Supplements: which are worth taking?

Numbers match the citations in the article. A working link isn't proof that a page supports a claim; check the quoted passage and date.

1. 1

   Multivitamin supplementation and COVID-19 incidence and symptom severity in the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomized trial.

   The American journal of clinical nutrition (Li et al.)Published Jun 9, 2026Checked Sep 30, 2026

   “Of the 18,205 participants who remained in the trial and were adherent to study pills as of 1 January, 2020, 382 were randomly assigned to MVM and 404 to MVM placebo (no MVM) and reported a COVID-19 infection through 31 December, 2020. In intention-to-treat analysis, the hazard ratio for MVM compared with placebo was 0.93 for COVID-19 incidence [95% confidence interval (CI): 0.81, 1.07]. Among the 338 cases with detailed symptom data, the odds ratio for symptomatic COVID-19 comparing MVM with placebo was 0.70 (95% CI: 0.44, 1.11). In per-protocol analyses, among participants compliant with study pills during 2020, the odds ratio for symptomatic COVID-19 was 0.60 (95% CI: 0.37, 0.99).ConclusionsA daily MVM supplement, compared with placebo, does not significantly reduce COVID-19 incidence among older adults but shows a promising signal lowering the odds of symptomatic COVID-19 illness. The Cocoa Supplement and Multivitamin Outcomes Study was registered at clinicaltrials.gov as NCT02422745.”

2. 2

   Long-Term Effect of Multivitamin Supplementation on Incident Self-Reported Hypertension and Blood Pressure Changes in the COSMOS Trial.

   American journal of hypertension (Hamaya et al.)Published Apr 1, 2026Checked Sep 30, 2026

   “Hypertension diagnosis was ascertained through self-reports. Additionally, in two substudies with BP measurements (N = 529 at clinic by research staff and 994 at home by technician), we evaluated the effects on 2-year BP changes.ResultsIncident hypertension was observed in N = 1034 (22.9%) in MVM arm and N = 1039 (23.6%) in placebo arm over a median of 3.4 years (IQR: 3.0, 3.9) of follow-up, with hazard ratio (HR) 0.98 [95% CI: 0.90, 1.06]. Effects differed according to baseline diet quality, with HRs of incident hypertension 0.81 [0.70, 0.95] and 1.14 [1.01, 1.28] among participants with lower and higher Alternate Mediterranean Diet score, respectively (P-interaction = .001). There was no effect of MVM on 2-year changes in systolic BP (4.4 mmHg in MVM; 4.5 mmHg in placebo), while pronounced effects were observed for baseline normal BP (P-interaction = .004).ConclusionsMVM supplementation versus placebo did not reduce hypertension incidence or lower BP overall. Exploratory analyses showed greater reduction in hypertension risk and BP changes among those with lower dietary quality and normal BP at baseline, respectively.Clinical trial registrationNCT02422745.”

3. 3

   Magnesium Supplementation for Sleep in Adults: A Systematic Review of Randomized Controlled Trials.

   Journal of dietary supplements (Lopresti et al.)Published Aug 28, 2026Checked Sep 30, 2026

   “Twelve randomized controlled trials were included in the main synthesis on sleep, insomnia, or poor sleep, and three additional trials involving movement- or cramp-related sleep disturbance were analyzed separately. One separately analyzed restless legs syndrome trial enrolled participants aged 15-50 years because adult-only data were unavailable. Populations, magnesium formulations, elemental doses, treatment durations, outcome roles, and measures varied substantially. Findings were inconsistent across subjective outcomes, EEG, research-grade actigraphy, consumer wearable-derived estimates, and physiological markers. Certainty of evidence was low to very low. Current evidence does not support oral magnesium as a routine treatment for insomnia. It may provide modest benefits for selected subjective outcomes in some adults, but the findings are inconsistent, and the certainty of the evidence is low or very low. Further well-designed, adequately powered trials are required before routine recommendations can be made.”

4. 4

   The Effect of Dietary Supplements on Female Infertility in Terms of Endometrial Thickness, Pregnancy, Live Birth and Miscarriage: A Systematic Review and Meta-Analysis.

   Nutrients (Michaelsen et al.)Published Jun 16, 2026Checked Sep 30, 2026

   “Twenty placebo-controlled and 20 no-treatment comparator studies were included. Most studies had some concerns in risk of bias. Primary analyses showed an improvement in endometrial thickness following N-acetyl-cysteine supplementation compared to placebo, while no effect was found for supplements on pregnancy-related outcomes. Certainty of evidence of primary analyses was low. Secondary analyses indicated positive differences in endometrial thickness and pregnancy-related outcomes following supplementation with different supplements compared to placebo and no treatment. This review found no high-certainty evidence that dietary supplements improve female infertility outcomes when compared with placebo. Secondary analyses combining placebo and no-treatment comparator studies generated hypotheses for myo-inositol, N-acetyl-cysteine, vitamin D, vitamin E, and ≥3 substance dietary supplements, but these are at higher risk of bias and require confirmation in adequately powered placebo-controlled trials with live birth as the primary outcome.”

