Advertisement
American Journal of Kidney Diseases

Use of a Decision Aid for Patients Considering Peritoneal Dialysis and In-Center Hemodialysis: A Randomized Controlled Trial

Open AccessPublished:April 03, 2019DOI:https://doi.org/10.1053/j.ajkd.2019.01.030

      Rationale & Objective

      Annually, about 100,000 US patients face the difficult choice between the most common dialysis types, in-center hemodialysis and peritoneal dialysis. This study evaluated the value of a new decision aid to assist in the choice of dialysis modality.

      Study Design

      A parallel-group randomized controlled trial to test the efficacy of the decision aid on decision-making outcomes.

      Setting & Participants

      English-speaking US adults with advanced chronic kidney disease and internet access enrolled in 2015.

      Intervention

      Participants randomly assigned to the decision aid intervention received information about chronic kidney disease, peritoneal dialysis, and hemodialysis and a value clarification exercise through the study website using their own electronic devices. Participants in the control arm were only required to complete the control questionnaire. Questionnaire responses were used to assess differences across arms in decision-making outcomes.

      Outcomes

      Treatment preference, decisional conflict, decision self-efficacy, knowledge, and preparation for decision making.

      Results

      Of 234 consented participants, 94 (40.2%) were lost to follow-up before starting the study. Among the 140 (70 in each arm) who started the study, 7 were subsequently lost to follow-up. Decision aid users had lower decisional conflict scores (42.5 vs 29.1; P < 0.001) and higher average knowledge scores (90.3 vs 76.5; P < 0.001). Both arms had high decisional self-efficacy scores independent of decision aid use. Uncertainty about choice of dialysis treatment declined from 46% to 16% after using the decision aid. Almost all (>90%) users of the decision aid reported that it helped in decision making.

      Limitations

      Limited generalizability from the study of self-selected study participants who had to have internet access, speak English, and have computer literacy. High postrandomization loss to follow-up. Evaluation of only short-term outcomes.

      Conclusions

      The decision aid improves decision-making outcomes immediately after use. Implementation of the decision aid in clinical practice may allow further assessment of its effects on patient engagement and empowerment in choosing a dialysis modality.

      Funding

      This study was funded through a Patient Centered Outcomes Research Institute (PCORI) award (#1109).

      Trial Registration

      Registered at ClinicalTrials.gov with study number NCT02488317.

      Index Words

      More than 120,000 US patients started dialysis for end-stage kidney disease in 2015.
      • Saran R.
      • Robinson B.
      • Abbott K.C.
      • et al.
      US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
      Approximately 90% of dialysis patients receive hemodialysis (HD) at a dialysis center (“in-center”), 10% receive peritoneal dialysis (PD), and 0.4% use home HD as their first treatment modality.
      • Saran R.
      • Robinson B.
      • Abbott K.C.
      • et al.
      US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
      Quality of life and mortality rates of patients treated with HD and PD are similar, yet PD use in the United States is much lower than in other countries.
      • Weinhandl E.D.
      • Foley R.N.
      • Gilbertson D.T.
      • Arneson T.J.
      • Snyder J.J.
      • Collins A.J.
      Propensity-matched mortality comparison of incident hemodialysis and peritoneal dialysis patients.
      • Quinn R.R.
      • Hux J.E.
      • Oliver M.J.
      • Austin P.C.
      • Tonelli M.
      • Laupacis A.
      Selection bias explains apparent differential mortality between dialysis modalities.
      Although clinical contraindications may limit modality choice, most patients are eligible for both treatment options and the treatment choice should reflect patient preferences.
      • Lee A.
      • Gudex C.
      • Povlsen J.V.
      • Bonnevie B.
      • Nielsen C.P.
      Patients' views regarding choice of dialysis modality.
      • Segall L.
      • Nistor I.
      • Van Biesen W.
      • et al.
      Dialysis modality choice in elderly patients with end-stage renal disease: a narrative review of the available evidence.
      Accumulating evidence suggests that treatment consistent with patient preferences may improve quality of life and medical outcomes.
      • O’Hare A.M.
      • Armistead N.
      • Funk Schrag W.L.
      • Diamond L.
      • Moss A.H.
      Patient-centered care: an opportunity to accomplish the “three aims” of the National Quality Strategy in the Medicare ESRD Program.
      • Mapes D.L.
      • Lopes A.A.
      • Satayathum S.
      • et al.
      Health-related quality of life as a predictor of mortality and hospitalization: the Dialysis Outcomes and Practice Patterns Study (DOPPS).
      Patients and their care partners must understand the choices and related impacts on daily life to actively engage in the decision-making process.
      • Covic A.
      • Bammens B.
      • Lobbedez T.
      • et al.
      Educating end-stage renal disease patients on dialysis modality selection: clinical advice from the European Renal Best Practice (ERBP) advisory board.
      • Robinski M.
      • Mau W.
      • Wienke A.
      • Girndt M.
      Shared decision-making in chronic kidney disease: a retrospection of recently initiated dialysis patients in Germany.
      However, studies have shown that many patients feel unprepared and ill-informed about initiating dialysis and available options.
      • Song M.
      • Lin F.
      • Gilet C.A.
      • Arnold R.M.
      • Bridgman J.C.
      • Ward S.E.
      Patient perspectives on informed decision-making surrounding dialysis initiation.
      Additionally, low health literacy and numeracy can be barriers to understanding differences in treatment options and involvement in decision making.
      • Seo J.
      • Goodman M.S.
      • Politi M.
      • Blanchard M.
      • Kaphingst K.A.
      Effect of health literacy on decision-making preferences among medically underserved patients.
      • Reyna V.F.
      • Nelson W.L.
      • Han P.K.
      • Dieckmann N.F.
      How numeracy influences risk comprehension and medical decision making.
      • Peters E.
      • Hibbard J.
      • Slovic P.
      • et al.
      Numeracy skill and the communication, comprehension, and use of risk-benefit information.
      Patient decision aids are used to facilitate patient decision making about health care options.
      • Joseph-Williams N.
      • Newcombe R.
      • Politi M.
      • et al.
      Toward minimum standards for certifying patient decision aids: a modified Delphi consensus process.
      They aim to provide unbiased information about available options, increase participation in the decision-making process, reduce perceived pressure in selecting treatment choice, and mitigate decisional conflict.
      • Joseph-Williams N.
      • Newcombe R.
      • Politi M.
      • et al.
      Toward minimum standards for certifying patient decision aids: a modified Delphi consensus process.
      Several studies have shown that decision aids can have a substantial impact on key outcomes, including satisfaction with and confidence in the decision made, consequently improving treatment self-management.
      • Stacey D.
      • Légaré F.
      • Col N.F.
      • et al.
      Decision aids for people facing health treatment or screening decisions (review).
      • O’Connor A.M.
      • Tugwell P.
      • Wells G.A.
      • et al.
      A decision aid for women considering hormone therapy after menopause: decision support framework and evaluation.
      • Rothert M.L.
      • Talarczyk G.J.
      Patient compliance and the decision-making process of clinicians and patients.
      We describe the collaborative development of a web-based decision aid in the Empowering Patients on Choices for Renal Replacement Therapy (EPOCH-RRT) Study, involving researchers and a multistakeholder advisory panel composed of patients, care partners (family members), and patient advocates. We report results of the randomized controlled trial conducted to test the decision aid for efficacy in improving decision-making outcomes.

