele‐medicine* #41 telehome* #42 tele‐home* #43 #33 or #34 or #35 or #36 or #37 or #38 or #39 or #40 or #41 or #42 #44 #43 and #32 from 2011 to 2013, in Trials	4
9. Clinicaltrials.gov (www.clinicaltrials.gov)	Interventional Studies | dementia OR alzheimer* | telephone OR telemedicine OR telecare OR phone OR helpline OR tele‐home OR telecar | Adult, Senior | received from 05/01/2011 to 02/24/2013	5
10. ICTRP Search Portal (http://apps.who.int/trialsearch) [includes: Australian New Zealand Clinical Trials Registry; ClinicalTrilas.gov; ISRCTN; Chinese Clinical Trial Registry; Clinical Trials Registry – India; Clinical Research Information Service – Republic of Korea; German Clinical Trials Register; Iranian Registry of Clinical Trials; Japan Primary Registries Network; Pan African Clinical Trial Registry; Sri Lanka Clinical Trials Registry; The Netherlands National Trial Register]	Telephone AND dementia	24
TOTAL before de‐duplication and first‐assess		811
TOTAL after de‐dupe and first‐assess		94

Appendix 23. Chang 2004 assessed with CASP

No	Criterion	Screening questions	Yes	Note
1	/	Was there a clear statement of the aims of the research? what the goal of the research was why it is important its relevance	X	The purpose of this study was stated First study to address that question
2	/	Is a qualitative methodology appropriate? if the research seeks to interpret or illuminate the actions and/or subjective experiences of research participants	X	Reasons for experiencing a helpful telephone call were examined
		Is it worth continuing?	X	/
3	Dependability	Was the research design appropriate to address the aims of the research? if the researcher has justified the research design (e.g. have they discussed how they decided which methods to use?)	No The study authors gave no justification for the research design
4	Dependability	Was the recruitment strategy appropriate to the aims of the research? if the researcher has explained how the participants were selected if they explained why the participants they selected were the most appropriate to provide access to the type of knowledge sought by the study if there are any discussions around recruitment (e.g. why some people chose not to take part)	Yes This is a sub‐study and so the sample was given by the main sampling strategy
5	Dependability Dependability Dependability	Were the data collected in a way that addressed the research issue? if the setting for data collection was justified if it is clear how data were collected (e.g. focus group, semi‐structured interview etc) if the researcher has justified the methods chosen if the researcher has made the methods explicit (e.g. for interview method, is there an indication of how interviews were conducted, did they used a topic guide?) if methods were modified during the study. If so, has the researcher explained how and why? if the form of data is clear (e.g. tape recordings, video material, notes etc.) if the research has discussed saturation of data	No No discussion about the method of data collection. Face‐to‐face interview data could have been tape‐recorded, but it would have influenced the main study, where face‐to‐face contacts were avoided Data were collected by semi‐structured interviews via telephone Only recorded by hand No discussion or justification of the methods chosen The authors stated very little about how they conducted the interview No modification of the methods during the study The authors named the data format The authors did not discuss saturation of data
6	Dependability	Has the relationship between researcher and participants been adequately considered? if the researcher critically examined their own role, potential bias and influence during: formulation of research questions; data collection, including sample how the researcher responded to events during the study and whether they considered the implications of any changes in the research design	Yes with restrictions The authors examined their own role critically and possible bias partly. They used one strategy to verify the collected data The authors did not report any events or changes in the research design
7	/	Have ethical issues been taken into consideration? if there are sufficient details of how the research was explained to participants for the reader to access whether ethical standards were maintained if the researcher has discussed issues raised by the study (e.g. issues around informed consent or confidentiality or how they have handled the effects of the study on the participants during and after the study) if approval has been sought from the ethics committee	Yes with restrictions The authors did not state how they explained the research to the participants The researchers did not discuss ethical issues raised by the study The study was approved by the Institutional Review Board
8	Dependability	Were the data analysis sufficiently rigorous? if there is an in‐depth description of the analysis process if thematic analysis is used. If so, is it clear how the categories/themes were derived from the data? whether the researcher explains how the data presented were selected from the original sample to demonstrate the analysis process if sufficient data are presented to support the findings to what extent contradictory data are taken into account whether the researcher examined their own role, potential bias and influence during analysis and selection of data for presentation critically	Yes with restrictions The researcher described the analysis process The authors named the different steps within the content analysis Authors show an intermediate stage of their categories The researchers describe their categories but quote for less than half of their categories correlated citations. The findings could have been presented in more detail Contradictory data were taken into account only sporadically The authors critically examine their own role and possible bias. They used different strategies to not influence the data analysis
9	Credibility	Is there a clear statement of findings? if the findings are explicit if there is adequate discussion of the evidence both for and against the researcher’s arguments if the researcher has discussed the credibility of their findings (e.g. triangulation, respondent validation, more than one analyst) if the findings are discussed in relation to the original research question	Yes with restrictions The findings were described but could have been explained in more detail The researchers discussed their findings The researchers did not discuss the credibility of their findings The authors discussed the findings in relation to the original research question
10	/	How valuable is the research? if the researcher discusses the contribution the study makes to existing knowledge or understanding (e.g. do they consider the findings in relation to current practice or policy, or relevant research‐based literature?) if they identify new areas where research is necessary if the researchers have discussed whether or how the findings can be transferred to other populations or considered other ways the research may be used	Valuable The researchers linked each theme of the findings to existing knowledge The authors showed new research areas The authors addressed the transferability

