Article Text

Original research
Shared decision-making in advanced kidney disease: a scoping review
  1. Noel Engels1,2,3,
  2. Gretchen N de Graav3,
  3. Paul van der Nat4,
  4. Marinus van den Dorpel3,
  5. Anne M Stiggelbout5,
  6. Willem Jan Bos2,4
  1. 1Department of Shared Decision-Making and Value-Based Health Care, Santeon, Utrecht, The Netherlands
  2. 2Internal Medicine, Leiden University Medical Center, Leiden, The Netherlands
  3. 3Internal Medicine, Maasstad Hospital, Rotterdam, the Netherlands
  4. 4Department of Value-Based Health Care, Sint Antonius Hospital, Nieuwegein, The Netherlands
  5. 5Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, The Netherlands
  1. Correspondence to Noel Engels; n.engels{at}santeon.nl

Abstract

Objectives To provide a comprehensive overview of interventions that support shared decision-making (SDM) for treatment modality decisions in advanced kidney disease (AKD). To provide summarised information on their content, use and reported results. To provide an overview of interventions currently under development or investigation.

Design The JBI methodology for scoping reviews was followed. This review conforms to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) checklist.

Data sources MEDLINE, Embase, Web of Science, Cochrane Library, Emcare, PsycINFO, PROSPERO and Academic Search Premier for peer-reviewed literature. Other online databases (eg, clinicaltrials.gov, OpenGrey) for grey literature.

Eligibility for inclusion Records in English with a study population of patients >18 years of age with an estimated glomerular filtration rate <30 mL/min/1.73 m2. Records had to be on the subject of SDM, or explicitly mention that the intervention reported on could be used to support SDM for treatment modality decisions in AKD.

Data extraction and synthesis Two reviewers independently screened and selected records for data extraction. Interventions were categorised as prognostic tools (PTs), educational programmes (EPs), patient decision aids (PtDAs) or multicomponent initiatives (MIs). Interventions were subsequently categorised based on the decisions they were developed to support.

Results One hundred forty-five interventions were identified in a total of 158 included records: 52 PTs, 51 EPs, 29 PtDAs and 13 MIs. Sixteen (n=16, 11%) were novel interventions currently under investigation. Forty-six (n=46, 35.7%) were reported to have been implemented in clinical practice. Sixty-seven (n=67, 51.9%) were evaluated for their effects on outcomes in the intended users.

Conclusion There is no conclusive evidence on which intervention is the most efficacious in supporting SDM for treatment modality decisions in AKD. There is a lot of variation in selected outcomes, and the body of evidence is largely based on observational research. In addition, the effects of these interventions on SDM are under-reported.

  • nephrology
  • internal medicine
  • end stage renal failure

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

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.

Strengths and limitations of this study

  • The search queries for this scoping review were conducted without time period restrictions and generated comprehensive results covering all possible interventions that support SDM for treatment modality decisions in AKD.

  • Two reviewers independently used predeveloped charting and data-extraction tables to screen the literature, select records for inclusion and extract the relevant data.

  • The interventions identified in the included records are presented based on the decisions they were developed to support, after which information is provided on their content, format, evidence and availability.

  • Included records were not formally assessed for quality; potential risks of bias in the reported outcomes remain undetermined.

  • Interventions and/or findings from records not written in English, or records inaccessible due to subscription limitations or internet protocol address geo-blocking, are not reported.

Introduction

Over 2 million patients with kidney failure currently rely on kidney replacement therapy (KRT) to stay alive.1 This number has been estimated to double by 2030,2 and many patients with advanced kidney disease (AKD) will have to make treatment modality decisions as their kidneys deteriorate over time.

Guidelines on the management of chronic kidney disease (CKD) emphasise the importance of timely kidney failure treatment modality education and decisional support as patients progress to the more advanced stages of kidney disease.3 4 Delays in the decision-making process can result in suboptimal dialysis initiation, which is associated with increased patient morbidity, mortality and healthcare costs.5

Shared decision-making (SDM) has been recognised as the preferred model to help patients with AKD understand their treatment options, and make informed decisions that align with their values and preferences.6 7 SDM requires that patients and clinicians proactively engage in a collaborative decision-making process.8–10 This process should be characterised by deliberation, during which patients become aware of their choice, understand all of their options and get to consider what matters most to them. A three-step framework has been developed to help guide this decision-making process with the following conversational steps in clinical practice: (1) team talk, (2) option talk and (3) decision talk.10 In addition, educational programmes (EPs) and decision support interventions such as patient decision aids (PtDAs) and prognostic tools (PTs) can be used to support deliberation and help patients and clinicians engage in SDM.10 Multiple efforts have been made to foster SDM across the international healthcare community,11 12 but there are still signs that patients experience a low degree of SDM,13 and efforts to incentivise SDM risk being limited to the promotion of PtDAs.14 15 A broader sense of awareness and knowledge of SDM is needed for it to become widely implemented,8 and stakeholders should share their experiences to speed up this process.16