      Methods

      Decision Aid Development

      Content for the decision aid was based on literature review, US Renal Data System data,
      • Saran R.
      • Robinson B.
      • Abbott K.C.
      • et al.
      US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
      and results from previous EPOCH-RRT studies.
      • Dahlerus C.
      • Quinn M.
      • Messersmith E.
      • et al.
      Patient perspectives on the choice of dialysis modality: results from the Empowering Patients on Choices for Renal Replacement Therapy (EPOCH-RRT) study.
      • Zee J.
      • Zhao J.
      • Subramanian L.
      • et al.
      Perceptions about the dialysis modality decision process among peritoneal dialysis and in-center hemodialysis patients.
      A multistakeholder panel reviewed and refined the decision aid through an iterative process. Usability testing further helped refine the content, design, and structure of the decision aid.
      Per the International Patient Decision Aids Standards (IPDAS) checklist, IPDASi version 3.0,
      • Joseph-Williams N.
      • Newcombe R.
      • Politi M.
      • et al.
      Toward minimum standards for certifying patient decision aids: a modified Delphi consensus process.
      • Lewis K.B.
      • Wood B.
      • Sepucha K.R.
      • Thomson R.G.
      • Stacey D.
      Quality of reporting of patient decision aids in recent randomized controlled trials: a descriptive synthesis and comparative analysis.
      the decision aid addressed all qualifying criteria and the following certification criteria: balanced information for both treatment options, references to the funding source, additional resources and information about research used to develop the decision aid, year of publication, terms of use, and privacy policy. The website will be updated as needed based on annual literature review.

      Intervention

      The decision aid contained sections on: (1) chronic kidney disease (CKD) and its progression; (2) information and comparison of PD and HD based on patient priorities, positive and negative features of each option, options for switching, potential associated lifestyle changes, and side effects of both; and (3) an interactive value clarification exercise (VCE) to engage in the deliberation process and possibly lead to less regret and better preparation for decision making.
      • O'connor A.M.
      • Rostom A.
      • Fiset V.
      • et al.
      Decision aids for patients facing health treatment or screening decisions: systematic review.
      • Feldman-Stewart D.
      • Tong C.
      • Siemens R.
      • et al.
      The impact of explicit values clarification exercises in a patient decision aid emerges after the decision is actually made: evidence from a randomized controlled trial.
      The VCE provided a visual mapping between dialysis modality type and a list of lifestyle factors previously identified in EPOCH-RRT interviews as important to dialysis patients (Item S1). The decision aid integrated quotes from patients and tips from health care professionals (Item S2). Printing options were included to support discussing dialysis options with medical staff.
      The study website enabled collection of questionnaire data and walked users through all sections without skipping ahead. We provided “hover over” definitions for commonly used terms and logged progress so users could resume where they left off when unable to explore the entire decision aid in a single session. However, questionnaires had to be completed and submitted in a single session.

      Participant Recruitment

      Participants were recruited through nationwide social media outreach and locally in clinics, with a total recruitment target of 150 (Fig 1). The national outreach involved e-mail blasts and postings on Facebook and Twitter from August to October 2015 in collaboration with the National Kidney Foundation and American Association of Kidney Patients. Only those who could be re-contacted by telephone, self-identified as patients with CKD, and met all inclusion criteria were tracked. Local recruitment was conducted in 3 University of Michigan CKD clinics and 4 Henry Ford Health System CKD clinics across Southeast Michigan. Clinic and research staff reviewed visit schedules to identify patients meeting clinical criteria to approach at the time of their clinic visit between May 2015 and January 2016. Study coordinators obtained informed consent either verbally after telephone screening or in person and provided login information to the study website. Participants received a $25 gift card upon completion or attempted completion of study questionnaires. Follow-up of participants who had consented but either not started or started but not completed the study involved up to 5 attempted telephone or e-mail contacts over a 2-week period. In January 2016, we attempted a final contact of those who had either not started or not completed the study. Those who did not respond to any such contact were considered not reachable.
      Figure thumbnail gr1
      Figure 1Consolidated Standards of Reporting Trials (CONSORT) flow diagram of participants describing recruitment and flow of study participants through the study. Abbreviation: CKD, chronic kidney disease.

      Study Design

      Inclusion criteria were: (1) 18 years or older, (2) estimated glomerular filtration rate < 25 mL/min/1.73 m2, (3) internet access through a computer or tablet, and (4) English language fluency. Immediately after obtaining informed consent from a participant, the study coordinator provided the participant with a unique user login and study ID. The list of IDs provided to each recruiter was randomly generated by an independent study programmer and each ID appeared as a random sequence of letters. The list alternated between the intervention and control arms to ensure parallel assignment to the intervention or control arms of consented participants. However, neither the study coordinator nor the participant could discern the assignment based on the ID and both were therefore blinded to treatment assignment before consent and before study start. The study coordinator also remained blinded to treatment assignment throughout the study because participants engaged in the study on their own time.
      Study coordinators also provided information for accessing and using the study website to consented participants. Participants could access the study website from their own computers or portable devices using the login credentials provided. Initiating login by participants defined the start of study participation. The study team at the data coordinating center could track task completion for each participant after login and followed up weekly to check on any technical issues and promote study completion.
      Participants in the control arm were only required to complete the control questionnaire (Item S3) and click the submit button, at which point participation was considered complete. They then had the option to access the decision aid, if interested. Participants in this arm were included in the analysis if they answered all questions in the control questionnaire. Participants in the intervention arm were required to click response options for all pretest questions (Item S4) and click the submit button to proceed to the decision aid. They then clicked a button on the last page of the decision aid to indicate that they had completed review of the decision aid, and this would enable them to proceed to the posttest (Item S5). Initiation of the posttest immediately after decision aid review allowed for assessment of immediate DA effects. We considered intervention arm participants to have completed the study if they answered all questions in the posttest and clicked the final submit button.
      All study procedures were approved by institutional review boards at each study site and the data coordinating center (Henry Ford Health System #8144; University of Michigan, IRBMED eResearch #HUM00073058; Ethical & Independent Review Services #13016-03B).

