[alister]

Several comments here mentioned shift work as a possible explanation.

The paper concedes that shift work is unhealthy[1] but claims that shift work doesn't explain their finding[2]. And their conclusion is "avoiding night light may be a promising approach for preventing cardiovascular diseases," but without telling us why. It's going to be fascinating if there's a mechanism by which sleeping with light can cause heart disease.

[1] "Evidence demonstrates higher risks of adverse cardiovascular events, coronary heart disease, heart failure, atrial fibrillation, and mortality due to cardiovascular disease in rotating shift workers."

[2] "Following separate adjustments for pre-existing diabetes, hypertension, high BMI, high cholesterol ratio, short, long, or inefficient sleep, and exclusion of shift workers, the relationships of night light with cardiovascular risks were attenuated but remained statistically significant for all outcomes except stroke."

[WD-101000]

> It's going to be fascinating if there's a mechanism by which sleeping with light can cause heart disease.

I suspect everyone in the field already knows the top-level answer: light at night blunts the output of the circadian pacemaker (SCN), with all sorts of downstream effects including control of various hormones. So the levels will be different with light at night. "at night" means biological night. If someone consistently sleeps on some schedule with bright enough light during their awake time, and it's dark during their sleep time, it's fine.

I'm not in the field. I read up on it at one point at a shallow level and talked to some researchers about it informally.