We previously set out to write a scoping review on interventions that can support SDM for treatment modality decisions in AKD after we showcased a lack of a comprehensive overview of these interventions in the literature.17 We performed an additional preliminary search on the MEDLINE database prior to the conduct of this review and did not identify previous scoping reviews on the topic aligning to the same concept. We did identify a scoping review on the information available for clinicians counselling older patients with kidney failure,18 a systematic review on PtDAs developed to support SDM between dialysis and conservative care management (CCM) pathways,19 a scoping review on predialysis EPs,20 a systematic review on PTs developed to predict kidney failure21 and a Cochrane review on the effects of PtDAs in people facing treatment or screening decisions.22

We conducted this scoping review to provide clinicians, researchers and other stakeholders with one comprehensive, but digestible source of information on interventions that can support SDM for treatment modality decisions in AKD. An overview of interventions currently under development or investigation is also provided. We hope that this review will facilitate the future implementation of SDM in clinical practice, as well as stimulate development and research on new and effective interventions by exploring and defining knowledge gaps on the subject.

Methods

We followed the JBI methodology for scoping reviews23 and our scoping review protocol17 when we conducted this scoping review. Our objectives, research questions and methods are specified in our protocol (see online supplemental appendix 1). In addition to the protocol, we also used: (1) more detailed inclusion criteria during the screening and inclusion process and (2) the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) checklist24 (see online supplemental appendix 2) when we completed the review. No other changes were made in the methodology described in our protocol.

Objectives

In brief, our objectives were to provide:

  1. a comprehensive overview of interventions that can support SDM for treatment modality decisions in AKD;

  2. summarised information on their contents, use and reported results;

  3. an overview of interventions currently under development or investigation.

Inclusion criteria

We searched the peer-reviewed and grey literature for records on interventions that support SDM for treatment modality decisions in AKD. We considered any intervention in standard care that can support deliberation and/or help patients and clinicians engage in SDM (eg, EPs, PtDAs, PTs) an SDM intervention. No time period restriction was used in an effort to be as comprehensive as possible. Records were eligible for inclusion if they were written in English, and if the study population consisted of patients >18 years of age with an estimated glomerular filtration rate <30 mL/min/1.73 m2. Records had to be on the subject of SDM or explicitly mention that the reported interventions could be used to support SDM for treatment modality decisions in AKD. Records that reported on interventions that could clearly be used to support SDM without explicitly mentioning it were also included.

Exclusion criteria

We excluded records if:

  • they only reported on interventions for advance care planning;

  • they only reported on interventions for the withdrawal of treatment.

Search methodology

We performed a three-step search strategy as explained in the JBI methodology for scoping reviews23 and in our review protocol17 (see online supplemental appendix 1). We searched MEDLINE, Embase, Web of Science, Cochrane Library, Emcare, PsycINFO, PROSPERO and Academic Search Premier for peer-reviewed literature. We searched OpenGrey, researchgate.net, clinicaltrials.gov, europepmc.org, Google Scholar and websites of the Kidney Disease Improving Global Outcomes Association, the Renal Physicians Association, the American Society of Nephrology, the Canadian Society of Nephrology, the National Institute of Health and Care Excellence, the European Renal Association—European Dialysis and Transplant Association, the Kidney Health Australia—Caring for Australians with Renal Impairment Association and the Ottawa Hospital Research Institute for grey literature.

A research librarian generated the search queries (see online supplemental appendix 3). The results were uploaded in RefWorks V.2.0.

Record selection and data extraction

We used previous publications, the International Patient Decision Aids Standards (IPDAS) minimum standards criteria25 and the Standards for UNiversal reporting of patient Decision Aid evaluation (SUNDAE) checklist26 to design charting and data-extraction tables used for record selection and data extraction. Two reviewers (NE and GNdG) independently performed the process of record selection and data extraction. Disagreements were resolved by discussion or consultation with the research team (PvdN, MvdD, WJB, AMS).

We initially screened the titles and abstracts of all identified records after which the charting table was used to register records selected for full-text analysis in Microsoft Excel V.16. We then performed full-text analysis of the selected records during which a final selection was made for data extraction. We also screened the references of this selection for additional records on the subject.

Selected records were categorised based on record type and on their scope and context as mentioned by the authors and developers. We categorised the interventions we identified in these records based on whether these interventions were PTs, EPs or PtDAs. Interventions were categorised as multicomponent initiatives (MIs) when two or more of these interventions were combined to support patients with AKD in treatment modality decisions, or implement SDM in clinical practice. We subsequently categorised the identified interventions based on the decisions they were developed to support.

Extracted data included: primary author, developer, date of publication, country of origin, type of record, study population/target demographic, study aims, study methods, sample size, study arms, intervention, format and context of the intervention, contents of the intervention, patient participation in development, comparator, study outcomes, reports on outcomes of SDM, use of International Consortium for Health Outcomes Measurement (ICHOM)27 or Standardised Outcomes in Nephrology (SONG)28 outcomes, main findings, implemented in clinical practice, recruitment status, date of completion and/or publication.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Results

Figure 1 illustrates a flow chart of the screening and inclusion process. We conducted the final search query in February 2021. We identified 1512 records and included a total of 158 records. Records were excluded because they were on another subject (n=1215, 80.3%), not available (n=127, 8.4%), not in English (n=57, 3.8%), duplicates (n=34, 2.2%), reviews (n=28, 1.9%) on the wrong population (n=27, 1.8%) or protocols for completed studies (n=24, 1.6%).

