Plan-switching patterns among Medicare beneficiaries who disenroll

Since 2013, Medicare has conducted an annual survey to understand the reasons why beneficiaries voluntarily disenroll from Medicare Advantage (MA) and Prescription Drug Plans (PDPs). Little is known about which types of plans beneficiaries choose when they switch and the factors associated with the types of switches beneficiaries make. We examined the switching patterns of Medicare beneficiaries (i.e., shifting to fee-for-service, to another plan of the same type, or to a different plan type) and the association between beneficiary choice of plan type and reasons for disenrollment, the plan’s Star rating, and beneficiary characteristics.   

We used data from the 2013 Medicare disenrollment survey, based on a random sample of 51,895 Medicare beneficiaries who voluntarily disenrolled from MA-PD or PDP plan coverage. We constructed five composite reasons for disenrolling: financial considerations, concerns about benefits and coverage, patient experiences with prescription drugs, patient experiences with the plan, and the plans’ coverage of specific doctors and hospitals. Beneficiary and plan characteristics were based on Medicare administrative sources and self-reported survey responses. Person-level multinomial logistic regression models estimated adjusted odds ratios for each predictor and each possible plan switch (e.g., MA-PD to PDP) relative to choosing another plan of the same type. 

For both MA-PD and PDP, most disenrollees (72-75%) switched to another plan of the same type. Only 16% switched from MA-PD to PDP and 23% from PDP to MA-PD; beneficiaries rarely switched to an MA-Only plan or into fee-for-service without PDP coverage. Financial reasons for disenrolling were particularly associated with retaining the same type of coverage. In contrast, citing disenrollment reasons related to patient experience was positively associated with switching to a different coverage type. Higher educational attainment (a proxy for higher socioeconomic status) was positively correlated with a switch to a PDP (from either MA or PDP), suggesting such beneficiaries preferred more flexible access to doctors and hospitals but with greater out-of-pocket expenses. Beneficiaries who disenrolled from plans with high Star quality ratings were more likely to have switched plan type, suggesting they may not like the type of coverage, whereas disenrollees from plans with low Star ratings tended to switch to another plan of the same coverage type. 

These findings may inform plan quality improvement efforts. Clear communication of cost, patient experience, and other quality data can improve plan choice. Nonetheless, some level of disenrollment may indicate activated patients responding appropriately to information only obtainable through interactions with plans.