Article Text

Download PDFPDF

Family group decision-making interventions in adult healthcare and welfare: a systematic literature review of its key elements and effectiveness
  1. Chantal F Hillebregt1,
  2. Eline W M Scholten1,
  3. Marcel W M Post1,2,
  4. Johanna M A Visser-Meily1,3,
  5. Marjolijn Ketelaar1
  1. 1Center of Excellence for Rehabilitation Medicine Utrecht, Brain Center University Medical Center Utrecht, and De Hoogstraat Rehabilitation, Utrecht, The Netherlands
  2. 2Department of Rehabilitation Medicine, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
  3. 3Department of Rehabilitation, Physical Therapy Science & Sports, UMC Utrecht Brain Center, University Medical Center Utrecht, Utrecht, The Netherlands
  1. Correspondence to Dr Marjolijn Ketelaar; M.Ketelaar{at}


  • Contributors CFH developed the first version of the manuscript, in collaboration with EWMS, JMAV-M, MWMP and MK. JMAV-M, MWMP and MK supervised the writing process and provided feedback on several drafts of the manuscript. EWMS and MK executed the double screening. JMAV-M, MWMP and MK provided methodological advice. CFH wrote the final version of the manuscript.

  • Funding This study is financially supported by ZonMw, the Netherlands Organisation for Health Research and Development, Fonds Nuts Ohra and Revalidatiefonds, grant number: 630000003.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The data that support the findings of this study are available from the corresponding author, (MK), upon reasonable request.

  • Patient consent for publication Not required.

View Full Text

Statistics from

Strengths and limitations of this study

  • This is the first review of the effectiveness of family group decision-making (FGDM) intervention studies in adult healthcare and welfare to date.

  • A search of 14 relevant electronic bibliographic databases was conducted which gives this review breadth and comprehensiveness.

  • Studies were only assessed when written in the English language.

  • Identification of FGDM studies only when all three FGDM key elements were present.

  • Narrative analyses because of heterogeneous studies and few high-quality studies.


Family group decision-making

The family group decision-making (FGDM) model is implemented in child care and welfare in more than 30 countries worldwide.1 The roots of FGDM originate from the family group conference (FGC) model in New Zealand1 where it was legislated as the decision-making process to be used in cases of child abuse, neglect and youth offending in 1989.2 3 Since then, the model is subject to change resulting in many variants with a range of models,4 which all can be referred to under the umbrella of the broader term FGDM.5 While outside the USA, meetings are referred to as FGCs, within the USA, different variants are named such as: family unity meeting, team decision-making meeting, family decision meeting, family team conferences and family team meeting.4 6–12

In FGDM the person in need and members of their family reflect on goals with respect to participation in daily activities through structured meetings. The term ‘family’ is widely interpreted with the inclusion of extended family members, friends, neighbours and significant others.13 During FGDM, the participants share ideas on possible solutions to achieve goals and decide on a concrete plan to support the person in need in the way he or she wants. The ownership of this plan remains with the family; they are responsible for contributing, controlling and executing the actions/goals of the plan including inviting people they want to the conference, with little contribution from professionals. In this way, FGDM aims to stimulate the decision-making process, with a rearrangement of tasks and responsibilities to shift the balance of power from the professional towards the person in need and the family within this decision-making process.14 FGDM is expected to facilitate the assessment of problems into a new perspective where the family and the person in need have the main voice in matters that concern them,15 enlarging their empowerment.16 Major advantages are that FGDM uses resources already existing within society namely the family and others,17 who are often better able to find workable solutions14 consistent with their own culture, lifestyle and history than professionals.18

Families often experience the FGCs as very positive and helpful.19 20 Various process evaluations in child care, for instance, show that participants feel listened to1 and are satisfied,21 even though a recent meta-analysis of FGC (one of the variants of FGDM) involving fourteen controlled studies, showed no effectiveness in child-related outcomes.7

Key elements of FGDM

All variants under the umbrella of FGDM have a common philosophy5 of engaging the family with a central role in the decision-making process,8 22–24 characterised by three key elements. These key elements are important because they indicate the difference between FGDM and more general family-centred meeting interventions. The first key element is the importance of drawing up an action plan serving as a basis for the action-oriented perspective and to clarify, monitor and achieve goals. Second, the understanding that FGDM decision models are ‘family driven’, meaning that the approach is not aimed at the family, but achieves results through the contributions of the family. The person in need together with the family set the agenda, develop and implement the plan. They possess the executive role rather than the professional. Here, FGDM distinguishes itself from ‘traditional’ approaches of family meetings that are often ‘family centred’15 where the professional remains in their executive role.

