Patient Cost-sharing Effects on Emergency Department Visits: Incentives and Alternatives

Objectives: We examined the impact of emergency department (ED) and office visit copayment on ED visit rates for low and high severity conditions in a prepaid, integrated delivery system (IDS).

Methodology: In a quasi-experimental longitudinal study, we classified the severity of ED visits using ICD-9 codes and the New York University (NYU) ED-coding system. We examined the association between the ED copayment amount and visits that were non-emergencies, emergencies, or indeterminate, and changes in the ED copayment effect on ED visits as office copayments increased. Using automated databases and the 2000 US Census, we collected data over a 36-month period (January 1999 through December 2001) among 2,257,445 commercially insured health system members. We applied negative binomial regression analysis with gamma–distributed patient random-effects to patients' repeated annual ED visit counts. We modeled secular trends with year terms and included an exposure variable for the months that each subject was enrolled in the health system. We adjusted for age, gender, neighborhood socio-economic status (SES), prior utilization levels, comorbidity levels, medical center, and drug cost-sharing levels.

Results: Among commercially insured subjects, the relative ED visit rates for non-emergencies decreased with higher copayment amounts, RR=0.98 [95% CI: 0.97–0.99] ($1–5 copayment/visit), 0.95 [0.93-0.97] ($10–15), 0.86 [0.85-0.88] ($20–35), and 0.73 [0.71-0.75] ($50–100), compared with no copayment. For visits classified as emergencies, the corresponding relative visit rates were $1-5: 1.02 [1.00-1.04], $10-15: 0.97 [0.94-1.00], $20-35: 0.95 [0.93-0.97], and $50-100: 0.86 [0.83-0.89]. Importantly, as the office visit copayment increased for a given ED copayment amount, the magnitude of the ED copayment effect on ED visit rates lessened; for non-emergency visits: RR=0.69 [0.65–0.72] with $50-100 ED-copayments and $0 Office-copayments; RR=0.83 [0.76-0.90] with $50-100 ED-copayments and $30 Office-copayments, compared with no copayment for both ED and Office visits. There was a similar trend for visits classified as emergencies: RR=0.78 [0.72-0.84] with $50-100 ED-copayment with $0 Office-copayments; RR=1.04 [0.93-1.17] with $50-100 ED-copayment with $30 Office-copayments, compared with no copayment.

Conclusions: As the cost of substitutes increases, the magnitude of cost-sharing effects decreases. Overall, patients who face ED copayments reduce their use of the ED for non-emergency conditions; high ED copayment amounts (e.g., $50-100) are associated with some reduction in ED use for emergency conditions.

Implications: Studies examining the effects of cost-sharing, benefits, and insurance coverage should include explicit information on the availability and costs of potential substitutes.