The Effectiveness of Family Health Conversations Delivered by Nurses: A Systematic Review

ABSTRACT

Aim

To extract and interpret quantitative data exploring the effectiveness of family health conversations (FHCs) on family functioning, perceived support, health‐related quality of life, caregiver burden and family health in families living with critical or chronic health conditions.

Background

Addressing the health of families affected by critical or chronic illnesses requires focused attention. The effective integration of FHCs is hampered by a scarcity of rigorous quantitative studies that provide solid evidence on best practices and outcomes.

Design

A systematic review following the Joanna Briggs Institute guidelines.

Methods

The review is reported according to the PRISMA 2020 checklist. Appropriate studies were searched in PubMed, CINAHL, PsycINFO, Scopus and Cochrane Databases. Results of the search were imported into the Covidence web‐based program. Included were studies with a quantitative research design, delivered to families with critical or chronic health conditions, describing FHCs based on the Calgary Family Assessment Model and/or the Calgary Family Intervention Model, and/or the Illness Beliefs Model, using reliable and validated instruments, published between 2008 and 2023, and written in English.

Results

In total, 24 papers met the inclusion criteria. Sixteen papers used a quasi‐experimental design, eight of which included a control group. Two papers used a mixed methods design, and six papers were randomised controlled trials (RCTs). A statistically significant effect of FHCs on family functioning was reported in two RCTs and three quasi‐experimental papers. We also found that a statistically significant effect of FHCs was reported on perceived support in 9 of 15 papers, quality of life in 4 of 11 papers and caregiver burden in 1 of 3 papers.

Conclusion and Implications for Clinical Practice

The interventions reviewed revealed variability and partial results concerning the effectiveness of FHCs on family functioning. More rigorous research about short‐term, intermediate‐ and long‐term effectiveness is needed before conclusions can be drawn.

Reporting Method

The study is reported according to the PRISMA 2020 (Preferred Reporting Items for Systematic reviews and Meta‐Analyses) (File S1).

Patient or Public Contribution

No patient or public contribution. Data were gathered from previously published studies.

Summary.

• What does this paper contribute to the wider global clinical community?

  • ○
    The review presents a summary of what is currently known about the effectiveness of FHCs on family functioning, perceived support, health‐related quality of life, caregiver burden and family health.
  • ○
    The literature shows variability and partial results concerning the effectiveness of FHCs on family functioning, perceived support, health‐related quality of life, caregiver burden and family health.
  • ○
    A deeper understanding of how FHCs improve the health and functioning of individuals and their families is required to further integrate FNCs into clinical practice.

1. Introduction

Globally, there is a growing recognition of the vital role that family care plays in monitoring, treating and managing health conditions, as well as in promoting patient safety (Meiers et al. 2020). This has led to the development of a substantial body of knowledge about the support of families dealing with illness in their family (Thürlimann et al. 2022; Wang et al. 2024; Østergaard et al. 2023). However, healthcare systems across the globe encounter difficulties in effectively integrating theoretical and evidence‐based knowledge to promote family health (Shajani and Snell 2023). The implementation of family systems nursing into clinical practice is lacking due to knowledge and contextual barriers (Thürlimann et al. 2022; Østergaard et al. 2023).

Within the family system nursing framework, communication with families through structured family conversations is considered a key intervention positively impacting families affected by chronic (Broekema et al. 2021) and/or critical illness (Kiwanuka et al. 2022). The systematic scoping review by Azcárate‐Cenoz et al. (2024) provided a broad overview of studies on Family Health Conversations (FHCs) aiming to describe the characteristics, feasibility and effectiveness of FHCs. Regarding effectiveness, the study highlighted family functioning and perceived support as the most significantly impacted outcomes. Recognising family functioning as the core outcome of FHCs within the family systems nursing framework, the present systematic review builds on the previous scoping review by systematically evaluating the available quantitative evidence for the effectiveness of FHCs.

2. Background

Within the nursing profession, therapeutic conversations—also known as FHCs (Benzein et al. 2008) or strength‐oriented therapeutic conversations (Petursdottir et al. 2021; Svavarsdottir et al. 2025)—have been based on the Calgary Family Assessment Model (CFAM), the Calgary Family Intervention Model (CFIM) and the Illness Belief Model (IBM). In this systematic review, we use the term FHCs as a common name for conversations built on the CFAM, CFIM and the IBM.

The CFAM is an integrated multidimensional framework used for assessing families and guiding interventions aimed at resolving family health‐related problems or issues, consequently maintaining or improving family functioning and patient outcomes by guiding healthcare professionals in their interventions (Mileski et al. 2022). The model consists of three major categories: (a) structural—for example, who belongs to the family, (b) developmental—for example, what developmental state the family is in and (c) functional—for example, how the family is functioning. At each point, it is important for nurses to decide which categories are most relevant and appropriate to explore and assess with each family.

The CFIM follows the CFAM (Shajani and Snell 2023) and the IBM (Wright and Bell 2021). The CFIM is a strength‐based model in which family interventions are identified and implemented, with the aim of promoting, improving and/or maintaining effective family functioning. FHCs interventions are based on the above family system models and are offered to families within relational and a collaborative context, such as through inviting, asking, encouraging and/or supporting families. In the Illness Beliefs model (Wright and Bell 2021) beliefs are defined as phenomena that develop over the life cycle through education, relationships and experiences and can change in response to new situations. In that way, illness beliefs become a template for how families respond to the illness and to each other. Through FHCs, illness beliefs are explored to determine whether they are helpful or not, to soften illness suffering and to invite healing and improved functioning.

These family system models serve as an organising framework that conceptualises how a specific FHC intervention targeting cognitive (e.g., evidence‐based information), affective (e.g., support), or behavioural (e.g., behavioural change) domains of family functioning is implemented by healthcare professionals. Skilled nurses play a crucial role in fostering family functioning through effective family conversations. FHC are a key component of the CFIM and the IBM. It emphasises respectful, strengths‐based dialogue, focusing on and acknowledging the family's experience and beliefs about illness. To ensure their effectiveness, FHC must be closely aligned with the key domains of family functioning identified by both nurses and families as areas of concern, thereby establishing the foundation and central focus of the intervention.

This review considers family functioning the primary outcome affected by FHCs. Family functioning pertains to the operational dynamics of a family unit and the interactions among its members (Shajani and Snell 2023). This concept encompasses multiple facets of family life, such as communication patterns, roles, relationships, decision‐making processes and overall dynamics. A thorough understanding of family functioning is crucial because it plays a pivotal role in the well‐being and development of individual family members (Ahlberg et al. 2023). Family functioning is not only a reflection of a functional unit but also contributes significantly to the overall growth and development of its members.

Family functioning as the primary outcome is closely interconnected to other concepts (secondary outcomes), such as perceived support, health‐related quality of life, caregiver burden, and family health. Family functioning was found to be a strong factor related to family health for both patients and their family members (Shamali et al. 2021). In addition, families who perceive more support from nurses are prone to have a high level of family health and enhanced family functioning (Shamali et al. 2019). On the other hand, weak family functioning and greater differentiation of perceived family functioning within dyads have been associated with higher levels of depressive symptoms among patients with acute ischaemic stroke (Li et al. 2023) and a significantly lower health‐related quality of life (HRQoL) in stroke survivors (Li et al. 2024). Lastly, individuals engaged in caregiving roles have increased tasks and responsibilities, inducing elevated levels of burden in the caregiving process (Ploeg et al. 2020; Rezaei et al. 2020). In a study among 5034 caregivers across 23 countries, the pooled prevalence of caregiver burden was estimated to be 32% among caregivers providing care to persons with mental illness. Notably, the prevalence was highest (36%) among caregivers of patients undergoing hospital treatment (Cham et al. 2022). In a recent meta‐analysis, the pooled prevalence of depression among caregivers of cancer patients was 42%, with a higher rate of depression among female caregivers (58%) when compared to their male counterparts (34%) (Bedaso et al. 2022). Caregiver burden is often overlooked by healthcare professionals, who typically emphasise how relatives can assist the patient rather than considering the support that relatives themselves may require from healthcare professionals (Allen et al. 2022). Furthermore, caregivers undergo significant life changes, bearing the weight of numerous responsibilities, overwhelming emotions, exhaustion, heightened stress and a lack of personal time (Ploeg et al. 2020).

Family intervention research, where the content of the intervention has been uncovered, has focused on the importance of family interactions. Several reviews have evidenced that family interventions in general can enhance the health and well‐being of families (Kiwanuka et al. 2022; Smith et al. 2020; Thürlimann et al. 2022; Wang et al. 2024; Østergaard et al. 2023). FHCs as a specific type of family intervention have been studied across a variety of clinical settings including hospital care (Svavarsdottir et al. 2020), critical care (Ahlberg et al. 2021) primary care (Broekema et al. 2021) and in families throughout the lifespan (Dorell et al. 2017; Sigurdardottir et al. 2013) dealing with diverse health conditions such as critical illness (Kynoch et al. 2016), cancer (Coyne et al. 2020) chronic illness (Broekema et al. 2021; Mahrer‐Imhof et al. 2022) and mental health disorders (Kim and Park 2023). Within the current systematic review, the focus was on adult caregivers; however, we did not target specific patient populations, nor specific age groups, disease categories, or health domains.

