University College London, Department of Epidemiology and Public Health, London, United Kingdom.
University College London, Department of Epidemiology and Public Health, London, United Kingdom.
Soc Sci Med. 2019 Apr;226:113-122. doi: 10.1016/j.socscimed.2019.02.038. Epub 2019 Feb 28.
Several studies have documented that older workers who live in areas with higher unemployment rates are more likely to leave work for health and non-health reasons. Due to tracking of area disadvantage over the life course, and because negative individual health and socioeconomic factors are more likely to develop in individuals from disadvantaged areas, we do not know at what specific ages, and through which specific pathways, area unemployment may be influencing retirement age. Using data from the MRC National Survey of Health and Development, we use structural equation modelling to investigate pathways linking local authority unemployment at three ages (4y, 26y and 53y) to age of retirement (right-censored). We explored five hypothesized pathways: (1) residential tracking, (2) health, (3) employment status, (4) occupational class, and (5) education. Initially, pathways between life course area unemployment, each pathway and retirement age were assessed individually. Mediation pathways were tested in the full model. Our results showed that area unemployment tracked across the life course. Higher area unemployment at ages 4 and 53 were independently associated with earlier retirement age [1% increase = mean -0.64 (95% CI: -1.12, -0.16) and -0.25 (95% CI: -0.43, -0.06) years]. Both were explained by adjustment for individual employment status at ages 26 and 53 years. Higher area unemployment at age 26 was associated with poorer health and lower likelihood of employment at aged 53; and these 2 individual pathways were identified as the key mediators between area unemployment and retirement age. In conclusion, these results suggest that interventions designed to create local employment opportunities for young adults should lead to extended working through improved employment and health at mid-life.
已有多项研究表明，居住在失业率较高地区的老年工人更有可能因健康和非健康原因提前退休。由于在整个生命过程中对地区劣势进行了跟踪，而且弱势地区的个人健康和社会经济因素更有可能恶化，因此我们不知道在什么特定年龄段，以及通过哪些特定途径，地区失业会影响退休年龄。本研究使用 MRC 国家健康与发展调查的数据，采用结构方程模型来研究将三个年龄（4 岁、26 岁和 53 岁）的地方当局失业率与退休年龄（右删失）联系起来的途径。我们探讨了五个假设途径：（1）居住轨迹；（2）健康；（3）就业状况；（4）职业类别；（5）教育。最初，单独评估了生命历程中地区失业与每个途径之间以及与退休年龄之间的途径。在全模型中测试了中介途径。结果表明，地区失业在整个生命过程中都存在轨迹。4 岁和 53 岁时的地区失业率较高与提前退休年龄独立相关[增加 1%意味着平均提前退休年龄 -0.64（95%CI：-1.12，-0.16）和-0.25（95%CI：-0.43，-0.06）年]。这两个结果都可以通过调整 26 岁和 53 岁时的个人就业状况来解释。26 岁时的地区失业率较高与 53 岁时的健康状况较差和就业可能性较低有关；这两个个体途径被确定为地区失业与退休年龄之间的关键中介途径。总之，这些结果表明，为年轻成年人设计的旨在创造当地就业机会的干预措施应该通过改善中年时的就业和健康状况，延长工作时间。
