Seasonality, Hospitalization, and Initiation of Home Dialysis: A Retrospective Cohort Study.

RATIONALE &

OBJECTIVE

Patients often initiate dialysis for end-stage kidney disease (ESKD) as inpatients, which may steer patients away from home dialysis. Because hospitalizations are often seasonal, this study examined whether inpatient initiation may contribute to the seasonality of home dialysis use.

STUDY

DESIGN

Retrospective cohort study.

SETTING &

PARTICIPANTS

US adults with fee-for-service Medicare insurance who initiated dialysis between January 1, 2007, and November 30, 2020.

EXPOSURE: Season of the year defined by equinoxes and solstices, and inpatient dialysis initiation as a mediator (hospital discharge within 2 weeks of the first outpatient dialysis treatment).

OUTCOME: Home dialysis use at dialysis initiation and at 3, 6, and 12 months after dialysis initiation. Time to first home dialysis use.

ANALYTICAL APPROACH: Multivariable logistic and cause-specific hazard models to examine whether dialysis initiation as an inpatient mediated seasonal differences in home dialysis use.

RESULTS

Among 548,530 adults, the mean age of the population was 71.0, 44% were female, 3% were Asian, 22% were Black, 73% were White, and 2% were other race or ethnicity. Compared with dialysis initiation in summer, initiation in winter was more likely to occur in the inpatient setting (odds ratio [OR], 1.13 [95% CI, 1.11-1.15]). Initiation in winter (vs summer) also had a lower odds of initiating dialysis in the home setting (OR, 0.89 [95% CI, 0.86-0.91]), a lower odds of receiving home dialysis 12 months after dialysis initiation (OR, 0.95 [95% CI, 0.92-0.97]), and a lower rate of home dialysis use over the follow-up period (HR, 0.94 [95% CI, 0.92-0.96]). Mediation analysis demonstrated that initiation of dialysis in the inpatient setting explained 40% of the total association between season and home dialysis use at initiation of dialysis, 77% of the association between season and home dialysis use 12 months after dialysis initiation, and 55% of the association between season and the overall rate of home dialysis use.

LIMITATIONS: Residual confounding from unobserved variables; results limited to patients with fee-for-service Medicare insurance; and analysis of data from before the COVID-19 pandemic.

CONCLUSIONS

Patients initiating dialysis in the winter were less likely to use home dialysis. Initiation of dialysis in the inpatient setting partially mediated these findings and may impede efforts to promote the use of home dialysis.

PLAIN-LANGUAGE SUMMARY: Home dialysis is underused in the United States despite having outcomes similar to in-center hemodialysis and, in many circumstances, being associated with a higher quality of life. Research studies have broadly shown seasonal effects on health care utilization. This study showed that patients starting dialysis in the winter were less likely to initiate home dialysis than during other seasons, and these differences in home dialysis use persisted throughout the first year of dialysis. When examining potential mechanisms for these differences, this study found that dialysis initiation during the winter was more likely to occur in the hospital, which, in turn, was associated with lower use of home dialysis. Initiation of dialysis during a hospitalization appears to be a factor that may impede efforts to promote the use of home dialysis.