Psychosocial interventions for prevention of depression in older people

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

1. To assess the effectiveness of psychosocial interventions in primary prevention of depressive symptoms and unipolar depressive disorders in people over the age of 65.

2. The secondary objective is to separately assess the effectiveness of different forms of preventive interventions that have an impact on the social capital (i.e. social network, social support, trust, social participation) of participants.

Background

Description of the condition

The incidence of depressive symptoms and depression is frequent and the prevalence is high; unipolar depression is expected to become the second‐ranked cause of disease burden in 2020 worldwide (WHO 2004). Among the ageing population, anxiety and depression are currently the most prevalent mental health problems (Luijendijk 2008; de Beurs 2005). With the growth of the older adult population worldwide, depressive disorders in older people are set to become a bigger mental health issue (European Commission 2004; Heikkinen 2004).

According to the Diagnostic and Statistical Manual of Mental Disorders (DSM‐IV) major depression (also named unipolar depression or clinical depression) is a mental illness characterized by symptoms such as depressed mood and diminished interest or pleasure in nearly all activities, which must have been prominent for a period of at least two weeks. Clinical depression can also include symptoms such as a decrease in the everyday functional level and feelings of despair, worthlessness or guilt, as well as thoughts of committing suicide. Among older people, cognitive symptoms (e.g. disorientation, memory loss) seem to occur more frequently than among the other age groups (DSM‐IV). Older adults also often report physical problems rather than emotional problems (DSM‐IV), resulting in under‐diagnosis of depressive disorders (DSM‐IV). The onset of depression and its recurrence is influenced by a wide range of risk and protective factors at different stages of the lifespan, including biological, psychological, family, social and societal factors (WHO 2004).

One risk factor for the onset of depression is gender (Smit 2006; Cole 2003; Steunenberg 2006); according to a large body of research, women are more at risk of depression than men. Related to this, greater levels of loneliness have frequently been found among older women than older men (Pinquart 2001). Both a negative (Jongenelis 2004) and positive (Beekman 1995) association between increasingly older age and the prevalence of depressive disorders have been reported. Various medical conditions and functional limitations are generally among the most common risk factors for depressive symptoms (Bisschop 2004) and for the incidence (Krishnan 2002; Smit 2006) and prevalence (Krishnan 2002; Beekman 1995) of depressive disorders among older people. The association between socio‐economic factors and the onset of late‐life depression has also been highlighted (Smit 2006).

Different types of individual psychosocial resources seem to have a significant effect on depressive symptoms, mentioning self‐esteem (Bisschop 2004) and mastery (Steunenberg 2006; de Beurs 2005) as illustrative examples. Studies have shown that associations exist between social capital and mental health indicators among the ageing population. Crucial components of the individual‐level social capital concept (Almedom 2005), such as social support (Lynch 1999; Jongenelis 2004), social network size (Steunenberg 2006), and loneliness (Bisschop 2004; Jongenelis 2004) were shown to be associated with depressive symptoms and depression. Research has highlighted that civic mistrust and lack of reciprocity or of social participation (i.e. low individual‐level social capital) are associated with depressive symptoms among the older adults (Pollack 2004).

The primary prevention of depressive symptoms and depression may therefore be regarded as crucial. There is a need for effective interventions for the prevention of depression and to promote mental health, implementing findings from available research in the area.

Description of the intervention

Research has demonstrated a significant reduction in depressive symptoms through adequate prevention programmes, but the number of studies targeting the incidence of depression and its prevention is limited (Jané‐Llopis 2003). Studies in the area have demonstrated that time‐limited, small group‐based psychoeducational and skill‐training interventions, theoretically based on cognitive behavioural principles, have an impact on depressive symptoms and quality of life (Coon 2003). For instance, skill training with a focus on emotional problem management, e.g. depression or anger management, may lead to decreased symptoms of depression, as well as significant positive increases in self‐efficacy in negative thought management among the older adult participants (Coon 2003). The importance of psychosocial interventions customized for older adults suffering from various functional limitations is emphasized in a growing body of research. These interventions encompass information about the available professional aids, with support as well as the promotion of personal resources and coping strategies being crucial components (Birk 2004). Studies on the social network and social support of older adults suffering from physical impairment have targeted interventions aimed at reducing loneliness among impaired older adults (Verstraten 2005). Group discussions and exchanges of experiences among the participants are also frequently applied in psychosocial intervention programmes (Birk 2004). Studies also show that psychosocial interventions aiming to increase the social contacts of the older participants tend to improve the mental well‐being as the feelings of loneliness are reduced (Stevens 2000). These friendship programmes often attract older women living alone and they are therefore targeting a major risk group for depression (Stevens 2000).

