"Multivitamins do not reduce risk of cardiovascular disease or prolong life, study finds," the Sun reports, while The Daily Telegraph warns that, "Some vitamin supplements can increase risk of a stroke".
Both headlines were prompted by a major new review looking at evidence from 277 trials on the effect of 24 different nutritional or dietary supplements in nearly 1 million people.
The resulting findings were very far from clear cut. There was tentative and weak evidence that some omega-3 fatty acid supplements may slightly reduce the risk of cardiovascular disease.
There was similarly weak evidence folic acid may decrease the risk of stroke.
And there was moderate evidence that taking a combination of calcium and vitamin D supplements can increase the risk of a stroke.
All other supplements studied, including multivitamins, antioxidants, and A and B vitamins, were not found to have any link with cardiovascular disease or early death.
But there were few studies for many of these supplements.
The evidence was of variable quality and it was difficult to be certain about the effects of different supplements.
Supplements may still be beneficial in certain circumstances, notably for people with specific vitamin or mineral deficiencies, so follow your doctor's advice.
The most effective way to reduce your risk of cardiovascular disease is through a combination of exercise, healthy eating, maintaining a healthy weight, not smoking, and drinking alcohol in moderation.
This study was carried out by researchers from West Virginia University, Johns Hopkins Bloomberg School of Public Health and School of Medicine, and East Carolina University.
Three of the authors were supported by the Blumenthal Scholars Fund in Preventive Cardiology at Johns Hopkins University.
One of the authors also declared affiliations with the National Heart, Lung, and Blood Institute, the Food and Drug Administration, the American Heart Association and various pharmaceutical companies.
The study was published in the peer-reviewed journal the Annals of Internal Medicine.
The UK media's reporting of the study was accurate.
This was a systematic review of other systematic reviews that aimed to see the effect of nutritional and dietary supplements on risk of mortality or cardiovascular disease.
The researchers focused on identifying reviews that had combined the results of individual randomised controlled trials.
Such an "umbrella" review, as termed by the researchers, is useful for gathering the evidence to date on a topic.
But the findings are only as good as the quality of the studies they included, in this case the individual reviews and the trials informing them.
The authors searched literature databases to identify systematic reviews and meta-analyses of trials assessing the effect of nutritional supplements in adults on mortality and cardiovascular outcomes.
Studies pooling observational studies or looking at intermediate markers of cardiovascular disease (such as cholesterol levels or inflammatory markers) were excluded.
The researchers assessed the quality of the individual systematic reviews and meta-analyses using standard recognised methodology.
They then created an "evidence map", a diagram displaying the effects of the supplements with a grading of the certainty of the evidence, ranging from very low- to high-level evidence.
This grading takes into account various factors, such as likelihood of bias and how consistent and precise the evidence of any effect is.
The researchers included 9 systematic reviews with 4 additional randomised control trials (RCTs).
In total, this included 277 individual trials, almost 1 million participants, and 105 meta-analyses covering 24 interventions.
These included 16 types of supplement and 8 dietary interventions.
There was moderate-certainty evidence that a lower salt intake reduces the risk of death in people with normal blood pressure (risk ratio [RR] 0.90, 95% confidence interval [CI] 0.85 to 0.95).
There was also moderate-certainty evidence that a lower salt intake reduces the risk of death from cardiovascular disease specifically in people with high blood pressure (RR 0.67, 95% CI 0.46 to 0.99).
No other supplements or nutritional interventions were found to affect the risk of early death.
There was low-certainty evidence that omega-3 long-chain polyunsaturated fatty acids (LC-PUFA) reduce the risk of a heart attack (RR 0.92, 95% CI 0.85 to 0.99).
Similarly, there was low-certainty evidence that they reduce the risk of coronary heart disease in general (RR 0.93, 95% CI 0.89 to 0.98).
There was low-certainty evidence folic acid reduces the risk of stroke (RR 0.80, 95% CI 0.67 to 0.96).
But there was moderate-certainty evidence that combined calcium plus vitamin D increases the risk of stroke (RR 1.17, 95% CI 1.05 to 1.30).
No other supplements or nutritional interventions were found to affect cardiovascular disease risk.
The researchers concluded: "Reduced salt intake, omega-3 LC-PUFA use, and folate supplementation could reduce risk for some cardiovascular outcomes in adults. Combined calcium plus vitamin D might increase risk for stroke."
This useful study gathered the reviews and meta-analyses published to date on the effect nutritional and dietary supplements have on cardiovascular and mortality outcomes.
It benefits from only looking at reviews that have combined the results of randomised controlled trials, rather than observational cohort studies.
With observational studies you can never be sure whether any outcome is the direct effect of the supplement, or of other health, dietary and lifestyle changes. Trials should balance such factors out.
But with such a wide mix of trials, with highly varied populations, interventions, methods of assessment and follow-up, the overall low certainty of the findings is an unavoidable limitation.
There was a lack of evidence for many of the interventions assessed, including multivitamins, antioxidants, and A and B vitamins.
The researchers also noted several indications of publication bias, where it's possible trials finding some evidence of an effect (benefit or harm) may have been more likely to have been published.
The researchers were not able to break down effects by specific doses or durations of use.
Another difficulty is that the reviews did not analyse effects within specific subgroups of people.
For example, it may be that people with folate deficiency benefit from folate supplements, but not the wider population.
Similarly, the researchers found signs of a link between calcium and vitamin D and potential stroke risk, but this needs to be interpreted carefully.
Our bodies need vitamin D, and some people can be at risk of vitamin D deficiency, particularly if their skin gets little exposure to natural daylight. That's why vitamin D supplements are recommended in certain groups.
At the same time, it's already well recognised that taking too much vitamin D and having too much calcium in your body can harm your health.
Overall, this review cannot tell us for certain whether all nutritional and dietary supplements are good, bad or ineffective for all people.
If a doctor has prescribed supplements, they will have assessed the risks and benefits, and decided that taking the supplement is likely to be of more benefit than harm in your individual circumstances.