Understanding how out-of-pocket expenses, treatment value, and patient characteristics influence treatment choices

Abstract

e20543 Background: Cost-sharing is a method to control “over-use” of healthcare services but may have the unintended consequence of reducing utilization of necessary services among vulnerable patients (pts). It is not known if cost influences pts’ choices of cancer treatments. Methods: A survey was administered to a convenience sample of pts on surveillance for localized cancer. Domains included demographic questions as well as clinical scenarios to elicit the maximum co-payment pts would be willing to pay for hypothetical cancer treatments. Scenario A described adjuvant therapy; questions varied reductions in recurrence rates from 5–20%. Scenario B described palliative therapy; questions varied 2-year survival between 19–34%. Scenario C described palliative therapy; questions varied improvement in median life expectancy from 11–20 months. GEE fit multiple logistic regressions examined pts’ characteristics associated with willingness to pay (WTP). Pts chose from 9 co-payment levels (highest: >$7,000/6 months for Scenario A; >$1,000/3 weeks for Scenarios B-C), which were dichotomized (highest 5 vs lowest 4 levels). Results: 60 pts completed the survey. Demographics: Female (78%); < 65 (83%), Caucasian (97%), college degree (58%). 13% reported making financial sacrifices to pay for treatment. Overall, pts showed a greater WTP for more effective treatments (p<0.05 for all 3 scenarios). In Scenario B, pts who were currently working demonstrated a greater WTP (OR 12.6 95% CI 2.0 -80.4) when controlling for dichotomized efficacy. In Scenario C, pts with a college degree showed greater WTP (OR 5.0, 96% CI 1.2–20.9). In addition, pts who reported having to make previous financial sacrifices for care showed lower WTP (OR 0.2 95%CI 0.04–0.6). Conclusions: Pts may assign a higher value to treatments of greater efficacy. Even among this relatively young, affluent and educated group, demographic variables were associated with WTP. Larger studies in more diverse populations are required to further elucidate how cost may influence pts'treatment choices and contribute to health disparities. These findings may inform health policy by suggesting benefit plans that use cost sharing to optimize use of limited health care resources without compromising access to needed care. [Table: see text]