Recently, a broad scoping review mapped the characteristics, feasibility and effectiveness of nurse‐led family systems conversations (Azcárate‐Cenoz et al. 2024). This review included 37 studies with both qualitative and quantitative research designs. The interventions were based on different theoretical frameworks and were found to vary in the content and duration of the sessions offered. Most of the interventions demonstrated benefits for the families regarding their family functioning and perceived support. Generally, when mixing qualitative and quantitative research designs, the interventions demonstrated a tendency towards improved family support and family functioning as well as positive adaptation strategies within the families. However, existing research is limited by small sample sizes, the absence of rigorous experimental designs such as randomised controlled trials (RCTs) and a lack of long‐term outcome assessments.

The present systematic review exclusively examines quantitative research assessing the effectiveness of FHCs on family functioning as the primary outcome, since this is considered the most important concept to be impacted by FHCs. Additionally, secondary outcomes in terms of perceived support, health‐related quality of life, caregiver burden and family health were considered. Conducting a review including merely quantitative data, alongside existing reviews, can yield valuable insights by enhancing generalisability and offering practical implications for policy and practice.

2.1. Aim of the Study

The overall aim of the systematic review was to extract and interpret quantitative data from published studies exploring the effectiveness of FHCs in families confronted with critical or chronic health conditions. The aim was converted into a primary and secondary research question.

Primary research question:

1. Are FHCs, compared to standard care or pre‐intervention care, effective in improving family functioning in families confronted with critical or chronic health conditions?

Secondary research question:

• 2Are FHCs, compared to standard care or pre‐intervention care, effective in improving perceived support, health‐related quality of life, caregiver burden and family health in families confronted with critical or chronic health conditions?

3. Method

3.1. Design and Protocol Registration

This systematic review was performed using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta‐Analysis) (Page et al. 2021) guidelines and reported using the PRISMA 2020 checklist (McGowan et al. 2020) (File S1). The protocol for the study was registered in PROSPERO with the registration number CRD42022333887 (https://www.crd.york.ac.uk/PROSPERO/view/CRD42022333887).

3.2. Search Strategy

The search for relevant studies for incorporation in the review was built based on the PICO (Population, Intervention, Comparison, Outcomes) components (Higgins et al. 2023) (see Table 1).

TABLE 1.

PICO strategy used in the systematic review.

Population	Patients of all ages diagnosed with a critical or chronic disease receiving nursing care services involving family members Care providers (licensed or unlicensed) and caregivers [family/friends (paid or unpaid)] involved in the process of care of a family member
Intervention	Family health conversations based on the Calgary Family Assessment Model, and/or the Calgary Family Intervention Model, and/or the Illness Beliefs Model
Comparison	Non‐exposed control group or another intervention
Outcome	Family functioning measured by validated and reliable measurements Perceived support, health‐related quality of life, caregiver burden, and family health measured by validated and reliable measurements

Abbreviation: PCC, population, concept, context.

We searched PubMed, CINAHL, PsycINFO, Scopus and Cochrane databases with the support of a research librarian who assisted in developing the search strategies. The following search strategy was used for PubMed: (“Family”[Mesh] OR famil*[tiab]).

AND (Conversat*[tiab] OR sotc[tiab] OR Calgary model*[tiab] OR Calgary family assessment*[tiab] OR Calgary family intervention*[tiab]) (search strategies for other databases are presented in Table 2). Selected terms were searched within the title, abstract, keywords and subject headings (e.g., MeSH) as appropriate. The Booleans “OR” and “AND” were used to combine search terms and develop the search strategy. The search was restricted to studies reported in English and published from 1 January 2008 and forward (Table 2). After completing the search, the results were imported into the Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia. Available at www.covidence.org. Duplicates were automatically removed.

TABLE 2.

Search strategies.

Databases	Records (n)
PubMed
“Family” OR famil* AND (Conversat* OR sotc OR Calgary model* OR Calgary family assessment* OR Calgary family intervention*)	n = 1922
CINAHL
“Family Nursing” OR nurs* OR AB nurs* AND (Conversat* OR sotc OR Calgary model* OR Calgary family assessment* OR Calgary family intervention* OR AB Conversat* OR sotc OR Calgary model* OR Calgary family assessment* OR Calgary family intervention*)	n = 2651
PsycINFO
TI nurs* OR AB nurs* AND (TI Conversat* OR sotc OR “Calgary model*” OR “Calgary family assessment*” OR “Calgary family intervention*” OR AB Conversat* OR sotc OR “Calgary model*” OR “Calgary family assessment*” OR “Calgary family intervention*”)	n = 1201
Scopus
TITLE‐ABS‐KEY (Nurs*) AND TITLE‐ABS‐KEY (Conversat* OR sotc OR “Calgary model*” OR “Calgary family assessment*” OR “Calgary family intervention*”) AND PUBYEAR AFT 2008	n = 2729
Cochrane
Nurs* AND (Conversat* OR sotc OR “Calgary model*” OR “Calgary family assessment*” OR “Calgary family intervention*”)	n = 450

Note: All searches were limited to articles written in English and published from 2008.

3.3. Inclusion and Exclusion Criteria

Included studies were papers (1) that delivered FHCs based on CFAM and/or the CFIM (Shajani and Snell 2023) and/or the Illness Beliefs Model (IBM) (Bell and Wright 2015); (2) in which FHCs were delivered to families with critical or chronic health conditions; (3) in which FHCs were delivered to any kind of patient‐family population with no restriction on age or health domain; (4) written in English and (5) featuring a quantitative research design (randomised controlled trials (RCTs) or quasi‐experimental) (in the case of a mixed‐method study, only quantitative data were used); (6) using reliable and validated instruments to measure family functioning and (7) that were published between 2008 and 2023.

We excluded theoretical papers, case control studies, prospective or retrospective cohort studies, qualitative studies and those in which the intervention was not based on the Calgary models.

3.4. Study Selection

After importing all identified citations into Covidence, the title and abstract of papers retrieved from the search strategy were independently screened in two rounds by two randomly selected team members. Subsequently, the full text of potentially eligible papers was independently assessed by another two randomly selected reviewers from the research team. Covidence provided all random allocations and kept records of who voted to ensure equal participation. Covidence also captured any disagreement and reasons for including/excluding the papers. Throughout the review, it was an integrative process where the research team regularly met online to discuss and resolve disagreements and decide whether to exclude or include the paper in the next step. Finally, a hand search of relevant references was performed to identify additional papers. The process and findings were validated through presentations and discussions within the research team.

3.5. Methodological Quality Assessment

Quality assessment was conducted on all included studies using the Joanna Briggs Institute (JBI) Critical Appraisal Checklists appropriate for the different quantitative study designs [https://jbi.global/critical‐appraisal‐tools]. The Mixed Method Appraisal Tool (MMAT) version 2018 was used for the mixed‐method studies (Hong et al. 2018). Quality assessment was performed independently by two random members of the research team. No cut‐off points are recommended in the user guides. However, in this review, we applied a 50% threshold in the quality assessment to determine study inclusion. This decision was informed by methodological precedents in the literature, which demonstrated the practical application of a 50% cut‐off using structured appraisal tools such as the JBI checklists (Glasgow et al. 2020). In question 3 (were treatment groups similar at baseline) and question 6 (were outcome assessors blind to treatment assignment) in the JBI quasi‐experimental studies, assessment was unclear in most of the studies. This might be a source of bias; however, they still met the inclusion threshold. Therefore, no studies were excluded based on the quality assessment. The results of the quality assessment are presented in Table 3.

TABLE 3.

Quality assessment of the 24 included studies.