How the intervention might work

As previously illustrated, there are certain vulnerable groups at risk for depressive symptoms and depression. By identifying the risk factors (Heikkinen 2004), screening the older adult population for risk factors (Cole 2003) and by targetting effective preventive interventions towards high‐risk groups, a cost‐effective approach for the prevention of depressive symptoms and depression can be achieved (Smit 2006).

Why it is important to do this review

Given the large population in question, there is a clear need for a systematic review of existing evidence on the effectiveness of psychosocial interventions for the primary prevention of depressive symptoms and depression in older people. A review and meta‐analysis on the prevention of the onset of depressive disorders in all age groups was recently published (Cuijpers 2008), but of the included studies, only one targeted older adults. There is already a Cochrane review on psychotherapeutic treatments for older people suffering from depression (Wilson 2008), and another review is ongoing looking at the continuation and maintenance treatments for depressive disorders among older adults (Wilkinson 2007). However, these two reviews deal with treatments for older people already suffering from depressive disorders. This review will target the primary prevention of depressive symptoms and depression among older adults.

Objectives

1. To assess the effectiveness of psychosocial interventions in primary prevention of depressive symptoms and unipolar depressive disorders in people over the age of 65.

2. The secondary objective is to separately assess the effectiveness of different forms of preventive interventions that have an impact on the social capital (i.e. social network, social support, trust, social participation) of participants.

Methods

Criteria for considering studies for this review

Types of studies

Only studies with a randomised controlled design (RCTs) will be considered for this review. Quasi‐randomised trials and all studies where inadequate allocation concealment has been applied will be excluded.

Types of participants

Older adults (population defined as people aged 65 and older or when referred to as elderly, geriatric, senior citizens or older adults) who do not meet the diagnostic criteria for any depressive disorder at the time of enrolment in the intervention. This includes studies targeting the general older adult population, studies with participants who might be at risk for depression, or those who already suffer from sub‐clinical symptoms of depression but who do not fulfil the diagnostic criteria for any depressive disorder (according to DSM‐IV or ICD‐10) at the time. Older adults with physical disease will be included in the review, but older populations suffering from any mental disorder (e.g. dementia) are not. Studies will be considered even if the age range begins under 65, if the mean age is notably over 65 (i.e. 70 or over). Also, if outcomes relevant for the review are reported separately for participants over the age of 65, the study will be included even though the mean age of the study population is under 70 years. There will be no upper age limit for the participants considered. The study setting can be institutions or in the community.

Studies that do not include a clear definition of participants or lack adequate assessment of participants at enrolment will be excluded from the review.

Types of interventions

Experimental interventions

For this review psychosocial interventions are defined as any intervention that emphasizes psychological or social factors rather than biological factors (Ruddy 2005). This definition allows for inclusion interventions with components of psychological therapies and health education, as well as interventions with a focus on social aspects, such as social support and networking. We will also include interventions that have a biological component in addition to a psychosocial component, e.g. exercise groups for older people. Interventions aim directly or indirectly to reduce or prevent depressive symptoms and/or the onset of a depressive disorder in the older adult population. The psychosocial interventions can appear in any format, e.g. in groups or individually, as long as they are described in the study and allow replication.

These criteria exclude secondary and tertiary preventive interventions, relapse prevention and all forms of pharmacological interventions. Interventions that target the organization of care will not be considered for this review.

