Observational meta-analysis links beverages to gout and hyperuricemia risk

A 2026 meta-analysis of 52 observational studies reports that alcohol and sugar-sweetened beverage consumption are associated with higher odds of hyperuricemia and gout, while coffee consumption is associated with lower odds of gout. The outcomes did not move in parallel: coffee was not clearly associated with hyperuricemia, and tea showed no significant overall association with either condition, although a female subgroup had higher odds of hyperuricemia among tea consumers. These are observational associations, not evidence that changing a beverage habit will prevent either disorder.

The review combined cross-sectional, cohort, and case-control studies published between 2003 and 2024. Searches of PubMed, Embase, the Cochrane Library, and Web of Science were updated through April 11, 2026, yielding 24,550 records, of which 52 met the eligibility criteria. The evidence base included 26 cross-sectional studies, 20 cohort studies, and six case-control studies conducted across Asia, North America, South America, Europe, and Australasia. Thirty-one studies assessed hyperuricemia and 21 assessed gout. Two reviewers independently screened studies and extracted data, with disagreements resolved by consensus or consultation with a third reviewer.

For categorical comparisons, alcohol consumption was associated with higher odds of gout, with a pooled odds ratio of 1.809 and a 95% confidence interval of 1.372 to 2.385. The same association appeared for hyperuricemia, for which the pooled odds ratio was 1.524, with a 95% confidence interval of 1.363 to 1.703. A separate analysis of three cohort studies using hazard ratios found a higher rate of gout among alcohol consumers during follow-up, with a pooled hazard ratio of 1.554 and a 95% confidence interval of 1.237 to 1.953. Positive associations with hyperuricemia were observed in both men and women, although differences between the sex-specific estimates were not statistically significant. The gout analysis showed substantial between-study heterogeneity, with an I² statistic of 97.8%.

Sugar-sweetened beverages showed a similar pattern. Their consumption was associated with higher odds of gout, with a pooled odds ratio of 1.371 and a 95% confidence interval of 1.059 to 1.776. For hyperuricemia, the pooled odds ratio was 1.478, with a 95% confidence interval of 1.342 to 1.627. The association was present in subgroup analyses for men and women. Funnel-plot asymmetry in the hyperuricemia analysis raised the possibility of small-study effects or publication bias, although the main association remained positive when individual studies were omitted in sensitivity analyses. The pooled hyperuricemia estimate was also heterogeneous, with I² of 47.6%.

Coffee showed the opposite categorical pattern for gout. Across three studies involving 266,268 participants, coffee consumption was associated with lower odds of gout, with a pooled odds ratio of 0.543 and a 95% confidence interval of 0.341 to 0.866. The estimate was highly heterogeneous, however: the I² statistic was 80.2%, indicating that variation among studies accounted for much of the observed difference. The evidence for hyperuricemia was less clear. Six studies involving 192,056 participants found no significant overall association, with a pooled odds ratio of 0.950 and a 95% confidence interval of 0.872 to 1.035. Sex-specific analyses did not detect a significant difference between men and women.

The dose-response models refined those comparisons by examining reported intake levels rather than simply contrasting consumers with non-consumers. Two prospective cohorts involving 135,302 participants found an inverse association between coffee intake and gout without evidence of nonlinearity. Compared with non-consumers, the model estimated an odds ratio of 0.820 at 592.5 millilitres per day, with a 95% confidence interval of 0.710 to 0.940; an odds ratio of 0.560 at 1,066.5 millilitres per day, with a 95% confidence interval of 0.440 to 0.710; and an odds ratio of 0.450 at 1,422 millilitres per day, with a 95% confidence interval of 0.340 to 0.600. At 2,133 millilitres per day, the estimated odds ratio was 0.380, with a 95% confidence interval of 0.190 to 0.740. Coffee intake was not significantly associated with hyperuricemia in the available dose-response data. These models are exploratory rather than proof that a particular coffee quantity is protective.

