My Moneybox predecessor Annie Lowrey did a great piece a while back reviewing some of the literature on the terrible consequences of long commutes, and now a new study gives us even more. It shows that when you control for sociodemographic characteristics, smoking, alcohol intake, family history of diabetes, and history of high cholesterol that commuting distance is negatively associated with physical activity and cardiorespiratory fitness and positively associated with BMI, waste circumference, systolic and diastolic blood pressure, and continuous metabolic score.
What's more, if you add physical activity and cardioresporatory fitness to the control model you still get a statistically significant correlation with high blood pressure.
In other words, a long commute is definitely bad for your blood pressure but it's positively disastrous if it ends up eating into your physical activity time. So a longish walk or bike ride to work probably isn't that bad, but a long drive is disastrous. I really firmly believe that the existence of persistent regular traffic jams is just about the most underrated problem in American public policy, especially because it's a problem we could almost certainly solve relatively easily with a mix of congestion pricing, demand-responsive pricing of street parking, and more bus service.