The psychosocial intervention studies will be categorized according to the experimental interventions' content into one of the following categories:

1. Skill training interventions (e.g. education interventions, cognitive training)

2. Physical exercise interventions (e.g. physical exercise programmes in groups)

3. Group support interventions (e.g. discussion groups)

4. Reminiscence interventions (i.e. interventions with components of life reviewing or life storytelling)

5. Social activities (e.g. visiting cultural sites, gardening, needlework)

6. Multicomponental interventions (i.e. interventions containing various combinations of several different components)

7. Other psychosocial interventions (i.e. any intervention that fulfils the inclusion criteria but cannot be considered in any of the other interventions groups listed)

Each category of psychosocial interventions will be compared with control groups and with other psychosocial interventions. Studies which have an increase of social capital amongst their stated aims will be considered for sub‐analysis. Such aims will be an increase in social network, social support, trust or social participation.

Comparator interventions

All included interventions will have a control condition that might be care as usual, a waiting list group or other comparison intervention or no intervention.

Types of outcome measures

Primary outcome

The primary outcome measure used in the review will be the occurrence of depressive symptoms, as measured by depression rating scales (such as the Geriatric Depression Scale). The primary outcome will be reported using continuous data.

Secondary outcomes

Secondary outcome measures are as follows:

1. Incidence of major depression (as defined by DSM‐IV, ICD‐10 or diagnostic cut‐off points on depression rating scales). 2. Suicidality 3. Functional level (e.g. various ADL measures) 4. Quality of life (e.g. WHOQOL, WHO 1997). Measures of health‐related quality of life (e.g. SF‐36 Health Survey, Hays 1997) will also be considered for this outcome. 5. Life satisfaction (e.g. SWLS, Diener 1985) 6. Cost‐effectiveness of the interventions (e.g. financial cost comparisons) 7. Acceptability of intervention (total attrition rates or attrition due to adverse effects) 8. Determinants (major risk and protective factors) of depressive symptoms and depression: a. Positive mental health (e.g. sense of coherence (SOC, Antonovsky 1987; Antonovsky 1993) and self‐esteem (RSES, Rosenberg 1965) b. Social capital (social network (e.g. UCLA Loneliness Scale, Russell 1996), social support (e.g. OSLO‐3, Brevik 1996), trust (e.g. Interpersonal Trust Scale (ITS), Rotter 1967).

Studies that do not include prevention of depression or reduction of depressive symptoms as an outcome will be excluded from the review.

The outcomes will be recorded immediately after the intervention and at follow‐up. Outcome at follow‐up will be grouped according to length of follow‐up in three categories: Up to six months, up to two years, and more than two years.

Search methods for identification of studies

Electronic searches

See: Cochrane Depression, Anxiety and Neurosis Group methods used in reviews.

ELECTRONIC SEARCHES

Study identification from electronic databases will include the following:

I) The Cochrane Central register of Controlled Trials (CENTRAL) searched in January 2009 The complete search strategy is listed in Appendix 1.

II) AgeLine (1978‐2009) searched in January 2009 The complete search strategy is listed in Appendix 2.

III) Applied Social Sciences Index and Abstracts (ASSIA) 1987‐2009 searched in January 2009 The complete search strategy is listed in Appendix 3.

IV) CINAHL (1982‐2009) searched in January 2009 The complete search strategy is listed in Appendix 4.

V) EMBASE (1980‐2009) searched in January 2009 The search strategy is based on Ccdan's search strategy listed in the CRG's specialized register. The complete search strategy is listed in Appendix 5.

VI) Medline (1950‐2009) searched in January 2009 The complete search strategy is listed in Appendix 6.

VII) PsycINFO (1887‐2009) searched in January 2009 The complete search strategy is listed in Appendix 7.

VIII) Sociological Abstracts (1952‐2009) searched in January 2009 The complete search strategy is listed in Appendix 8.

IX) Web of Science (ISI) (1900‐2009) searched in January 2009 The complete search strategy is listed in Appendix 9.

X) Social Services Abstracts (1979‐2009) searched in January 2009 The complete search strategy is listed in Appendix 10.

XI) OpenSIGLE (1980‐2009) searched in January 2009 The complete search strategy is listed in Appendix 11.

