Effects of Employer-Offered HDHPs on Low-Value Spending in the Privately Insured Population

Background: With healthcare costs rising and overtreatment being identified as a major contributor, stakeholders have sought ways to reduce wasteful spending without negatively affecting quality of care. One avenue that has gained considerable popularity is imposing demand-side cost-sharing in the form of high-deductible health plans (HDHPs). Proponents of these plans argue that they reduce wasteful spending by providing patients with incentives to limit their receipt of low-value services. On the other hand, there is concern that patients may respond to these incentives by reducing their consumption of medical services indiscriminately, which may negatively impact health outcomes.

Objective: To estimate the intent-to-treat (ITT) effect of a firm offering a HDHP and the local-average treatment effect (LATE) on average low-value expenditures per employee.

Data: Individual-level data comes from Truven Health MarketScan® Research Databases, which provides claims data for non-elderly, privately insured employees from a nationally representative set of large US employers.

Methods: The study uses an intent-to-treat (ITT) design to analyze the effect of a firm offering a HDHP on various spending outcomes, and an instrumental-variable (IV) design to determine the local average treatment effect (LATE) of enrolling in a HDHP on spending. For the ITT, we employ a generalized difference-in-differences design with fixed effects for firm and calendar year and estimate differences in spending through a two-part model. For the IV analysis, we use a two-stage least-squares model, using firm offer of a HDHP as an instrument for enrollment in a HDHP. The outcomes of interest in both analyses are low-value spending and total spending in the outpatient setting, which is further grouped by imaging and laboratory spending in order to test for differences in specific subcategories. Low-value spending is proxied using payments for an index of services defined based on measures from the American Board of Internal Medicine Foundation’s Choosing Wisely campaign.

Results: Firm offer of a HDHP is causally associated with across the board reductions of 7.0% (SE 0.021) in low-value outpatient spending, 7.6% (SE 0.018) in total outpatient spending, 21.3% (SE 0.034) in low-value imaging spending, 17.6% (SE 0.065) in total imaging spending, 6.9% (0.052) in low-value laboratory spending, and 4.4% (SE 0.041) total laboratory spending. Enrolling in a HDHP reduces spending by $10.19 (SE 4.82) for low-value outpatient services, $806.64 (SE 202.72) for total outpatient services, $16.80 (SE 6.87) for low-value imaging services, $182.69 (SE 97.44) for total imaging services, $5.44 (SE 3.81) for low-value laboratory services, and $45.00 (SE 49.81) for total laboratory services. Despite the marginal reductions, there is no significant difference in the HDHP effect within a category between total and low-value spending, indicating that while HDHPs reduce spending, they may not lead to a more efficient use of care.

Conclusions: HDHPs may represent too blunt an instrument to bring about a targeted reduction in low-value spending. This could mean that the rapid uptake of HDHPs in recent years is a risky endeavor that could potentially threaten quality of care as a consequence of reducing spending.