The third important key element is the three phases characterising the FGDM process8 25 26; the preparation phase, the conference phase and the evaluation phase (see figure 1). Through these meetings, a gradual shift takes place in the balance of power from professional towards increased empowerment of the person in need and their family. In the preparation phase, the referral commences and the concept of FGDM is explained to the person in need and their family and a conference coordinator is appointed. In the original FGC model dating from the 1980s, the key values of having an independent coordinator and private family time1 are quite fixed. However, these can vary between the different FGDM types.4 9 Depending on the type of FGDM, this coordinator can be a member of an independent agency or the family may decide, based on their preference for example to a member of the professional team.21 27 In the conference phase, the actual FGDM takes place in which the person in need and the family members reflect on their goals and the support needed to make these goals achievable. Depending on the FGDM type, the family can choose if they want to schedule private family time, without professionals being present, to develop a concrete action plan.21 27 In the evaluation phase, the achievements resulting from the goals and action plan are evaluated.

Figure 1

Key elements of family group decision-making.26

FGDM in the adult healthcare/welfare

Involving the family in adult healthcare and welfare is not new and is implemented in a number of settings. Predischarge and follow-up family meetings are conducted in geriatric medicine, rehabilitative medicine, palliative care and psychiatry,28–32 for instance. These meetings are often held to improve communication between the multidisciplinary team and family members discussing patient’s health condition and progress.33 In some healthcare settings like rehabilitation medicine, a goal-directed programme, using an action plan and a predetermined set of preparation and evaluation meetings, is already part of the regular rehabilitation programme. Although two out of three key elements characterising the FGDM process8 25 26 (action plan and the meeting set-up in three phases) are present, these meetings do not qualify as FGDM. One of the important differences with FGDM is that these family meetings lack the family-driven key element. Despite the fact that the family is often engaged in the decision-making process setting up a plan; the coordination of tasks and the ownership of the plan remains with the professionals.29 34 35 First, more awareness is needed to understand the importance of distinguishing the more commonly implemented family-centred meetings and FGDM by its key elements. Second, with FGDM studies slowly increasing within different settings among adults in public mental healthcare,8 36–39 rehabilitation medicine,40 general social welfare41–43 and social welfare for elderly,44 45 it is important to systematically examine the effectiveness of FGDM. To our knowledge, such a review has not been published in adult healthcare and welfare to date.


The objectives of the present study are: (1) to systematically review the presence of the three key elements of FGDM in family-centred interventions in adult care and welfare, and (2) to evaluate the effectiveness of the FGDM interventions.


Research design and methodology

We followed the systematic review methodology using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement46 and the Cochrane Handbook for Systematic Reviews of Interventions47 which specify different phases concerning the search strategy and screening (preparation phase, title/abstract screening, full-text screening and extraction).47

Patient involvement

As this is a review screening literature, no patients/clients are involved.

Data sources and search strategy

A comprehensive search of 14 relevant electronic bibliographic databases and 1 academic search engine (Google Scholar), was conducted (PubMed/Medline, CINAHL, PsycINFO, Cochrane Library, International Bibliography of the Social Sciences, Campbell library, Social Science Research Network, ASSIA Applied Social Sciences Index and Abstracts, Social Care Online, sociological abstracts/social services abstracts, Web of Science, Scopus, SocINDEX, ERIC). It was a deliberate choice to only include the intervention eligible criteria into the search strategy, including most common variants of the intervention FGDM (see online supplementary appendix 1, example full electronic search strategy). Articles not eligible to the other criteria were screened out within the title abstract phase. This resulted in a broad screening process aiming to be as thorough and accurate as possible. The search string was adapted for use with different bibliographic databases. All articles were searched until 21 February 2018, with no further predetermined date limitation. Only studies in the English language were included.