Appendix 24. Salfi 2005 assessed with CASP

No	Criterion	Screening questions	Yes	Note
1	/	Was there a clear statement of the aims of the research?	X	/
2	/	Is a qualitative methodology appropriate? if the research seeks to interpret or illuminate the actions and/or subjective experiences of research participants	X	Dimensions and experiences were examined
		Is it worth continuing?	X	/
3	Dependability	Was the research design appropriate to address the aims of the research? if the research has justified the research design (e.g. have they discussed how they decided which methods to use?)	Yes The authors discuss the research design
4	Dependability	Was the recruitment strategy appropriate to the aims of the research? if the researcher has explained how the participants were selected if they explained why the participants they selected were the most appropriate to provide access to the type of knowledge sought by the study if there are any discussions around recruitment (e.g. why some people chose not to take part)	Yes with restrictions The authors describe their sampling procedures (criterion and convenience) with inclusion and exclusion criteria via gatekeepers (directors of the agencies) The authors did not explain why they selected these participants The authors critiqued their sampling strategy
5	Dependability Dependability Dependability	Were the data collected in a way that addressed the research issue? if the setting for data collection was justified if it is clear how data were collected (e.g. focus group, semi‐structured interview etc) if the researcher has justified the methods chosen if the researcher has made the methods explicit (e.g. for interview method, is there an indication of how interviews were conducted, did they used a topic guide?) if methods were modified during the study. If so, has the researcher explained how and why? if the form of data is clear (e.g. tape recordings, video material, notes etc.) if the research has discussed saturation of data	Yes with restrictions The setting for data collection was given by the research question and general situation The authors stated the data collection method The authors did not justified the chosen method The used method (interview guide) was explained The authors stated deviations from the study protocol and explained why The data format was clearly stated The authors stated that they reached saturation. Comment by the review authors: To reach saturation with the interview data of only eight caregivers and four counsellors is difficult
6	Dependability	Has the relationship between researcher and participants been adequately considered? if the researcher critically examined their own role, potential bias and influence during: formulation of research questions; data collection, including sample how the researcher responded to events during the study and whether they considered the implications of any changes in the research design.	Yes Strategies were used to help validate the data and to ensure the credibility and trustworthiness of the research findings The authors did not report any events or changes in the research design
7	/	Have ethical issues been taken into consideration? if there are sufficient details of how the research was explained to participants for the reader to access whether ethical standards were maintained if the researcher has discussed issues raised by the study (e.g. issues around informed consent or confidentiality or how they have handled the effects of the study on the participants during and after the study) if approval has been sought from the ethics committee	Yes No detailed description on how the research was explained to participants The authors discussed issues raised by the study around informed consent Approval was received from an ethics committee prior to conducting this study
8	Dependability	Was the data analysis sufficiently rigorous? if there is an in‐depth description of the analysis process if thematic analysis is used. If so, is it clear how the categories/themes were derived from the data? whether the researcher explains how the data presented were selected from the original sample to demonstrate the analysis process if sufficient data are presented to support the findings to what extent contradictory data are taken into account whether the researcher critically examined their own role, potential bias and influence during analysis and selection of data for presentation	Yes No in‐depth data are available The authors describe the analysis process in detail Data were collected from the caregivers and the support providers The authors present sufficient data to support the findings Only few contradictory data were reported The authors discussed their own potential bias and implemented many strategies to prevent bias
9	Credibility	Is there a clear statement of findings? if the findings are explicit if there is adequate discussion of the evidence both for and against the researcher’s arguments if the researcher has discussed the credibility of their findings (e.g. triangulation, respondent validation, more than one analyst) if the findings are discussed in relation to the original research question	Yes with restrictions Yes, but one result is only represented in the discussion section “importance of debriefing” [end of page 717] and not in the results section The authors discuss their findings, but not very intensively The researcher did not discuss the credibility of the findings Yes
10	/	How valuable is the research? if the researcher discusses the contribution the study makes to existing knowledge or understanding (e.g. do they consider the findings in relation to current practice or policy, or relevant research‐based literature?) if they identify new areas where research is necessary if the researchers have discussed whether or how the findings can be transferred to other populations or considered other ways the research may be used	Yes They tried to do it, but the study is one of the first for this research question not exactly, but they showed that new themes should be noted in offering telephone counselling and maybe some more research should be conducted to confirm the findings They discussed the lack of statistical data on all clients as a limit of the study

