Deductibles and Diabetes Care

As higher deductibles and other forms of cost-sharing spread, concern properly arises over whether these benefit features deter necessary as well as unnecessary care.  A study in the journal Diabetes Care attempts to look at differences in care receipt according to benefit plan type.   (Diabetes Article)   Diabetes currently affects about 30 million Americans and there is a recommended care program that involves regular testing and physician visits, and drugs are the mainstay of treatment.  It is reasonable to wonder if higher cost-sharing amounts are keeping some patients from getting care that might benefit them.  The authors used data from the 2011-13 Medical Expenditure Panel Survey (so the usual cautions about survey data apply) to identify a subcategory of people with diabetes and covered by private health insurance.  The sample is pretty small, only around 1460 people.  Three groups were created–people with no deductible, people with a deductible of less than $1200 a person/$2400 a family and those with deductibles higher than that.  In addition, the respondents were separated by income, 17% were lower-income and the remainder were considered higher income.  Among lower-income people, 29% had no deductible, 49% were in a low deductible plan and 22% had a high deductible.  For the higher-income group, 31% had no deductible, 45% had a low deductible and 24% were covered by a high-deductible plan.

Overall use of health care as well as diabetes-specific care were analyzed.  Within the lower-income group, receipt of primary care, specialty care, and checkups generally declined from no deductible to low deductible to high deductible.  For emergency room use and hospitalizations, the low-deductible group had higher utilization than the no deductible one and the high deductible category.  In the higher income group, The same general pattern existed.  Between income groups, those with low-income and no deductible generally had higher outpatient use but lower ER and inpatient use.  The same pattern was found for low deductibles.  But in the high deductible category, higher income patients had more utilization in almost every category.  Most tellingly, there were no significant differences by deductible or income status in measures related to diabetes care or in health status.  So maybe patients can make a good judgment about prioritizing care in light of their health conditions, or maybe some of the supposedly necessary diabetes care doesn’t actually make a difference in health outcomes.  One deficiency in the study is the failure to separately analyze for presence and use of a health savings account, which limits impact of a higher deductible.  All persons with such an account were just lumped in the high deductible category.  It would be interesting to see if the presence of a savings account led to more use of what is currently viewed as necessary diabetes care.