	Studies	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9	Q10	Q11	Q12	Q13
	JBL checklist for critical appraisal of RCTs a
1	Østergaard et al. 2018	y	y	y	na	na	y	y	y	y	y	y	y	y
2	Ågren et al. 2019	y	u	y	na	na	u	y	n	u	y	y	y	y
3	Østergaard et al. 2021	y	y	y	na	na	y	y	y	y	y	y	y	y
4	Rosenstrøm et al. 2023	y	y	y	na	na	u	y	n	y	y	y	y	y
5	Petersen et al. 2022	y	y	y	na	na	y	y	n	u	y	y	y	y
6	Lee et al. 2018	y	y	y	na	na	u	y	n	u	y	y	y	y
	JBL checklist for critical appraisal of quasi‐experimental studies b
7	Halldórsdóttir and Svavarsdóttir 2012	y	y	u	y	y	na	y	y	y
8	Svavarsdottir et al. 2012	y	y	u	n	y	na	y	y	y
9	Sigurdardottir et al. 2013	y	y	u	n	y	na	y	y	y
10	Kamban and Svavarsdottir 2013	y	y	u	y	y	n	y	y	y
11	Svavarsdottir and Sigurdardottir 2013	y	y	u	n	y	n	y	y	y
12	Konradsdottir and Svavarsdottir 2013	y	y	u	n	y	n	y	y	y
13	Sveinbjarnardottir et al. 2013	y	y	u	y	y	na	y	y	y
14	Ragnarsdóttir and Svavarsdottir 2014	y	y	u	y	y	n	y	y	y
15	Svavarsdottir et al. 2014	y	y	u	n	y	na	y	y	y
16	Lämås et al. 2016	y	y	y	y	y	n	y	y	y
17	Gisladottir and Svavarsdottir 2017	y	y	u	n	y	n	y	y	y
18	Petursdottir and Svavarsdottir 2019	y	y	u	n	y	na	y	y	y
19	Svavarsdottir and Gisladottir 2019	y	y	u	n	y	na	y	y	y
20	Faarup et al. 2019	y	u	y	y	y	n	y	y	y
21	Svavarsdottir et al. 2020	y	y	u	n	y	na	y	y	y
22	Broekema et al. 2021	y	y	y	y	y	n	y	y	y
	JBL checklist for mixed methods studies c
23	Dorell et al. 2017	n	y	y	y	y
24	Ahlberg et al. 2021	y	y	y	n	y

Abbreviations: n, no; na, not applicable; Q, question; RCT, randomised controlled trial; u, unclear; y, yes.

^a Q1, was true randomisation used for the assignment of participants to treatment groups? Q2, was allocation to treatment groups concealed? Q3, were treatment groups similar at baseline? Q4, were participants blind to treatment assignment? Q5, were those delivering treatment blind to treatment assignment? Q6, were outcome assessors blind to treatment assignment? Q7, were treatment groups treated identically except for the intervention of interest? Q8, was follow‐up complete, and if not, were differences between groups in terms of their follow‐up adequately described and analysed? Q9, were participants analysed in the groups to which they were randomised? Q10, were outcomes measured in the same way for all treatment groups? Q11, were outcomes measured in a reliable way? Q12, was appropriate statistical analysis used? Q13, was the trial design appropriate, and were any deviations from the standard RCT design (individual randomisation, parallel groups) accounted for in the conduct and analysis of the trial?

^b Q1 is it clear in the study what is the ‘cause’ and what is the ‘effect’ (i.e., there is no confusion about which variable comes first)? Q2, were the participants included in any comparisons similar? Q3, were the participants included in any comparisons receiving similar treatment/care, other than the exposure or intervention of interest? Q4, was there a control group? Q5, were there multiple measurements of the outcome both pre and post the intervention/exposure? Q6, was follow up complete and if not, were differences between groups in terms of their follow up adequately described and analysed? Q7, were the outcomes of participants included in any comparisons measured in the same way? Q8, were outcomes measured in a reliable way? Q9, was appropriate statistical analysis used?

^c Q1, is there an adequate rationale for using a mixed methods design to address the research question? Q2, are the different components of the study effectively integrated to answer the research question? Q3, are the outputs of the integration of qualitative and quantitative components adequately interpreted? Q4, are divergences and inconsistencies between quantitative and qualitative results adequately addressed? Q5, do the different components of the study adhere to the quality criteria of each tradition of the methods involved?

3.6. Data Extraction and Summary/Synthesis

A pre‐specified Excel form was used to extract data for the synthesis of the evidence. The extracted information included authors, publication year, country, aim of the study, research design, participants' characteristics, description of the intervention delivery (and alternative in comparator group), target population/healthcare context, outcome measures and instruments used, risk of bias, results, or effect measures.

There was a limited scope for meta‐analysis because of the small number of existing trials and because of differences in intervention delivery and the range of different outcome measures used. Therefore, a narrative synthesis (Ryan 2013) of the findings was obtained from the included papers, structured around the aim of the study, type and dose of intervention delivery, characteristics of the participants, target population and type of outcome measure. This process was guided by recommendations of the PAGER (Patterns, Advances, Gaps, Evidence for practice and Research recommendations) framework for improving the quality of reporting of scoping reviews (Bradbury‐Jones et al. 2022) (Table 4).

TABLE 4.

PAGER framework.

Author, year	Family functioning				Perceived support				Health‐related quality of life				Caregiver burden				Family health
	D	P	Po	O	D	P	Po	O	D	P	Po	O	D	P	Po	O	D	P	Po	O
Ahlberg et al. 2021	x	x	x	x					x	x	x	x
Broekema et al. 2021	x	x	x	x					x	x	x	x	x	x	x	x
Dorell et al. 2017									x	x	x	x
Faarup et al. 2019	x	x	x	x					x	x	x	x
Gisladottir and Svavarsdottir 2017	x	x	x	x	x	x	x	x
Halldórsdóttir and Svavarsdóttir 2012					x	x	x	x
Kamban and Svavarsdottir 2013	x	x	x	x	x	x	x	x
Konradsdottir and Svavarsdottir 2013					x	x	x	x
Lee et al. 2018	x	x	x	x									x	x	x	x	x	x	x	x
Lämås et al. 2016									x	x	x	x
Petersen et al. 2022					x	x	x	x
Petursdottir and Svavarsdottir 2019					x	x	x	x					x	x	x	x
Ragnarsdóttir and Svavarsdottir 2014					x	x	x	x
Rosenstrøm et al. 2023	x	x	x	x	x	x	x	x	x	x	x	x
Sigurdardottir et al. 2013					x	x	x	x	x	x	x	x
Svavarsdottir et al. 2012	x	x	x	x	x	x	x	x
Svavarsdottir and Sigurdardottir 2013	x	x	x	x	x	x	x	x
Svavarsdottir et al. 2014	x	x	x	x	x	x	x	x	x	x	x	x
Svavarsdottir et al. 2020					x	x	x	x	x	x	x	x
Sveinbjarnardottir et al. 2013	x	x	x	x	x	x	x	x
Østergaard et al. 2018 and Østergaard et al. 2021	x	x	x	x	x	x	x	x	x	x	x	x					x	x	x	x
Ågren et al. 2019	x	x	x	x					x	x	x	x

Abbreviations: D, dose of intervention; O, outcomes; P, participants; Po, population.

3.7. Ethical Considerations

Only secondary data not involving direct data collection with human beings were used in this systematic review. Thus, ethical approval was not needed.

4. Results

4.1. Study Selection

The initial search yielded 8046 citations, including two papers from a hand search and an updated search from June 2023 to December 2023 which yielded another 999 citations. Thus, 9046 citations were imported into Covidence for screening. After removing 4350 duplicates, 4696 citations remained. Title and abstracts were screened, and 4635 citations were found irrelevant, leaving 61 papers for full‐text review. Thirty‐seven papers were excluded due to a qualitative design (n = 18), descriptive form (n = 11), none of the predefined outcomes were included (n = 6), results of intervention not described (n = 1) and not CFAM/CFIM/IBM‐based (n = 1), leaving a total of 24 papers to be included in the review (Figure 1).

FIGURE 1.

Covidence PRISMA flow chart. CFAM, Calgary Family Assessment Model; CFIM, Calgary Family Intervention Model; IBM, Illness Beliefs Model. [Colour figure can be viewed at wileyonlinelibrary.com]

4.2. Participants and General Study Characteristics

In all, 1368 patients and 1103 family members were recorded in the 24 papers. Additionally, four papers (Ahlberg et al. 2021; Dorell et al. 2017; Lämås et al. 2016; Ågren et al. 2019) only reported the number of participants at family level, including the patient (101 participants). More than half of the patients and family members (58%, n = 1422) came from three of the papers (Sveinbjarnardottir et al. 2013; Østergaard et al. 2021, 2018), all other papers having small sample sizes, ranging from 112 patients and family members (Broekema et al. 2021) to 19 caregivers (Svavarsdottir and Sigurdardottir 2013). Patients and family members were recruited from hospitals (n = 16), outpatient clinics (n = 4), a rehabilitation clinic (n = 1), home care (n = 1), a residential home (n = 1) and a community centre (n = 1). Reported ages ranged from 0 to 84 years for patients, and 19 to 84 years for family members (see Tables 5 and 6).

TABLE 5.

Charting table on primary outcome: Family functioning.

