Your sleep score, a one-time cholesterol edit, and coffee beyond caffeine

Hours asleep are the number most sleep apps put front and centre. But a large UK Biobank analysis suggests those hours are not all interchangeable. Researchers used wrist monitors to follow more than 95,000 people for a median of 8.9 years and compared estimated sleep patterns with later diagnoses.

More REM sleep—the stage most associated with vivid dreaming—was associated with lower risk across 83 later disease outcomes. More deep sleep was associated with lower risk across seven. Short total sleep mattered too: sleeping under five hours was associated with higher risk of 37 conditions, while many outcomes had their lowest observed risk in the six-to-eight-hour range.

That is a fascinating map of associations, not a prescription for chasing a particular REM score. Wrist monitors estimate sleep stages less precisely than a sleep laboratory, and the direction of travel can run both ways: underlying health problems may alter sleep. Still, the study is a useful nudge away from treating one nightly total as the whole story.

One of the hardest questions in cancer screening is not whether mammograms find cancers early—they do—but how often they find a cancer that would never have caused symptoms during a person’s lifetime. That is called overdiagnosis, and published estimates have ranged from very low to nearly 50%.

A new meta-analysis revisited randomized mammography trials with close attention to timing. The researchers accounted for how long people were followed and for screening that happened after trials ended—details that can make a temporary early increase in detected cancers look like lasting overdiagnosis. Using that framework, they concluded the patterns were compatible with overdiagnosis below 5%, rather than estimates approaching 50%.

It is not the final word on screening. This was a reanalysis of older trials, not a new trial, and other researchers and guidelines may reasonably interpret the trade-offs differently. But it usefully shows why a headline percentage can conceal important choices about how a study counts, and when it stops counting.

The idea is striking: instead of taking a cholesterol-lowering medicine repeatedly, receive one infusion designed to switch off a liver gene. In a first-in-human phase 1 study, 15 people whose lipid disorders were not adequately controlled by medication received CTX310, a CRISPR-Cas9 treatment targeting the gene ANGPTL3.

At the highest dose, LDL cholesterol was 52.5% lower and triglycerides were 47.8% lower than baseline after 12 months. No serious treatment-related adverse events were reported during that year. Those numbers are an intriguing early signal, especially because the edit is intended to be permanent.

But this is the moment for the large asterisk. Fifteen people is far too few to establish long-term safety, and the study was not designed to show whether the treatment prevents heart attacks or other clinical outcomes. It was funded by CRISPR Therapeutics, and extended follow-up matters especially here: unlike a medicine that can be stopped, gene editing is meant to last.

When people talk about losing strength with age, muscle size usually gets the attention. This study points to another possible piece of the puzzle: the neuromuscular junction, the tiny handoff point where a nerve tells a muscle to move.

Researchers found evidence that this handoff was less reliable in older adults and aged rodents. They linked the problem to reduced expression of NaV1.4, a sodium channel involved in the signal at the junction. In preclinical experiments, inhibiting another channel, ClC-1, reversed a related deficit.

The compelling part is the shift in perspective. Loss of strength and function may not be explained only by how much muscle remains; the quality of the nerve-to-muscle conversation may matter as well. The treatment angle remains early science, though. It was tested before clinical trials in older people, and the findings still need confirmation in people with clinically defined sarcopenia, the age-related loss of muscle and function.

Coffee research often gets flattened into a simple question: is it good for you or bad for you? A small new study makes that framing feel less useful. It suggests that coffee’s effects may involve more than caffeine—and that the microbes in the gut may be part of the story.

Researchers compared 31 regular coffee drinkers with 31 non-drinkers. The regular drinkers then went without coffee for two weeks before caffeinated or decaffeinated coffee was reintroduced. The study reported differences linked with habitual coffee intake in gut microbial and metabolic profiles, alongside differences in measures of stress, mood and cognition. Testing regular and decaf coffee separately after the break was a particularly interesting design choice: it looks beyond caffeine alone.

That does not mean coffee caused the mood, cognition or microbiome differences. With only 62 participants, the study is too small to settle cause and effect, and people who habitually drink coffee may differ from non-drinkers in many ways that matter. It was also sponsored by the Institute for Scientific Information on Coffee. For now, think of this as a glimpse of a complicated system—coffee compounds, caffeine, diet, expectations and gut microbes—not a verdict on your daily cup.