Appendix 25. Main reported findings of included qualitative studies

Study	Reported main findings of the study
Chang 2004	1. Sharing thoughts and feelings 2. Expressing feelings of being overwhelmed 3. Discussing physical and psychosocial problems 4. Forgetting the situation 5. Seeking reassurance 6. Asking for information 7. Lacking in helpfulness
Salfi 2005	1. Telephone support meets the need for information and education 2. Telephone support meets the need for referral and assistance to navigate the system 3. Telephone support meets the need for emotional support 4. Telephone support is a convenient way to meet caregiver needs 5. The need for increased hours of availability 6. The need for a multilingual service 7. Companionship 8. Helplessness and frustration 9. Empowering caregivers

Data and analyses

Comparison 1. Telephone counselling.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Depressive symptoms	3	163	Std. Mean Difference (IV, Random, 95% CI)	0.32 [0.01, 0.63]
2 Burden	4	165	Std. Mean Difference (IV, Random, 95% CI)	0.45 [‐0.01, 0.90]
3 Support	2	67	Std. Mean Difference (IV, Random, 95% CI)	0.25 [‐0.24, 0.73]

1.1. Analysis.

Comparison 1 Telephone counselling, Outcome 1 Depressive symptoms.

1.2. Analysis.

Comparison 1 Telephone counselling, Outcome 2 Burden.

1.3. Analysis.

Comparison 1 Telephone counselling, Outcome 3 Support.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Chang 1999.

Methods	Randomised controlled trial
Participants	Country: USA n = 65 Inclusion criteria: caregiving dyads: English‐speaking, access to a videotape player and a telephone Care recipient: significant dressing and eating problems as screened by the Zarit's memory/behavioral scale (3 or 4 on tasks of dressing and eating), Mini‐Mental State Examination (MMSE) score < 21
Interventions	Intervention: the Nurseline video‐assisted modelling program (NVAMP) intervention consisted of two parts: (a) videotapes demonstrating assisted modelling behaviour (eating and dressing); and (b) a Nurseline support program to reinforce the video information and assist the caregiver to explore coping strategies Control: attention‐only (A‐O) placebo telephone calls Length: 2 months
Outcomes	Carer: depressive symptoms, anxiety, burden, satisfaction and coping strategies
Notes	Funding source: National Alzheimer's Disease and Related Disorders Association, Chicago
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Most outcomes were self‐reported
Incomplete outcome data (attrition bias) All outcomes	High risk	Detailed comparison between completers and drop‐outs, reasons for drop‐out explained; but MMSE differed between groups and no intention‐to‐treat analysis (ITT)
Selective reporting (reporting bias)	High risk	Only reported that caregiver burden was not significant; no data stated
Other bias	Low risk	None

Chang 2004.

