For how we live.
A review of 41 studies found that the great majority of AI medication-adherence prediction models carried high risk of bias, and that fancier algorithms did not reliably predict better.
JAMA published paired systematic reviews for the coming respiratory season. Both found vaccines reduced hospitalisation across age groups, with no new safety signals, though protection wanes over time.
A meta-analysis of 78 studies found the menstrual cycle has at most a trivial effect on performance. A randomised trial found timing strength training to cycle phase built no more muscle than training consistently.
An updated Annals review now covers 38 randomised trials and 25,816 people. Tirzepatide leads the marketed drugs at 19.0% placebo-subtracted weight loss; the experimental multiagonists run higher.
Most breast pain is self-limited, and cancer is rarely the cause — though it should be ruled out. A 2025 meta-analysis found no clear benefit from vitamin B6, a widely recommended remedy.
Twenty-three studies, mostly from the US and China, found mental health risk rising with temperature. Nowhere in the Pacific or Africa was represented.
A systematic review of 43 studies judged cups a safe option, with leakage similar to or lower than pads and tampons. For the newer reusable products, published research is mostly about acceptability, not safety.
PMDD is the severe, mostly psychological form of premenstrual illness, affecting about 3.6% of women in one study. For it, SSRIs are the best-evidenced treatment, with side effects that are common and dose-related.
A systematic review located 19 articles on melatonin in children aged 0 to 6. Trials covered 167 children, all with neurologic conditions. There was no efficacy data for typically developing children.
Black cohosh, Chinese herbal medicine and vitamin D reached moderate certainty for specific outcomes. Most of the rest of the field — 158 studies deep — sits at low or very low certainty.
Seventy-three trials, and a familiar pattern: the benefit versus no treatment shrinks by a third when only the best-conducted studies are counted. The head-to-head comparisons rest on a few hundred people.