Sharing the Burden: Are Cost Sharing Initiatives as Effective among Individuals with Lower Incomes?

The RAND health insurance experiment completed in the mid-1980s remains the seminal work on health insurance and cost sharing, showing that increasing patients’ share of the cost through copays or higher co-insurance reduces health care spending without negative effects on patient health. Although the RAND study did not find substantial differences between income groups in patient responses to these changes in price for health services, health care costs have increased substantially since then, and people may now be more sensitive to these higher costs and may have difficulties in making decisions about where to cut back on utilization. Coupled with the higher rate of chronic disease among individuals with lower incomes, there is also a distinct possibility that foregoing prescriptions, lab tests, or other procedures because of financial pressures could lead to higher overall health care costs via emergency services.

In a 2013 study, economists Amitabh Chandra, Jonathan Gruber, and Robin McKnight take advantage of a 2006 Massachusetts program in order to update the RAND findings with current data. The program provided completely subsidized health insurance to residents with incomes up to 150 percent of the federal poverty line and partially subsidized insurance for residents with incomes between 150 and 300 percent. In July 2008, the copayments (the part of the cost that patients pay directly) were increased for all individuals. This discontinuity is exploited by the authors to estimate elasticities of demand for health care products and services for these low-income populations. (The elasticity of demand is a measure of how consumers respond to a price change and is measured by the percent change in quantity demanded divided by the percent change in price. In this setting, negative elasticities indicate that patients responded to the higher prices by purchasing less health care.)

The authors estimate that the elasticity of demand for health spending for all enrollees—having incomes between 100 and 300 percent of the federal poverty line—is -0.16, which indicates that a ten percent increase in cost to the patient was associated with to a 1.6 percent decrease in overall health care use. This was combined with a statistically insignificant change in hospital-specific usage, meaning that patients did not seem to begin putting off primary care and routine services only to later show up at hospitals with serious health problems.

Importantly, individuals with a chronic disease, such as hypertension, high cholesterol, diabetes, asthma, or arthritis, who constitute 31 percent of the sample, are far less sensitive to the price increases. When costs rise, these high-need patients cut back on medical services only a third as much as others. Patients with asthma actually increase their utilization of health care services as the cost rises (elasticity of 0.37). Similarly, individuals in the top quartile of spending are not at all sensitive to the increase in prices and use more health care (elasticity of 0.13).

While these results reaffirm the RAND study’s conclusions that higher copays lead to reduced health care spending overall without serious adverse effects, it is important to note how this conclusion reverses for individuals in the top quartile of spending and individuals with chronic conditions. The Massachusetts Health Policy Commission’s 2013 report on cost trends indicates that five percent of patients account for almost half of all health care spending in Massachusetts and that 29 percent of these patients remain persistently high-cost over multiple years. If these individuals fall within the parts of the population with the lowest sensitivity to price while accounting for the majority of spending, any cost-sharing initiatives are likely to disproportionately punish the sickest patients while having limited ability to rein in overall health care spending.

The policies of the Affordable Care Act, especially its “silver” plans, attempt to restrain overall demand through cost sharing, while limiting these negative effects on the sickest low-income individuals by providing them additional government assistance. Individuals buying coverage on the marketplace appear to have responded, as silver plans have been the most popular option since the Act took effect, constituting 60 percent of all plans sold in the state and federal health exchanges.

Feature Photo: cc/(martinak15)