      Outcomes and Measures

      Decision aids are evaluated based on improvements in the quality of the decision and the decision-making process based on underlying theories in decision making that provide a strong rationale for assessment before and immediately after exposure.
      • Sepucha K.R.
      • Borkhoff C.M.
      • Lally J.
      • et al.
      Establishing the effectiveness of patient decision aids: key constructs and measurement instruments.
      • O'Connor A.M.
      • Bennett C.L.
      • Stacey D.
      • et al.
      Decision aids for people facing health treatment or screening decisions.
      Outcomes used to evaluate the decision aid were treatment preference, decisional conflict, decision self-efficacy, preparation for decision making, and knowledge. Participants selected current treatment preference (HD, PD, unsure, and other) at baseline and postintervention.
      The Decisional Conflict Scale measures perceived uncertainty in choosing options and satisfaction with effective decision making.
      • O’Connor A.M.
      Validation of a decisional conflict scale.
      This is a 16-item scale with a 5-point Likert response format with scores reversed on negative statements. Items are summed, averaged, and multiplied by 25 to obtain scores ranging from 0 (no decisional conflict) to 80 (high decisional conflict).
      The Decision Self-efficacy Scale measures confidence in making an informed decision.
      • O’Connor A.M.
      User Manual – Decision Self-efficacy Scale.
      We used 10 of the 11 items scored on a 5-point response format because deferral of decision might not be an option for those facing kidney failure. Items are summed, averaged, and multiplied by 25 to obtain scores ranging from 0 (not at all confident) to 100 (very confident). These first 3 primary outcomes were included in the control arm test and both pre- and posttests in the intervention arm so that differences between those using and not using the decision aid, as well as changes from before to after using the decision aid, could be assessed. The latter was of particular interest in case there were any unobserved confounding variables between the control and intervention arms.
      The validated Preparation for Decision Making Scale measures patients’ readiness for communicating with practitioners and making a health decision.
      • Bennett C.
      • Graham I.D.
      • Kristjansson E.
      • Kearing S.A.
      • Clay K.F.
      • O'Connor A.M.
      Validation of a preparation for decision making scale.
      It was assessed only postintervention because it specifically relates to the decision tool. It is a 10-item scale with a 5-point response format in which items are summed, averaged, subtracted by 1, and multiplied by 25 to get a score from 0 to 100, with higher scores indicating higher preparation for decision making.
      For the fifth primary outcome of knowledge, questions were adapted from the Chronic Hemodialysis Knowledge Survey (CHeKS),
      • Gheewala P.A.
      • Peterson G.M.
      • Zaidi S.T.R.
      • Jose M.D.
      • Castelino R.L.
      Public knowledge of chronic kidney disease evaluated using a validated questionnaire: a cross-sectional study.
      previously shown to correlate with clinical outcomes.
      • Cavanaugh K.L.
      • Wingard R.L.
      • Hakim R.M.
      • Elasy T.A.
      • Ikizler T.A.
      Patient dialysis knowledge is associated with permanent arteriovenous access use in chronic hemodialysis.
      For intervention arm participants, knowledge questions were asked only during the posttest to prevent focus on content related to the knowledge questions while reviewing the decision aid and prevent positive bias during the posttest due to previous viewing of the questions.
      At baseline, both the control and intervention group (pretest) participants provided demographic information (age, sex, race, ethnicity, and education) and were assessed for subjective literacy
      • Chew L.D.
      • Griffin J.M.
      • Partin M.R.
      • et al.
      Validation of screening questions for limited health literacy in a large VA outpatient population.
      and numeracy (Table 1).
      • Fagerlin A.
      • Zikmund-Fisher B.J.
      • Ubel P.A.
      • Jankovic A.
      • Derry H.A.
      • Smith D.M.
      Measuring numeracy without a math test: development of the Subjective Numeracy Scale.
      • Zikmund-Fisher B.J.
      • Smith D.M.
      • Ubel P.A.
      • Fagerlin A.
      Validation of the subjective numeracy scale: effects of low numeracy on comprehension of risk communications and utility elicitations.
      because these are known barriers to involvement in decision making. The posttest within the intervention arm also included questions on the user experience related to usability, satisfaction with the decision aid, adequacy, relevance, and quality of content, as well as open-ended questions for positive and negative feedback. Thematic analysis of free-text responses across participants identified distinct motifs aggregated for frequency.
      • Braun V.
      • Clarke V.
      Using thematic analysis in psychology.
      Table 1Questionnaire Design, Distribution of Sections in Control and Intervention Arms
      SectionControlIntervention
      PretestPosttest
      Treatment preference
      Decisional conflict
      • O’Connor A.M.
      Validation of a decisional conflict scale.
      Decision self-efficacy
      • O’Connor A.M.
      User Manual – Decision Self-efficacy Scale.
      Knowledge
      • Zikmund-Fisher B.J.
      • Smith D.M.
      • Ubel P.A.
      • Fagerlin A.
      Validation of the subjective numeracy scale: effects of low numeracy on comprehension of risk communications and utility elicitations.
      Literacy
      • Gheewala P.A.
      • Peterson G.M.
      • Zaidi S.T.R.
      • Jose M.D.
      • Castelino R.L.
      Public knowledge of chronic kidney disease evaluated using a validated questionnaire: a cross-sectional study.
      Numeracy
      • Cavanaugh K.L.
      • Wingard R.L.
      • Hakim R.M.
      • Elasy T.A.
      • Ikizler T.A.
      Patient dialysis knowledge is associated with permanent arteriovenous access use in chronic hemodialysis.
      • Chew L.D.
      • Griffin J.M.
      • Partin M.R.
      • et al.
      Validation of screening questions for limited health literacy in a large VA outpatient population.
      Demographics
      Preparation for decision making
      • Bennett C.
      • Graham I.D.
      • Kristjansson E.
      • Kearing S.A.
      • Clay K.F.
      • O'Connor A.M.
      Validation of a preparation for decision making scale.
      User experience

      Statistical Analysis

      Analysts not blinded to treatment assignment tested for demographic differences between the intervention and control arms using baseline responses. Age, race, sex, education, ethnicity, and numeracy were compared using t, Pearson χ2, and Fisher exact tests. We compared the intervention posttest with the control arm using unpaired t, Wilcoxon rank sum, and Pearson χ2 tests. We compared outcomes between the pre- and posttest within the intervention arm using paired t tests, Wilcoxon signed rank tests, and tests for marginal homogeneity.
      We tested whether differences between intervention posttest and control arm and differences between intervention pre- and posttest responses differed across age (continuous), sex (female, male), education level (college graduate or above vs some college or under), or race groups (black vs nonblack) for all primary outcomes. We used generalized estimating equation logistic or linear regression models for these tests by including an interaction term between subgroups and different arms in models.
      • Liang K.Y.
      • Zeger S.L.
      Longitudinal data analysis using generalized linear models.
      Models accounted for correlations within subjects when comparing pre- and posttest intervention arm responses using an exchangeable correlation structure and sandwich-type estimator for standard errors. These P values were corrected for multiple hypothesis testing using the Benjamini-Hochberg procedure.
      • Benjamini Y.
      • Hochberg Y.
      Controlling the false discovery rate: a practical and powerful approach to multiple testing.