5. 5

   Dietary Polyphenols in Type 2 Diabetes: A Metabolite-Centric Review of Human Evidence.

   Nutrients (Andrés et al.)Published Sep 1, 2026Checked Oct 4, 2026

   “Using randomised controlled trials and meta-analyses published between 2015 and 2025, we review the human evidence for anthocyanins, catechins, stilbenes, cocoa flavanols and olive phenolics, with briefer coverage of isoflavones, curcuminoids and ellagitannins. Anthocyanins show the most consistent glycaemic signal, with a best available pooled estimate of about 0.3% for HbA1c (-0.31%, median exposure eight weeks) derived from trials that did not verify metabolite exposure, an effect that is modest and lies at the lower bound of what is generally regarded as clinically meaningful. Green tea catechins and high-phenolic extra-virgin olive oil are associated with comparable postprandial gains, whereas cocoa flavanols and resveratrol more reliably improve vascular and inflammatory intermediates than glycaemia itself. Most interventions were well tolerated at the nutritional doses tested. Polyphenols are better regarded as meal-timed dietary adjuncts to standard T2D care than as primary glycaemic agents.”

6. 6

   The Role of Anthocyanins, Curcumin, and Resveratrol in the Prevention and Management of Metabolic Disorders: A Systematic Review.

   Molecules (Basel, Switzerland) (Gazda & Glibowski)Published May 26, 2026Checked Oct 4, 2026

   “Databases such as PubMed and Embase were searched, and the final selection included 105 studies that met the inclusion criteria. The analyzed studies demonstrated that anthocyanin supplementation (up to 320 mg/day) was associated with reductions in inflammatory markers such as IL-6 and TNF-α, improvements in HDL cholesterol, and modest reductions in HbA1c (~0.3-0.5%). Curcumin supplementation was associated with decreases in body weight (up to 0.82 kg), BMI (up to 0.30 kg/m2), triglycerides, total cholesterol, and fasting glucose levels. Resveratrol showed mixed but potentially beneficial effects on insulin sensitivity, oxidative stress markers, and lipid metabolism, although the clinical outcomes remained inconsistent across studies. These findings suggest that the antioxidant effects of anthocyanins, curcumin, and resveratrol may be related to their ability to suppress oxidative stress and inflammatory processes, thereby contributing to improvements in glucose and lipid metabolism. The conclusions from this analysis may contribute to a better understanding of the role of antioxidants in the management of metabolic health and indicate directions for future research in this area.”

7. 7

   Consumer-Relevant Endpoints in Nutritional Supplement Research: A Case of Multivitamin and Mineral Supplements.

   Current developments in nutrition (Sheldon et al.)Published May 2, 2026Checked Sep 30, 2026

   “They are one of only a few interventions the public actively embraces, despite clear guidelines that do not recommend routine MVM use for healthy people, because a balanced diet should provide all essential nutrients. This public adoption of MVMs creates a paradox when people say they take MVMs to support their well-being, yet scientific studies do not capture or measure what that actually entails. It also presents an opportunity to explore why this divergence exists, and in doing so, explore how people understand health and well-being. This review argues that trials of MVM supplementation in the general adult population are methodologically misaligned with the outcome consumers value, relying on endpoints and instruments that are neither representative nor sufficiently sensitive. Beyond any placebo effect, the summative effect of addressing multiple subclinical micronutrient deficiencies represents a potential, plausible biological basis for improved quality of life (QoL). Future MVM studies should be designed to comprehensively evaluate QoL and related outcomes (e.g., cognition, fatigue, and resilience) to align with reasons why adults take MVM supplements.”

8. 8

   Emerging Patterns in Dietary Supplement Use Among US Adults, 1999-2023.

   JAMA network open (Lam et al.)Published Jun 1, 2026Checked Sep 30, 2026

   “Trends were evaluated across cycles and by subgroups (age, sex, race and ethnicity, education, and income-to-poverty ratio). Prepandemic and early-pandemic (2017 to March 2020) and later-pandemic and postpandemic (August 2021 to August 2023) periods were compared. Survey-weighted binomial regression models were used to estimate linear trends across cycles.ResultsThis nationally representative study included 63 442 adults (weighted mean [SD] age, 47.3 [17.0] years; 33 127 women [52%]). Overall supplement use increased from 51% (95% CI, 49%-54%) in 1999 to 2000 to 60% (95% CI, 57%-63%) in August 2021 to August 2023 (P for trend, Conclusions and relevanceIn this repeated cross-sectional analysis of NHANES 1999 to 2023, supplement use among US adults increased over time, with diversification beyond MVMMs toward products marketed for immune, anti-inflammatory, gut, skin, or joint health. A substantial increase was observed in older adults. Ongoing monitoring of supplement use trends is needed to understand evolving public health implications.”