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) flow chart of screening and inclusion process.

Figure 2 illustrates the included records stratified by type, scope and context. The majority of these records are observational (n=68, 43.0%) and experimental studies (n=39, 24.7%). A smaller proportion are study protocols (n=17, 10.8%), meeting abstracts (n=16, 10.1%), mixed-methods studies (n=12, 7.6%) and websites (n=6, 3.8%). Most records report on EPs (n=62, 39.2%), followed by PTs (n=42, 26.6%), PtDAs (n=37, 23.4%) and MIs (n=17, 10.8%).

Figure 2

Included records stratified by record type, scope and context.

We identified 145 interventions in the included records. Figure 3 illustrates the distribution of these interventions. The majority of these records are PTs (n=52, 35.9%) and EPs (n=51, 35.2%), followed by PtDAs (n=29, 20.0%) and MIs (n=13, 8.9%). Some of these interventions were only identified in meeting abstracts (n=14, 9.7%). A minority were novel interventions that we identified in study protocols (n=16, 11.0%). Figures 4 and 5 illustrate the implementation and evaluation rates of the identified interventions. About one-third of the interventions (n=46, 35.7%) were reported to have been implemented in clinical practice. About half of the interventions (n=67, 51.9%) were evaluated for their effects on outcomes in the intended users. PTs were the interventions with the least information on implementation status and were the least evaluated interventions, followed by PtDAs, EPs and MIs. Interventions were generally evaluated on health-related outcomes and on knowledge, decisional quality, communication and patient activation. Patients that were exposed to the interventions generally had better outcomes than patients that were not exposed to the interventions.

Figure 3

Distribution of the different types of interventions.

Figure 4

Stacked bar count of the interventions stratified by implementation status. *Interventions currently under investigation (n=16, 11.0%) are not shown here.

Figure 5

Stacked bar count of the interventions stratified by evaluation status. *Interventions currently under investigation (n=16, 11.0%) are not shown here.

Prognostic tools

We identified 52 PTs. All PTs were identified in peer-reviewed articles.29–59

Table 1 provides an overview of the identified PTs with their characteristics and performance metrics. Table S1 in online supplemental appendix 4 provides additional details on the identified PTs (eg, sources for publicly available PTs). Nineteen PTs predict the risk of progression to kidney failure (no. 1–19) and help patients and clinicians decide whether or not patients should start with preparations for kidney failure. One PT also predicts the risk of cardiovascular disease and death (no. 19). Twenty-eight PTs predict the risk of death after starting dialysis (no. 20–47) and help patients and clinicians decide whether or not patients should choose to start dialysis. Two PTs predicts the risk of death after starting CCM (no. 48, 49) and help patients and clinicians decide whether or not patients should choose to start CCM. One PT predicts and compares the risk of death after starting dialysis or transplantation (no. 50) and helps patients and clinicians decide between dialysis and transplantation options. One PT predicts the risk of deceased donor kidney graft failure (no. 51) and helps patients and clinicians decide whether or not patients should accept a deceased donor kidney transplantation offer. One PT predicts the risk of living donor kidney graft failure (no. 52) and helps patients and clinicians decide whether or not patients should accept a living donor kidney transplantation (LDKT) offer.

Table 1

Overview of the identified PTs, their characteristics and performance metrics

A relatively large proportion (n=19, 36,5%) of the identified PTs were developed to be used in elderly patients with AKD (no. 4, 25, 26, 28–31, 33, 34, 40–49). The remaining PTs can be used in the general population of patients with AKD.

The majority of PTs (n=32, 61.5%) are publicly available as formulas (no. 2–5, 7–19, 22, 24, 35–39, 46–52), eight of which (no. 2, 3, 19, 38, 39, 50–52) can be used on interactive websites. One of these PTs (no. 50) has been designed as a PtDA. Point-based scoring systems (no. 1, 20, 21, 25–34, 45) and nomograms (no. 44) were also used, although less frequently (n=14, 26.9%, n=1, 1.9%). A minority of PTs (n=6, 11.5%) are not publicly available (no. 6, 23, 40–43) and depend on computer software to be used.

Not all PTs were completely validated (assessed for performance) during development. About a quarter (n=11, 21.2%) were not evaluated on calibration outcomes (no. 22, 23, 26, 30, 34, 40–43, 51, 52), and some (n=2, 3.8%) were not validated at all (no. 20, 21). Most of them (n=37, 71,2%) were developed and validated with the same cohort of patients (no. 1, 6, 7, 9–18, 22–26, 28–33, 40–52). A quarter (n=13, 25.0%) were developed and validated with different patient cohorts (no. 2–5, 8, 19, 27, 34–39). The discriminatory power of the PTs was generally acceptable, one-fourth (n=13, 25.0%) had C-statistics below 0.7 on all, or a subset of, predictions (no. 9, 28–30, 34, 39, 41, 43, 46, 48, 50–52). The remaining PTs had better C-statistics.