      Results

      Study Sample

      Of 556 patients initially screened and selected, 83 were not reachable and 195 no longer met inclusion criteria when approached for consent (Fig 1). Some patients did not have internet connectivity or access to a computer. Of patients who declined to participate (n = 44), reasons included poor eyesight for the study tasks or being too ill or fatigued to participate. A total of 234 patients consented to the study and were given randomized login information for the study website; 51 from social media outreach and 183 from local clinic recruitment. Notably, 40.2% (94 total; 1 from social media outreach and 93 from local clinic recruitment) were lost to follow-up before starting the study. These participants did not log into the study site to start the study and 78 were not reachable by telephone or e-mail for follow-up. The remaining 140 participants, 70 in each arm, started the study.
      Seven participants in the intervention arm started the study and completed the pretest but did not go on to complete the posttest questionnaire. All these participants were originally recruited through local clinics. Sensitivity analyses with and without these participants suggest that these departures did not affect our study results. Fifty of the 63 (79%) intervention arm participants completed the pre- and posttests within 1 week, with 60% having completed both on the same day. Only 5 participants took more than 1 month to review the decision aid and complete the posttest. Sensitivity analysis with and without the 13 participants who had a gap of more than 7 days between completion of the 2 questionnaires did not change any of the mean values of measured outcomes or reported statistical differences.

      Patient Characteristics

      Demographic information was self-reported in the control and pretest questionnaires (Table 1); participant composition was found to be similar in both groups (Table 2) and was unavailable for those who never started the study. Our study sample was younger than the US CKD stages 4 to 5 population in 2014 (estimated glomerular filtration rates < 30 mL/min/1.73 m2, mean age of 76.8 years, 46.2% men, and 77.6% white) but with similar proportions of white and male participants
      • Saran R.
      • Robinson B.
      • Abbott K.C.
      • et al.
      US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
      . Almost all had graduated high school (96%) and considered English their native language (94%). Literacy (mean literacy = 2.81) and subjective numeracy (mean numeracy = 3.88) were similar in both groups at the start of the study.
      Table 2Participant Characteristics, by Control and Intervention Arms
      Patient CharacteristicsControlInterventionP
      No. of patients7070
      Age, y59 ± 1459 ± 150.9
      Race0.8
       White79%74%
       Black14%17%
       Other7%9%
      Male sex50%43%0.4
      Hispanic or Latino/Latina3%3%0.9
      High school graduate94%99%0.9
      English as native language96%91%0.5
      Ability to understand
       Reading materials
      Mean score±standard deviation of answer choices from all the time (0) to none of the time (4); higher score indicates greater ability to understand.
      2.94 ± 1.253.67 ± 1.450.3
       SNS
      Mean score for answers (not at all good [1]–extremely good [6]) of the 3 questions: How good are you at working with fractions? How good are you at figuring out how much a shirt will cost if it is 25% off? How often do you find numerical information to be useful?
      3.83 ± 1.113.92 ± 0.990.8
       SNS ability
      Mean score for answers (not at all good [1]–extremely good [6]) of the 2 questions: How good are you at working with fractions? How good are you at figuring out how much a shirt will cost if it is 25% off?
      3.88 ± 1.183.92 ± 1.100.9
       SNS preference
      Mean score for answers (not at all good [1]–extremely good [6]) of the question: How often do you find numerical information to be useful?
      3.74 ± 1.263.93 ± 1.080.5
      Abbreviation: SNS, Subjective Numeracy Scale.
      a Mean score ± standard deviation of answer choices from all the time (0) to none of the time (4); higher score indicates greater ability to understand.
      b Mean score for answers (not at all good [1]–extremely good [6]) of the 3 questions: How good are you at working with fractions? How good are you at figuring out how much a shirt will cost if it is 25% off? How often do you find numerical information to be useful?
      c Mean score for answers (not at all good [1]–extremely good [6]) of the 2 questions: How good are you at working with fractions? How good are you at figuring out how much a shirt will cost if it is 25% off?
      d Mean score for answers (not at all good [1]–extremely good [6]) of the question: How often do you find numerical information to be useful?

      Efficacy of the Decision Aid

      Reduction in Uncertainty

      The control and intervention participants’ pretest scores suggest that both arms had similar baseline uncertainty on treatment choice, 40% and 47%, respectively. After the intervention, the proportion of “not sure” responders was 24 percentage points lower than the corresponding proportion in the control group (Table 3). Within the intervention group, the proportion of not sure responders was 30 percentage points after, versus before, using the decision aid (Table 3). Of the 29 participants in the intervention arm who selected not sure before the intervention, 8 remained unsure, 15 selected HD, 5 selected PD, and 1 selected other. Of those who had selected HD or PD initially, only 1 participant each switched to the other option and another switched to not sure.
      Table 3Outcome Measures for Decision Aid Efficacy
      Comparison of Control and Intervention Arm OutcomesControlInterventionP
      We compared the intervention posttest with the control arm using unpaired t, Wilcoxon rank sum, and Pearson χ2 tests. We compared outcomes between the pre- and posttest within the intervention arm using paired t tests, Wilcoxon signed rank tests, and tests for marginal homogeneity.
      No. of participants7063
      Which dialysis type do you think you might choose?<0.001
       Hemodialysis23 (16%)43 (27%)
       Peritoneal dialysis31 (22%)37 (23%)
       Not sure40 (28%)16 (10%)
       Other6 (4%)5 (3%)
      Decisional Conflict Score
      Score ranges from 1 to 80.
      (higher = more conflict)
      42.5 ± 17.129.1 ± 13.7<0.001
      Decisional Self-efficacy Score
      Score ranges from 0 to 100.
      (higher = more confident)
      79.9 ± 17.682.0 ± 18.40.4
      Knowledge
      Score ranges from 0 to 100.
      (higher = more correct answers chosen)
      76.5 ± 15.390.3 ± 11.9<0.001
      Comparison of Before (Pre-) and After (Post-) DA Use Outcomes for Intervention ArmPretestPosttestP
      We compared the intervention posttest with the control arm using unpaired t, Wilcoxon rank sum, and Pearson χ2 tests. We compared outcomes between the pre- and posttest within the intervention arm using paired t tests, Wilcoxon signed rank tests, and tests for marginal homogeneity.
      No. of participants6363
      Which dialysis type do you think you might choose?<0.001
       Hemodialysis21 (13%)43 (27%)
       Peritoneal dialysis29 (18%)37 (23%)
       Not sure46 (29%)16 (10%)
       Other5 (3%)5 (3%)
      Decisional Conflict Score
      Score ranges from 1 to 80.
      (higher = more conflict)
      43.6 ± 15.929.1 ± 13.7<0.001
      Decisional Self-efficacy Score
      Score ranges from 0 to 100.
      (higher = more confident)
      82.2 ± 18.082.0 ± 18.40.9
      Note: Results shown as count (percentage) or mean ± standard deviation.
      Abbreviation: DA, decision aid.
      a We compared the intervention posttest with the control arm using unpaired t, Wilcoxon rank sum, and Pearson χ2 tests. We compared outcomes between the pre- and posttest within the intervention arm using paired t tests, Wilcoxon signed rank tests, and tests for marginal homogeneity.
      b Score ranges from 1 to 80.
      c Score ranges from 0 to 100.
      Based on generalized estimating equation models with interactions between subgroups and arm, older age was initially associated with a larger difference between the control and intervention posttest arms (raw P = 0.01). However, this association became non–statistically significant (at significance level 0.05) after correction for multiple hypothesis testing (corrected P = 0.2). Similarly, while those in the intervention arm who were college graduates or above had a nominally larger decrease in uncertainty after using the decision aid (raw P = 0.04), this finding lost statistical significance after correction (P = 0.3). There was no evidence of effect modification by other subgroups (raw P range, 0.3-0.8; corrected P range, 0.8-0.9).