Author(s), publication year, country	Study aim	Design	Healthcare context, target population, participants' characteristics	Intervention	Alternative intervention	Instruments used	Results
Ahlberg et al. 2021, Sweden	To compare responses from 12 different types of follow‐up interventions for families of critically ill persons, focusing on individual hopes, HRQOL, family functioning, and ability to cope with challenges	Mixed methods including quasi‐experimental, controlled pre‐post‐test comparison	ICU patients and family members' post‐discharge IG (FHC); 7 patients and 10 family members (n = 17) CG (SGC): 21 family members Mean age; 53 years in IG and 59 years in CG	Three FHCs post‐discharge + closing letter after final session provided by 4 trained ICU nurses, duration not reported	Support Group Conversations (SGC) 1½ hour, 2 months after ICU discharge On demand a second SGC	FAD‐GF	Significant improvement of family functioning (total score) for patients and family members together at 3 months post intervention in (mean IG 1.5 vs. CG 2.8; p = 0.001) for family members
Broekema et al. 2021, Netherlands	To assess the effects of Family Nursing Conversations of family caregiver burden, family functioning, patient's quality of life and the amount of professional home care	Quasi‐experimental Controlled pre‐post‐test comparison	Patients and family members receiving homecare IG: 21 patients (mean age 80 years) and 34 family members (mean age 68 years) CG: 30 patients (mean age 79 years) and 27 family members (mean age 66 years)	Two FHCs over a period of max 3 months provided by 9 trained (6‐day education) homecare nurses, duration not reported	Standard care	FAD‐GF	Significant improvement of family functioning (total score) (IG vs. CG) at 6 months post intervention patients: (mean IG 3.46 vs. CG 3.18; p = 0.001) for family members and (mean IG 3.58 vs. CG 3.04; p = 0.004) for patients
Faarup et al. 2019, Denmark	To assess the impact of FHCs as a supplement to conventional care on HRQOL, family functioning and family hardiness among patients with glioblastoma multiforme and their family members	Quasi‐experimental Two group pre‐post‐test comparison	Patients receiving in‐hospital care at a neurosurgical ward/outpatient clinic for glioblastoma 67 patients (mean age 67 years) Pre‐test CG patients (n = 40) Post‐test IG patients (n = 20)	One to three FHCs provided by two trained project nurses (not involved in care) in the outpatient clinic or at home, duration 60–90 min	Standard care	FFSS	No significant differences in family functioning (total score) 14 weeks post intervention for neither patients nor family members
Gisladottir and Svavarsdottir 2017, Iceland	To evaluate the effectiveness of TCI in caregivers of adolescents with ADHD regarding strengthening the supportive role	Quasi‐experimental Single group, pre‐post‐test comparison	Parents of children with ADHD 60 parents (37 mothers, 21 fathers, 1 unknown) of adolescents aged 13–17 years	TCI was part of a support programme with – Group sessions (2) – FHC (private) (1) – Booster session (1) Provided in a 2‐month period by 2 study nurses from the research team (master's and PhD degree)	NA	ICE‐EFFQ	No significant differences in family functioning (total score) at 1 or 2 months post intervention
Kamban and Svavarsdottir 2013, Iceland	To measure the benefit of a short‐family therapeutic conversation intervention in an acute paediatric unit	Quasi‐experimental Controlled pre‐post‐test comparison	Parents of children with bronchiolitis caused by RSV IG: n = 21 (age 19–> 31 years/14 female) CG: n = 20 (age 19–> 31 years/11 female)	One FHC in the hospital setting by a trained study nurse duration 15–50 min (average 31 min)	Standard care	ICE‐EFFQ	Significant decline in family functioning (total score) at 11 days post intervention in mothers (mean IG 4.0 vs. CG 4.1; p = 0.007)
Lee et al. 2018, Taiwan	To measure the effects of a brief family‐centred care program for hospitalised patients with bipolar disorder and their family caregivers	RCT	Hospitalised patients with bipolar disorders and their caregivers IG: n = 36 (age 21 to ≥ 66 years) CG: n = 36 (age 21 to ≥ 66 years)	Four sessions (in 2 weeks) during hospitalisation provided by an experienced psychiatric nurse (master's degree in psychiatric nursing), duration 90 min	Standard care	FFS	Significant effects in family function (total score) after the intervention was completed (mean IG 150.19 vs. CG 134.42; p = 0.03)
Svavarsdottir et al. 2012, Iceland	To evaluate the effectiveness of a short‐term therapeutic conversation intervention with families who were receiving health care services related to childhood/adolescent acute and chronic illnesses	Quasi‐experimental Controlled pre‐post‐test comparison	Parents of children with chronic or acute illness in the hospital setting IG: 41 parents (37 mothers and 4 fathers) CG: 35 parents (30 mother and 5 fathers)	One FHC by a study nurse (master's degree), within the hospital setting (in‐patient, out‐patient, acute care, and daycare), duration 15–50 min (average 30 min)	Standard care	ICE‐EFFQ	Total score not reported
Sveinbjarnardottir et al. 2013, Iceland	To evaluate the effectiveness of implementing a therapeutic conversation in acute inpatient psychiatry with families, by evaluating family perceived support, expressive family function and general well‐being	Quasi‐experimental Controlled pre‐post‐test comparison	Patients with psychiatric illness in inpatient psychiatry IG: 68 patients and 69 family members CG: 74 patients and 74 family members	Two to five FHCs by 12 trained (1‐day education) clinical nurses, duration 30–60 min	Standard care	ICE‐EFFQ	No significant differences in family functioning (total score) at 42 days post intervention for neither patients nor family members
Svavarsdottir and Sigurdardottir 2013, Iceland	To test the effectiveness of a two‐to‐three session FAM‐TCI for primary and partner caregivers of children and adolescents in active cancer treatment on perceived family support and expressive family functioning	Quasi‐experimental Single group, pre‐post‐test comparison	Parents of children in active cancer treatment (in‐hospital and day treatment) 19 parents (age 31–60 years) of children and adolescents (age 0–18 years) Primary caregivers (n = 10/9 female) and partner caregivers (n = 9/9 male)	Two to three FHCs within a period of 10–12 weeks by a trained study nurse (master's degree) in collaboration with a paediatric nurse, duration 45–90 min (average 60 min)	NA	ICE‐EFFQ	Significant improvement of family functioning (total score) at 1 week post intervention for primary caregivers (mean 37.5 vs. 48.2; p = 0.021)
Svavarsdottir et al. 2014, Iceland	To evaluate the benefits of a two‐session FAM‐TCI for families of children diagnoses with asthma, cancer or diabetes	Quasi‐experimental Single group, pre‐post‐test comparison (secondary analysis of pooled data)	Children with chronic illnesses in the hospital setting 35 mothers and 25 fathers (age 31–50 years) of children (age 0–18 years)	Two FHCs within a period of 10–12 weeks by a trained study (ANP) nurse (master's degree), duration 45–90 min (average 60 min)	NA	ICE‐EFFQ	No significant differences in family functioning (total score) at 5–7 days post intervention
Rosenstrøm et al. 2023, Denmark	To evaluate a family‐focused intervention for patients with AF in addition to conventional care and to establish its effect on HRQOL, anxiety, depression, AF symptoms, family functioning, and family support	RCT	Patients newly diagnosed with AF. IG: 33 patients (mean age 68 years/22 males) CG: 35 patients (mean age 68 years/22 males)	One to three FHCs +2‐h group education within a 2‐month period by three trained AF specialised clinical nurses	Standard care	ICE‐EFFQ	No significant differences in family functioning (total score) at 6 months post intervention
Østergaard et al. 2021, Denmark	To evaluate the effect of family nursing therapeutic conversations for patients with CHF and their family members on HRQOL, family functioning, family health and social support	RCT multicentre	Patients with CHF and their family members receiving care from three HF‐clinics. IG: patients (n = 232; mean age 67 years) and family members (n = 165; mean age 59 years) CG: patients (n = 236; mean age 67 years) and family members (n = 160; mean age 64 years)	Three FHCs during an interval of 6–12 weeks by nine trained CHF specialised clinical nurses	Standard care	FAFHES family functioning scale	No significant differences in family functioning (total score) at 3, 6, and 12 months post intervention
Ågren et al. 2019, Sweden	To investigate outcomes of a pilot RCT, nurse led Health Promoting Conversations Intervention	Pilot RCT	ICU patients and family members' post discharge 17 families; 10 IG (n = 28) and 7 CG (n = 17) Mean age; 61 years in IG and 60 years in CG	Three FHCs 4–8 weeks after discharge with 2‐week interval + closing letter after final session provided by trained research nurses (not involved in care for the patient), duration 2 h (first) + 30–40 min	Standard care	FAD‐GF	Significant improvement of family functioning (total score for patients and family members together) at 12 months post intervention in favour of IG (mean IG 1.5 vs. CG 1.8; p = 0.03)

Abbreviations: ADHD, Attention Deficit Hyperactivity Disorder; AF, Atrial Fibrillation; ANP, Advanced Nursing Practitioners; CG, Control Group; CHF, Congestive Heart Failure; FAD‐GF, Family Assessment Device‐General Functioning; FAFHES, Family Functioning, Family Health and Social Support questionnaire; FAM‐TCI, Family Therapeutic Conversation Intervention; FFS, Family Function Scale; FFSS, Family Function Style Scale; FHCs, Family Health Conversations; HF, Heart Failure; HRQOL, Health‐Related Quality of Life; ICE‐EFFQ, Iceland Expressive Family Functioning Questionnaire; ICU, Intensive Care Unit; IG, Intervention Group; NA, Not Applicable; RCT, Randomised Controlled Trial; RSV, Respiratory Syncytial Virus; TCI, Therapeutic Conversation Intervention.