Eligibility criteria

The PICOCS process framework48 was used to formulate the eligibility criteria according the concepts: Population, Interventions, Comparators, Outcomes, Context and Study design.

Studies were included if they met the following eligibility criteria:

  • Types of population (P): The target population of the study is aged eighteen or older.

  • Types of interventions (I): We used a two-step selection approach. Aim 1: family-centred interventions including FGDM. Aim 2: only FGDM interventions. For the first aim, we selected all family-centred studies, defined as studies of interventions to engage family/friends in the decision-making process. Furthermore, we screened these family-centred studies for the presence of the three FGDM key elements ((1) plan with goals/actions, (2) family-driven, (3) three phases). For the second aim, we included only interventions qualified as FGDM with all three key elements present.

  • Comparators (C): Treatment as usual or no treatment was identified as comparator.

  • Types of outcomes (O): All outcomes were included. Studies must have included at least one quantifiable outcome measure.

  • Context (C): Any setting that serves for adult healthcare or welfare.

  • Types of study design (S): Studies using controlled trial designs (randomised and non-randomised) with a comparison group.


Preparation phase

The citation management software programme Mendeley was used to import studies from the 14 electronic bibliographic databases. One masterfile was created merging the studies of the 14 databases after removing all duplications. All articles were uploaded to the systematic literature review software Rayyan49 where studies were categorised and labelled according a predetermined coding scheme.

Title abstract and full-text screening

All articles were first selected in which some kind of engagement of family/friends in the decision-making process was described. We have named these ‘family-centred studies’. All family-centred studies were further screened for eligibility according to the other eligibility criteria (eg, population, study design, etc). When a decision could not be made based on title and abstract, the full paper was obtained for detailed assessment.

One researcher performed the title abstract screening (CFH) with 10% double-screening (EWMS). Inconsistencies were discussed with a third researcher (MK) until consensus was reached. The full-text screening was conducted by two researchers (CFH 100%; EWMS and MK 50% each). Inconsistencies were discussed until consensus was reached.

Data extraction, methodological quality and level of evidence

The remaining studies were extracted and 100% double assessed (CFH/EWMS) concerning their (1) characteristics, (2) quality/level of evidence and (3) presence of the key elements of FGDM: (1) plan, (2) family driven, (3) three phases. The identification of the FGDM key elements was deliberately postponed to the data extraction phase instead of the full-text screening phase. Consequently, an overview of family-centred outcome studies versus FGDM intervention studies can be presented as support for our first research aim.

A standardised data extraction form from the Cochrane Collaboration47 was used to extract the following data from the eligible studies: author, year, country, design, sample, cohort, intervention, measures, main findings, limitations.

The quality of the studies was determined using the Quality Assessment Tool for Quantitative Studies50 judged to be suitable in systematic reviews of effectiveness.47 50

The tool assesses six domains: (1) selection bias, (2) study design, (3) confounders, (4) blinding, (5) reliability and validity of data collection method; and (6) withdrawals/drop-outs. With use of the tools guideline,51 a global quality score can be calculated ranging from 1 to 3, with 1 being the best score reflecting no ‘weak’ domain rating; 2 reflecting one ‘weak’ rating and 3 reflecting two or more ‘weak’ ratings50 (see table 1).

Table 1

Quality Assessment Tool for Quantitative Studies50

The level of evidence was checked using the classification developed by the Oxford Centre for Evidence-Based Medicine52 53 ranking the studies conform their probability of bias54 (conform the study design criteria, only level 1 and 2 studies are included) (see table 2).

Table 2

Level of evidence54




The search yielded a total of 3356 hits retrieved from the database searches. After removing duplicate citations (n=1676), a total of 1680 unique records were left. The articles were published in the time period ranging from 1948 to 2018 (see figure 2).

Figure 2

Flow chart of search and selection process.46 FGDM, family group decision-making.