Table 2 provides an overview of the PTs (no. 2–5, 19, 20, 25, 26, 30, 33, 51, 52) that were externally validated in independent external validation studies.31 32 46 55 60–69 Table S2 in online supplemental appendix 4 provides additional details on these external validation studies. One PT (no. 2) was externally validated in six different studies,32 60–63 67 two other PTs (no. 3, 25) were externally validated in three31 62 67 and four different studies,46 55 64 66 respectively. The other nine (no. 4, 5, 19, 20, 26, 30, 33, 51, 52) were externally validated less frequently.

Table 2

Overview of PTs validated in independent external validation studies

The majority of the PTs (n=10, 83.3%) were externally validated in different patient populations (no. 2, 3, 4, 5, 19, 20, 25, 26, 30, 33) than the ones they were developed for. Three PTs had poor discriminatory power in these patient populations (no.2, 30, 33), with C-statistics between 0.5 and 0.6.63 66 Performance metrics were generally comparable between developmental and external validation studies when similar patient populations were used (see tables 1 and 2). Only one PT (no. 50) was reported to have been implemented in clinical practice. This PT was designed as a PtDA57 and is in that regard the only PT that has been evaluated for its effects on outcomes in the intended users.70

Educational programmes

We identified 41 EPs (excluding ten currently under investigation). Thirty-five were identified in peer-reviewed articles.71–104 Six were identified in meeting abstracts found in the grey literature.105–110

Table 3 provides an overview of the identified EPs and their characteristics. Table S3 in online supplemental appendix 4 provides additional details on the identified EPs (eg, sources for publicly available EPs). One EP was developed to promote peritoneal dialysis (PD) and helps patients choose whether to start with PD or not (no. 1). Eleven EPs help patients choose between dialysis options (no. 2–11), with two promoting a particular treatment modality (no. 5, 11). Ten help patients choose between dialysis and transplantation options (no. 12–21), with two promoting a particular treatment modality (no. 13, 14). Seven EPs were developed to promote LDKT (no. 2–28) and help patients decide whether to pursue LDKT or not. Two EPs help patients choose between dialysis and CCM options (no. 29, 30). Seven help patients choose between transplantation, dialysis and CCM options (no. 31–37), one of which promotes home therapy modalities (no. 37). Four EPs help patients choose between transplantation options (no. 38–41).

Table 3

Overview of the identified EPs and their characteristics

Most EPs were developed for the general population of patients with AKD, however some (n=5, 12.2%) were specifically developed for Hispanic and African-American patients (no. 20, 26–28, 40), and some (n=3, 7.3%) were specifically developed for suboptimal dialysis initiation patients (no. 13, 14, 31).

About one-third of the EPs (n=14, 34.1%) consist of a single medium format (no. 1, 6–11, 17, 20, 21, 28, 35–37). The remaining programmes consist of a combination of different medium formats. About half of the EPs (n=20, 48.8%) use coaches to guide patients through the programme (no. 3, 12–14, 18, 19, 22, 23, 25, 27, 28, 30–35, 39–41). A minority (n=7, 17.0%) were developed with the input of patients (no. 1, 10, 20, 26, 38–40) and even less (n=3, 7.3%) describe a reading level (no. 25, 39, 40). Only a few EPs (n=2, 4.9%) are publicly available (no. 20, 32).

More than half of the EPs (n=26, 63.4%) were reported to have been implemented in clinical practice (no. 2–6, 8, 11–19, 21, 26–34, 36). All but one (no. 33) have been evaluated for their effects on outcomes in the intended users.

Table 4 provides an overview of the studies71–86 88–118 that evaluated the identified EPs. Table S4 in online supplemental appendix 4 provides additional details on these studies. The majority of these EPs (n=19, 47.5%) were evaluated in experimental studies (no.1, 3, 5, 7, 16, 19, 20, 22–27, 29, 35, 38–41), more than half of which (n=14, 73,7%) were randomised controlled trials (RCTs).71 74 86 88 90–92 101–103 111–113 115 Less (n=16, 40.0%) were evaluated in observational studies (no.2, 4, 6, 11–15, 17, 18, 21, 28, 30–32, 34), a minority of which (n=4, 25.0%) were prospective cohort studies.76 94 95 117 Five (n=5, 12.5%) EPs (no.8–10, 36, 37) were evaluated in studies presented in meeting abstracts.105–107 109 110

Table 4

Overview of studies evaluating the identified EPs

EPs were generally evaluated for their effects on health-related outcomes and on knowledge, communication and patient activation. None of the EPs were evaluated for their effects on SDM. Thirteen (n=13, 31.7%) EPs (no.1, 8, 9, 10, 11, 21, 34, 35, 36, 37, 39, 40, 41) were evaluated in studies published after the standardised outcome sets for CKD, dialysis and transplantation were published by ICHOM and SONG. None of these EPs were evaluated with these outcomes. EPs that promote particular treatment modalities (no.1, 5, 11, 13, 14, 22–28) appear to increase the number of patients planning to start with the promoted modalities (see table 4). Patients exposed to EPs generally had more favourable health-related outcomes than patients that were not exposed to EPs (see table 4). They were also more knowledgeable about their treatment options, better equipped to communicate about their treatment options and more active in choosing and requesting a preferred treatment modality (see table 4).

