The purpose of this study is reduce episodes of intradialytic hypotension, low blood pressure during a hemodialysis session, in patients with End Stage Renal Disease (ESRD). Recruitment will take place on the clinic level rather than the patient level.

When a person's kidneys stop working, he or she has end-stage renal disease (ESRD). Individuals with ESRD cannot live without either dialysis therapy-in which a machine performs the functions of the kidneys-or a kidney transplant. Dialysis must remove fluid as well as toxins in the blood. People with ESRD have a high risk for death, and the usual cause is cardiovascular disease.

Most people in the United States who have ESRD get hemodialysis therapy in a clinic for four hours at a time, three times a week. The stability of hemodialysis sessions varies, and many sessions become unstable from low blood pressure and other complications. Unstable dialysis sessions can result in negative symptoms, like fatigue.

Dialysis instability is an important problem. Session instability is linked to injury to the heart and other organs. Patients who have unstable dialysis sessions are more likely to end up in the hospital or die than are those who have stable sessions. Session instability is preventable. The main causes of instability are removal of fluid from a patient too fast or removal of too much fluid. Session instability results from many factors: decisions made by patients, and decisions by healthcare providers.

Presently, the way to best improve the stability of dialysis is not clear. Dialysis clinics approach this problem differently, and there is variation among clinics in how often hemodialysis sessions become unstable.

In partnership with the National Kidney Foundation and Fresenius Medical Care North America, the investigators will test two interventions designed to increase the stability of patient dialysis. One intervention, multimodal provider education, focuses on dialysis facility care teams. It includes team training, online education, and a checklist. Another intervention, patient activation, focuses on patients. It includes peer mentoring by trained ESRD patients. Mentors will hold with other patients multimedia-aided meetings that include skills instruction and role modeling. These interventions have been successful in hospital care and in chronic disease care, and the investigators will adapt them to dialysis safety.

The investigators will then conduct a study in 20 dialysis facilities in different parts of the United States. Five facilities will get the provider education only; five will get the patient activation intervention only; five will get both interventions; and five will get no interventions. The investigators will test whether session stability improves in the facilities that get either intervention over the 48-week study period. This study is expected to clarify whether these interventions can make dialysis safer for ESRD patients. This will inform hemodialysis care providers on whether to pursue provider-focused or patient-focused safety interventions, or both. People on hemodialysis will also have information to help them decide whether to become engaged in their session stability, and the intervention will help them learn how to do so.

Selection is at both the clinic level and individual patient level.

Clinic Inclusion Criteria:

• outpatient hemodialysis facilities
• at least 70 adult (>21 years old) patients to guarantee sample size

Clinic Exclusion Criteria:

• facilities involved in another study
• facilities in immediate jeopardy
• facilities with 1-star quality ratings
• facilities designated as COVID-19 isolation facilities

Individual Patient Exclusion Criteria:

• individual patients who are currently incarcerated
• individual patients who have poor cognition or cognitive impairment
• individual patients unable to comprehend the patient information sheet due to lack of facility in English or Spanish
• individual patients who have opted out of data collection

These are all participants who were analyzed for baseline data, i.e. all enrolled participants regardless of how long they were enrolled in the study.

Adverse events were monitored throughout all study periods: baseline (12 weeks), intervention (24 weeks), and follow up (12 weeks). Baseline monitoring was limited to counts of all-cause mortality because data for this time period were collected retrospectively.

This was a pragmatic, low-risk trial with no changes to prescribed care. All-cause mortality was tracked. Adverse events relating to a breach in confidentiality were not observed. Other adverse events were not systematically tracked because the patient is very ill at baseline.

Results displayed for the primary and secondary outcomes are based on data from the 24-week intervention period, with the exception of quality of life measures based on the post-intervention quality of life survey conducted annually in each clinic.