Title abstract and full-text screening

A total of 1680 articles were screened in the title/abstract search conform all eligible criteria, including the first step of the intervention eligible criteria (family-centred studies). In total, 1669 articles were excluded on studies in another context (consumer purchase decision-making, marital decision-making, etc) that did not meet the study design criteria of pertaining a controlled trial, did not meet the population criteria of the participant being adult, were not family centred, had no abstract and full text (inaccessible information of meetings, conferences, etc) and were non-English. The 10% double screening (CFH/EWMS) showed good inter-rater agreement (85% consensus; kappa=0.61).55

A total of 11 articles were double checked on the eligibility criteria (CFH/EWMS/MK). Non-consensus articles were discussed until consensus was reached. In total, five studies did not meet the following eligibility criteria: study design (n=3),31 56 57 population (n=1)58 and no full text (n=1).59

Data extraction and quality assessment

Of the remaining six family-centred articles,41 42 60–63 data were extracted and summarised concerning their (1) characteristics, (2) quality/level of evidence and (3) presence of the key elements of FGDM (see table 3). 

Table 3

Summary of included studies, quality assessment, level of evidence

Characteristics of extracted family-centred studies

All six articles reported information conform the PICOCS process frame work48 eligibility criteria. However, the heterogeneity in outcome, context, intervention and participants was high and the results varied greatly. Therefore, it was not possible to analyse the articles quantitatively. Instead, the data extraction tool47 formed the basis of a narrative synthesis.

The articles were published between 2011 and 2017. The six articles originated in total from three studies, emanated in Norway from Malmberg-Heimonen and Johansen,41 42 the Netherlands from Joling et al.60–62 and the USA from de Havenon et al.63 Two studies had a randomised controlled trial (RCT) design, one study was a pilot controlled trial. Different context fields were described: mental health, dementia care and intensive care unit.

There was a widespread of sample sizes: ranging between 22 and 192 participants. There were three groups of study participants who were questioned on different outcome variables: social welfare recipients, caregivers and patients dyads and family members. The comparison condition for the studies was either treatment as usual or no treatment. The outcome measures for the Malmberg-Heimonen and  Johansen study41 42 were prominently self-reported questionnaires measuring social support, life satisfaction and mental health of social recipients. The study conducted by Joling et al60–62 reported nursing home placements, depression, anxiety and cost-effectiveness. The study by de Havenon et al63 described a family meeting survey reporting the satisfaction with audio-visual family meetings.

Quality/level of evidence of the extracted family-centred outcome studies

With use of the Quality Assessment tool for quantitative studies,50 six domains were assessed from the three studies: (1) selection bias, (2) study design, (3) confounders, (4) blinding, (5) reliability and validity of data collection method and (6) withdrawals/drop-outs (see online supplementary appendix 2 for detailed information). The 100% double screening (CFH/EWMS) resulted in 88% consensus with a very good inter-rater agreement of 0.81 (kappa score).55

Although two of the three eligible studies had a strong study design (RCT), the overall quality of the three studies was low. All studies reported challenges with a low participation rate (<60%) of the initial eligible group participants or presented no documentation of the participation rate. The low participation rate also affected the level of evidence,52 53 resulting in level 2B studies (n=2) and level 2C (n=1) (see table 3).

Furthermore, blinding of participants was in all studies not possible due to the nature of the intervention. To conclude, limited information was available on the reliability and validity of used data collection methods.

Presence of FGDM key elements in family-centred interventions

To examine the presence of the three key elements of FGDM in family-centred interventions in adult care and welfare (see table 4); all six articles of the three studies were 100% double screened (CFH/EWMS) and resulted in a good inter-rater agreement (83% consensus with kappa of 0.67).55 64 65 Some differences in agreement were found in labelling the key elements of the Joling et al study.60–62 It was not instantly clear if the study was family driven and had a set-up of a plan with actions/goals.