Patient decision aids

We identified 27 PtDAs (excluding two currently under investigation). Fourteen were identified in peer-reviewed articles.57 119–131 Thirteen were identified in the grey literature,132–141 seven of which in meeting abstracts.132–135

Table 5 provides an overview of the identified PtDAs and their characteristics. Table S5 in online supplemental appendix 4 provides additional details on the identified PtDAs (eg, sources for publicly available PtDAs). One PtDA helps patients choose whether or not they should start with dialysis (no. 1) and one helps them decide when to start with dialysis if they decide to do so (no. 2). Nine help patients choose between dialysis options (no. 3–11) and five help them choose between transplantation and dialysis options (no. 12–16). One PtDA helps patients decide whether or not they want to accept an infectious risk donor kidney donation offer (no. 17). Four help patients choose between dialysis and CCM options (no. 18–21). Six help patients choose between transplantation, dialysis and CCM options (no. 22–27).

Table 5

Overview of the identified PtDAs and their characteristics

Most PtDAs were developed for the general population of patients with AKD, a minority (n=3, 11.1%) were specifically developed for elderly patients with AKD (no.18, 19, 21).

A large proportion of PtDAs (n=23, 85.2%) consist of a single medium format (no.1–8, 10–17, 19–23, 26, 27). Most of these PtDAs (n=11, 47.8%) are interactive websites (no.1–5, 11–14, 17). The remaining PtDAs consist of a combination of different medium formats. A majority of the PtDAs (n=17, 62.9%) contain values-clarification and preference-elicitation exercises (no.1–6, 8–11, 15, 18–20, 23–25).

Two-thirds of the PtDAs (n=18, 66.6%) were developed with the input of patients (no.3, 5–9, 11–14, 16, 19, 20, 22, 24–27), one of which was largely developed with the input of African-American patients (no.24). A minority (n=6, 22.2%) describe a reading level (no.5, 6, 16, 18, 24, 27). Only a few PtDAs (n=5, 18.5%) are not publicly available (no.7, 8, 18, 21, 24).

Ten PtDAs (n=10, 37.0%) were reported to have been implemented in clinical practice (no.6–9, 13, 15, 16, 25–27). The majority (n=17, 62.9%) have been evaluated for their effects on outcomes in the intended users (no.6–11, 13, 14, 16–18, 21, 22, 24–27).

Table 6 provides an overview of the IPDAS minimum standards component scores for 26 of the identified PtDAs. Table S6 in online supplemental appendix 4 provides these scores in greater detail. One PtDA (no. 21) could not be scored according to these criteria because the accompanying documentation did not provide enough information. Decision support interventions have to meet six qualifying criteria to qualify as PtDAs. Just about half (n=13, 48.1%) met all qualifying criteria (no.1, 4–7, 9, 11, 14, 18–20, 23–25). The sixth qualifying criterium (‘the PtDA describes what it is like to experience the consequences of the options’) was the least met criterium by the other PtDAs.

Table 6

IPDAS minimum standards component scores of the identified PtDAs

Table 7 provides an overview of the studies70 121–127 129 130 132 135 142–145 that evaluated 17 PtDAs (no.6–11, 13, 14, 16–18, 21, 22, 24–27). Table S7 in online supplemental appendix 4 provides additional details on these studies. Most (n=8, 47.0%) were evaluated in experimental studies (no.6, 11, 13, 17, 18, 21, 24, 25), three-quarter of which (n=6, 75.0%) were RCTs.70 124 126 127 129 145 Five (n=5, 29.4%) PtDAs (no.7, 8, 16, 26, 27) were evaluated in studies presented in meeting abstracts.132 135 The remaining four (n=4, 23,5%) PtDAs (no.9, 10, 14, 22) were evaluated in observational studies142 144 and mixed-methods studies,122 123 125 130 143 two of which included pilot evaluations.122 125

Table 7

Overview of studies evaluating the identified PtDAs

PtDAs were generally evaluated for their effects on health-related outcomes, and on knowledge, decisional quality and patient activation. Only one PtDA (no.7) was evaluated for its effects on SDM.122 143 One meeting abstract presented a study that evaluated whether SDM scores differed among hospitals if PtDAs (no.16, 26, 27) were used or not.135 Ten (n=10, 37.0%) PtDAs (no.9, 10, 11, 13, 16, 18, 21, 24, 26, 27) were evaluated in studies published after the standardised outcome sets for CKD, dialysis and transplantation were published by ICHOM and SONG. None of these PtDAs were evaluated with these outcomes. Patients that used PtDAs were generally more knowledgeable about their treatment options, and had better scores of decisional quality and patient activation than patients that did not use PtDAs (see table 7). The two studies that evaluated the PtDA (no.7) on outcomes of SDM showed that patients experienced the intervention as SDM (see table 7). The meeting abstract that presented a study that evaluated whether SDM scores differed among hospitals if PtDAs were used or not showed that hospitals that used these PtDAs (no.16,26,27) generally had better scores for SDM compared with hospitals that did not use them (see table 7). However, these differences were not significant (see table 7).