TABLE 6.

Charting table on secondary outcomes: Perceived support, health‐related quality of life, caregiver burden, and family health.

Author(s), year of publication, country	Design	Healthcare context, target population, participants' characteristics	Intervention	Alternative intervention	Instruments used	Results
Perceived support
Gisladottir and Svavarsdottir 2017, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison	Parents of children with ADHD 60 parents (37 mothers, 21 fathers, 1 unknown) of adolescents aged 13–17 years	FHCs was part of a support programme with – Group sessions (2) – FHC (private) (1) – Booster session (1) Provided by study nurses from the research team (master's and PhD degree)	NA	ICE‐FPSQ	No significant improvement in perceived support
Halldórsdóttir and Svavarsdóttir 2012, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Patients with COPD admitted to respiratory hospital ward IG: 15 family members CG: 15 family members	One FHC during hospital stay by a trained study nurse (researcher), duration 15–27 min (average 23 min)	Standard care	ICE‐FPSQ	Significant improvement in perceived support (total score) in favour of the IG for family members: IG (M = 3.31, SD = 1.030), CG (M = 2.61, SD = 1.140, p = 0.045) after receiving either FHC or standard care
Kamban and Svavarsdottir 2013, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Parents of children with bronchiolitis caused by RSV IG: n = 21 (age 19 to > 31 years/14 female) CG: n = 20 (age 19 to > 31 years/11 female)	One FHC in the hospital setting by a trained study nurse duration 15–50 min (average 31 min)	Standard care	ICE‐FPSQ	Total score perceived support not reported
Konradsdottir and Svavarsdottir 2013, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison	Parents of children receiving care in an outpatient paediatric diabetic clinic 14 families/24 parents (13 mothers and 11 fathers)	One FHC in the hospital setting by a trained advanced practice nurse (master's degree) duration 15–50 min (average 31 min)	NA	ICE‐FPSQ	No significant improvement in perceived support
Petersen et al. 2022, Denmark	Pilot RCT	Elderly patients undergoing major emergency abdominal surgery and their family members IG: 9 patients (mean age 79 years) and 8 family members (mean age 67 years) CG: 8 patients (mean age 77 years) and 8 family members (mean age 63 years)	Three FHCs (two in‐hospital and one at home), by trained (3‐day education) hospital and homecare nurse together, duration	Standard care	ICE‐FPSQ	No significant improvement in perceived support
Petursdottir and Svavarsdottir 2019, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison	Family caregivers of patients with advanced cancer receiving palliative home care 55 family caregivers (48 in the analyses)	Two FHCs at home by specialised palliative home care unit/nurses, duration 60–90 min	NA	ICE‐FPSQ	Significant improvement of perceived support post‐intervention in family caregivers: pretest (M = 44.7, SD = 8.4), post‐test (M = 58.3, SD =8.4, p = 0.001) 2–4 weeks post intervention
Ragnarsdóttir and Svavarsdottir 2014, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Parents of children with rare chronic illnesses 30 parents (aged 19–60 years) IG: 15 parents CG: 15 parents	One FHCs in the outpatient hospital setting by a nurse duration 20–45 min (average 30 min)	Standard care	ICE‐FPSQ	Significant improvement in parents perceived support: IG (M = 2.21, SD = 1.02), CG (M = 1.63, SD = 0.47, p = 0.049) in favour of the IG, 3–5 days after baseline
Rosenstrøm et al. 2023, Denmark	RCT	Patients newly diagnosed with AF. IG: 33 patients (mean age 68 years/22 males) CG: 35 patients (mean age 68/22 males)	One to three FHCs +2‐h group education within a 2‐month period by one AF specialised clinical nurse	Standard care	ICE‐FPSQ	Significant improvement in patients perceived support: IG (M = 64.00, SD = 54.00–66.00), CG (M = 28.00, SD = 21.50–35.50, p < 0.001) in favour of the IG at 6 months post intervention
Sigurdardottir et al. 2013, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Parents of children with asthma 31 families (31 mothers and 15 fathers/mean age 19–60 years) IG: 15 families CG 16 families	Two FHCs within a 2‐week period in outpatient hospital setting provided by clinical nurses (master's degree), duration 45–90 min (average 60 min)	Standard care	ICE‐FPSQ	Significant improvement in perceived support in fathers post intervention: IG (M 37.6, SD = 16.7), CG (19.5, SD = 11.1, p = 0.03)
Svavarsdottir et al. 2012, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Parents of children with chronic or acute illness in the hospital setting IG: 41 parents (37 mothers and 4 fathers) CG: 35 parents (30 mother and 5 fathers)	One FHC by a trained study nurse, duration 15–50 min (average 30 min)	Standard care	ICE‐FPSQ	Total score perceived support not reported
Svavarsdottir and Sigurdardottir 2013, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison	Parents of children in active cancer treatment (in‐hospital and day treatment) 19 parents (age 31–60 years) of children and adolescents (age 0–18 years) Primary caregivers (n = 10/9 female) and partner caregivers (n = 9/9 male)	Two to three FHCs within a period of 10–12 weeks by a trained study nurse (master's degree), duration 45–90	NA	ICE‐FPSQ	Significant improvement in perceived support in primary caregivers at 1 week post intervention: pretest (M = 37.5, SD = 5.6), post‐test (M = 48.2, SD = 11.05, p = 0.021)
Svavarsdottir et al. 2014, Iceland	Quasi‐experimental Single group, pre‐post‐test comparison (secondary analysis of pooled data)	Children with chronic illnesses (cancer, diabetes, and asthma) in the hospital setting 35 mothers and 25 fathers (age 31–50 years) of children (age 0–18 years)	Two FHCs within a period of 10–12 weeks by a trained study nurse (master's degree), duration 45–90 min (average 60 min)	NA	ICE‐FPSQ	Significant improvement in perceived support post intervention in mothers 5–7 days post intervention: pretest (median = 38.00, range = 19.00) post‐test (median = 54.00, range = 27.00, p = 0.028) but not in fathers
Svavarsdottir et al. 2020, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison	Mothers of children and adolescents with newly diagnosed chronic illnesses (arthritis, epilepsy, diabetes, ADHD) 31 mothers (aged 18–60 years)	Two FHCs, duration 45–90 min within 4–10 weeks by three APN (master's degree) with graduate courses in Family Systems Nursing.	NA	ICE‐FPSQ	Significant improvement of perceived support after second FHCs: pretest (M = 33.24, SD = 17.63), post‐test (M = 51.38, SD = 17.01, p = 0.000)
Sveinbjarnardottir et al. 2013, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Patients with psychiatric illness in inpatient psychiatry IG: 68 patients and 69 family members CG: 74 patients and 74 family members	Two to five FHCs by 12 trained (1‐day education) clinical nurses, duration 30–60 min	Standard care	ICE‐FPSQ	Total score perceived support not reported
Østergaard et al. 2021, Denmark	RCT multicentre	Patients with CHF and their family members receiving care from HF‐clinic IG: patients (n = 232) and family members (n = 136) CG: patients (n = 236) and family members (n = 160)	Three FHCs during interval 6–12 weeks by six nurses specialised in HF having 3 days education and training in FHCs	Standard care	FAFHES social support scale	Significant improvement in patients perceived support after 3 months: IG (M = 79.60, SD = 21.49), CG (M = 71.96, SD = 23.78, p = 0.002), 6 months: IG (M = 78.98, SD = 21.25, p = 0.008), and 12 months: IG (M = 78.02, SD = 23.52), CG (M = 71.11, SD = 24.77, p = 0.018) in favour of the IG
Health‐related quality of life
Ahlberg et al. 2021, Sweden	Mixed methods including quasi‐experimental, controlled pre‐ post‐test comparison	ICU patients and family members' post‐discharge IG: 7 patients and 10 family members (n = 17) CG: 13 patients and 8 family members (n = 21)	Three FHCs post‐ discharge + closing letter after final session provided by trained ICU nurses	NA	SF36	No significant improvement in HRQoL between groups
Broekema et al. 2021, Netherlands	Quasi‐experimental Controlled pre‐ post‐test comparison	Patients and family members receiving homecare IG: 21 patients (mean age 80 years)/34 family members (mean age 68 years) CG: 30 patients (mean age 79 years) and 27 family members (mean age 66 years)	Two FHCs over a period of max 3 months provided by trained (6‐day education) homecare nurses	Standard care	Euro‐QoL 5D EQ5D index EQ5D VAS	No significant improvement in HRQoL
Dorell et al. 2017, Sweden	Mixed methods Single group, pre‐ post‐test comparison	Family members of elderly people in a residential home 10 families (22 family members: 20 females and 2 males)	Three FHCs (twice weekly) by a trained ‘clinical’ nurse working in nursing home and a research nurse, mean duration 60 min	NA	SWED‐QUAL	No significant improvement in HRQoL
Faarup et al. 2019, Denmark	Quasi‐experimental Single group pre‐ post‐test comparison	Patients receiving in‐hospital care at a neurosurgical ward for glioblastoma 67 patients (mean age 67 years)	One to three FHCs provided by two trained project nurses (not involved in care) in the outpatient clinic or at home, duration 60–90 min	Standard care	WHO‐QOL‐BREF	No significant improvement in HRQoL
Lämås et al. 2016, Sweden	Quasi‐experimental Controlled pre‐post‐test comparison	Family members of patients (< 65 years) suffering from stroke receiving rehabilitation N = 38 IG: n = 17 CG: n = 21	Three FHCs at home starting 3–12 months after diagnosis by 2 clinical nurses, duration average 60 min	Standard care	EQ‐5D SF‐6D	Significant improvement in HRQOL in favour of the IG 6 months after the FHCs were completed EQ‐5D index: IG (M = 0.63, SD = NA), CG (M = 0.73, SD = NA), p = 0.050 EQ‐VAS: IG (M = 70.87, SD = NA), p = 0.001 SF 6D: IG (M = 0.70, SD = NA), CG (M = 0.74, SD = NA), p = 0.010
Rosenstrøm et al. 2023, Denmark	RCT	Patients newly diagnosed with AF IG: 33 patients (mean age 68 years/22 males) CG: 35 patients (mean age 68 years/22 males)	One to three FHCs +2‐h group education within a 2‐month period by one AF specialised clinical nurse.	Standard care	AFEQT	No significant improvement in HRQoL
Sigurdardottir et al. 2013, Iceland	Quasi‐experimental Controlled pre‐ post‐test comparison	Parents of children with asthma 31 families (31 mothers and 15 fathers/mean age 19–60 years) IG: 15 families CG: 16 families	Two FHCs within a 2‐week period in outpatient hospital setting provided by clinical nurses (master's degree), duration 45–90 min (average 60 min)	Standard care	PedsQL Asthma module	No significant improvement in HRQOL
Svavarsdottir et al. 2014, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison (secondary analysis of pooled data)	Parents of children with chronic illnesses (cancer, diabetes, and asthma) in the hospital setting 35 mothers and 25 fathers (age 31–50 years) of children (age 0–18 years)	Two FHCs within a period of 10–12 weeks by a trained study nurse (master's degree), duration 45–90 min (average 60 min)	NA	PedsQL modules for asthma, cancer, diabetes, and paediatrics	No significant improvement in HRQOL
Svavarsdottir et al. 2020, Iceland	Quasi‐experimental Single group, pre‐post‐test comparison	Mothers of children and adolescents with newly diagnosed chronic illnesses (Arthritis, epilepsy, diabetes, ADHD) 31 mothers (aged 18–60 years)	Two FHCs, duration 45–90 min within 4–10 weeks by three APN (master's degree) with graduate courses in Family Systems Nursing.	NA	PedsQoL Family Impact Module	Significant improvement in QoL after second FHCs score: pretest (M = 64.43, SD = 23.33), post‐test (M = 71.40, SD = 22.73), p = 0.004
Østergaard et al. 2018, Denmark	RCT	Patients with CHF and their family members receiving care from HF‐outpatient clinics IG: patients (n = 180) (mean age 67 years) CG: patients (n = 167) (mean age 67 years)	Three FHCs during an interval of 6–12 weeks by six nurses specialised in HF and having 3 days education and training in FHCs	Standard care	KCCQ	No significant improvement in HRQoL, but clinical important improvement in seven domains of KCCQ compared to one domain in the control group after 3 months
Ågren et al. 2019, Sweden	RCT pilot study	ICU patients and family members' post‐discharge 17 families; 10 IG (n = 28) and 7 CG (n = 17) Mean age; 61 years in IG and 60 years in CG	Three FHCs 4–8 weeks after discharge with 2 weeks interval + closing letter after final session provided by trained project nurses, duration 2 h + 30–40 min		SF36	Significant improvement in HRQoL in favour of the intervention in social functioning: IG (M = 82.9, SD = 16.1), CG (M = 84.1, SD = 25.1), p = 0.02 and mental health: IG (M = 87.3, SD = 12.6), CG (M = 76.4, SD = 15.6), p = 0.01 at 12 months post intervention
Caregiver burden
Broekema et al. 2021 Netherlands	Quasi‐experimental Controlled pre‐ post‐test comparison	Patients and family members receiving homecare IG: 21 patients (mean age 80 years)/34 family members (mean age 68 years) CG: 30 patients (mean age 79 years) and 27 family members (mean age 66 years)	Two FHCs over a period of max 3 months provided by trained (6‐day education) homecare nurses.	Standard care	CarerQoL: CarerQoL. 7D CarerQoL VAS	Significant decrease of caregiver burden in favour of the IG at 6 months post intervention in CarerQoL. 7D scores: IG (M = 84.49, SD = 12.27), CG (M = 81.58, SD = 11.65), p = 0.018
Lee et al. 2018, Taiwan	RCT	Family caregivers of hospitalised patients with bipolar disorders IG: n = 16 (age 21 to ≥ 66 years) CG: n = 20 (age 21 to ≥ 66 years)	Four sessions twice a week by a therapist and an experienced psychiatric nurse with a master's degree	Standard care	CBI	No significant decline in caregiver burden
Petursdottir and Svavarsdottir 2019, Iceland	Quasi‐experimental Single group, pre‐ post‐test comparison	Family caregivers of patients with advanced cancer receiving palliative care 55 family caregivers (48 in the analyses)	Two FHCs at home by specialised palliative home care nurses, duration 60–90 min	NA	BASC	Significant decrease of caregiver burden 2–4 weeks postintervention: pretest (M = 12.3, SD = 7,1), post‐test (M = 10.4, SD = 6,4), η 2ₚ = 0.090, p = 0.007
Family health
Lee et al. 2018, Taiwan	RCT	Family caregivers of hospitalised patients with bipolar disorders IG: n = 16 (age 21 to ≥ 66 years) CG: n = 20 (age 21 to ≥ 66 years)	Four sessions twice a week by a therapist and an experienced psychiatric nurse with a master's degree	Standard care	CHQ‐12	No significant improvement in caregiver health
Østergaard et al. 2021, Denmark	RCT multicentre	Patients with CHF and their family members receiving care from HF‐clinics IG: patients (n = 232) and family members (n = 136) CG: patients (n = 236) and family members (n = 160)	Three FHCs during an interval of 6–12 weeks by six nurses specialised in HF and having 3 days education and training in FHCs	Standard care	FAFHES family health scale	No significant improvement in Family Health