Table 4

Overview of eligible studies to the FGDM key elements

The family meetings intervention consisted of two individual sessions with the primary caregiver and four family counselling sessions including family members and friends.60–62 Aims were to improve emotional and instrumental support given to the patient by relieving the burden of caregiving through mobilising the existing family networks of the patient and primary caregiver; teach problem-solving techniques as well as offer psychoeducation to the primary caregiver. The counsellor (professional) led the family meetings in a directing role. Consequently, we cannot speak of a family-driven approach. Although the extended family is mobilised, increasing tasks, commitment and family support, it was not clarified what exactly the support consisted of. Therefore, we examined additional information in a manual written for family meetings counsellors.66 In the manual, a plan was mentioned but the ownership of the plan remained with the professionals. They set the agenda, leading the meeting. In summary, although the set-up of an action plan and three phases of the meetings could be identified (preparatory meetings with the caregiver, family meetings with family and friends and an evaluation meeting), the study did not meet with the FGDM key element of a family-driven approach.

de Havenon et al63 compared audio-visual family meetings versus in-person family meetings with both a set-up initiated by professionals to enhance discourse and decision-making with the patients family members to facilitate medical decisions. Consequently, we cannot speak of a family-driven approach. The other FGDM key elements (action plan, three phases) were also not present.

The Malmberg-Heimonen and Johansen study41 42 implemented a FGC model with the following characteristics: (1) it is the participant’s meeting, he/she is in charge and decides who is invited; (2) the participant gets assistance from an independent FGC facilitator (not employed by the same organisation); (3) the extended network of the participant is invited; (4) private time: extended network discusses alone in the second part of the meeting without members of public organisations; (5) The FGC process results in an action plan.

One meeting is divided into three parts: (1) information part; (2) the meeting between the participant and his or her private network; (3) the concluding part. In summary, we can speak of a family-driven intervention where all FGDM key elements are met.

Evaluation of the effectiveness of FGDM interventions

For the purpose of evaluating the effectiveness of FGDM interventions, only two papers of the Malmberg-Heimonen and Johansen study were included,41 42 since the Joling et al and de Havenon et al studies were excluded for having not all60–62 or none63 of the FGDM key elements present. The study described the effectiveness of FGC among social welfare recipients regarding the outcomes life satisfaction, social support, mental distress, anxiety and depression and employment.41 42 Although there were significant increases regarding the outcome measures of the intervention group after a follow-up period of 16–23 weeks; the 1-year follow-up identified neutral effects of the intervention compared with the control group. It can be concluded that the positive effect faded away with time.41

Discussion and implications for future research

The first aim of this review was to systematically review the presence of the three key elements of FGDM26 (plan with actions/goals, family driven, three phases of meetings) in family-centred interventions in adult care and welfare. An extensive search of 1680 articles from 14 databases resulted in six articles on three studies of family-centred interventions. One study41 42 possessed all the key elements required to qualify as FGDM. Of the two other family-centred interventions,60–63 key elements of FGDM (in particular the family-driven element) were missing. This might be exemplary for family meetings in adult healthcare/welfare. Although the patient and family may be engaged in the decision-making process, the way these meetings are organised fails to really put the patient and family in the drivers seat. Often there are also multifactorial reasons related to decision-making processes such as personal factors as well as, clinician/professional and organisational factors.67 More research is needed in the requirements and the desire to ‘upgrade’ these meetings to a next level from a collaborative to a family-driven model where the professional remains low profile and supports the family in order for them to take over the executive role.

The second aim was to evaluate the effectiveness of FGDM interventions in adult healthcare/welfare. While the one intervention study41 42 described short-term effectiveness of FGDM among social welfare recipients, this effect was neutralised after the 1-year follow-up.41 Reciprocal interaction seemed difficult to maintain over time, indicating the importance of offering follow-up meetings shortly after the first FGC, sharing responsibility and accountability to fulfil the action plan.38 41