      Reduction in Decisional Conflict

      The intervention group scored 13.4 points less than the control group in decisional conflict (Table 3) on average, while the decision aid was effective in decreasing the average decisional conflict score by 15 points, from 44 to 29 among those in the intervention group (Table 3). No effect modification by age, sex, or race (raw P range, 0.2-0.9; corrected P range, 0.8-0.9) was observed. While those in the intervention arm who were college graduates or above had an average decrease in decisional conflict scores after using the decision aid of 8.6 points more than those with some college education or below in initial analyses (raw P = 0.01), this finding was not statistically significant after correction (P = 0.2). The average decisional conflict score among the control and pretest responders was similar, at 43 and 44, respectively.

      No Change in Decision Self-efficacy

      Decision self-efficacy scores were high on the 0 to 100 scale (∼80) for the control group and the intervention group at pretest (Table 3). There was little change in this score after use of the decision aid (Table 3) and no evidence of effect modification by subgroups (raw P range, 0.1-0.9; corrected P range, 0.5-0.9).

      Improving Knowledge

      The control arm on average answered 77% of the knowledge questions accurately. After going through the decision aid, the intervention arm correctly answered 90% of these questions (Table 3). Black participants in the control arm had lower baseline knowledge scores compared with nonblack participants (62.2 vs 78.9; P = 0.02). The difference in knowledge scores between the control and intervention arms was nominally greater for black participants compared with nonblack participants (26 vs 12; raw P = 0.02), but this finding did not retain statistical significance after correction (P = 0.3). Our study was not powered for comparing other minority groups. We did not observe effect modification by age, sex, or educational level on the knowledge outcome (raw P range, 0.4-0.6; corrected P = 0.8).

      User Experience

      The mean intervention arm score on the Preparation for Decision Making Scale was 76.4 ± 18.9 (standard deviation) with > 90% of participants indicating that the decision aid helped somewhat to a great deal, both for preparing for dialysis and for follow-up with care providers (Figs 2 and 3). The majority (92%) of participants found the content balanced and not slanted toward either option, 88% trusted it, 87% agreed/strongly agreed that it was relevant to them, and 89% said they would recommend it to others, with 49% agreeing that the decision aid was extremely helpful in understanding dialysis options. Intervention arm participants provided free-text feedback summarized in Table 4. Only 1 person did not like the website at all and 2 people would definitely not recommend the decision aid to others. Although the distribution of responses on questions related to preparing for dialysis and follow-up were viewed positively by all 3, dissatisfaction about the absence of information related to home HD was expressed in the open-ended feedback.
      Figure thumbnail gr2
      Figure 2Summary of responses related to factors related to supporting decision making. Ten items with responses on a Likert scale from “not at all” to “a great deal.”
      Figure thumbnail gr3
      Figure 3Perceived benefit of the decision aid in decision making, understanding options, and knowledge. Three items with responses on a Likert scale from “not at all” to “extremely.”
      Table 4Summary of Open-Ended Feedback on the Decision Aid: Most Frequently Cited Responses
      Most Frequent Topic/ThemeFrequency
      Frequency indicates the proportion of posttest respondents (n=63) whose responses for each question included these themes. Some responses coded to more than 1 theme; therefore, frequencies do not add up to 100%.
      Sample Quotes
      What you think about the decision aid website you have just reviewed?
       Informative65%“There was a lot of very good information to assist me to make a very serious decision when and if the time comes.…”

      “I was knowledgeable already on 80% of the information, but it was helpful… I hope all treatments improve.”
       Helpful40%
       Good40%
       Not good/something missing22%
       Not sure3%
      What did you like about the decision aid website?
       Easy to use/accessible29%“I like the 'feel' perspective… the facts of each treatment can be found everywhere, but not often do you see the feelings of the patient, put in consideration. I had mixed feelings about which way to go, but this site helped a great deal.”
       Explanation of options21%
       Informative17%
       Helpful for decision making10%
       Comprehensive8%
       Patient testimonials8%
       Graphics8%
       Interactivity (VCE decision tool)6%
      What suggestions do you have to improve the decision aid website?
       None41%“… what about home hemo? And, what about info about having to change dialysis types if, for instance, your abdominal wall becomes tough and can no longer filter?”
       More information29%
       Improve content format10%
       Improve features8%
       Improve quotes6%
      Abbreviation: VCE, value clarification exercise.
      a Frequency indicates the proportion of posttest respondents (n = 63) whose responses for each question included these themes. Some responses coded to more than 1 theme; therefore, frequencies do not add up to 100%.

