To develop a quasi‐experimental method for estimating opulation ealth anagement () program savings that mitigates common sources of confounding, supports regular updates for continued program monitoring, and estimates model precision.
Administrative, program, and claims records from anuary 2005 through une 2009.
Data are aggregated by member and month.
Study participants include chronically ill adult commercial health plan members. The intervention group consists of members currently enrolled in , stratified by intensity level. Comparison groups include (1) members never enrolled, and (2) participants not currently enrolled. Mixed model smoothing is employed to regress monthly medical costs on time (in months), a history of enrollment, and monthly program enrollment by intensity level. Comparison group trends are used to estimate expected costs for intervention members. Savings are realized when participants' costs are lower than expected.
This method mitigates many of the limitations faced using traditional pre‐post models for estimating savings in an observational setting, supports replication for ongoing monitoring, and performs basic statistical inference.
This method provides payers with a confident basis for making investment decisions.
Data Collection/Extraction Methods