The lack of rigorous studies of effectiveness of FGDM in adult healthcare/welfare limited answering the second aim. Regardless, this evaluation still adds to our body of knowledge of FGDM in adult healthcare and welfare. Primarily, it highlights the need for high-quality controlled trials. To gain further understanding in the reasoning behind the lack of outcome research, it might be wise to look at the development of FGDM outcome research conducted in child healthcare, which has overcome its own share of obstacles. First, we need to recognise that outcome research of FGDM interventions only started off in the last decennia. It was not until the year 2000 when one of the first controlled studies68 was published internationally, comparing FGC intervention versus regular treatment.21 Second, multiple studies reported challenges detecting the efficacy of FGDM through time.69 Researchers faced difficulties including enough family group conferencing cases to evaluate, resulting in a small sample size.21 Furthermore, there was a lack of ability to follow people and their outcomes over a long period of time, this made them unable to measure the long-term effect.69 Third, researchers faced challenges conducting high-quality controlled trials using scientific research methods in a socially complex environment with many unpredictable influencing factors as is often the case in a clinical practice setting.69 70 Lastly, the only conducted meta-analysis of FGDM in child care included 14 controlled trials from 2000 to 2016, overall showing no overall effectiveness in child-related outcomes.7 The majority of the included articles were low in quality and were limited in information on programme fidelity. The authors stressed the importance of gathering information on programme fidelity making it able to understand a lack of effect due to possible poor implementation.7 Indeed, this review also both reveals the urgency of high-quality controlled trials but simultaneously highlights the need to evaluate their implementation to gain more understanding of influencing factors such as possible barriers and facilitators. Exploring experiences and opinions of both patients, family and professionals who are already participating in the FGDM-meetings is essential and highly needed. The fact is that despite the low quantity and evidence of FGDM outcome research, there is a widespread use and enthusiasm for the implementation of FGDM worldwide in more than thirty countries to date.1 These future directions in research can accelerate new comprehension and acknowledgement that shifting the balance of power from professional towards the person in need and their family is most essential in democratising healthcare and welfare.


Several limitations to this systematic review need to be acknowledged. First is the inability to assess studies written in languages other than English, leading to potential selection bias. Nevertheless, we think that the majority of the studies could be assessed, only a small amount of studies were written in a language different than English.

The second limitation concerns the operationalisation used to identify the FGDM meetings.

We only identified FGDM studies when demonstrating meetings involving extended family in the decision making process with the three key elements of FGDM present ((1) plan, (2) family driven, (3) three phases). We have made the decision to integrate the most described key elements with general consensus found in literature, which can frame the outcome of the review. Third, all extracted studies scored low on their quality assessment score and the heterogeneousness was high in both types of population, intervention, context and outcomes. This made mutual comparison of the studies difficult.

Last, it was quite a challenge to identify the intervention eligibility criteria of the articles. The majority of the articles provided limited information about the actual intervention; there was often a lack of detail reporting specific characteristics and content of the intervention. For example, no articles were found to have implemented the Template for Intervention Description and Replication (TIDieR) guidelines,71 a 12-item template to explain and elaborate on the intervention improving quality.


This is the first review to systematically differentiate FGDM interventions from other family-centred interventions and evaluate its effectiveness within the adult population. Although the studies are still low in both quantity and quality, we have gained an understanding of the differences of FGDM and family-centred intervention studies by means of the identification of the FGDM key elements. The majority of the extracted studies (2/3) are still using the collaborative model of engaging patients and family in the decision-making process instead of the family-driven perspective whereby the professional remains low profile and supports the family in order for them to take over the executive role. These results might be exemplary for a lot of family meetings in the adult healthcare/welfare.

We could not sufficiently answer our second aim regarding the effectiveness of FGDM among adults due to the lack of outcome studies in the adult healthcare/welfare. Further high-quality intervention studies are required to evaluate the impact of FGDM on adults in need, including their families. Moreover, insight into barriers and facilitators influencing FGDM will support our understanding how to empower persons in need, further democratising healthcare and welfare.


We are grateful to Sana Fazel, Neuroscience and Experimental Psychology at University of Sussex, Brighton, England, for the English proofreading of our manuscript.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68.
  69. 69.
  70. 70.
  71. 71.
View Abstract


  • Contributors CFH developed the first version of the manuscript, in collaboration with EWMS, JMAV-M, MWMP and MK. JMAV-M, MWMP and MK supervised the writing process and provided feedback on several drafts of the manuscript. EWMS and MK executed the double screening. JMAV-M, MWMP and MK provided methodological advice. CFH wrote the final version of the manuscript.

  • Funding This study is financially supported by ZonMw, the Netherlands Organisation for Health Research and Development, Fonds Nuts Ohra and Revalidatiefonds, grant number: 630000003.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The data that support the findings of this study are available from the corresponding author, (MK), upon reasonable request.

  • Patient consent for publication Not required.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.