Multicomponent initiatives

We identified nine MIs (excluding four currently under investigation). Eight were identified in peer-reviewed articles,146–155 one was identified in two meeting abstracts found in the grey literature.156 157

Table 8 provides an overview of the identified MIs and their characteristics. Table S8 in online supplemental appendix 4 provides additional details on the identified MIs (eg, sources for publicly available MIs). One MI was developed to promote PD (no.1) and helps patients choose whether to start with PD or not. One MI as developed to help patients choose between dialysis options (no.2). Two MIs help patients choose between transplantation and dialysis options (no.3,4), one of which promotes transplantation and strives to reduce racial disparities in access to kidney transplantation (no.3). Five help patients choose between transplantation, dialysis and CCM options (no.5–9), with three promoting a particular treatment modality (no.5,7,9). Almost all MIs were developed for the general population of patients with AKD, one (no.8) was specifically developed for suboptimal dialysis initiation patients.

Table 8

Overview of the identified MIs and their characteristics

All MIs include an educational component for patients. One MI also educates clinicians (no.3). Two-thirds (n=6, 66.7%) include decision support interventions or other tools that help patients make values-based and preferences-based decisions (no.2, 3, 6–9). One MI (no.3) includes a previously identified PtDA (no.13) and components of a previously identified EP (no.26). The majority (n=7, 77.8%) of MIs (no.2–4, 6–9) use coaches to support patients and guide them through the programme. Almost half of the MIs (n=4, 44.4%) were developed as quality improvement initiatives and include nationwide, or facility level, policy and protocol changes (no.1, 3, 5, 8). Some MIs (n=3, 33.3%) were specifically developed to implement SDM in clinical practice (no.4, 6, 9).

Only one MI (no.4) was developed with the input of patients. None contain reading level information. Only one MI (no.3) consists of components that are publicly available.

All but one MI (no.4) were reported to have been implemented in clinical practice. All MIs have been evaluated for their effects on outcomes in the intended users, most of which (n=8, 88.9%) were patients (no.1, 2, 4–9). One MI (no.3) was evaluated for its effects on a dialysis facility level.

Table 9 provides an overview of the studies146–157 that evaluated the identified MIs. Table S9 in online supplemental appendix 4 provides additional details on these studies. Two-thirds of MIs (n=6, 66.6%) were evaluated in observational studies (no.1, 5–9), three of which (n=3, 33.3%) were prospective cohort studies.146 149–153 155 A minority (n=2, 22.2%) were evaluated in experimental studies (no.3, 4), only one of which was an RCT.147 One MI (no. 2) was evaluated in two sequential studies presented in two meeting abstracts.156 157

Table 9

Overview of studies evaluating the identified MIs

MIs were generally evaluated for their effects on health-related outcomes and on patient activation. None of the MIs were evaluated for their effects on SDM. Four (n=4, 44.4%) MIs (no.2, 3, 4, 9) were evaluated in studies published after the standardised outcome sets for CKD, dialysis and transplantation were published by ICHOM and SONG. None of these MIs were evaluated with these outcomes. Patients exposed to MIs generally had more favourable health-related outcomes (see table 9). They were also more active in choosing and requesting treatments (see table 9). MIs that promote particular treatment modalities (no.1, 3, 5, 7, 9) appear to increase the number of patients planning to start with these modalities (see table 9). The MI that was evaluated for its effects on a dialysis facility level (no.3) reported significant changes in the proportion of patients referred for transplantation147 and high fidelity to the intervention.154 It also appeared to reduce racial disparities in access to kidney transplantation.147

Interventions currently under development or investigation

We identified 16 interventions currently under investigation. The majority of these interventions (n=12, 75.0%) were identified in the grey literature,158–169 the remaining interventions (n=4, 25.0%) were identified in published protocol papers.170–173

Table 10 provides an overview of these interventions. Table S10 in online supplemental appendix 4 provides additional details on these interventions. More than half (n=10, 62.5%) are EPs (no.1, 2, 4, 5, 7–11, 16), one-quarter (n=4, 25.0%) are MIs (no.3, 6, 14, 15) and one-eighth (n=2, 12.5%) are PtDAs (no.12,13). Four interventions help patients choose between dialysis options (no.1–4), two of which promote a particular treatment modality (no.1, 4). Seven interventions promote LDKT and help patients choose whether or not to pursue LDKT (no.5–11). Two help patients choose between dialysis or CCM options (no.12, 13). Only one helps patients choose between transplantation, dialysis and CCM options (no.14). Two help patients choose between transplantation options (no.15, 16).

Table 10

Overview of interventions currently under development or investigation

A quarter of these interventions (n=4, 25.0%) included components of previously identified interventions (no.8, 11, 14, 16), and two (n=2, 12.5%) are modifications (no.6, 7) of another intervention under investigation (no.5). Two interventions (n=2, 12.5%) have been previously evaluated and are undergoing additional evaluation (no.9, 10).

Half of the interventions (n=8, 50.0%) have been developed for specific patient populations, most of which (n=5, 62.5%) were developed for African-American, Hispanic or non-white race patients (no.5–8,11). Less (n=3, 37.5%) were developed for elderly patients (no.2, 12, 13). The remaining interventions can be used by the general population of patients with AKD.

