The Hidden Cost of Stress at Work

According to data from the Organisation for Economic Co-operation and Development, Americans work more hours per year than those in similar countries. Working is, in a sense, an American pastime. Work brings benefits, but it also brings costs. Many jobs are stressful or physically demanding, imposing a toll on both mental and physical health. The question of how work affects our health is incredibly important but hard to answer. Of course, there are some easily understood cases. For example, coal miners face an increased risk in mortality due to respiratory diseases, and firefighters are at increased risk of cancer from exposure to asbestos. Most of us aren’t mining coal or putting out fires. Our jobs are more mundane, but they may still impact our health.In a recent study published in the Journal of Public Economics, we sought to understand the effect of working in a stressful and physically demanding occupation on our health. We focused on the birth outcomes of mothers who work in two occupations that require extensive education and training: physicians and lawyers. We compared mothers who were physicians to those who were lawyers, since physicians and lawyers have similar years of education, similar income, and similar access to health care. You might expect their birth outcomes to be similar as well. However, while both groups work long hours (physicians more than lawyers in our data), physicians, especially those in hospitals, spend a lot more time working on their feet and are more likely to experience periods of intense stress. This is particularly true in the early years of physician training, where 80-hour work weeks and overnight shifts are common. So, despite the health benefits that tend to come with greater education and wealth, the stress and difficulty of a physician’s work may actually worsen health. We used birth certificate data from Texas, which is one of only a few states that includes the parents’ occupation on the birth certificate. Using these responses, we identified mothers who listed their occupation as a physician or lawyer. For physician mothers, we were also able to determine whether the mother was a surgeon. We found that physician mothers had worse birth outcomes than lawyer mothers. For example, physician mothers were 16% more likely to have a low birth weight infant and were 10% more likely to deliver pre-term. When we looked at surgeons specifically—an analysis motivated by the additional physical exertion and stress that comes with being a surgeon—we found that surgeon mothers had the worst birth outcomes compared to both lawyers and other doctors Importantly, our analysis accounted for the fact that physician mothers tend to have children at later ages, which increases the risk for negative birth outcomes.These findings, while interesting, couldn’t tell us if being a physician itself caused worse birth outcomes. To help answer that part of the question, we used a “natural experiment” that was designed to improve the working environment for some physicians. Natural experiments are instances in which people are exposed, by chance, to one path or another that allows researchers to study cause and effect (in this case, studying the impact of improved working conditions on health, a question that would be difficult to study in a controlled, randomized study). In 2011, the Accreditation Council for Graduate Medical Education enacted a reform that limited the number of hours that first-year residents (physicians in training) could consecutively work to be less than 16 consecutive hours. The reform was intended to improve the working conditions for first-year residents. Since the work reform only affected physicians, we could compare the birth outcomes of physician mothers to those of lawyer mothers before and after the reform to understand how an improvement in working conditions affected physician mothers’ birth outcomes. (Lawyer mothers should be unaffected.) In addition, because the reform was targeted at first-year residents, we focused on physicians who were 26 to 30 years old at the time the reform was enacted. We found that the work reform improved birth outcomes for these young physician mothers. Compared to babies born to physician mothers before the work reform, babies born to physician mothers after the reform had higher birth weights and a lower chance of being born small for their gestational age. The improvement was concentrated among young surgeon mothers. The reform, as we expected, did not have an effect on lawyer mothers or older physician mothers, whose work hours weren’t subject to the reform.Our findings suggest that workplace conditions can impact the health of not only workers themselves, but also their children. In other words, the effects of work may reach beyond an individual employee and into the next generation.The improvements in birth outcomes that we observed following the physician work-hour reform demonstrate that negative health outcomes aren’t inevitable. Changes to workplace environments—such as limiting excessively long shifts, which are common among physicians in training—could improve health. This suggests that policymakers, employers, and professional organizations have opportunities to design work environments that make us healthier, whether by limiting excessively long shifts, improving options for greater physical activity and healthier food choices in the workplace, or reducing stress.The connection between work and well-being is often discussed in terms of productivity, burnout, or job satisfaction. Our findings highlight another dimension: the health of future generations. The way we organize work today may have consequences that extend far beyond the workplace itself.