      Discussion

      Decision aids improve knowledge, enhance perception of risks, lessen decisional conflict, and increase participation in shared decision making.
      • Stacey D.
      • Légaré F.
      • Col N.F.
      • et al.
      Decision aids for people facing health treatment or screening decisions (review).
      A small number of decision aids related to dialysis treatment modality choice incorporate value clarification tools: 3 in the United States and some developed outside the United States or in other health care contexts.
      Healthwise, Cochrane Patient Decision Aid.
      • Winterbottom A.E.
      • Gavaruzzi T.
      • Mooney A.
      • et al.
      Patient acceptability of the Yorkshire Dialysis Decision Aid (YoDDA) Booklet: a prospective non-randomized comparison study across 6 predialysis services.
      • Schatell D.
      • Agar J.
      • Witten B.
      • Bauer M.
      • Klicko K.
      Medical Education Institute, Inc
      My Life, My Dialysis Choice.
      The main difference between our choosingdialysis.org decision aid and the others developed in the United States was the emphasis and consistent application of patient-centered approaches in the development of decision aid content and testing.
      Similar to other decision aids used in different health decisions,
      • Stacey D.
      • Légaré F.
      • Col N.F.
      • et al.
      Decision aids for people facing health treatment or screening decisions (review).
      those who completed review of our decision aid indicated improved knowledge, better preparation for decision making, and reduced decisional conflict but no significant improvement in decision self-efficacy shortly after its use. Our results suggest that the extent of benefit from the decision aid on reducing decision uncertainty might vary by age and education level, while the reduction in decisional conflict might vary based on education. These factors might be indicators of differences in engagement with the decision aid, and future decision aid implementation studies could explore this further.
      Improving knowledge through CKD education has been proposed as one solution to overcoming identified barriers such as patients’ awareness of choices and disparities in shared decision making and improved patient-centered care.
      • Ayanian J.Z.
      • Cleary P.D.
      • Weissman J.S.
      • Epstein A.M.
      The effect of patients' preferences on racial differences in access to renal transplantation.
      • Norris K.C.
      • Agodoa L.Y.
      Unraveling the racial disparities associated with kidney disease.
      • Binik Y.M.
      • Devins G.M.
      • Barre P.E.
      • et al.
      Live and learn: patient education delays the need to initiate renal replacement therapy in end-stage renal disease.
      • Perestelo-Perez L.
      • Rivero-Santana A.
      • Sanchez-Afonso J.A.
      • et al.
      Effectiveness of a decision aid for patients with depression: a randomized controlled trial.
      Differences in the knowledge test between black and nonblack participants may be related to racial disparities in treatment choices that are well documented, although we saw no differences in treatment preferences by race across treatment arms in this study.
      • Saran R.
      • Robinson B.
      • Abbott K.C.
      • et al.
      US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
      • Ayanian J.Z.
      • Cleary P.D.
      • Weissman J.S.
      • Epstein A.M.
      The effect of patients' preferences on racial differences in access to renal transplantation.
      • Sheu J.
      • Ephraim P.L.
      • Powe N.R.
      • et al.
      African American and non-African American patients' and families' decision making about renal replacement therapies.
      This decision aid was effective at helping patients with advanced CKD who were still unsure of their treatment option to be less unsure but did not make people who had selected an option to become less sure or to change their selection. This could be ascribed to the baseline high self-efficacy scores suggesting greater confidence of participants in being able to make a decision. Multiple factors may contribute to decision self-efficacy, which would require a holistic socioecologic approach to move the needle on this indicator.
      Some participants in our study provided feedback that they found the VCE helpful, although the decision aid literature is uncertain of the benefits of VCE on decision-making outcomes.
      • Nelson W.L.
      • Han P.K.
      • Fagerlin A.
      • Stefanek M.
      • Ubel P.A.
      Rethinking the objectives of decision aids: a call for conceptual clarity.
      • Feldman-Stewart D.
      • Tong C.
      • Siemens R.
      • et al.
      The impact of explicit values clarification exercises in a patient decision aid emerges after the decision is actually made: evidence from a randomized controlled trial.
      Evaluation shortly after decision aid use might have limited opportunity to adequately assess the use of different features of the decision aid reflective of real-world use, such preparing for a doctor’s visit.
      There was high loss to follow-up before study start among participants recruited at CKD clinics. Higher participation among social media recruitment respondents suggests greater prior interest in the study among these participants compared with those approached at local clinics. Because these participants dropped out after randomization and despite our findings of similar participant characteristics across study arms, there may be unobserved factors that were unbalanced and could have biased our comparisons between the control and intervention arms. Future studies may benefit from a study design involving randomization after website login. This also implies that the effect of the decision aid observed in this study may not apply to decision aid use in real-world settings. However, similar results when comparing pre- and posttest responses in the intervention arm provide additional evidence of decision aid efficacy.
      Our study has several other limitations. First, participants were on average younger than the US CKD population
      • Saran R.
      • Robinson B.
      • Abbott K.C.
      • et al.
      US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
      and based on US Census data, more educated than the US general population. Study participants had high literacy scores, almost all graduated high school, and the percentage of native English speakers was very high. This likely is a reflection of the success of the online recruitment strategies among people with these attributes. Patients with stage 4 CKD are often dealing with a heavy medication burden and a multitude of physical and mental symptoms related to kidney failure before the start of renal replacement therapy.
      • Weinhandl E.D.
      • Foley R.N.
      • Gilbertson D.T.
      • Arneson T.J.
      • Snyder J.J.
      • Collins A.J.
      Propensity-matched mortality comparison of incident hemodialysis and peritoneal dialysis patients.
      • Clarke A.L.
      • Yates T.
      • Smith A.C.
      • Chilcot J.
      Patient's perceptions of chronic kidney disease and their association with psychosocial and clinical outcomes: a narrative review.
      These factors may have negatively influenced willingness to participate in a research study and contributed to the low participation rate, potentially resulting in a less generalizable participant cohort despite similar race and sex mix as the US CKD population. We envisioned a web-based format as the ideal way to quickly disseminate the decision aid to the broadest possible audience. However, lack of internet access and computer literacy limitations challenged recruitment efforts, also contributing to a less representative cohort in terms of age and educational attainment.
      Second, while we did not find evidence that our randomization strategy was unsuccessful, we were unable to operationalize a truly random process across different sites and recruitment efforts. Because the list of study IDs given to each study coordinator alternated between the intervention and control groups, random assignment relies on the assumption that the order of participant consent is random. Furthermore, if a participant revealed his or her treatment assignment to the study coordinator after starting the study and the study coordinator retained a copy of the full study ID list, the coordinator could have discerned other participants’ treatment assignments.
      Third, decision-making outcomes were only assessed immediately after exposure to the decision aid. This was done to evaluate the short-term effects of the decision aid and is consistent with theoretical constructs and IPDAS guidelines.
      • Sepucha K.R.
      • Borkhoff C.M.
      • Lally J.
      • et al.
      Establishing the effectiveness of patient decision aids: key constructs and measurement instruments.
      Efforts are currently underway to develop a study to implement the decision aid in nephrology clinics that will allow for better understanding of long-term effects of the decision aid and in real-world settings.
      We incorporated feedback such as the need for information about home HD from participants to refine the decision aid further. The final decision aid is now available at http://choosingdialysis.org/. Our work suggests that this decision aid, developed through a stakeholder-engaged process, informs and supports patients with CKD in making the difficult choice of dialysis modality. The broader implementation of this decision aid could complement current CKD education in clinical practice and could support both care providers and patients in shared decision making by facilitating communication about treatment options. Additionally, the decision aid could also become a resource for disseminating end-stage kidney disease knowledge with the potential for improving health outcomes through more active engagement in care.
      • Covic A.
      • Bammens B.
      • Lobbedez T.
      • et al.
      Educating end-stage renal disease patients on dialysis modality selection: clinical advice from the European Renal Best Practice (ERBP) advisory board.
      • King K.
      Patients’ perspective of factors affecting modality selection: a National Kidney Foundation patient survey.
      • Golper T.
      Patient education: can it maximize the success of therapy?.

      Article Information

      Authors’ Full Names and Academic Degrees

      Lalita Subramanian, PhD, Junhui Zhao, PhD, Jarcy Zee, PhD, Megan Knaus, MPH, Angela Fagerlin, PhD, Erica Perry, LMSW, June Swartz, MA, Margie McCall, BA, Nicole Bryant, and Francesca Tentori, MD.

      Authors’ Contributions

      Conception and design of the study: LS, FT, MK, AF, EP, JS, MM, NB; data acquisition: LS, FT, MK, AF, EP, JS; data analysis and interpretation: LS, FT, JZh, JZe. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved.