Table 11 provides an overview of the studies158–173 evaluating these interventions. Table S10 in online supplemental appendix 4 provides additional details on these studies. Almost all of the interventions (n=15, 93.8%) will be evaluated in experimental studies, the majority (n=14, 93.3%) of which (no.1–9, 11, 12, 14–16) in RCTs.158–164 166 168–171 173 Selected outcomes include health-related outcomes, knowledge, communication, patient activation and decisional quality. None of the authors report evaluating these interventions for their effects on SDM. Nine (n=9, 56.3%) interventions (no.2, 3, 4, 8, 10, 11, 12, 14, 16) were identified in registries or protocol papers published after the standardised outcome sets for CKD, dialysis and transplantation were published by ICHOM and SONG. The authors for these interventions do not report evaluating their interventions with these outcomes. Six of the studies have completed their recruitment procedures,159 161 163 164 171 173 five are recruiting,158 160 162 165 172 three are active but not recruiting,166 168 169 one has an unknown recruitment status170 and one does not provide a recruitment status.167

Table 11

Overview of current studies evaluating the novel interventions

Discussion

We identified a considerable number of interventions that can be used to support SDM for treatment modality decisions in AKD. We observed that there are interventions that support a decision between a limited number of treatment modalities, and that there are interventions that support a decision between all treatment modalities for kidney failure. Almost all PTs that we identified make predictions that support decisions encompassing a single action or a single treatment modality (eg, start or delay preparations for kidney failure, accept or reject LDKT offer, start or forego dialysis). One PT compares prognostic information of two treatment options (transplantation and dialysis), which can be used to help patients make a treatment modality decision based on their personal risks.57 Similarly, most of the EPs, PtDAs and MIs we identified either provide information on a single treatment modality (eg, dialysis) or on two different treatment modalities (eg, dialysis and transplantation). A minority of the identified interventions provide information on all treatment modalities for kidney failure, including CCM. We recommend that clinicians use interventions appropriate to the decisions their patients have to make. With regard to treatment modality decisions in AKD, patients go through a ‘hierarchy of nested decisions rather than a choice between discrete options’.19 For example, patients will first have to decide whether they would prefer KRT or CCM, after which those that opt for KRT have to choose between their treatment options (eg, transplantation or dialysis modalities). With the right tools, clinicians can help patients: (1) navigate this decision-making process and (2) make choices based on their values and preferences. Values-clarification and preference-elicitation are important in the deliberation phase of SDM, and may ultimately lead to making a shared decision based on what ‘matters most to patients’.9 10 Most of the PtDAs we identified contain exercises that help patients in this process, and in that regard offer more than just education alone. Ideally, clinicians should use a combination of EPs, PtDAs and PTs to support and engage their patients in the decision-making process and make treatment modality decisions according to the principles of SDM. MIs cater to this idea, but most of the MIs we identified combine education with decision aids or other tools that help patients make values-based and preference-based decisions. We identified one novel MI (no.14) that combines all three of these tools. This MI is currently being evaluated in an RCT, which may provide some of the first evidence on the effectiveness (on patient-reported, biomedical and health system outcomes) of an intervention that combines education with prognostic information and decision support for patients with AKD.173

We observed considerable variation in the level of detail provided regarding the content of the interventions we identified. Most researchers and developers gave information on the structure and the medium that were used for the interventions, but they generally left the topics that were discussed unmentioned. They also varied in how they called treatment modalities (eg, independent vs self-care dialysis), and in the level of detail in which they described treatment modalities (eg, PD vs CAPD and APD). We found most of this variation in EPs, which may be explained by a lack of standardisation to specify and report on their contents.20 On the contrary, for PtDAs the IPDAS criteria were developed to standardise and improve their contents, development, implementation and evaluation.174 These criteria are widely accepted, and researchers and developers use them to develop and score their PtDAs.121 126 131 We also scored the PtDAs we identified with these criteria and found that just about half officially qualified as PtDAs. There is a need for a similar set of criteria that can be used for the development, implementation and evaluation of EPs. The variation in the literature makes it difficult to understand causal relationships between the interventions and the reported outcomes. It also hampers the development of new and effective interventions because it limits the possibility of synthesising evidence and replicating effective interventions.20

About one-third of the interventions we identified were reported to have been implemented in clinical practice. We found that PTs were the interventions with the least information on implementation status, followed by PtDAs, EPs and MIs. Most of the PTs could not be used by both patients and clinicians because their developers only presented them as formulas. This has been noted before and limits their usability in clinical practice.21 In fact, the only PT that was reported to have been implemented was designed as a PtDA that can be used by both patients and clinicians. On the contrary, PtDAs are almost always either printed materials or interactive websites that patients can use at the convenience of their own time. This enhances their usability and makes them an important supplemental resource for patients to learn about their treatment options.175 EPs tend to vary in their components,20 but generally consist of multiple outpatient consultations or educational sessions that are supplemented with printed and/or audio-visual materials, websites and coaches that guide patients through the programme. Hospitals often develop their own proprietary EPs, which may explain why more than half of the EPs we identified were reported to have been implemented in clinical practice. Most of the MIs we identified were also reported to have been implemented in clinical practice, presumably because they were often part of quality improvement initiatives that used multifaceted implementation strategies to support the implementation process. Implementation is important because only this ultimately leads to patients actually using these interventions. It can also provide real-world evidence on their effects and on the effectiveness of different implementation strategies. The implementation of future interventions should be facilitated by developing them with usability in mind, and by offering them through implementation strategies that combine different approaches.176 In addition, involving stakeholders and end-users (eg, patients and clinicians) will also facilitate the implementation process.177 Overall, end-user participation was low in the development of the interventions we identified. It seems plausible that this might have affected their implementation rates, but we cannot support this with the available data.

