Gumpertz Marcia L, Pickle Linda Williams, Miller Barry A, Bell B Sue
Department of Statistics, North Carolina State University, Raleigh, 27695-8203, USA.
Cancer Causes Control. 2006 Apr;17(3):325-39. doi: 10.1007/s10552-005-0513-1.
Examination of patterns of advanced breast cancer may provide evidence needed to direct health care resources to those communities or population groups in greatest need. We assessed to what degree biologic, ethnic, and sociodemographic factors could explain such patterns within Los Angeles County.
The proportion of cases of advanced disease among all breast cancer cases identified during 1992-1996 were analyzed using generalized linear mixed models with random census tract effects. Models included characteristics of the individual and her tumor, census tract of residence, and aggregated health districts.
Approximately 6% of cases, ranging from 4% for Asian to 10% for Black women, were diagnosed as advanced, exhibiting striking geographic patterns. Tumor histology and hormone receptor status were most predictive of advanced disease. Sociodemographic variables such as marital status, median income, and distance to nearest mammography unit showed additional association with risk.
These models explain most of the geographical patterns and eliminate differences between White and Hispanic but not Asian or Black women, identify subpopulations at high risk of advanced disease, and suggest cancer control opportunities.
对晚期乳腺癌模式的研究可为将医疗资源导向最有需求的社区或人群提供所需证据。我们评估了生物学、种族和社会人口统计学因素在多大程度上能够解释洛杉矶县内的此类模式。
采用具有随机普查区效应的广义线性混合模型，分析了1992年至1996年期间确诊的所有乳腺癌病例中晚期疾病病例的比例。模型包括个体及其肿瘤的特征、居住普查区和综合健康区。
约6%的病例被诊断为晚期，亚洲女性为4%，黑人女性为10%，呈现出显著的地理模式。肿瘤组织学和激素受体状态最能预测晚期疾病。婚姻状况、收入中位数和到最近乳腺摄影单位的距离等社会人口统计学变量显示出与风险的额外关联。
这些模型解释了大部分地理模式，消除了白人和西班牙裔女性之间的差异，但未消除亚洲或黑人女性之间的差异，识别出晚期疾病高风险亚人群，并提示了癌症控制机会。
