Since the 1980s, the US follows a different growth path than other developed economies: the reduction in hours worked stopped, and life expectancy started to increase at a slower pace despite a surge in health expenditure as a share of GDP. There are many specific explanations for each phenomenon. For example, Case and Deaton (2015) put into the spotlight the opioid epidemic increasing mortality among non-Hispanic white Americans. We investigate the plausibility of a global link between these three phenomena. For some reasons (taxes, change in preferences, change in returns, increasing inequality), Americans are willing to work a lot, thereby deteriorating their health capital at a higher rate than Europeans. They offset their health problems by purchasing costly drugs and medical care, but because curative medicine is less efficient than a preventive life-style, the medical treatment may not be fully operative. As a result, Americans experience lower gains in life expectancy and a higher share of medical spending. We build an exogenous growth model with a dynamic equation of accumulation of health capital (Grossman 1972), which depreciation rate increases with work effort. We are able to reproduce the three stylized facts at the steady state: lower preferences for leisure as in the US lead to both a higher number of hours worked and a higher share of health expenditure and, provided that medical technology exhibits strong enough diminishing returns, a deterioration of the health capital stock which translates into a lower life expectancy. This provides a restriction to test the external validity of our global explanation.