Abstract：
Evidence on the effects of neighborhood socioeconomic disadvantage on dementia risk in racially and ethically diverse populations is limited. Our objective was to evaluate the relative extent to which neighborhood disadvantage accounts for racial/ethnic variation in dementia incidence rates. Secondarily, we evaluated the spatial relationship between neighborhood disadvantage and dementia risk.
In this retrospective study using electronic health records (EHR) at two regional health systems in Northeast Ohio, participants included 253,421 patients aged >60 years who had an outpatient primary care visit between January 1, 2005 and December 31, 2015. The date of the first qualifying visit served as the study baseline. Cumulative incidence of composite dementia outcome, defined as EHR-documented dementia diagnosis or dementia-related death, stratified by neighborhood socioeconomic deprivation (as measured by Area Deprivation Index) was determined by competing-risk regression analysis, with non-dementia-related death as the competing risk. Fine-Gray sub-distribution hazard ratios were determined for neighborhood socioeconomic deprivation, race/ethnicity, and clinical risk factors. The degree to which neighborhood socioeconomic position accounted for racial/ethnic disparities in the incidence of composite dementia outcome was evaluated via mediation analysis with Poisson rate models.
Increasing neighborhood disadvantage was associated with increased risk of EHR-documented dementia diagnosis or dementia-related death (most vs. least disadvantaged ADI quintile HR = 1.76, 95% confidence interval = 1.69-1.84) after adjusting for age and sex. The effect of neighborhood disadvantage on this composite dementia outcome remained after accounting for known medical risk factors of dementia. Mediation analysis indicated that neighborhood disadvantage accounted for 34% and 29% of the elevated risk for composite dementia outcome in Hispanic and Black patients compared to White patients, respectively.
Neighborhood disadvantage is related to the risk of EHR-documented dementia diagnosis or dementia-related death and accounts for a portion of racial/ethnic differences in dementia burden, even after adjustment for clinically important confounders.
In this retrospective study using electronic health records (EHR) at two regional health systems in Northeast Ohio, participants included 253,421 patients aged >60 years who had an outpatient primary care visit between January 1, 2005 and December 31, 2015. The date of the first qualifying visit served as the study baseline. Cumulative incidence of composite dementia outcome, defined as EHR-documented dementia diagnosis or dementia-related death, stratified by neighborhood socioeconomic deprivation (as measured by Area Deprivation Index) was determined by competing-risk regression analysis, with non-dementia-related death as the competing risk. Fine-Gray sub-distribution hazard ratios were determined for neighborhood socioeconomic deprivation, race/ethnicity, and clinical risk factors. The degree to which neighborhood socioeconomic position accounted for racial/ethnic disparities in the incidence of composite dementia outcome was evaluated via mediation analysis with Poisson rate models.
Increasing neighborhood disadvantage was associated with increased risk of EHR-documented dementia diagnosis or dementia-related death (most vs. least disadvantaged ADI quintile HR = 1.76, 95% confidence interval = 1.69-1.84) after adjusting for age and sex. The effect of neighborhood disadvantage on this composite dementia outcome remained after accounting for known medical risk factors of dementia. Mediation analysis indicated that neighborhood disadvantage accounted for 34% and 29% of the elevated risk for composite dementia outcome in Hispanic and Black patients compared to White patients, respectively.
Neighborhood disadvantage is related to the risk of EHR-documented dementia diagnosis or dementia-related death and accounts for a portion of racial/ethnic differences in dementia burden, even after adjustment for clinically important confounders.
摘要：
关于种族和道德多样化人群中邻里社会经济劣势对痴呆症风险影响的证据有限。我们的目的是评估邻里劣势在痴呆发病率中种族/民族差异的相对程度。其次，我们评估了邻域劣势与痴呆风险之间的空间关系.
在这项回顾性研究中，使用俄亥俄州东北部两个地区卫生系统的电子健康记录（EHR），参与者包括在2005年1月1日至2015年12月31日期间进行了初级护理门诊就诊的253,421例年龄>60岁患者.第一次合格访视的日期作为研究基线。复合痴呆结局的累积发生率，定义为EHR记录的痴呆诊断或痴呆相关死亡，通过竞争风险回归分析确定了按邻里社会经济剥夺分层的(通过面积剥夺指数衡量)，与非痴呆相关的死亡作为竞争风险。确定了社区社会经济剥夺的精细灰色子分布危险比，种族/民族,和临床危险因素。通过泊松比率模型的中介分析，评估了社区社会经济地位在复合痴呆结局发生率中的种族/族裔差异的程度。
邻里劣势的增加与EHR记录的痴呆诊断或痴呆相关死亡的风险增加有关（大多数与在调整了年龄和性别后，最不利的ADI五分位数HR=1.76，95％置信区间=1.69-1.84）。在考虑了已知的痴呆症医学风险因素后，邻里劣势对这种复合痴呆症结局的影响仍然存在。中介分析表明，与白人患者相比，西班牙裔和黑人患者的复合痴呆结果的高风险分别占34％和29％。分别。
邻里劣势与EHR记录的痴呆症诊断或痴呆症相关死亡的风险有关，并且是痴呆症负担中种族/民族差异的一部分。即使在调整了临床上重要的混杂因素之后。
在这项回顾性研究中，使用俄亥俄州东北部两个地区卫生系统的电子健康记录（EHR），参与者包括在2005年1月1日至2015年12月31日期间进行了初级护理门诊就诊的253,421例年龄>60岁患者.第一次合格访视的日期作为研究基线。复合痴呆结局的累积发生率，定义为EHR记录的痴呆诊断或痴呆相关死亡，通过竞争风险回归分析确定了按邻里社会经济剥夺分层的(通过面积剥夺指数衡量)，与非痴呆相关的死亡作为竞争风险。确定了社区社会经济剥夺的精细灰色子分布危险比，种族/民族,和临床危险因素。通过泊松比率模型的中介分析，评估了社区社会经济地位在复合痴呆结局发生率中的种族/族裔差异的程度。
邻里劣势的增加与EHR记录的痴呆诊断或痴呆相关死亡的风险增加有关（大多数与在调整了年龄和性别后，最不利的ADI五分位数HR=1.76，95％置信区间=1.69-1.84）。在考虑了已知的痴呆症医学风险因素后，邻里劣势对这种复合痴呆症结局的影响仍然存在。中介分析表明，与白人患者相比，西班牙裔和黑人患者的复合痴呆结果的高风险分别占34％和29％。分别。
邻里劣势与EHR记录的痴呆症诊断或痴呆症相关死亡的风险有关，并且是痴呆症负担中种族/民族差异的一部分。即使在调整了临床上重要的混杂因素之后。