Abbreviations: ADHD, Attention Deficit Hyperactivity Disorder; AF, Atrial Fibrillation; AFEQT, Atrial Fibrillation Effect on Quality‐of‐Life Questionnaire; APN, Advanced Practice Nurse; BASC, Brief Assessment Scale for Caregivers; CarerQoL 7D, Care‐related Quality of Life 7D; CarerQoL VAS, Care‐related Quality of Life Visual Analogue Scale; CarerQoL, Care‐related Quality of Life; CBI, Caregiver Burden Inventory; CG, Control Group; CHF, Congestive Heart Failure; CHQ‐12, Chinese Health Questionnaire‐12; COPD, Chronic Obstructive Pulmonary Disease; EQ‐5D VAS, EQ‐5D Visual Analogue Scale; EQ‐5D, EuroQol‐5D; FAFHES, Family Functioning, Family Health, and Social Support questionnaire; FHCs, Family Health Conversations; HF, Heart Failure; ICE‐FPSQ, Iceland Family Perceived Support Questionnaire; ICU, Intensive Care Unit; IG, Intervention Group; KCCQ, Kansas City Cardiomyopathy Questionnaire; M, mean; NA, Not Available; PedsQL, Paediatric Quality of Life; RCT, Randomised Controlled Trial; RSV, Respiratory Syncytial Virus; SD, Standard deviation; SF‐36, Medical Outcomes Study Short‐form 36; SF‐6D, Medical Outcomes Study Short‐form 6; SWED‐QUAL, Swedish Health‐Related Quality of Life Survey; WHO‐QOL‐BREF, World Health Organisation‐Quality of Life (short version).

Sixteen of the 24 included papers used a quasi‐experimental design, eight of which included a control group. Two papers used a mixed‐methods design, and six papers were RCTs. The studies were conducted in Iceland (13 papers), Denmark (5 papers), Sweden (4 papers), the Netherlands (1 paper) and Taiwan (1 paper).

4.3. Characteristics of the Intervention

The intervention was referred to under different names, such as Family Health Conversation (Ahlberg et al. 2021; Faarup et al. 2019), Family Nursing Conversations (Broekema et al. 2021), Therapeutic Conversation Intervention (Gisladottir and Svavarsdottir 2017), Short‐family Therapeutic conversation (Kamban and Svavarsdottir 2013; Rosenstrøm et al. 2023; Svavarsdottir et al. 2012; Sveinbjarnardottir et al. 2013), family‐centred care (Lee et al. 2018), Family Therapeutic Conversation Intervention (Svavarsdottir and Sigurdardottir 2013; Svavarsdottir et al. 2014) Family Nursing Therapeutic Conversation (Østergaard et al. 2021, 2018) and Health Promoting Conversation (Ågren et al. 2019).