Higher reported alcohol intake was associated with higher estimated odds of both conditions, but the modeled shape differed by outcome. A dose-response analysis of two case-control studies and one cohort, involving 186,305 participants, found a positive and statistically significant nonlinear association with gout. For hyperuricemia, one cohort and two cross-sectional studies involving 22,369 participants showed a positive overall association without significant evidence of nonlinearity. The modeled range was 0 to 75 grams of ethanol per day. Sugar-sweetened beverage analyses also associated greater intake with higher estimated odds. Two cohorts involving 125,299 participants showed a positive association with gout without significant nonlinearity, while the hyperuricemia model, based on one cross-sectional study and one cohort involving 8,005 participants, suggested a nonlinear relationship. These findings cannot define safe consumption thresholds.

Tea produced the least definitive results. Two studies involving 135,302 participants found no significant overall association with gout, with a pooled odds ratio of 1.307 and a 95% confidence interval of 0.880 to 1.940. Nine studies involving 222,383 participants also found no significant overall association with hyperuricemia, with a pooled odds ratio of 1.119 and a 95% confidence interval of 0.961 to 1.304. A female subgroup had a statistically significant association, with an odds ratio of 1.294 and a 95% confidence interval of 1.081 to 1.549, but the male estimate was not significant and the difference between sexes was not statistically significant. The female estimate came from an unadjusted model that did not account for several potential confounders, including sex, body mass index, alcohol use, and high-purine diet, so the review authors cautioned against a firm interpretation. The overall tea-hyperuricemia analysis also showed substantial heterogeneity, with I² of 76.8%.

Several biological mechanisms could be consistent with the observed patterns, but the epidemiological findings do not prove that the proposed pathways cause the outcomes. Alcohol metabolism consumes cellular adenosine triphosphate and increases purine degradation, while lactate can compete with urate for renal excretion. Fructose-rich drinks can rapidly consume hepatic adenosine triphosphate during phosphorylation and may also affect insulin sensitivity and urate clearance. Coffee contains caffeine, chlorogenic-acid polyphenols, and diterpenes, all of which have been investigated for possible effects on xanthine oxidase, insulin sensitivity, renal urate transport, and inflammation. The review also discussed possible mechanisms involving tea compounds such as caffeine, theophylline, polyphenols, and oxalate, but noted that the available tea evidence was limited and inconsistent.

The analysis has important constraints that shape how the findings should be read. Every included study was observational, so residual confounding remains possible. Beverage habits were linked to broader dietary patterns, physical activity, alcohol use, and other lifestyle characteristics, and adjustment varied among studies. Exposure measurement also varied, with researchers relying on questionnaires, self-reported consumption, differing serving sizes, and categories that sometimes could not be converted accurately. Funnel plots and Egger’s tests had limited power in analyses containing only a few studies, making small-study effects and selective reporting difficult to exclude. Leave-one-out analyses generally preserved the direction of the main associations, but statistical significance sometimes changed when particular studies were removed. The authors therefore described the dose-response results as exploratory and called for large prospective cohorts with clearly defined beverage types, standardized exposure assessment, and more comprehensive adjustment for dietary and lifestyle confounders.

Subject of Research: Beverage intake associations with gout and hyperuricemia in observational meta-analysis

Article Title: Association between intake of different beverages and risk of hyperuricemia and gout: a risk factor and dose–response meta–analysis

Article References: Li, M., Wang, J., Wang, Z., Li, J., & Li, W. (2026). Association between intake of different beverages and risk of hyperuricemia and gout: a risk factor and dose–response meta–analysis. Frontiers in Nutrition, 13, Article 1956907. https://doi.org/10.3389/fnut.2026.1956907

Image Credits: AI Generated

DOI: 10.3389/fnut.2026.1956907

Keywords: gout, hyperuricemia, beverage consumption, coffee, alcohol, sugar-sweetened drinks, tea, dose-response analysis, meta-analysis, observational studies, uric acid, Association