Associations of Patient Activation with Outcomes Among Patients on Chronic Hemodialysis: A Retrospective, Longitudinal, Observational Study.

Background: Patient activation, defined as an individual's readiness, willingness, and ability to manage their own health and health care, is associated with healthy behaviors and improved outcomes. Patients undergoing in-center hemodialysis (HD) have low activation, but the association of patient activation with clinical outcomes among dialysis patients is unclear. We investigated the association between patient activation and outcomes in HD patients.

Methods: This cohort included 925 prevalent, in-center HD patients among 10 facilities in a mid-sized dialysis provider. All patients who completed the Patient Activation Measure 13-item (PAM-13) survey during a previous study were included, and their records were cross-referenced with data from the electronic heath system. Patients were followed for 180 days after completion of the survey for the primary outcomes of time to (1) death and (2) time to hospitalization. Markers of non-adherence during the month before and the month after completion of the PAM survey were examined as secondary outcomes, including (1) serum potassium >5.0 mEq/L; (2) serum phosphorus >5.5 mg/dL; (3) missed dialysis treatment due to absence (not hospitalization); and (4) interdialytic weight gain >4.0%. Univariate and adjusted regression models were fit to estimate associations of a 3-point increment in PAM-13 score with the outcomes of interest; adjustment factors comprised age, sex, dialysis vintage, serum albumin, diabetes, and hospitalization history.

Results: A 3-point increment in PAM score was associated with lower hazard of death (univariate HR=0.89, 95% CI: 0.84-0.94; adjusted HR=0.90, 95% CI: 0.85-0.96), but not with hospitalization (univariate HR=0.99, 95% CI: 0.96-1.02; adjusted HR=0.99, 95% CI: 0.96-1.02). Higher scores were associated with increased odds of having high phosphorus levels in the unadjusted analysis, but this was attenuated and not significant in adjusted models. There were no significant relationships between a 3-point increment in PAM score and any of the other secondary outcomes in univariate and adjusted analyses.

Conclusion: In a cohort of prevalent, in-center HD patients, low activation was associated with mortality but not with hospitalization or measures of non-adherence.