Popescu Ioana, Schrag Deborah, Ang Alfonso, Wong Mitchell
*Division of General Internal Medicine and Health Services Research, David Geffen School of Medicine, Los Angeles, CA †Division of Population Sciences, Dana Farber Cancer Institute and the Harvard Medical School, Boston, MA.
Med Care. 2016 Aug;54(8):780-8. doi: 10.1097/MLR.0000000000000561.
Despite a large body of research showing racial/ethnic and socioeconomic disparities in cancer treatment quality, the relative role of physician-level variations in care is unclear.
To examine the effect of physicians on disparities in breast and colorectal cancer care.
Linked SEER Medicare data were used to identify Medicare beneficiaries diagnosed with colorectal and breast cancer during 1995-2007 and their treating physicians.
We identified treating physicians from Medicare claims data. We measured the use of NIH guideline-recommended therapies from SEER and Medicare claims data, and used logistic models to examine the relationship between race/ethnicity, socioeconomic status, and cancer quality of care. We used physician fixed effects to account for between-physician variations in treatment.
Minority and low socioeconomic status beneficiaries with breast and colorectal cancer were less likely to receive any recommended treatments as compared with whites. Overall, between-physician variation explained <20% of the total variation in quality of care. After accounting for between-physician differences, median household income explained 14.3%, 18.4%, and 13.2% of the variation in use of breast-conserving surgery, chemotherapy, and radiation for breast cancer, and 13.7%, 12.9%, and 12.6% of the within-physician variation in use of colorectal surgery, chemotherapy, and radiation for colorectal cancer, whereas race and ethnicity explained <2% of the within-physician variation in cancer care.
Between-physician variations partially explain racial disparities in cancer care. Residual within-physician disparities may be due to differences in patient-provider communication, patient preferences and treatment adherence, or unmeasured clinical severity.
尽管大量研究表明在癌症治疗质量方面存在种族/族裔和社会经济差异,但医生层面护理差异的相对作用尚不清楚。
研究医生对乳腺癌和结直肠癌护理差异的影响。
利用关联的监测、流行病学和最终结果(SEER)医疗保险数据,确定1995 - 2007年间被诊断为结直肠癌和乳腺癌的医疗保险受益人和他们的治疗医生。
我们从医疗保险理赔数据中确定治疗医生。我们从SEER和医疗保险理赔数据中衡量美国国立卫生研究院(NIH)指南推荐疗法的使用情况,并使用逻辑模型来研究种族/族裔、社会经济地位与癌症护理质量之间的关系。我们使用医生固定效应来解释医生之间治疗的差异。
与白人相比,患有乳腺癌和结直肠癌的少数族裔和社会经济地位较低的受益人接受任何推荐治疗的可能性较小。总体而言,医生之间的差异解释了护理质量总差异的不到20%。在考虑医生之间的差异后,家庭收入中位数解释了乳腺癌保乳手术、化疗和放疗使用差异的14.3%、18.4%和13.2%,以及结直肠癌手术、化疗和放疗医生内部使用差异的13.7%、12.9%和12.6%,而种族和族裔解释了医生内部癌症护理差异的不到2%。
医生之间的差异部分解释了癌症护理中的种族差异。医生内部残留的差异可能是由于患者与提供者沟通的差异、患者偏好和治疗依从性,或未测量的临床严重程度。