No language or data restrictions will be applied within the limitations of the search tools.

Searching other resources

Reference lists of all retrieved articles and included studies will be further checked for relevant material.

The journals the Gerontologist and Journal of the American Geriatrics Society will be manually searched for the years 2006‐2008.

Experts in the field, identified as first authors of the included studies, will be contacted and consulted.

Data collection and analysis

Selection of studies

The abstracts of the publications identified through the search strategy described above will be inspected by two persons, who individually will compose a list of potentially eligible studies. Based on these first drafts, the study selection will be discussed between the reviewers. In case of disagreements regarding specific studies, a third reviewer will be consulted for final decision.

Data extraction and management

Data will be extracted from each included intervention study according to the following model

Study design (i.e. randomisation, allocation concealment) Participants (i.e. number of participants, mean age, context and setting of the intervention) Intervention (i.e. intervention content and set‐up) Measured outcomes (primary/secondary) Results (i.e. the results of the conducted study according to the outcomes considered) Follow‐up Country of origin

Data will be extracted by the principal reviewer and 20% independently repeated by a second reviewer (if additional resources become available, all data extraction will be independently duplicated).

Assessment of risk of bias in included studies

The assessment of risk of bias in the included studies will be done according to the Cochrane Collaboration Handbook (Higgins 2008).

The following six domains will be considered when applying the Cochrane Collaboration 'risk of bias' tool: 1) Sequence generation: Was the allocation sequence adequately generated? 2) Allocation concealment: Was allocation adequately concealed? 3) Blinding of participants, personnel and outcome assessors for each main outcome or class of outcomes: Was knowledge of the allocated intervention adequately prevented during the study? 4) Incomplete outcome data for each main outcome or class of outcomes: Were incomplete outcome data adequately addressed? 5) Selective outcome reporting: Are reports of the study free of suggestion of selective outcome reporting? 6) Other sources of bias: Was the study apparently free of other problems that could put it at a high risk of bias? A short description of the reported procedure will be provided from each included study. Assessments on the risk of bias will be made for each domain within and across studies, based on the following three categories: A. Yes (low risk of bias) B. Unclear C. No (high risk of bias).

The principal author will assess the risk of bias in selected studies according to the six individual domains listed above. Summary assessments of the risk of bias will be performed for the outcomes according to the Cochrane Handbook recommendations (Higgins 2008) The entries will be independently checked by a second author. Any disagreement will be discussed between the reviewers. If necessary, the authors of the included studies will be contacted for further information.

Studies that will be assessed as having a high risk of bias (category C) according to the summary assessments performed, will not be included in the review.

Measures of treatment effect

Data will be entered into Revman 5 software by the principal reviewer. Intention‐to‐treat (ITT) data will be used when available.

Analysis of continuous data

For continuously distributed outcomes (e.g. depressive symptoms, positive mental health, quality of life, functional level), the weighted mean difference or standardised weighted mean difference (WMD or SMD) will be calculated as appropriate using a random effects model. Where ITT data is not available, end‐point data for trial completers will be used (WMD or SMD). The preferred continuous measure of effect is WMD. SMD will be calculated when outcomes have been measured using different rating scales. Substantially skewed data (where the standard deviation is more than twice the mean value) will not be entered in the formal meta‐analysis, but will be described in the text.

Analysis of dichotomous data

For binary efficacy outcomes (e.g. depression, acceptability), the Mantel‐Haenszel random effects model for calculating odds ratio (OR) will be applied.

Unit of analysis issues

If any multi‐arm studies are included in the meta‐analysis, the intervention condition groups will be collapsed into a single arm to be compared with the control condition group, or the control group will be equally divided into two.

Regarding any cluster‐randomised trials included, a summary measurement from each cluster will be used, conducting the analysis on the same level as the allocation. In the case of a large amount of cluster‐randomised trials being included, the direct estimate of the required effect measure will be calculated (e.g. OR and CI) using an analysis model appropriate for the cluster design.

Dealing with missing data

In order to perform an ITT analysis, the method of last observation carried forward (LOCF) will be used. This way of dealing with missing data introduces uncertainty about the reliability of the results and it will therefore be clearly stated in the review when LOCF data are used in the analysis.

