Comparative Effectiveness of Communication Strategies for Medicare Beneficiaries
Keywords: Cost effectiveness, comparative effectiveness, communication
Communicating benefits and motivating enrollment in the Medicare Low Income Subsidy (LIS) program that assists low-income Medicare beneficiaries by covering their prescription drug costs and Part D premiums has proven to be challenging. Recent estimates suggest that 2 million beneficiaries may be eligible for but are not enrolled in this program. To date little empirical evidence has been accumulated about the differential impact of various approaches to outreach and communication. Such studies could help policy makers improve upon current outreach approaches for this vulnerable population. We have recently presented results from a field experiment evaluating the impact of several direct marketing strategies in comparison with standard community-based efforts and have discussed how using multiple imputation procedures can help avoid biases in the evaluation due to item non-response to income questions. This report extends that work to include preliminary cost-effectiveness data and compares estimates based on self-reported application for the benefit to those obtained from administrative data on study participants during the four months following the interventions.
Details of the experimental design have been presented previously (Funderburk et al., 2009a,b). Participants were randomly assigned to one of four direct marketing strategies that varied along a continuum of intensity or to a control condition. Application rates among those in the target audience control group ranged from 0.08% (self-report, within 2 weeks of intervention) and 0.51% (administrative data within 4 months of intervention), while rates among those in the lowest cost direct marketing strategy ranged from 2.84 (self-report) and 2.51% (administrative). Rates among those in the most intensive direct marketing strategy ranged from 3.72% (self-report) to 4.76% (administrative). The cost-effectiveness ratios favored the lowest cost direct marketing strategy (e.g., agency letter @ $54/application vs. outbound personal assistance with online application @ $207/application, based on validated administrative data). Findings were robust under sensitivity analysis and in scale-ups involving estimates for campaigns of 1 and 3 million individuals.
These results suggest that relatively low-cost direct marketing strategies can significantly enhance outreach effectiveness for the LIS, perhaps by effectively driving individuals to community based programs or other resources.