Bailey Katharine, Ryan Andy, Apostolidou Sophia, Fourkala Evangelia, Burnell Matthew, Gentry-Maharaj Aleksandra, Kalsi Jatinderpal, Parmar Max, Jacobs Ian, Pikhart Hynek, Menon Usha
BMC Public Health. 2015 Mar 17;15:253. doi: 10.1186/s12889-015-1609-5.
Evidence is mounting that area-level socioeconomic indicators are important tools for predicting health outcomes. However, few studies have examined these alongside individual-level education. This nested cohort study within the control arm of the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) assesses the association of mutually adjusted individual (education) and area-level (Index of Multiple Deprivation-IMD 2007) socioeconomic status indicators and all-cause female mortality.
Participants resident in England who had completed both baseline (Wave 1) and follow up (Wave 2) questionnaires were included. Follow-up was through the Health and Social Care Information Centre with deaths censored on 31st December 2012. IMD, education and a range of covariates were explored. Cox regression models adjusted for all covariates were used. Sensitivity analysis using imputation was performed (1) including those with missing data and (2) on the entire cohort who had completed the baseline questionnaire.
Of the 54,539 women resident in England who completed both Wave 1 and Wave 2 questionnaires, 4,510 had missing data. The remaining 50,029 women were included in the primary analysis. Area-level IMD was positively associated with all-cause mortality for the most deprived group compared to the least deprived (HR=1.42, CI=1.14-1.78) after adjusting for all potential confounders. Sensitivity analyses showed similar results with stronger associations in the entire cohort (HR=1.90, CI=1.68-2.16). The less educated an individual, the higher the mortality risk (test for trend p=<0.001). However, the crude effect on mortality of having no formal education compared to college/university education disappeared when adjusted for IMD rank (HR=1.08, CI=0.93-1.26).
Women living in more deprived areas continue to have higher mortality even in this less deprived cohort and after adjustment for a range of potential confounders.
This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN22488978.
越来越多的证据表明，地区层面的社会经济指标是预测健康结果的重要工具。然而，很少有研究将这些指标与个体层面的教育情况一起进行考察。这项在英国卵巢癌筛查协作试验（UKCTOCS）对照组中开展的嵌套队列研究，评估了相互调整后的个体（教育程度）和地区层面（多重剥夺指数-2007年IMD）社会经济地位指标与全因女性死亡率之间的关联。
纳入居住在英格兰且完成了基线（第1波）和随访（第2波）问卷的参与者。通过健康与社会护理信息中心进行随访，截至2012年12月31日统计死亡情况。对IMD、教育程度和一系列协变量进行了探究。使用对所有协变量进行调整的Cox回归模型。进行了使用插补法的敏感性分析：（1）纳入有缺失数据的人群；（2）对完成基线问卷的整个队列进行分析。
在居住在英格兰且完成了第1波和第2波问卷的54,539名女性中，有4,510人存在缺失数据。其余50,029名女性被纳入主要分析。在对所有潜在混杂因素进行调整后，与最不贫困组相比，最贫困组的地区层面IMD与全因死亡率呈正相关（风险比=1.42，可信区间=1.14-1.78）。敏感性分析显示了相似的结果，在整个队列中的关联更强（风险比=1.90，可信区间=1.68-2.16）。个体受教育程度越低，死亡风险越高（趋势检验p<0.001）。然而，在对IMD排名进行调整后，与接受大学教育相比，未接受正规教育对死亡率的粗略影响消失了（风险比=1.08，可信区间=0.93-1.26）。
即使在这个贫困程度较低的队列中，且在对一系列潜在混杂因素进行调整后，生活在更贫困地区的女性死亡率仍然较高。
本研究注册为国际标准随机对照试验，编号为ISRCTN22488978。