If standard deviations are not available and cannot be calculated from available continuous data, the authors will be contacted. If no standard deviations can be found through any of these steps, the study will be excluded from the review.

In the case of missing dichotomous data it will assumed that the negative outcome was experienced at the end of the intervention (e.g. depression).

Assessment of heterogeneity

The impact of statistical heterogeneity on the meta‐analysis will be assessed by quantifying inconsistence among the studies with the I² test (Higgins 2008).

Assessment of reporting biases

The existence of publication bias will be examined graphically by the use of a funnel plot. If asymmetry is present, likely reasons will be explored. The limits of the funnel plots (i.e. that a funnel plot may not necessarily indicate publication bias and that publication bias does not always cause asymmetry in a funnel plot) will be taken into consideration.

Data synthesis

The Mantel‐Haenszel random effect model will be applied for the analysis.

Subgroup analysis and investigation of heterogeneity

Preventive intervention studies that have the aim of increasing the documented social capital will be separately assessed for their effectiveness. In order to investigate heterogeneity, sub‐group analyses will be performed, based on sufficient data. Trials that originate from the European Region will be considered as a sub‐group to be analysed separately, assessing possible differences between interventions performed and implemented in Europe and globally.

Sensitivity analysis

One sensitivity analysis that excludes trials where outcomes have not been assessed by masked investigators is planned. A sensitivity analysis will also be performed on the impact of imputation of missing data for both the dichotomous and continuous outcomes.

Appendices

Appendix 1. The Cochrane Central register of Controlled Trials (CENTRAL) search strategy

random$ control$ trial$.mp. randomi?ed trial$.mp. randomi?ed stud$.mp. control$ clinical trial$.mp. randomized controlled trials/ randomized controlled trial.pt. (random$ adj alloc$).ti,ab. (random$ adj assign$).ti,ab. "AGED, 80 AND OVER"/ or AGED/ (elderly or geriat$ or older$ or old age or late life or senil$).ti,ab. psychosocial.ti. (psychosocial$ adj intervent$).mp. (psychosocial$ adj program$).mp. (psychosocial$ adj support$).mp. (psychosocial$ adj treatment$).mp. (psychosocial$ adj educati$).mp. (psychosocial$ adj training).mp. social support.mp. depres$.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] or/1‐8 or/9‐10 or/11‐18 and/19‐22

Appendix 2. AgeLine search strategy

((kw=random* assign* or kw=randomized controlled trials or ti=random*) and ((de=old old or older adults or 65+ or 70+) or (kw=geriatr* or kw=elder* or kw=senile* or kw=late life or kw=old age or kw=Aged, 80‐And‐Over)) and (ab=depressi* or ab=mood or ab=mental status or ab=psychological well‐being or de=psychological well being or ti=psychosocial intervention*)) not (kw=antidepress* or kw=drug therapy or de=depression or de=anxiety disorders or de=anxiety)

Appendix 3. Applied Social Sciences Index and Abstracts (ASSIA) search strategy

((kw=random* within 1 assign* or kw=random* within 1 stud* or kw=random* within 1 trial*) or (kw=randomized controlled trial*)) and ((de=elderly women or de=elderly men or de=elderly people or de=very old) or (ab=elderly or kw=geriatr* or kw=senile* or kw=old age or kw=late life or kw=aged, 80‐and‐over)) and ((de=quality of life) or (de=life satisfaction or ab=depressi* or ab=mental health index or ab=satisfaction or ab=psychological function* or kw=psychosocial interventi* or ab=social participation or de= wellbeing or ab=well‐being or ab=wellbeing or ab=mood or ab=self‐esteem)) not ((de=antidepressant drugs or de=medical treatment or kw=drugs or kw=serotonin reuptake inhibitors) or (kw=antidepres*))

