The purpose of this study is to test the effect of the "Best Case/Worse Case" (BC/WC) communication tool on receipt of palliative care and intensity of treatment at the end of life, quality of life, and quality of communication for older patients with end-stage renal disease (ESRD) receiving outpatient care at ten nephrology clinics. The intervention was developed and tested with acute care surgical patients at the University of Wisconsin (UW) and is now being testing to see if the intervention will work in a different setting.

The intervention will be tested with 320 older adults who have end-stage renal disease (ESRD) and are receiving care from a nephrologist enrolled in the study. Randomly assigned nephrologists within each site will receive the intervention (training to use the BC/WC tool) or to be in the waitlist control, meaning that they will not be offered BC/WC training until the end of the study, when all participants have been enrolled. Participants will be on follow up with surveys and chart review for up to two years after study enrollment. Caregivers will also be invited to participate and complete surveys.

This study will test the effect of the Best Case/Worst Case intervention on receipt of palliative care and intensity of treatment at the end of life, quality of life, and quality of communication for older patients with end-stage renal disease (ESRD). This multi-site cluster randomized trial will enroll 320 participants who are making a dialysis initiation decision and receive care from a nephrologist trained to use the Best Case/Worst Case tool, or care from a nephrologist who has not been trained to use this tool (usual care). Randomly assigned nephrologists within each site will receive the intervention or waitlist control (upon study completion). Participants will be on follow up for up to two years after study enrollment via regular surveys and chart review. This study has three aims:

Aim 1: To test the effect of the Best Case/Worst Case intervention on (1) receipt of palliative care and (2) intensity of treatment at the end of life for older patients with ESRD. Chart reviews will be used to determine whether participants have received at least one outpatient or inpatient palliative care consultation within 12 months of enrollment in the study. These consultations must be clearly marked as palliative care, provided by a clinician with palliative care training and have documented discussion of goals clarification, advance care planning, symptom management, coping, spiritual needs, or end-of-life care. To measure intensity of treatment received at the end of life, it will be determined whether participants have had an ICU admission within 30 days of death as a primary outcome and ICU admission, emergency room (ER) visit, or hospital admission within 30 days of death as a composite secondary outcome.

Aim 2: To test the effect of the Best Case/Worst Case intervention on quality of life. The primary outcome for Aim 2 is quality of life as measured by the Functional Assessment of Chronic Illness Therapy -Palliative Care (FACIT-Pal Version 4) at baseline, and every three months for up to 2 years after study enrollment. The hypothesis is that the overall quality of life will decline over time as participants become more infirm. The average change in health-related quality of life over time which has been shown to decline less with the receipt of concurrent palliative care will be compared.

Aim 3: To test the effect of the Best Case/Worst Case intervention on the quality of communication. To evaluate participant's assessment of nephrologist communication, the Quality of Communication (QOC) scale developed by Randy Curtis will be used. Unlike other measurements of physician communication that have high ceiling effects and limited ability to measure change, the QOC includes 7 items specific to end-of-life communication, which, if not performed by the clinician, are scored as zero. This will allow us to discriminate between quality of communication attributable to participant satisfaction (with high ceiling effects) versus content.

Inclusion Criteria:

• Estimated glomerular filtration rate (eGFR) of less than or equal to 24
• Not currently on dialysis (participants are eligible if they have had intermittent dialysis in the past or have dialysis access in place but are not currently on dialysis)
• Participants must meet one or more of the following criteria: age greater than 80, evidence from the medical record that the patient has comorbid illness such that the modified Charlson score is 4 or greater, or a negative response to the standard "Surprise Question" ("Would you be surprised if this patient died in the next year?") from the participant's nephrologist.

Exclusion Criteria:

• Currently on dialysis
• Lack decision-making capacity
• Do not speak English

A de-identified data set from this project will be stored in the Palliative Care Research Cooperative Group (PCRC) De-identified Data Repository (DiDR).

One year after study completion or two months after initial manuscript has been accepted for publication (whichever comes first), UW will securely transfer a de-identified dataset to the PCRC Data Repository.

Investigators may request access to de-identified study data for secondary analyses using the PCRC Data Repository webform. Requests must include a summary of the following: research questions, objectives, hypotheses, statistical analysis plan, and project timeline. Requestors must provide a list of specific variables they would like included in the secondary data set. Each request must also include a data security plan and explanation of how the data will be stored and who will have access to it. The PCRC staff and UW PI will review all requests assessing the scientific integrity, feasibility and analytic rigor.