A year of diet and exercise support made metabolic syndrome reversal four times likelier. The raw result was 10 people versus 3 — after nearly half disappeared

Metabolic syndrome is not one diagnosis from one test. It is the name given to a cluster of five risk factors: raised blood glucose, high blood pressure, high triglycerides, low HDL cholesterol and a large waist. Meeting at least three counts as the syndrome. A study published on 22 October 2026 asked whether primary-care clinics could help enough people fall back below that line with a structured year of diet and exercise support.

Sixteen primary-care networks in Alberta, Canada, were randomized as whole sites rather than person by person. Eight sites, with 320 patients, offered the CHANGE programme; eight, with 354 patients, continued usual care. The programme was intensive. A dietitian made an individual food plan, an exercise specialist prescribed aerobic, strength and flexibility work, and patients met both professionals weekly for 12 weeks and monthly for the next nine months. Their family doctor checked in every three months.

Among the 350 people with the measurements needed at both ends, 10 of 177 in the programme no longer met three criteria at 12 months, compared with 3 of 173 under usual care: 6% against 2%. That simple comparison just missed the conventional statistical threshold (p = 0.053). After the researchers adjusted for how many metabolic-syndrome criteria a person had at the start, the programme was associated with 4.37 times the odds of reversal (95% confidence interval 1.14 to 16.67).

The enormous-looking multiplier needs its denominator beside it. Only 13 people reversed the syndrome, so moving one or two people would shift the estimate substantially; the confidence interval runs from a modest advantage to more than sixteenfold odds. Starting severity was an even bigger divider. Compared with people who began with all five criteria, those who began with three had almost 20 times the odds of reversal, but that estimate was also very imprecise (95% CI 2.37 to 163.28). Four starting criteria did not have a statistically clear advantage over five.

The more serious problem is who remained to be counted. Of 674 randomized participants, about 48% lacked the complete 12-month data required for the main analysis. The researchers had planned an intention-to-treat analysis, which preserves the protection of randomization, but switched to complete cases because so much information was missing. Loss was 50.0% in the intervention arm and 45.9% in usual care. The pandemic interrupted the 2016–2023 trial and pushed visits online, but a demanding programme can also lose people who find it hardest to continue.

There were other reasons to resist a simple cause-and-effect headline. The sites, not patients, were randomized, no one could be blinded, and the complete-case intervention group was younger. Baseline diastolic pressure and waist size differed too. At 12 months the groups did not differ significantly in the separate averages reported for blood pressure, waist circumference, HDL, triglycerides, fasting glucose or HbA1c. The primary result instead depended on a small number of people crossing a three-out-of-five boundary.

Still, this was not a supplement or a motivational leaflet dressed up as lifestyle care. It tested whether ordinary primary-care networks could organize repeated, individualized contact from diet, exercise and medical professionals. The answer is promising enough for a better-retained trial, especially for people newly meeting three criteria. It is not evidence that everyone can 'reverse' metabolic syndrome with willpower, nor that medication should be stopped. It is evidence that support has to be treated as an intervention in its own right — and measured with the same scepticism as a drug.