Center for Population and Health, Georgetown University, Washington, DC.
Department of Sociology, University of California, Riverside.
JAMA Netw Open. 2022 Apr 1;5(4):e226547. doi: 10.1001/jamanetworkopen.2022.6547.
The association between wealth and mortality is likely to be nonlinear and may result from selection and reverse causality.
To compare the magnitude of mortality disparities by wealth relative to other measures of socioeconomic status (SES).
DESIGN, SETTING, AND PARTICIPANTS: This population-based cohort study began in 1995 to 1996, with approximately 18 years of mortality follow-up. These analyses were completed in November 2021. Data were derived from a population-based sample that targeted noninstitutionalized, English-speaking adults aged 25 to 74 years in the contiguous US. The response rate for the telephone interview ranged from 60% (twin subsample) to 70% (main sample). A self-administered questionnaire was completed by 89% of those interviewed by telephone.
Net assets of the respondent and spouse or partner in 1995 to 1996.
All-cause mortality.
Among 6320 respondents (mean [SD] age at baseline, 46.9 [12.9] years; 3318 women [52.5%]), 1000 (15.8%) died by May 31, 2013. Adjusted for age, sex, and race, the mortality disparity by wealth was larger than the disparities by education, occupation, income, or childhood SES, especially at the oldest ages. After age 65 years, the hazard ratio [HR] was 2.69 (95% CI, 2.00-3.62) for those with no assets relative to those with at least $300 000 of wealth (in 1995 dollars), which translated into a 31 percentage point differential in estimated probability of surviving from age 65 years to 85 years (40% vs 71%). Additional wealth greater than $500 000 was not associated with lower mortality. In fully adjusted models, there was still a sizeable wealth disparity in mortality after age 65 years (HR, 1.89; 95% CI, 1.33-2.67). After adjustment for confounders, the estimated probability of surviving from age 65 to 85 years was 19 percentage points higher for persons with at least $300 000 in wealth (70%) than for those with no assets (51%), but there was a much larger 37 percentage point differential between never smokers (70%) and current smokers (33%).
In this cohort study, the fully adjusted disparity in mortality associated with wealth beyond age 65 years remained sizeable but was much smaller than the smoking differential.
财富与死亡率之间的关系可能是非线性的，可能是由于选择和反向因果关系所致。
比较财富与其他社会经济地位（SES）衡量指标相比，死亡率差异的程度。
设计、地点和参与者：本基于人群的队列研究始于 1995 年至 1996 年，进行了大约 18 年的死亡率随访。这些分析于 2021 年 11 月完成。数据来自于一项以美国大陆的非机构化、讲英语的 25 至 74 岁成年人为目标的基于人群的样本。电话访谈的回复率在双胞胎子样本中为 60%（双胞胎子样本）到主样本中的 70%（主样本）。通过电话采访的受访者中有 89%完成了自我管理的问卷调查。
1995 年至 1996 年受访者及其配偶或伴侣的净资产。
全因死亡率。
在 6320 名受访者（基线时的平均[SD]年龄为 46.9[12.9]岁；3318 名女性[52.5%]）中，截至 2013 年 5 月 31 日，有 1000 人（15.8%）死亡。调整年龄、性别和种族因素后，财富差异导致的死亡率差异大于教育、职业、收入或儿童时期 SES 导致的差异，尤其是在年龄最大的人群中。65 岁以后，与至少拥有 300000 美元财富（以 1995 年美元计算）的人相比，没有资产的人（HR，2.69[95%CI，2.00-3.62]）的死亡风险更高，这意味着从 65 岁到 85 岁的估计生存概率差异为 31 个百分点（40%比 71%）。额外的超过 500000 美元的财富与较低的死亡率无关。在完全调整的模型中，65 岁以后仍然存在相当大的财富差异（HR，1.89[95%CI，1.33-2.67]）。在调整混杂因素后，至少拥有 300000 美元财富的人（70%）比没有资产的人（51%）从 65 岁到 85 岁的预期生存概率高 19 个百分点，但从不吸烟者（70%）和吸烟者（33%）之间的差异要大得多，为 37 个百分点。
在这项队列研究中，65 岁以上财富与死亡率之间的完全调整差异仍然相当大，但比吸烟差异小得多。
