The association between tea consumption and hypertension: a systematic review and meta-analysis

The association between habitual tea consumption and hypertension remains controversial. This systematic review and meta-analysis aimed to synthesize observational evidence on tea consumption and hypertension and to explore whether this association differed according to tea type, sex, age, region, and hypertension diagnostic threshold.

PubMed, Embase, Cochrane Library, Web of Science, CNKI, Wanfang, and VIP were systematically searched from database inception to December 2025 for observational studies in adults. Studies reporting a distinct hypertension outcome and an extractable effect estimate for tea consumption were eligible. Study quality was assessed using the Newcastle-Ottawa Scale or AHRQ criteria, as appropriate. The most fully adjusted ORs, RRs, or HRs were extracted. Because these measures are not mathematically identical, pooled estimates were interpreted conservatively as relative association estimates rather than directly interchangeable causal effect measures. Dose-response analyses were conducted for studies with at least three quantitative exposure categories.

Twenty studies involving 319,565 participants were included. In 16 cross-sectional studies, tea consumption was weakly associated with lower hypertension prevalence (OR = 0.92, 95% CI: 0.87-0.97, I²  = 62.8%). This inverse association was stronger for black tea drinkers, Asians, women, participants aged < 60 years, and studies with a hypertension cutoff of ≥140/90 mmHg; these subgroup results are hypothesis-generating, not confirmatory. Pooled ORs of the four cohort studies indicated a 13% lower hypertension risk (OR = 0.87, 95% CI: 0.84-0.92, I²  = 31.9%). Dose-response analysis suggested a significant nonlinear association, with an initially steeper inverse pattern at lower intake levels followed by an apparent plateau at higher reported intake.

Habitual tea consumption was associated with a modestly lower likelihood of hypertension in observational studies. Owing to the observational design and marked inter-study heterogeneity, a definite clinical protective effect cannot be confirmed.

https://www.crd.york.ac.uk/PROSPERO/, CRD420261288902.