Appendix 4. CINAHL search strategy

exp Clinical Trials/ clinical trial.pt. exp Clinical Trials/ clinical trial.pt. (clinic$ adj trial$1).tw. ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$3 or mask$3)).tw. allocat$ random$.tw. random$ alloc$.tw. random$ control$ trial$.mp. randomi?ed trial$.mp. randomi?ed stud$.mp. control$ clinical trial$.mp. "AGED, 80 AND OVER"/ or AGED/ (elderly or geriatric or older adult$ or old age or late life or senil$).tw. Support, Psychosocial/ psychosocial support.tw. (psychosocial$ adj2 intervent$).mp. (psychosocial$ adj2 support$).mp. (psychosocial$ adj2 treatment$).mp. (psychosocial$ adj2 training$).mp. psychosocial$.mp. depres$.mp. [mp=title, subject heading word, abstract, instrumentation] or/1‐12 or/13‐14 or/15‐21 and/22‐25

Appendix 5. EMBASE search strategy

1 controlled study.de. 2 clinical trial.de. 3 major clinical study.de. 4 randomized controlled trial.de. 5 double blind procedure.de. 6 clinical article.de. 7 random$.mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 8 compar$.mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 9 control$.mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 10 follow up.mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 11 ((singl$ or doubl$ or tripl$ or trebl$) adj (blind$ or mask$ or dummy)).mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 12 placebo$.mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 13 (clinic$ adj (trial$ or stud$ or studies$)).mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name] 14 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 15 human.de. 16 DEPRESSI$.mp. 17 psychosocial$.mp. 18 psychosocial support.mp. 19 psychosocial$ adj (support or education or training or intervention$).mp. 20 17 or 18 or 19 21 14 and 15 and 16 and 20

Appendix 6. Medline search strategy

exp Clinical Trial/ randomized controlled trial/ clinical trial.pt. random$ control$ trial$.mp,pt. (clinic$ adj trial$1).mp. (random$ adj trial$).mp. (random$ adj allocat$).mp. randomi?ed stud$.mp. ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$3 or mask$3)).tw. double blind method/ Single‐Blind Method/ "AGED, 80 AND OVER"/ or AGED/ (elderly or geriatr$ or senil$ or older adult or old age or late life).ti,ab. (psychosocial$ adj intervent$).mp. psychosocial.ti. social support.mp. (psychosocial$ adj support$).mp. (psychosocial$ adj treatment$).mp. (psychosocial$ adj educati$).mp. (psychosocial$ adj training).mp. depres$.mp. [mp=title, original title, abstract, name of substance word, subject heading word] or/1‐11 or/12‐13 or/14‐20 and/21‐24

Appendix 7. PsycINFO search strategy

S1 SU drug therapy or SU pharmacology or SU serotonin reuptake inhibitors or SU electroconvulsive or TX antidepressive* or TX antidepressant* or SU drug* or KW medication* or KW medical patient* or KW medical therapy or KW medical treatment* S2 SU elderly or SU aging or SU elder care or SU geriatrics or SU geriatric patient* or AB older people or AB old people or AB older persons or AB elderly persons or AB old age or AB aged people or AB geriatr* OR AB late life or AB senil* or AB elderly population or AB older adult* or AB aging or AB aged, 80‐and‐over or TI aging or TI elderly OR TI geriatr* S3 AB random* stud* or AB RCT or AB random* controlled trial* or AB random* trial* or AB random* control* stud* or AB random* control* trial* or AB random* assign* or AB random* allocat* or TI random* or TI RCT S4 AB quality of life or AB life quality or AB depress* or AB mood or AB satisfaction or AB emotional well* or AB mental well* S5 S4 and S3 and S2 S6 S5 not S1

Appendix 8. Sociological Abstracts search strategy

((ti=random* or ab=randomized or kw=random* within 1 assign* or kw=random* within 2 stud* or kw=random* within 2 trial* or kw=random* control* trial* or kw= random* assign* or kw=random* allocat* or de=intervention*) and (de=elderly or de=aging or de=older people or de=elder care or ab=elderly or ab=older adults or ab=old old or ab=old age or ab=late life or ab=senil* or ab=geriatr* or ab=aged, 80‐and‐over or ab=late adulthood) and (de=social support or ab=quality of life or ab=mental health index or ab=geriatric depression scale or ab=social behavio* or ab=mental well* or ab=satisfaction or ab=psychological function* or ab=social participation or ab=social activity or ab=social relation* or ab=well‐being or ab=wellbeing or ab=mood or ab=self‐esteem)) not (de= medications or de=medical model or de=antidepress* or de=medical treatment)