Methods	Qualitative process evaluation
Participants	Country: USA n = 83 Inclusion: spoke English, had access to a VCR (video) player and a telephone and lived with an individual with dementia who had problems with dressing or eating
Interventions	Method: delivery of cognitive behavioural therapy (CBT) via semi‐structured interviews and open‐ended questions Methodology: not reported Research question: the purpose of the study was to determine the perceived helpfulness of telephone calls to 83 caregivers of family members with dementia Used data: counsellors' notes recorded whilst delivering CBT‐based counselling by phone (recording by hand as nearly verbatim as possible during the telephone call) and memos
Outcomes	Analysis: content analysis as proposed by Strauss 1990. Open Coding, line‐by‐line analysis and development of concepts and themes
Notes	Funding source: Alzheimer’s Association

Davis 2004.

Methods	Randomised controlled trial
Participants	Country: USA n = 71 Inclusion criteria Carers Live with the care recipient. Have primary caregiving responsibility for an average of 4 hours in the home each day. Be able to speak and read English. Not be participating in any other caregiver study. Be available by phone. Care recipients Be 50 or older. Have a physician‐determined diagnosis of dementia. Have a cognitive function (MMSE) score of 24 or less. Have 4 or more behavior problems, as determined by the problem frequency scale of the Revised Memory and Behavior Problem Checklist.
Interventions	Intervention: telephone‐based skills training plus conversation about finding and using local community resources Control: friendly calls Length: 3 months
Outcomes	Carer: burden, distress, depressive symptoms, social support, life satisfaction Care recipient: instrumental independence, physical independence, cognitive status
Notes	Funding source: The National Institute of Nursing Research (KG1 NR 00095) and the REACH Project at the University of Alabama and the University of Alabama at Birmingham (5U01NR 04261)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated; stratified by race of caregiver
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	5 of 8 outcomes were self‐reported measures of caregivers, 2 of these 8 outcomes were caregiver‐scored; blinding of caregivers not possible, so likely that the blinding could have been broken
Incomplete outcome data (attrition bias) All outcomes	High risk	Drop‐outs described, but unclear from which group(s) they came; not possible to judge, whether there was an imbalance between groups. In combination with per protocol‐analysis
Selective reporting (reporting bias)	High risk	Only significant outcomes were reported
Other bias	Low risk	None

Finkel 2007.

Methods	Randomised controlled trial
Participants	Country: USA n = 46 Inclusion criteria Caregivers: providing care for a relative with Alzheimer disease or related dementia for a minimum of 4 hours per day for at least the past 6 months; over the age of 21 years; living with or in the same geographic vicinity as the patient; having a telephone; planning to remain in the geographic area for at least 6 months; and competent in English. Care recipient: had a physician diagnosis of Alzheimer disease or another type of dementia and a MMSE score < 23 Exclusion criteria Dyads were excluded if the caregiver or care recipient had a terminal illness with life expectancy < 6 months; was in active treatment for cancer; was blind or deaf; or if the care recipient had an MMSE score of 0 and was bed bound (or nursing home admission was planned within 6 months).
Interventions	Intervention: modelled after the intervention developed at the Miami, Florida site of the Resources for Enhancing Alzheimer’s Caregiver Health (REACH) program and focused on provision of information about the disease and community resources and strategies to enhance safety, communication, self‐care, social support, and management of problem behaviours. The intervention differed from the REACH trial in that in that the customized computer‐telephone integration system (CTIS) was the primary vehicle for intervention Control: caregivers in the control group received basic educational materials, 2 brief (15 minute) telephone check‐in calls at 3 and 5 months postrandomisation, and were invited to participate in a workshop following the 6‐month assessment. Length: 6 months
Outcomes	Carer: depressive symptoms, burden, self‐efficacy, social support
Notes	Funding source: Braun Fund
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcome assessment by certified assessors who were masked with respect to treatment condition, but the outcomes were self‐reported by the unblinded participants
Incomplete outcome data (attrition bias) All outcomes	Low risk	36 of 46 patients completed study, reasons for drop‐out were stated, it seems that the drop‐outs and the reasons for drop‐out were nearly balanced. Per protocol‐analysis
Selective reporting (reporting bias)	Low risk	No hints of selective reporting
Other bias	Low risk	None