      Support

      Research reported in this article was funded through a Patient-Centered Outcomes Research Institute (PCORI) Award (1109) to Dr Tentori. Dr Tentori was supported in part by National Institute of Diabetes and Digestive and Kidney Diseases grant K01DK087762 . The funders did not have a role in study design, data collection, analysis, reporting, or the decision to submit for publication.

      Financial Disclosure

      Dr Tentori is an employee of DaVita HealthCare Partners, Inc. She was employed by Arbor Research Collaborative for Health, which administers the Dialysis Outcomes and Practice Patterns Study (DOPPS) Program, which is funded by a consortium of private industry, public funders, and professional societies. Principal funders: Amgen, Kyowa Hakko Kirin, and Baxter Healthcare. Additional support for specific DOPPS projects and/or program activities in specific countries provided by: Amgen , Association of German Nephrology Centres (Verband Deutsche Nierenzentren e.V.), AstraZeneca , European Renal Association-European Dialysis and Transplant Association, German Society of Nephrology, Hexal AG, Janssen, Japanese Society for Peritoneal Dialysis, Keryx, Proteon, Relypsa, Roche , Società Italiana di Nefrologia, Spanish Society of Nephrology, and Vifor Fresenius Medical Care Renal Pharma. Public funding and support is provided for specific DOPPS projects, ancillary studies, or affiliated research projects by: Australia: National Health & Medical Research Council; Canada: Canadian Institutes of Health Research and Ontario Renal Network; France: Agence Nationale de la Recherche; Thailand: Thailand Research Foundation , Chulalongkorn University Matching Fund, King Chulalongkorn Memorial Hospital Matching Fund, and the National Research Council of Thailand; United Kingdom: National Institute for Health Research via the Comprehensive Clinical Research Network; and United States: National Institutes of Health and PCORI. All support is provided without restrictions on publications. The remaining authors declare that they have no relevant financial interests.

      Acknowledgements

      This work would not have been possible without the contributions of all the participants who contributed to the EPOCH-RRT study. Valerie Kahn contributed to the decision aid development and usability testing. Jennifer McCready-Maynes, an employee of Arbor Research Collaborative for Health, provided editorial assistance.

      Data Sharing

      Deidentified data that underlie the results reported in this article are available to qualified researchers for approved scientific uses immediately following publication with no end date. Data access proposals should be directed to the corresponding author.

      Peer Review

      Received May 29, 2018. Evaluated by 2 external peer reviewers, with direct editorial input from a Statistics/Methods Editor, an Associate Editor, and the Editor-in-Chief. Accepted in revised form January 29, 2019.