All papers stated that their intervention was based on the theoretical frameworks of the CFAM and CFIM models. Furthermore, most studies included the IBM as part of their theoretical background. Following this theoretical framework, the papers described elements of the intervention such as exploring the family structure (e.g., genograms); jointly prioritising problems for discussion; ensuring that all members of the family could share their perspectives; using reflexive and circular questioning to challenge beliefs; and offering commendations while acknowledging emotions and suffering. Four studies offered a closing letter with a written reflection by the nurse after the final conversation (Ahlberg et al. 2021; Dorell et al. 2017; Lämås et al. 2016; Ågren et al. 2019).

The number of conversations offered ranged from one to five sessions, and the duration of each conversation ranged from 15 min to 2 h. In the paper by Svavarsdottir and Gisladottir (2019), the conversation was an element of a more extensive support programme, and in another paper (Rosenstrøm et al. 2023), the conversation was combined with a 2‐h group education. The conversations were provided by nurses in all papers. In most of the papers, nurses from clinical practice were trained to provide the intervention. In a few cases, the intervention was offered by project nurses or by nurses who were members of the research team (not working in clinical practice). However, the length of the training programmes offered ranged from one day (Sveinbjarnardottir et al. 2013) to six days (Broekema et al. 2021). Some papers mentioned training without specifying its duration or extent. Before being trained as a part of the intervention, the educational level of the nurses was master's and PhD (1 study), master's (6 studies), not specified for nurses (12 studies) and not specified for researchers (2 studies) (see Tables 5 and 6).

4.4. Primary Outcome

Family functioning was assessed in 13 papers by using various measures: the ICE Expressive Family Functioning Questionnaire (ICE‐EFFQ) in seven papers; the Family Assessment Device (subscale general functioning) in three papers; the Family Function Style Scale (FFSS) in one paper; the Family Function Scale (FFS) in one paper; and the family functioning scale from the family functioning, family health and social support (FAFHES) questionnaire in one paper. The reported fluctuation in time points of efficacy testing ranged from 3.5 days to 12 months after the intervention.

The four papers that used the 12‐item general functioning sub‐scale from the McMaster Family Assessment Device, along with the Chinese version of the FFS, showed statistically significant improvements in family functioning scores in favour of the intervention immediately following its implementation: IG (M = 150.19, SD = 23.12), CG (M = 134.42, SD = 20.05, p = 0.03) for patients and caregivers (Lee et al. 2018). At the 3‐month follow‐up, Ahlberg et al. (2021) found a statistically significant improvement in family functioning scores in favour of the intervention: IG (M = 1.5, SD = 0.5), CG (M = 2.8, SD = 0.5), p = 0.001 for family members. At the 6‐month follow‐up, Broekema et al. (2021) demonstrated a statistically significant improvement: IG (M = 3.46, SD = 0.44), CG (M = 3.18, SD = 0.57), p = 0.001 for family members, and for patients: IG (M = 3.58, SD = 0.47), CG (M = 3.04, SD = 0.38), p = 0.004 in favour of the intervention (Broekema et al. 2021). Finally, after a period of 12 months follow‐up, a statistically significant improvement was seen in family functioning scores in favour of the intervention for patients and family members: IG (M = 1.5, SD = 0.6), CG (M = 1.8, SD = 0.5, p = 0.03) (Ågren et al. 2019) (see Table 5).

The ICE‐EFFQ was used in seven papers employing a quasi‐experimental design. Among these, one single‐group study conducted within‐groups analyses and found a statistically significant improvement in total family functioning score: pretest (M = 37.5, SD = 5.6), post‐test (M = 48.2, SD = 11.05, p = 0.021), measured one week post‐intervention compared to baseline (Svavarsdottir and Sigurdardottir 2013). Conversely, between groups analysis: IG (M = 4.0, SD = 0.3), CG (M = 4.1, SD = 0.6, p = 0.007), a statistically significant decline in total ICE‐EFFQ family functioning scores was reported among mothers of children with bronchiolitis 11 days after receiving the 15‐min or less family interview (Kamban and Svavarsdottir 2013). None of the remaining papers were able to demonstrate the benefit of FHCs on family functioning regardless of dose, instruments used, or time of follow‐up (Faarup et al. 2019; Gisladottir and Svavarsdottir 2017; Halldórsdóttir and Svavarsdóttir 2012; Rosenstrøm et al. 2023; Svavarsdottir et al. 2014, 2012; Sveinbjarnardottir et al. 2013; Østergaard et al. 2021) (Table 4).

4.5. Secondary Outcomes

4.5.1. Perceived Family Support

Perceived family support was assessed in 15 studies using two measures: the Icelandic Perceived Family Support Questionnaire (ICE_PFSQ), which was used in 14 papers, and the family social support scale from the FAFHES questionnaire which was used in one paper (Table 5). Of the 14 papers using the ICE_PFSQ, eight papers showed a statistically significant improvement in the total score in favour of the intervention compared to a control group (Halldórsdóttir and Svavarsdóttir 2012; Petursdottir and Svavarsdottir 2019; Ragnarsdóttir and Svavarsdottir 2014; Rosenstrøm et al. 2023; Sigurdardottir et al. 2013; Svavarsdottir et al. 2020, 2014; Svavarsdottir and Sigurdardottir 2013). Moreover, one multicentre RCT study reported the effect size showing statistically significant improvement in FAFHES family support scale scores in favour of the intervention after 3 months (t(3.084) = 2.440, p = 0.002, Cohen's d = 0.3335, 95% CI [0.1207, 0.5463]), after 6 months (t(2.66) = 4.046, p = 0.008, Cohen's d = 0.3007, 95% CI [0.0785, 0.5230]) and after12 months follow‐up (t(2.438) = 1.663, p = 0.015, Cohen's d = 0.2907, 95% CI [0.0560, 0.5253]). The effect size was considered small (Østergaard et al. 2021). One study reported significant improvement in the two sub‐scales of ICE_PFSQ among parents of children with an acute illness who had participated in one family conversation compared to a control group receiving standard care (Svavarsdottir et al. 2012). One study did not report any results from using the ICE_PFSQ (Kamban and Svavarsdottir 2013), and three studies did not report the total scale scores (Kamban and Svavarsdottir 2013; Svavarsdottir et al. 2012; Sveinbjarnardottir et al. 2013).

When using a single group pre‐ and post‐test, six studies showed statistically significant improvement in the total score of the ICE_PFSQ post‐intervention compared to the baseline (Petursdottir and Svavarsdottir 2019; Svavarsdottir and Gisladottir 2019; Svavarsdottir et al. 2020, 2014; Svavarsdottir and Sigurdardottir 2013) among family members, but not among patients (Sveinbjarnardottir et al. 2013). Three studies did not find any benefit of the intervention in terms of perceived support (Gisladottir and Svavarsdottir 2017; Konradsdottir and Svavarsdottir 2013; Petersen et al. 2022). Furthermore, in the study by Petursdottir and Svavarsdottir (2019), the treatment effect was calculated to be F(2, 51.24) = 0.522, p = 0.001, indicating that over half of the variance in the outcome was explained by the treatment (Table 6).

4.5.2. Health‐Related Quality of Life

Eleven papers reported data on health‐related quality of life (Table 6). Two papers showed statistically significant improvement after the intervention using the SF‐6D classification system and the EQ VAS visual analogue scale (Lämås et al. 2016), and the Paediatric Quality of Life Inventory (PedsQL) family impact module (Svavarsdottir et al. 2020). Furthermore, one pilot RCT showed significant improvement in mental health and social functioning scores on the Medical Outcomes Study Short‐Form 36 among family members of former critically ill patients when compared to a control group (Ågren et al. 2019). Among patients diagnosed with heart failure, one RCT showed clinically important improvement in seven domains of the Kansas City Cardiomyopathy Questionnaire compared to one domain in the control group (Østergaard et al. 2018).

The remaining seven studies could not demonstrate any effect of the intervention regardless of study design, target population, or instruments used (Ahlberg et al. 2021; Broekema et al. 2021; Dorell et al. 2017; Faarup et al. 2019; Petersen et al. 2022; Rosenstrøm et al. 2023; Svavarsdottir et al. 2014) (Table 6).

4.5.3. Caregiver Burden and Family Health

Two out of three studies investigating the effect of the FNCs on caregiver burden showed statistically significant decreased caregiver burden among caregivers of patients receiving long‐term home health care (Broekema et al. 2021) and among caregivers of patients in the advanced/final phases of cancer (Petursdottir and Svavarsdottir 2019) (Table 6).

Family health was measured in two papers, both of which found no differences between the intervention group and the control group in scores on either the FAFHES family health questionnaire (Østergaard et al. 2021) or the Chinese health questionnaire‐12 (Lee et al. 2018) (Table 6).