Gant 2007.

Methods	Randomised controlled trial
Participants	Country: USA n = 32 Inclusion criteria Carers: male; willingness to be assigned by chance to either group (video vs instructional booklet); primary caregiver for a family member diagnosed with a dementing illness; living in same residence as care recipient; age 30‐85 years; no plans to place the family member in a nursing home or hospice within next 6 months; visual acuity sufficient to read a book; able to hear a voice on the telephone and TV; no current alcohol abuse or serious suicidal ideation; no history of suicide attempt; no current involvement in another caregiver intervention project; access to telephone, TV, and VCR; an established relationship with a clinic or primary care physician whom the caregiver had visited within the last year; at least 2 care recipient memory or behaviour problems that occurred in the past week, and at least a moderate level of distress (upset or annoyance) following these Care recipients: no lifetime history of schizophrenia, bipolar disorder, suicidality, Huntington’s disease, Korsakoff’s disease, multiple sclerosis, human immunodeficiency virus (HIV), or alcohol abuse; an established relationship with a clinic or primary care physician whom care recipient had visited within the last year; and a diagnosis of dementia confirmed by a physician
Interventions	Intervention: video series plus a workbook and telephone calls Control: education (booklet with instruction) plus telephone calls Length: 3 months
Outcomes	Carer: burden (upset, annoyance), distress, self‐efficacy, affect (positive and negative emotions)
Notes	Funding source: The Missouri Alzheimer’s Disease and Related Disorders Research Board and the National Institute of Mental Health
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Online randomisation (email correspondence)
Allocation concealment (selection bias)	Low risk	Sealed envelopes, research assistant not further involved in study (email correspondence)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcome were self‐reported measures of carers, blinding of carers was not possible
Incomplete outcome data (attrition bias) All outcomes	Low risk	Primary outcomes reported, ITT, reported withdrawals, stated reasons for withdrawals
Selective reporting (reporting bias)	Low risk	No hints of selective reporting
Other bias	Low risk	None

Glueckauf 2007.

Methods	Randomised controlled trial
Participants	Country: USA n = 36 Inclusion criteria Carers: provided care to their care recipient for a minimum of 6 hours per week for at least 6 months; reported specific caregiving problems that were amenable to short‐term cognitive behavioural intervention; reported no difficulties in hearing over the telephone; did not have a terminal illness Care recipients: a medical diagnosis of Alzheimer's disease or another type of progressive dementia; at least one limitation in basic activities of daily living (ADLs) or 2 dependencies in instrumental ADLs Exclusion criteria Carer: their care recipient had a terminal condition and death was anticipated within 6 months or had a severe illness other than progressive dementia; or they reported significant psychological difficulties (e.g. marked depression or suicidal ideation) for which immediate intervention was required
Interventions	Intervention: telephone counselling Control: routine education and support, plus written educational materials Length: 3 months
Outcomes	Carer: burden, depressive symptoms and self‐efficacy
Notes	Funding source: Johnnie B Byrd, Sr, Alzheimer's Center and Research Institute
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Most outcomes were self‐reported and a blinding of participants was not possible
Incomplete outcome data (attrition bias) All outcomes	High risk	14 of 36 family carers completed all phases of counselling program (12 weeks), drop‐outs reported, but not for all and no differentiation was made between groups.
Selective reporting (reporting bias)	Low risk	No hints of selective reporting
Other bias	Low risk	None

Salfi 2005.

