Decision Aid for Renal Therapy (DART)

Decision Aid for Renal Therapy: Promoting Knowledge and Autonomy in Chronic Kidney Disease Patients and Their Care-Partners

Good communication among patients, their families and loved ones, and their medical care providers is important when figuring out how to treat chronic diseases like kidney disease. A lot of people may not know all of their choices for how to treat kidney disease, and this can lead to rushed decisions or even a sense that there weren't any choices to make. In this study, the investigators are trying to find out if a decision-aid program on a computer can help people with kidney disease have more confidence in their decisions and have better agreement about their decisions with their families and loved ones. The DART study will be conducted at four sites in different areas of the country: Boston, Massachusetts; Portland, Maine; Chicago, Illinois; and San Diego, California. The study will enroll a total of 400 people with kidney disease at these four sites.

Aligning patient preferences (goals of care and values) with treatment is essential for quality health care. Treatment of life-limiting illness is especially preference-sensitive, where high-intensity care often offers marginal survival benefit but can worsen quality of life. Elderly persons with advanced chronic kidney disease (CKD) may face a choice between high-intensity dialysis and low-intensity conservative management (CM). This decision is of special import because it is often irreversible as dialysis itself can diminish residual kidney function. In adults over 70 years-old with advanced CKD, dialysis on average confers only marginally better survival than medical management while reducing independence and mobility and increasing medical procedures. Poor communication about benefits and risks of treatment options available to these patients results in decisional conflict: a state of uncertainty associated with making a choice that best reflects values and preferences. However, it is unclear how to best educate patients and their care-partners about their treatment choices and the importance of advance care planning. The objective of this trial is to compare the effectiveness of two widely used strategies, in-person education alone versus in-person education plus an interactive web-based decision aid, in: 1) reducing decisional conflict and empowering patients and care-partners to select treatment aligned with patient preferences; and 2) improving care-partners' ability to confidently and accurately express patients' preferences when patients are unable (proxy decision-making). The hypothesis is that, compared to traditional in-person education, use of the Decision Aid for Renal Therapy (DART) will decrease decisional conflict, increase completion of advance directives, improve patient and caregiver satisfaction with treatment (quantitative outcomes), and contribute to greater patient engagement, satisfaction with decision-making, and care-partner concordance (qualitative and quantitative outcomes). DART is a web-based multimedia decision-aid that is designed to be accessible to individuals with limited health literacy. The investigators will evaluate whether use of DART results in greater patient understanding of options, leading to better discussions with care providers, and ultimately lower decisional conflict and greater completion of advanced directives compared with the comparator, in-person education. DART is replicable, consistent, can be shared with care-partners, and can be viewed in the comfort of the patient's home. This randomized clinical trial targeting 400 older adults with advanced kidney disease and as many as 400 of their care-partners compares the effectiveness of DART plus in-person education to in-person education alone for reducing decisional conflict and increasing completion of advance care plans (ACPs) among older adults with advanced CKD and their care-partners. Patients and patient-care-partner pairs will be surveyed at baseline for goals of care, life goals, health literacy, patient activation, end-of-life (EOL) preferences and baseline scores on other study outcome measures, and followed at 3- to 6-month intervals for up to 18 months, censoring at dialysis, death or study end, to collect data on decisional conflict and completion of advance directives as well as QOL, satisfaction, dyad concordance and medical events, such as dialysis initiation.

Chronic Kidney Disease, Chronic Kidney Failure, Palliative Care, Dialysis, End-of-life care, Decision making

Usual Care as in the 'no intervention arm' below plus access to an web-based decision aid, the Decision Aid for Renal Therapy to patients and their care-partners

In-person education as would be done at study sites plus 'Choosing a Treatment for Kidney Failure', an educational booklet published by the National Kidney Foundation

DART is a web-based decision aid that informs older adults with advanced kidney disease of kidney disease treatment options and prompts them to consider their preferences and raise questions to discuss with their kidney disease providers.

The decisional conflict scale (DCS) measures personal perceptions of: Uncertainty in choosing options; and Modifiable factors contributing to uncertainty. The low literacy version of the DCS contains 10 items scored on a 3 point scale, with 0 points given for an answer of yes, 2 for an answer of unsure, and 4 for an answer of no. The total DCS score is the sum of the 10 items, multiplied by 2.5. Scores range from 0 (no decisional conflict) to 100 (extremely high decisional conflict) The DCS can be conceptualized as subscores, which will be examined as a part of this study, and are normalized to a 0 to 100 score as above. These include the following: Uncertainty subscore: 2 items (questions 9 and 10) Informed subscore: 3 items (questions 1, 2, and 3) Values clarity subscore: 2 items (questions 4 and 5) Support subscore: 3 items (questions 6, 7, and 8) Reference: https://decisionaid.ohri.ca/docs/develop/User_Manuals/UM_Decisional_Conflict.pdf

The decisional conflict scale (DCS) measures personal perceptions of: Uncertainty in choosing options; and Modifiable factors contributing to uncertainty. The low literacy version of the DCS contains 10 items scored on a 3 point scale, with 0 points given for an answer of yes, 2 for an answer of unsure, and 4 for an answer of no. The total DCS score is the sum of the 10 items, multiplied by 2.5. Scores range from 0 (no decisional conflict) to 100 (extremely high decisional conflict) The DCS can be conceptualized as subscores, which will be examined as a part of this study, and are normalized to a 0 to 100 score as above. These include the following: Uncertainty subscore: 2 items (questions 9 and 10) Informed subscore: 3 items (questions 1, 2, and 3) Values clarity subscore: 2 items (questions 4 and 5) Support subscore: 3 items (questions 6, 7, and 8) Reference: https://decisionaid.ohri.ca/docs/develop/User_Manuals/UM_Decisional_Conflict.pdf

Completion of an advance directive, Physician Orders for Life-Sustaining Treatment (POLST), or both among patient-participants ascertained by participant query and/or chart review.

