Science

There's no way the study design can prove a causal relationship, which to their credit, the authors freely admit. It seems they controlled for other CV risk factors (e.g. blood pressure) which would be expected. There's an interesting disclosure that their study cohorts are global, but their method of identifying patients for the melatonin cohort relied on prescription records... but in the US melatonin is OTC so USians would only be included in the non-melatonin group even if they were using it regularly. I can't be fucked to read into more detail, so my hot take is that people with insomnia severe enough to warrant treatment are de facto at higher risk of adverse health events and all-cause mortality just by virtue of how debilitating insomnia is, period. Maybe future studies will shed more light.