Methods	Qualitative descriptive study
Participants	Country: Canada n = 12 (8 carers, 4 counsellors) Inclusion criteria Carer: 18 years of age or older; had assumed the majority of caregiving duties for a family member with dementia; was able to communicate in English; had received telephone support at some point within the previous year; and resided within south central Ontario, Canada. Telephone support providers: healthcare professionals affiliated with 1 of 2 community organisations; able to communicate in English; and had provided telephone support to caregivers of persons with dementia within the previous year. Exclusion criteria Carer: caring for a person with a dementia related to malignancy or medication, or both. Telephone support providers: nonprofessional telephone support providers (i.e. friends and peers from support groups)
Interventions	Method: face‐to‐face interviews or telephone interviews with the caregivers and telephone support providers Methodology: not reported Research questions: What are the dimensions of the telephone support intervention? What are the experiences of caregivers of people with dementia who receive telephone support? What are the experiences of telephone support providers who provide telephone support to caregivers of people with dementia? Data used: contextual data (researcher field notes and memos, information from contact summary sheets, descriptive details regarding the agencies and telephone service providers) and interview data
Outcomes	Analysis: categorical aggregation, direct interpretation, patterns and naturalistic generalisation
Notes	Funding source: Alzheimer Society of Hamilton and Halton and the Wellington‐Dufferin Community Care Access Centre

Steffen 2000.

Methods	Randomised controlled trial
Participants	Country: USA n = 33 Inclusion criteria: carers self‐reported to be providing at least 5 hours per week of face‐to‐face direct care for a cognitively impaired older relative diagnosed with Alzheimer's disease, multi‐infarct dementia, or some other dementing illness; be able to describe at least 1 patient behaviour or caregiving situation in the past 2 weeks that was upsetting or stressful for them; have access to a working television and VCR weekly for 8 weeks; not exhibit evidence of psychosis, cognitive impairment, alcoholism, immediate suicidal risk, or bipolar disorder; not concurrently in psychotherapy; if receiving psychotropic medication, stabilised on same dose for at least 12 weeks; willing to accept random assignment to interventions; and patient had to be living in the community (i.e. not in a nursing home or an assisted‐living facility)
Interventions	Intervention: video plus workbook and telephone calls Control: waiting list Length: 2 months
Outcomes	Carer: anger intensity, depressive symptoms, self‐efficacy
Notes	Funding source: not stated
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Online randomisation (email correspondence)
Allocation concealment (selection bias)	Low risk	Sealed envelopes, research assistant not further involved in study (email correspondence)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcomes were self‐reported measures of carers, blinding of carers not possible
Incomplete outcome data (attrition bias) All outcomes	Low risk	Per protocol analyses, but primary outcomes reported, incomplete outcome data addressed, reasons stated, no imbalance
Selective reporting (reporting bias)	Low risk	No hints of selective reporting
Other bias	Low risk	None

Tremont 2008.

Methods	Randomised controlled trial
Participants	Country: USA n = 60 Inclusion criteria Carer: age 21 or older; lived with a relative with dementia in the community; provided a minimum of 4 hours of supervision or direct care per day for at least 6 months prior to enrolment Care recipients: age 50 or older; a medical diagnosis of dementia as indicated by a formal Diagnostic and Statistical Manual of Mental Disorders diagnosis of dementia confirmed by a physician; Clinical Dementia Rating (CDR) score of 1 (mild) or 2 (moderate) Exclusion criteria Carer: significant psychiatric illness (e.g. schizophrenia, bipolar disorder) or cognitive impairment (MMSE score < 25)
Interventions	Intervention: telephone counselling plus information about local resources and educational materials Control: information about local resources and educational materials Length: 12 months
Outcomes	Primary outcomes: carers ‐ depressive symptoms, burden and revised memory and behaviour problems Secondary outcomes: Alzheimer’s Disease Knowledge Test, quality of life/general health, self‐efficacy, Family Assessment Device, social support
Notes	Funding source: The National Institute of Mental Health (MH62561)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Urn randomisations (balancing on dementia severity, carer gender and relationship type)
Allocation concealment (selection bias)	Unclear risk	Not stated; possibly not concealed because of urn randomisations
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcomes were self‐reported measures of carers, blinding of carers not possible
Incomplete outcome data (attrition bias) All outcomes	High risk	60 carers enrolled at baseline, 33 carers had data for analysis at 12‐month assessment point; number of deaths differed between groups. Drop‐outs in the control group had higher depression scores
Selective reporting (reporting bias)	High risk	Study investigators reported only the significant outcomes
Other bias	Low risk	None

Wilz unpublished.

