Population aging is an important public health issue: 8.5% of the total population worldwide is aged 65 years and older, and this number is expected to increase to 20% by 2050.¹ Aging is usually accompanied by adverse health conditions, including chronic diseases, cognitive decline, psychological disorders, and physical function limitations, causing a huge burden on individuals, families, and society.² Approaches to achieve healthy aging, typically defined as being disease-free and physically, mentally, and cognitively healthy,^3,4 are urgently needed. However, only 10% to 35% older adults achieve healthy aging.^5,6 Identifying modifiable factors for healthy aging can inform interventions to promote this outcome.

Propensity score matching is commonly used in observational studies to control for confounding and estimate the causal effects of a treatment or exposure. Frequently, in observational studies data are clustered, which adds to the complexity of using propensity score techniques. In this article, we give an overview of propensity score matching methods for clustered data, and highlight how propensity score matching can be used to account for not just measured confounders, but also unmeasured cluster level confounders. We also consider using machine learning methods such as generalized boosted models to estimate the propensity score and show that accounting for clustering when using these methods can greatly reduce the performance, particularly when there are a large number of clusters and a small number of subjects per cluster. In order to get around this we highlight scenarios where it may be possible to control for measured covariates using propensity score matching, while using fixed effects regression in the outcome model to control for cluster level covariates. Using simulation studies we compare the performance of different propensity score matching methods for clustered data across a number of different settings. Finally, as an illustrative example we apply propensity score matching methods for clustered data to study the causal effect of aspirin on hearing deterioration using data from the conservation of hearing study.

Night shift work is becoming increasingly common worldwide, and 15% to 33% of the working population^1-3 are engaged in night shift work, especially among health care workers.² Previous studies have suggested that night shift work may result in circadian disruption, sleep disturbances, and other behavioral changes,⁴ leading to increased risk of chronic diseases,^4-11 mental disorders,^1,12,13 cognition impairment,^14,15 and mortality.^14,16 The World Health Organization’s International Agency for Research on Cancer concludes that shift work is probably carcinogenic for humans.^17,18