5. Discussion

5.1. Results

Within this systematic review, our main aim was to collect and map the quantitative evidence for the effectiveness of FHCs, primarily on family functioning in families confronted with critical or chronic health conditions. We identified 13 papers that measured family functioning as the primary outcome of FHCs, yielding mixed results: five papers reported significant positive effects, one paper showed a decline and seven papers found no significant effects post‐intervention. Our findings are in line with the recent scoping review by Azcárate‐Cenoz et al. (2024), who reported a statistically significant improvement in seven out of 12 studies measuring family functioning. There are, however, discrepancies in the interpretation of the results presented by Azcárate‐Cenoz et al. (2024), as they report a significant positive effect of the intervention on family functioning in two of the studies included in their review (Gisladottir and Svavarsdottir 2017; Svavarsdottir et al. 2014), Azcárate‐Cenoz et al. (2024). That being said, in the present systematic review, we found there were no significant effects on total scores reported post‐intervention in these two papers. In the study by Kamban and Svavarsdottir (2013), there was a significant decline in mothers' perceptions of family functioning measured by the ICE‐EFFQ. Conversely, in the study by Azcárate‐Cenoz et al. (2024), the authors reported this as a significant improvement in family functioning. Additionally, a meta‐analysis of two studies measuring the effectiveness of family‐centred interventions for family caregivers showed no effect on family functioning (Wang et al. 2024).

Three out of the five papers showing a statistically significant effect of the intervention (Ahlberg et al. 2021; Broekema et al. 2021; Ågren et al. 2019) used the General Functioning scale, which is a subscale of the McMaster Family Assessment Device (FAD) (Epstein et al. 1983). The FAD is one of the most common instruments used to measure global family functioning (Pritchett et al. 2011). The remaining papers showing no statistically significant improvement in family functioning post‐intervention used the ICE‐EFFQ (n = 6), FAFHES (n = 1) and FFSS (n = 1). The lack of a persuasive effect of the FHC on family functioning in the present review might be related to the instruments used. Self‐reported family functioning measures may be limited by concerns about how effective the family functioning measures are in examining the family level rather than the individual level, and whether an average score from individual family members will be reflected by the overall score (Pritchett et al. 2011).

Remarkably, most of the papers included in the present systematic review used a quasi‐experimental design. Quasi‐experiments do not rely on random assignment, which limits their ability to establish causal associations between interventions and outcomes (Higgins et al. 2023). Of the 24 papers, only six were RCTs and two of these were pilot studies with a limited sample size. High‐quality RCTs have conventionally been considered the best study design for examining the effectiveness of interventions. However, quasi‐experiments offer practicality and external validity, while RCTs provide strong internal validity but may be resource‐intensive and less applicable to real‐world contexts (Miller et al. 2020). Therefore, researchers should choose a design based on their specific research questions and constraints.

Also, careful study design and appropriate sample sizes are crucial to minimise statistical errors. Generally, the included papers were limited by small sample sizes and lack of power calculations. A larger sample size captures more variability and increases the chances of detecting real effects. Conversely, smaller samples may lack sufficient evidence to confidently reject the null hypothesis even if there is a true effect (Type II error). On the other hand, Type I errors might occur if we conclude there is an effect when, in fact, there is none (Smith 2012). Therefore, more high‐quality intervention studies are needed before we can conclude whether FHCs are effective for the improvement of family functioning.

Furthermore, factors such as gender, socio‐economic status and ethnic background have been found to affect self‐reported measures (Zhang 2018). It could also be that FHCs are more associated with factors other than family functioning. In that case, mediation analysis can provide evidence on the causal mechanisms of health exposures in clinical trials and observational studies (Cashin et al. 2023). Hence, Shamali et al. (2019, 2021) found that factors such as age, education and the work situation of patients and family members as well as health of the close relatives mediated the association between social support and family functioning. Factors facilitating the associations between FHCs, perceived support and family functioning should be investigated in future research. Consequentially, there is a need for greater precision in describing the intervention as well as more research evaluating the effect of family nursing care on family functioning (Thürlimann et al. 2022). Especially desirable are large‐scale RCTs with appropriate stratification for age, gender and socio‐demographic data (Østergaard et al. 2018, 2021). One way to produce more RCT research projects is to focus on multicentre or multinational FHCs research projects using sensitive, standard and validated instruments to measure family functioning to enable comparability and synthesis across studies.

5.2. Strengths and Limitations

A key strength of this systematic review is that, to our knowledge, it is the first review with strict inclusion criteria for quantitative studies using the CFAM and CFIM models as the theoretical framework and a focus on family functioning as the primary outcome. Another strength of this review is that we followed the JBI methodology and guidance for the conduct of systematic reviews, and a we conducted a broad search in numerous databases. In addition, the methodological quality of the studies was critically considered using the JBI Critical Appraisal Checklists.

The quality assessment revealed that the presence of methodological limitations in several studies, particularly regarding sample size and adequately addressing potential confounding variables, may be a source of selection bias. Furthermore, the relatively small number of included papers, the majority of which were conducted in the Nordic countries and especially in Iceland, may also be a selection bias that affects the ability to generalise to a global community. Therefore, our findings may not be representative of families' and clinicians' experiences in other countries. This restriction might be explained by the fact that leaders from practice and education in Iceland have been at the forefront when it comes to the implementation of FHCs (Svavarsdottir 2008).

This review may also be subject to language bias, as limited resources prevented the inclusion of articles published in languages other than English. However, English has become the official language of the scientific world, and restricting systematic reviews to English‐language publications appears to have little impact on the effect estimates and conclusions of systematic reviews (Dobrescu et al. 2021).

The heterogeneity of the papers included in this systematic review might further contribute to blurring the picture of the effectiveness of FHCs. Due to the differences in study designs, populations, the numbers and dosage of FHCs provided, clinical diversity, and the risk of bias in the individual papers, conducting a meta‐analysis was not feasible (Higgins et al. 2023). Variations in health conditions can initiate heterogeneity, making it challenging to compare and synthesise results across studies (Bouton et al. 2023). In our review, the proportion of studies showing a positive effect on family functioning included post‐discharge ICU patients and family members (two studies); patients and family members receiving homecare (one study); hospitalised patients with bipolar disorders and their caregivers (one study); and parents of children in active cancer treatment (one study). Thus, researchers should be encouraged to perform future studies based on comparative patient populations. Future studies should also take the family members' roles and responsibilities into account (e.g., parents vs. relatives vs. caregivers) to ensure that the findings are relevant and applicable to various patient populations (Bonsel et al. 2024). The variability in qualifications (researchers vs. project nurses vs. clinical nurses) and training (from 1 to 6 days) of the nurses to perform FHCs can affect how consistently the conversations are implemented across studies (Bell 2016). Well‐trained nurses are more likely to effectively engage families, address their concerns and facilitate meaningful conversations, leading to better patient and family outcomes (Kim and Kim 2024). Finally, the different follow‐up periods and completeness of family member participation in our review can introduce heterogeneity, making it difficult to compare and synthesise results. This variability can affect the consistency and reliability of the findings.

Thus, more solid homogeneous studies are needed before we can build an evidence‐based knowledge ground, and before a conclusion can be drawn about whether FHCs can maintain or improve the functioning of families confronted with critical or chronic health conditions.

6. Conclusions

This systematic review provides an overview of the current state of knowledge regarding the effectiveness of FHCs. Whereas earlier reviews primarily included qualitative studies, we were able to present 24 papers employing quantitative research designs, 13 of which measured family functioning. Of the 24 included papers, six presented an RCT, which is considered the gold standard in studying interventions. The literature concerning the effectiveness of FHCs on family functioning is sparse and varies in measurement techniques used across diverse clinical settings and populations. Perceived support may be a more sensitive outcome to FHC but should be tested in future studies. Therefore, drawing firm conclusions remains challenging. To attain superiority, scientific evidence regarding whether FHCs are beneficial for health outcomes of families, more high‐quality studies with larger sample sizes and more homogeneous study populations are needed.

7. Relevance to Clinical Practice

The evidence regarding the effectiveness of FHCs and how they improve the health and functioning of individuals and their families is rather sparse. More evidence is needed and consequently, further research is required on both the implementation and integration of FHCs into clinical practice. A uniform and validated method for assessing family functioning in clinical settings would enable comparisons of different approaches aimed at improving the functioning of individuals and their family members. Moreover, rigorous and high‐quality research is needed to verify the long‐term effects of these interventions on individuals and their family members.

Author Contributions

All authors contributed to the concept and design of the review. B.Ø., M.L.L., E.K.‐S., A.B. and S.R. analysed and interpreted the data, wrote the original draft and conducted the review and editing. All authors were involved in reviewing the manuscript drafts and gave approval for this version to be published.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

File S1: PRISMA 2020 Checklist.

Data Availability Statement

Data openly available in a public repository that issues datasets with DOIs: The data that support the findings of this study are openly available in the reference list.