Methods	Randomised controlled trial
Participants	Country: Germany n = 229 Inclusion criteria Carer: full‐time carer; no simultaneous psychotherapy; no cognitive impairment; no severe acute mental or physical illness, or both Care recipients: being cared for; diagnosis of Alzheimer’s disease; a GDS‐score > 3 (rated by the screening person based on the carer’s report) Exclusion criteria: Care recipients: being cared for in day care for more than 3 days per week
Interventions	Intervention: telephone counselling Controls: 2 control groups: 1 received progressive muscle relaxation and the other group received treatment as usual (untreated control group). Only the untreated control group were considered in the review Length: 3 months
Outcomes	Depressive symptoms, GBB‐24 (24 items to measure physical complaints: i.e. exhaustion, stomach trouble, rheumatic pain, heart trouble), emotional well‐being, perceived health status (by thermometer scaling 0‐100), GOAL attainment
Notes	Funding source: The German Federal Ministry of Health (LTDEMENZ‐44‐092)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomisation stratified by carers' relationship to patient, and gender; independent data management and biometry centre ensured randomisations, but method of sequence generation not described
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Blinded assessors, but most outcomes were self‐reported (and blinding of carers not possible)
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Unclear whether method of analysis was ITT or per protocol
Selective reporting (reporting bias)	Low risk	No hints of selective reporting
Other bias	Low risk	None

Winter 2007.

Methods	Randomised controlled trial
Participants	Country: USA n = 103 Inclusion criteria: females; 50 years of age or older; providing care for a minimum of 6 months to a relative with a physician’s diagnosis of ADRD (Alzheimer's disease or related disorders); and having weekly access to a telephone for at least 1 hour.
Interventions	Intervention: telesupport groups; 5 carers per group for an hour weekly Control: usual care Length: 6 months
Outcomes	Carer: depressive symptoms, burden and personal gains
Notes	Funding source: The Alzheimer’s Association
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding not possible, this could have influenced additional interventions, and thus most of the (subjective) outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcomes were self‐reported measures of carers, blinding of carers not possible
Incomplete outcome data (attrition bias) All outcomes	High risk	Primary outcome reported, but per‐protocol analyses, and no differentiation of drop‐out‐rates or reasons for dropping out stated.
Selective reporting (reporting bias)	Low risk	No hints of selective reporting
Other bias	Low risk	None

Abbreviations

ADLs = activities of daily living
 ADRD = Alzheimer's disease or related disorders
 A‐O = attention only
 CBT = cognitive behavioural therapy
 CDR = Clinical Dementia Rating
 CTIS = customized computer‐telephone integration system
 GBB = Gießen Body Complaint List
 GOAL = goal attainment scaling
 ITT = intention‐to‐treat analysis
 MMSE = Mini Mental State Examination
 NVAMP = Nurseline video‐assisted modelling program

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Glueckauf 2012	No control group with usual care or friendly calls
McHugh 2012	The counselling was not performed by a professional
van Mierlo 2012	No randomised controlled trial
Waldorff 2012	Intervention included regular face‐to‐face contact

Characteristics of ongoing studies [ordered by study ID]

Nichols 2005.

