Survival advantage of marriage in uterine cancer patients contrasts poor outcome for widows: A Surveillance, Epidemiology and End Results study

Abstract

BackgroundMarriage confers a survival advantage for many cancers but has yet to be evaluated in uterine cancer patients. We sought to determine whether uterine cancer survival varied by self-reported relationship status. MethodsData were downloaded from the Surveillance, Epidemiology, and End Results program for women diagnosed with uterine cancer (between 1991 and 2010 in nine geographic regions). Patients with complete clinical data for analysis were categorized as married, single, widowed or other (divorced or separated). Differences in distributions were evaluated using Chi-square, exact and/or Mantel–Haenszel test. Uterine cancer survival was analyzed by Kaplan–Meier method with log-rank test and multivariate Cox regression analysis. ResultsOf 47,420 eligible patients, 56% were married, 15% were single and 19% were widows. Married vs. non-married women had a higher likelihood of having low risk (grade 1/2 endometrioid) endometrial cancer and local disease (p<0.0001), and a reduced risk of cancer death (HR=0.8, 95% CI=0.77–0.84). Multivariate evaluation of uterine cancer survival by relationship type indicated that widows consistently had significantly worse uterine cancer survival than single, married and other women in all patients and subset analyses (p<0.0001). ConclusionWhile marital status is associated with differential uterine cancer survival, evaluation of self-reported relationship by type indicated that the poor outcome observed in widows explained most of the benefit attributed to marriage. This report identifies widows as a new high-risk subpopulation with significantly inferior outcomes potentially benefiting from personalized care and social support.