Appendix 9. Web of Science (ISI) search strategy

1 TS=(random* OR trial* OR control* OR (singl* OR doubl* OR trebl* OR tripl*) AND (blind* OR mask*)) 2 TS=(randomized controlled trial* OR controlled clinical trial* OR clinical trial* OR (clinical AND trial) OR random* allocat* OR comparative stud* OR comparative trial*) 3 TS=(aged OR elder* OR geriatric OR older adult* OR senil* OR older OR old age OR late life) 4 TS=((psychosocial* AND intervent*) OR psychosocial* OR social support OR (psychosocial* AND support) OR (psychosocial* AND treatment*) OR (psychosocial* AND educati*) OR (psychosocial* AND training)) 5 TS=(depress*) 6 #1 OR #2 7 #1 AND #2 8 #6 AND #7 9 #8 AND #3 AND #4 AND #5

Appendix 10. Social Services Abstracts search strategy

((ti=random* or ab=randomized or kw=random* within 1 assign* or kw=random* within 2 stud* or kw=random* within 2 trial* or kw=random* control* trial* or kw= random* assign* or kw=random* allocat* or de=intervention*) and (de=elderly or de=aging or de=older people or de=elder care or ab=elderly or ab=older adults or ab=old old or ab=old age or ab=late life or ab=senil* or ab=geriatr* or ab=aged, 80‐and‐over or ab=late adulthood) and (de=social support or ab=quality of life or ab=mental health index or ab=geriatric depression scale or ab=social behavio* or ab=mental well* or ab=satisfaction or ab=psychological function* or ab=social participation or ab=social activity or ab=social relation* or ab=well‐being or ab=wellbeing or ab=mood or ab=self‐esteem)) not (de= medications or de=medical model or de=antidepress* or de=medical treatment)

Appendix 11. OpenSIGLE search strategy

The database of grey literature was searched using the following keywords in various combinations

(prevent*) and (depress*) and (elder* or older adults or geriatr* or aged) and (psychosocial intervention or psychosocial* or intervention)

or

(prevent*) and (depress*) and (elder* or older adults or geriatr* or aged)

or

(elder* or older adults or geriatr* or aged) and (psychosocial intervention or psychosocial* or intervention) and (depression)

or

(random* or RCT) and (elder* or older adults or geriatr* or aged) and (psychosocial intervention or psychosocial* or intervention)

What's new

Date	Event	Description
23 October 2018	Amended	Protocol for a Cochrane Review withdrawn from publication.

History

Protocol first published: Issue 2, 2009

Date	Event	Description
28 April 2009	Amended	Contact details amended to indicate first/contact author is employed at two institutions in different countries (Sweden and Finland)

Contributions of authors

Co‐reviewer Dr. Eva Jané‐Llopis (EJ) took the initiative to conduct this review.

The contact reviewer (AF) has written the protocol of the review with guidance from co‐reviewers Prof. Kristian Wahlbeck (KW), Dr. Eva Jané‐Llopis (EJ), and Dr. Isabell Schierenbeck (IS).

The contact reviewer will be primarily responsible for construction of the review. The co‐reviewers will assist with their involvement in study selection, data extraction and data analysis.

Sources of support

Internal sources

• THL National Institute for Health and Welfare (formerly, STAKES), Finland.

• Nordic School of Public Health, Sweden.

• Vaasa Hospital District, Finland.

• University of Helsinki, Finland.

External sources

• DataPrev, Other.

Declarations of interest

Contact reviewer (AF): None Co‐reviewer (KW): None Co‐reviewer (EJ): None Co‐reviewer (IS): None

Notes

The planned review outlined in this protocol has not been successfully converted into a full Cochrane Review within established timelines and for this reason has been withdrawn by the Common Mental Disorders CRG from the CDSR.

Withdrawn from publication for reasons stated in the review