The Effect of Insurance Coverage on Breast Cancer Patients' Treatment and Hospital Choices
分析保险类型（HMO、其他私人保险、医疗补助或无保险）如何影响非老年乳腺癌女性的治疗方式（保乳手术 vs 全乳切除）和医院选择（最近癌症医院 vs 更远医院），对研究医疗激励和患者决策的学者有用。
Enrollment in health maintenance organizations (HMO's) has grown rapidly in recent years as both employers and policymakers attempt to constrain health-care spending. Despite the potential cost savings linked to managed-care insurance, critics contend that the financial incentives of such plans will elicit undertreatment and possibly constrain treatment options for persons with costly medical conditions. For example, one study found that heart-attack patients enrolled in HMO's were less likely to undergo either arteriography or coronary bypass surgery than were persons with indemnity insurance (Gary Young and Bruce Cohen, 1991). However, no prior study has investigated whether enrollment in a managed-care plan influences the choice of a particular hospital for inpatient procedures. Given the growth of HMO' s, it is important to assess whether this form of insurance affects treatment and hospital choices. We analyze whether type of insurance (HMO, other private, Medicaid, or no insurance) affects treatment and hospital choices for nonelderly women with breast cancer. We estimate a model of the choice between breastconserving surgery (BCS) and modified radical mastectomy (MRM), and the decision to seek care at the nearest hospital with a program approved by the American College of Surgeons (hereafter called cancer hospital) versus a more distant hospital. Our analysis treats these choices as jointly determined and examines the influence of type of insurance on both decisions. We study breast for two reasons. First, this disease is the most prevalent form of among women and ranks as the second major cause of death in the female population. Second, in 1990 the National Institutes of Health issued guidelines stating that BCS accompanied by lymph-node dissection and radiation therapy is the recommended treatment protocol for women with early-stage breast (NIH Consensus Conference, 1991). Yet, despite the change in guidelines, the use of BCS remains relatively low.