There are significant knowledge gaps when it comes to the effects these interventions have on patients, on the decision-making process, on SDM, and on the effects that SDM has on both patients and the decision-making process. First of all, only about half of the interventions were evaluated for their effects. The majority of these interventions were MIs, EPs and PtDAs. The only PT that was evaluated was the PT that was designed as a PtDA.57 70 This PT was also evaluated as part of an MI, but its use was not mandatory so its contribution to the reported effects remains undetermined.147 More evidence is needed on how PTs can be used to support patients and clinicians in the decision-making process, and there are multiple validated PTs that can be used in future research projects.

Patients who were exposed to EPs, PtDAs and MIs generally had better outcomes than patients that were not exposed to these interventions. It is difficult to say which intervention was the most effective because the majority were evaluated in observational studies and the exact ‘ingredients’ that elicit the reported outcomes are often unknown.20 Moreover, there was considerable variation in the selection of outcomes used to evaluate these interventions. Also, none of the interventions that were evaluated in studies published after the standardised outcome sets for CKD, dialysis and transplantation from ICHOM and SONG were published were evaluated with these outcomes. This highlights a need for: (1) more experimental research and (2) a standardisation in the selection of outcome measures for health-related outcomes (eg, the ICHOM or SONG standard sets), as well as for knowledge, decisional quality and SDM.

Finally, there is a clear lack of evidence when it comes to the effects that these interventions have on SDM. Notably, none of the MIs that were explicitly developed to enhance SDM in clinical practice were evaluated for their effects on SDM.148 150 153 Only a minority of PtDAs were evaluated for their effects on SDM, and SDM outcomes were generally better in patients that used them.122 135 143 Interventions should be evaluated on outcomes of SDM, especially if the intention is to support and implement SDM through these interventions. It is unfortunate that none of the researchers and developers of the interventions currently under evaluation report evaluating these interventions for their effects on SDM.

This scoping review is unique in the fact that we did not limit ourselves to a certain type of intervention, or to interventions that were developed to support or promote a particular treatment modality decision. We also did not identify reviews that categorised interventions based on the treatment modality decision(s) they support like we did. Even though we limited our search query to records published in English, the proportion of excluded records published in another language was small. We did exclude a larger number of records due to the fact that they were not available to us, either because of: (1) subscription limitations, (2) internet protocol address geo-blocking or (3) stakeholders that did not reply to our emails requesting access to their contents. Nevertheless, we feel that the selection of interventions presented in this review is a realistic reflection of the current state of developments in this field. We hope it stimulates clinicians to use these interventions in clinical practice, and that it incentivises researchers and developers to address the knowledge gaps we identified.

Conclusions

This scoping review provides clinicians, researchers and other stakeholders with one comprehensive, but digestible source of information on interventions that can be used to support SDM for treatment modality decisions in AKD. The usability of these interventions for SDM largely depends on whether patients can use them to compare all their treatment options, and whether they contain questions or exercises that help patients make decisions based on their values and preferences. It also depends on whether patients can access them at the convenience of their own time, and on how easily they can be used during healthcare encounters. Clinicians interested in SDM can select interventions from this review based on these properties, and ideally combine interventions that complement each other.

The implementation of the identified interventions in clinical practice was moderate and most likely depended on usability and the presence or absence of an implementation strategy.

No conclusive advice can be given on which intervention is the most efficacious in supporting SDM for treatment modality decisions in AKD. Outcomes seem to be better in patients exposed to these interventions, but this is largely based on observational research. In addition, the effects of these interventions on SDM are under-reported. There is a definite need for more experimental research and a standardisation in the development, implementation and evaluation of these interventions.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.

Acknowledgments

The authors thank Dr J W Schoones of the Leiden University Medical Centre for his help in generating the search queries, and Drs J van Meel of the Maasstad Hospital for his help in managing the RefWorks database and in gaining access to paywalled articles. The authors would also like to thank Dr C F van Uden Kraan for her insight and recommendations regarding the design and content of this scoping review.

References

Supplementary materials

Footnotes

  • Twitter @AMStiggelbout

  • Contributors NE is the primary and corresponding author and was responsible for the first and all subsequent drafts of this scoping review. GNdG was the second reviewer and participated in the process of study selection and data extraction. MvdD, PvdN, WJB and AMS all participated in discussions on the study design, the design of data extraction forms and critically revised drafts for improvements before publication. The ICMJE criteria for authorship have been met, and all authors approved the final version to be published. NE is responsible as guarantor for the overal content of this scoping review.

  • Funding This work was supported by ZonMW (registration no. 516007001) as part of the ‘Experiment Uitkomstindicatoren Santeon’.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

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

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.