Can light at night cause type 2 diabetes?
Research shows that light at night and disrupted day-night rhythms can harm blood sugar regulation. Whether this directly leads to type 2 diabetes is not yet certain.
Light at night does not directly cause type 2 diabetes, but there is clear evidence that disrupted day-night rhythms harm blood sugar regulation. People who work nights on a regular basis face a higher risk of type 2 diabetes, probably because weight gain and the body's reduced ability to respond to insulin both increase when the internal rhythm is thrown off for long periods. Whether this is a true cause, however, has not yet been proven.
Broader research shows that disrupted sleep and an out-of-sync day-night rhythm go hand in hand with the body becoming less responsive to insulin. Hormones involved include leptin (which signals fullness), ghrelin (which drives hunger), cortisol, and melatonin. This biological picture makes it plausible that light at night affects blood sugar, but firm conclusions about cause and effect are not yet available.
Artificial light at night is associated with a higher risk of high blood pressure, and a higher fasting blood glucose level turned out to be a small, partial explanation for that link, accounting for roughly 5% of the effect. This is not proof that light at night causes diabetes, but it does point to an effect on blood sugar control.
Whether actively reducing light at night in people with type 2 diabetes also improves blood sugar control is currently being investigated in a large study that also looks at bright light therapy during the day and eating earlier. Results are not yet available.
This answer draws on a review of shift work and diabetes, several reviews on sleep and disrupted day-night rhythms, and one large long-running study on artificial light at night and blood pressure. The evidence specifically linking light at night to blood sugar is limited: the available study measures blood pressure, not diabetes, as its main outcome, and the blood glucose effect is only a small intermediate factor. None of the available studies can make firm statements about cause and effect; they measure associations, not proven causes.