Trial name or title	Testing the effectiveness of telephone support for dementia caregivers (CONNECT)
Methods	Randomised clinical trial
Participants	154 caregivers (Black/African American, White/Caucasian, rural, urban)
Interventions	Behavioural: telephone support: each telephone support group of five caregivers and a group leader met 14 times over a year. The hour calls were semi‐structured with an educational component and a support component, led by a trained Group Leader. Topics included knowledge of dementia, safety, caregiver health and well being, communication, managing behavioural challenges and caregiver stress and coping
Outcomes	Primary: general well‐being, caregivers' levels of distress with care recipient behaviours. Secondary: cost‐effectiveness, time spent providing care
Starting date	2005
Contact information	Principal Investigator: Linda Olivia Nichols, PhD
Notes	ClinicalTrials.gov Identifier: NCT00119561

Tremont 2011.

Trial name or title	Relationship between expectancy/credibility and early response to telephone‐based dementia caregiver interventions
Methods	Randomised controlled trial
Participants	133 caregivers
Interventions	Family Intervention: telephone tracking (FITT)
Outcomes	Caregiver burden, depressive symptoms and others
Starting date	Not available
Contact information	Geoffrey Tremont, PhD
Notes	Will be published soon

Differences between protocol and review

Evaluation of efficacy

In the protocol we said that the qualitative data extraction form would be checked by two researches. However, it was checked by only one researcher. Furthermore, we wanted to conduct one meta‐analysis comparing the intervention with no treatment and another meta‐analysis comparing the intervention with general friendly calls, however, because we found so little evidence, we did not undertake this differentiation. Additionally, we presented the results of non‐pooled trials in narrative form, instead of as forest plots without pooling.

Evaluation of experience

A sensitivity analysis was not conducted for the experience aspect of the review due to the limited number of qualitative studies.

Clarification of methods

Evaluation of efficacy

Efficacy now includes the use of R for extracting standard deviations from P values. A sensitivity analysis with respect to the differences in methodological quality assessed by The Cochrane Collaboration's tool for assessing risk of bias was not conducted due to the limited number of quantitative studies. If sufficient additional data become available in future, we will complete this comparison.

Evaluation of experience

After consultation with the Cochrane Qualitative and Implementation Methods Group, we expanded the area of experience to include the perspective of the counsellors, and added the second part of the sentence: “Synthesis of qualitative studies to explore carers’ experiences of receiving telephone counselling and counsellors’ experiences of conducting telephone counselling” to Objectives. Furthermore, we adapted the inclusion criteria under Types of participants to read:

“For evaluation of experience, counsellors’ experiences of telephone counselling were considered and involved in the synthesis as well, if their experiences were investigated in combination with caregivers’ experiences.”

The counsellors’ perspectives are an enrichment for the synthesis because effective counselling also depends on the interaction between the carer and the counsellor. The experiences of the counsellors can assist in interpretation of carers’ needs and experiences.

Additionally, we used the CerQual for qualitative assessment in the review.

Contributions of authors

SL: correspondence; drafting review versions; search for trials, selection of RCTs; extraction of data; assessing risk of bias; data entry, data analysis; interpretation of data/analyses, updating review

DHB: selection of RCTs; extraction of data (qualitative studies); quality assessment (qualitative studies), data synthesis, interpretation of data synthesis, updating review

GR: data analysis, interpretation of data/analyses, updating review

EM: search for trials, obtaining copies of trial reports, updating review

GL: assessing risk of bias, interpretation of data/analyses

GA: interpretation of data/analyses

GM: selection of RCTs; interpretation of data/analyses, updating review

Sources of support

Internal sources

• German Cochrane Centre, Institute of Medical Biometry & Medical Informatics, University Medical Centre Freiburg, Germany.

• German Center for Evidence‐based Nursing, Martin Luther University Halle‐Wittenberg, Halle, Germany.

• Department of Nursing Science, Faculty of Health, Witten/Herdecke, Germany.

External sources

• Ministry of Education and Research, Germany.

  Financial support (funding number: 3832/03.07_2)

• Cochrane Qualitative and Implementation Methods Group, Other.

  Comments on the draft Cochrane protocol and review

Declarations of interest

SL: no conflicts of interest to declare
 DHB: no conflicts of interest to declare
 GR: no conflicts of interest to declare
 EM: no conflicts of interest to declare
 GA: no conflicts of interest to declare
 GM: no conflicts of interest to declare
 GL: no conflicts of interest to declare

New