Change in score on the CANHELP Lite Patient Questionnaire version 11 Nov 2014 for patient-participants. The CANHELP Lite Patient Questionnaire includes 21 items. One is a general question while the other 20 items inform 5 domains: Domain #1: Relationship with doctors: Questions 2, 3, and 4 Domain #2: Illness Management: Q5 - 13 Domain #3: Communication: Q13 - 16 Domain #4: Decision Making: Q17 - 20 Domain #5: Feeling at Peace: Q21 Questions are scored on a 5-point scale, ranging from 'Not at All' to 'Completely'. The Overall Score is the unweighted average of all answered questions. Individual domain scores are the unweighted average of non-missing questions specific to each domain. All scores are subsequently re-scaled to range between 0 (worst possible value) and 100 (best possible value). The Overall Score will be the focus of analyses. Ref: Heyland et al. J Pain Symptom Manage. 2013

Change in score on the CANHELP Lite Caregiver Questionnaire 11 Nov 2014 for carepartner participants. The CANHELP Lite Caregiver Questionnaire includes 23 items. Two are general questions while the remaining 21 items inform 5 domains: Domain #1: Relationship with doctors: Questions 3 - 5 Domain #2: Characteristics of Doctors and Nurses: Q6-7 Domain #3: Illness Management: Q8-16 Domain #4: Communication and Decision Making: Q17-20 Domain #5: Your Involvement: Q21-23 Questions are scored on a 5-point scale, ranging from 'Not at All' to 'Completely'. The Overall Score is the unweighted average of all answered questions. Individual domain scores are the unweighted average of non-missing questions specific to each domain. All scores are subsequently re-scaled to range between 0 (worst possible value) and 100 (best possible value). The Overall Score will be the focus of analyses. Ref: Heyland et al. J Pain Symptom Manage. 2013

Change in score on the CANHELP Lite Patient Questionnaire version 11 Nov 2014 for patient-participants. The CANHELP Lite Patient Questionnaire includes 21 items. One is a general question while the other 20 items inform 5 domains: Domain #1: Relationship with doctors: Questions 2, 3, and 4 Domain #2: Illness Management: Q5 - 13 Domain #3: Communication: Q13 - 16 Domain #4: Decision Making: Q17 - 20 Domain #5: Feeling at Peace: Q21 Questions are scored on a 5-point scale, ranging from 'Not at All' to 'Completely'. The Overall Score is the unweighted average of all answered questions. Individual domain scores are the unweighted average of non-missing questions specific to each domain. All scores are subsequently re-scaled to range between 0 (worst possible value) and 100 (best possible value). The Overall Score will be the focus of analyses. Ref: Heyland et al. J Pain Symptom Manage. 2013

Change in score on the CANHELP Lite Caregiver Questionnaire 11 Nov 2014 for carepartner participants. The CANHELP Lite Caregiver Questionnaire includes 23 items. Two are general questions while the remaining 21 items inform 5 domains: Domain #1: Relationship with doctors: Questions 3 - 5 Domain #2: Characteristics of Doctors and Nurses: Q6-7 Domain #3: Illness Management: Q8-16 Domain #4: Communication and Decision Making: Q17-20 Domain #5: Your Involvement: Q21-23 Questions are scored on a 5-point scale, ranging from 'Not at All' to 'Completely'. The Overall Score is the unweighted average of all answered questions. Individual domain scores are the unweighted average of non-missing questions specific to each domain. All scores are subsequently re-scaled to range between 0 (worst possible value) and 100 (best possible value). The Overall Score will be the focus of analyses. Ref: Heyland et al. J Pain Symptom Manage. 2013

A reversal of the post-intervention decision (dialysis or conservative management) from the first visit after DART/3month follow-up and study completion

Inclusion Criteria: CKD stages 4 or 5 (non-dialysis) without an established dialysis start or transplant date within three months of expected randomization; Age >70 (with no upper limit); English speaking; Willingness to be randomized to DART; and Able to sign informed consent. Exclusion Criteria: death, dialysis initiation or transplant deemed highly likely within the next three months by the patient's nephrologist

Porteny T, Gonzales KM, Aufort KE, Levine S, Wong JB, Isakova T, Rifkin DE, Gordon EJ, Rossi A, Di Perna G, Koch-Weser S, Weiner DE, Ladin K; Stakeholder Advisory Board. Treatment Decision Making for Older Kidney Patients during COVID-19. Clin J Am Soc Nephrol. 2022 Jul;17(7):957-965. doi: 10.2215/CJN.13241021. Epub 2022 Jun 7.