      Supplementary Material

      References

        • Saran R.
        • Robinson B.
        • Abbott K.C.
        • et al.
        US Renal Data System 2017 Annual Data Report: epidemiology of kidney disease in the United States.
        Am J Kidney Dis. 2018; 71: S1-S676
        • Weinhandl E.D.
        • Foley R.N.
        • Gilbertson D.T.
        • Arneson T.J.
        • Snyder J.J.
        • Collins A.J.
        Propensity-matched mortality comparison of incident hemodialysis and peritoneal dialysis patients.
        J Am Soc Nephrol. 2010; 21: 499-506
        • Quinn R.R.
        • Hux J.E.
        • Oliver M.J.
        • Austin P.C.
        • Tonelli M.
        • Laupacis A.
        Selection bias explains apparent differential mortality between dialysis modalities.
        J Am Soc Nephrol. 2011; 22: 1534-1542
        • Lee A.
        • Gudex C.
        • Povlsen J.V.
        • Bonnevie B.
        • Nielsen C.P.
        Patients' views regarding choice of dialysis modality.
        Nephrol Dial Transplant. 2008; 23: 3953-3959
        • Segall L.
        • Nistor I.
        • Van Biesen W.
        • et al.
        Dialysis modality choice in elderly patients with end-stage renal disease: a narrative review of the available evidence.
        Nephrol Dial Transplant. 2017; 32: 41-49
        • O’Hare A.M.
        • Armistead N.
        • Funk Schrag W.L.
        • Diamond L.
        • Moss A.H.
        Patient-centered care: an opportunity to accomplish the “three aims” of the National Quality Strategy in the Medicare ESRD Program.
        Clin J Am Soc Nephrol. 2014; 9: 2189-2194
        • Mapes D.L.
        • Lopes A.A.
        • Satayathum S.
        • et al.
        Health-related quality of life as a predictor of mortality and hospitalization: the Dialysis Outcomes and Practice Patterns Study (DOPPS).
        Kidney Int. 2003; 64: 339-349
        • Covic A.
        • Bammens B.
        • Lobbedez T.
        • et al.
        Educating end-stage renal disease patients on dialysis modality selection: clinical advice from the European Renal Best Practice (ERBP) advisory board.
        Nephrol Dial Transplant. 2010; 25: 1757-1759
        • Robinski M.
        • Mau W.
        • Wienke A.
        • Girndt M.
        Shared decision-making in chronic kidney disease: a retrospection of recently initiated dialysis patients in Germany.
        Patient Educ Couns. 2016; 99: 562-570
        • Song M.
        • Lin F.
        • Gilet C.A.
        • Arnold R.M.
        • Bridgman J.C.
        • Ward S.E.
        Patient perspectives on informed decision-making surrounding dialysis initiation.
        Nephrol Dial Transplant. 2013; 28: 2815-2823
        • Seo J.
        • Goodman M.S.
        • Politi M.
        • Blanchard M.
        • Kaphingst K.A.
        Effect of health literacy on decision-making preferences among medically underserved patients.
        Med Decis Making. 2016; 36: 550-556
        • Reyna V.F.
        • Nelson W.L.
        • Han P.K.
        • Dieckmann N.F.
        How numeracy influences risk comprehension and medical decision making.
        Psychol Bull. 2009; 135: 943-973
        • Peters E.
        • Hibbard J.
        • Slovic P.
        • et al.
        Numeracy skill and the communication, comprehension, and use of risk-benefit information.
        Health Aff (Millwood). 2007; 26: 741-748
        • Joseph-Williams N.
        • Newcombe R.
        • Politi M.
        • et al.
        Toward minimum standards for certifying patient decision aids: a modified Delphi consensus process.
        Med Decis Making. 2014; 34: 699-710
        • Stacey D.
        • Légaré F.
        • Col N.F.
        • et al.
        Decision aids for people facing health treatment or screening decisions (review).
        Cochrane Database Syst Rev. 2014; 1: CD001431
        • O’Connor A.M.
        • Tugwell P.
        • Wells G.A.
        • et al.
        A decision aid for women considering hormone therapy after menopause: decision support framework and evaluation.
        Patient Educ Couns. 1998; 33: 267-279
        • Rothert M.L.
        • Talarczyk G.J.
        Patient compliance and the decision-making process of clinicians and patients.
        J Compliance Health Care. 1987; 2: 55-71
        • Dahlerus C.
        • Quinn M.
        • Messersmith E.
        • et al.
        Patient perspectives on the choice of dialysis modality: results from the Empowering Patients on Choices for Renal Replacement Therapy (EPOCH-RRT) study.
        Am J Kidney Dis. 2016; 68: 901-910
        • Zee J.
        • Zhao J.
        • Subramanian L.
        • et al.
        Perceptions about the dialysis modality decision process among peritoneal dialysis and in-center hemodialysis patients.
        BMC Nephrol. 2018 Oct 29; 19: 298
        • Lewis K.B.
        • Wood B.
        • Sepucha K.R.
        • Thomson R.G.
        • Stacey D.
        Quality of reporting of patient decision aids in recent randomized controlled trials: a descriptive synthesis and comparative analysis.
        Patient Educ Couns. 2017; 100: 1387-1393
        • O'connor A.M.
        • Rostom A.
        • Fiset V.
        • et al.
        Decision aids for patients facing health treatment or screening decisions: systematic review.
        BMJ. 1999; 319: 731-734
        • Feldman-Stewart D.
        • Tong C.
        • Siemens R.
        • et al.
        The impact of explicit values clarification exercises in a patient decision aid emerges after the decision is actually made: evidence from a randomized controlled trial.
        Med Decis Making. 2012; 32: 616-626
        • Sepucha K.R.
        • Borkhoff C.M.
        • Lally J.
        • et al.
        Establishing the effectiveness of patient decision aids: key constructs and measurement instruments.
        BMC Med Inform Decis Making. 2013; 13: S12
        • O'Connor A.M.
        • Bennett C.L.
        • Stacey D.
        • et al.
        Decision aids for people facing health treatment or screening decisions.
        Cochrane Database Syst Rev. 2009; 3
        • O’Connor A.M.
        Validation of a decisional conflict scale.
        Med Decis Making. 1995; 15: 25-30
        • O’Connor A.M.
        User Manual – Decision Self-efficacy Scale.
        Ottawa Hospital Research Institute, Ottawa, Canada1995 ([modified 2002]) (Accessed August 20, 2017)
        • Bennett C.
        • Graham I.D.
        • Kristjansson E.
        • Kearing S.A.
        • Clay K.F.
        • O'Connor A.M.
        Validation of a preparation for decision making scale.
        Patient Educ Couns. 2010; 78: 130-133
        • Gheewala P.A.
        • Peterson G.M.
        • Zaidi S.T.R.
        • Jose M.D.
        • Castelino R.L.
        Public knowledge of chronic kidney disease evaluated using a validated questionnaire: a cross-sectional study.
        BMC Public Health. 2018; 18: 371
        • Cavanaugh K.L.
        • Wingard R.L.
        • Hakim R.M.
        • Elasy T.A.
        • Ikizler T.A.
        Patient dialysis knowledge is associated with permanent arteriovenous access use in chronic hemodialysis.
        Clin J Am Soc Nephrol. 2009; 4: 950-956
        • Chew L.D.
        • Griffin J.M.
        • Partin M.R.
        • et al.
        Validation of screening questions for limited health literacy in a large VA outpatient population.
        J Gen Intern Med. 2008; 23: 561-566
        • Fagerlin A.
        • Zikmund-Fisher B.J.
        • Ubel P.A.
        • Jankovic A.
        • Derry H.A.
        • Smith D.M.
        Measuring numeracy without a math test: development of the Subjective Numeracy Scale.
        Med Decis Making. 2007; 27: 672-680
        • Zikmund-Fisher B.J.
        • Smith D.M.
        • Ubel P.A.
        • Fagerlin A.
        Validation of the subjective numeracy scale: effects of low numeracy on comprehension of risk communications and utility elicitations.
        Med Decis Making. 2007; 27: 663-671
        • Braun V.
        • Clarke V.
        Using thematic analysis in psychology.
        Qual Res Psychol. 2006; 3: 77-101
        • Liang K.Y.
        • Zeger S.L.
        Longitudinal data analysis using generalized linear models.
        Biometrika. 1986; 73: 13-22
        • Benjamini Y.
        • Hochberg Y.
        Controlling the false discovery rate: a practical and powerful approach to multiple testing.
        J R Stat Soc B. 1995; 57: 289-300
      1. Healthwise, Cochrane Patient Decision Aid.
        • Winterbottom A.E.
        • Gavaruzzi T.
        • Mooney A.
        • et al.
        Patient acceptability of the Yorkshire Dialysis Decision Aid (YoDDA) Booklet: a prospective non-randomized comparison study across 6 predialysis services.
        Perit Dial Int. 2016; 36: 374-381
        • Schatell D.
        • Agar J.
        • Witten B.
        • Bauer M.
        • Klicko K.
        • Medical Education Institute, Inc
        My Life, My Dialysis Choice.
        https://mydialysischoice.org/
        Date accessed: September 1, 2017
        • Ayanian J.Z.
        • Cleary P.D.
        • Weissman J.S.
        • Epstein A.M.
        The effect of patients' preferences on racial differences in access to renal transplantation.
        N Engl J Med. 1999; 341: 1661-1669
        • Norris K.C.
        • Agodoa L.Y.
        Unraveling the racial disparities associated with kidney disease.
        Kidney Int. 2005; 68: 914-924
        • Binik Y.M.
        • Devins G.M.
        • Barre P.E.
        • et al.
        Live and learn: patient education delays the need to initiate renal replacement therapy in end-stage renal disease.
        Nerv Ment Disord. 1993; 181: 371-376
        • Perestelo-Perez L.
        • Rivero-Santana A.
        • Sanchez-Afonso J.A.
        • et al.
        Effectiveness of a decision aid for patients with depression: a randomized controlled trial.
        Health Expect. 2017; 20: 1096-1105
        • Sheu J.
        • Ephraim P.L.
        • Powe N.R.
        • et al.
        African American and non-African American patients' and families' decision making about renal replacement therapies.
        Qual Health Res. 2012; 22: 997-1006
        • Nelson W.L.
        • Han P.K.
        • Fagerlin A.
        • Stefanek M.
        • Ubel P.A.
        Rethinking the objectives of decision aids: a call for conceptual clarity.
        Med Decis Making. 2007; 27: 609-618
        • Feldman-Stewart D.
        • Tong C.
        • Siemens R.
        • et al.
        The impact of explicit values clarification exercises in a patient decision aid emerges after the decision is actually made: evidence from a randomized controlled trial.
        Med Decis Making. 2012; 32: 616-626
        • Clarke A.L.
        • Yates T.
        • Smith A.C.
        • Chilcot J.
        Patient's perceptions of chronic kidney disease and their association with psychosocial and clinical outcomes: a narrative review.
        Clin Kidney J. 2016; 9: 494-502
        • King K.
        Patients’ perspective of factors affecting modality selection: a National Kidney Foundation patient survey.
        Adv Ren Replace Ther. 2000; 7: 261-268
        • Golper T.
        Patient education: can it maximize the success of therapy?.
        Nephrol Dial Transplant. 2001; 16: 20-24