Counts describe this library, not the strength of evidence.
How to read this evidence
Separate prevention outcomes, risk-factor changes and screening performance. A drug trial in established disease, an exercise blood-pressure analysis and a smartwatch alert study cannot be treated as equivalent evidence for a healthy individual.
In a pooled analysis of 837 population-based studies with 7.5 million adults aged 20 to 64 in eight world regions, BMI and waist-to-height ratio (WHtR) separated people with and without hypertension equally well (C-statistics 0.72 to 0.81). At the same BMI, people in south Asia carried more abdominal fat, but the analysis is cross-sectional and cannot show that either measure predicts future hypertension.
Pooling 16 cohort studies of 1,044,035 people followed for 2.4 to 18 years, short sleep (compared with 7-8 hours) was linked to a higher risk of developing hypertension (HR 1.07, 95% CI 1.06-1.09), and the link was stronger under 5 hours (HR 1.11). The effect is small, and the abstract does not define short sleep beyond that subgroup.
In 74 adults with stage 1 primary hypertension, 6 weeks of the DASH diet within an 8-hour eating window lowered systolic and diastolic blood pressure by 8.459 and 9.459 mm Hg, versus 5.595 and 5.351 mm Hg on DASH alone. The trial is small and short, and the abstract gives no P values or confidence intervals for the comparison.
Across 20 trials with 1,192 adults with type 2 diabetes and overweight or obesity, combined aerobic and resistance training (CART) improved glycated hemoglobin, blood pressure, inflammatory markers, fitness and quality of life compared with standard treatment. The authors rate the evidence as poor quality with uncertain risk of bias, and the abstract gives no effect sizes.
In 242,763 matched pairs, estimated lifetime atrial fibrillation (AF) risk from age 45 to 85 was 20.9% in White Europeans versus 15.4% in East Asians (p < 0.001), but similar (13.4% vs 12.9%, p = 0.575) when lifestyle factors were all non-adverse. The two groups come from different cohorts, so the gap may not be only lifestyle.
Among 3067 US adults with hypertension, those with a neutrophil-to-lymphocyte ratio (NLR) above 3.5 had higher all-cause (HR 1.96) and cardiovascular mortality (HR 2.33) over a median 92.0 months. This is observational, and the ratio only modestly predicted survival (areas under the curve 0.64 to 0.70).
In hypertensive UK adults followed up to 21 years, visit-to-visit systolic blood pressure variability predicted cardiovascular events independent of average pressure (HR 1.22 per 5 mmHg), and amlodipine-based treatment beat atenolol-based for stroke (HR 0.82). The claim that variability mediated the benefit rests on post-trial observational modeling.
In medical records of more than 200,000 adults from a primary care cooperative, new hypertension diagnoses rose from 2.11 per 100 person-years in 2017-2019 to 5.20 in 2020-2022 (RR 2.46) and 6.76 in 2023. Diagnoses are not the same as true new disease, and the data cannot show that COVID-19 itself caused the rise.
In 342 Chinese adults with prehypertension, 12 months of supervised Tai Chi lowered office systolic blood pressure by 2.40 mm Hg more than aerobic exercise (95% CI -4.39 to -0.41; P=0.02). Both groups improved, there was no non-exercise comparison group, and the extra benefit is small.
Across 19 randomised trials (12,418 people) in low- and middle-income countries, digital health interventions lowered systolic blood pressure by 4.43 mmHg (95% CI -6.19 to -2.67) and more than doubled the odds of blood pressure control (OR 2.20, 95% CI 1.64-2.94) versus usual care. Results varied widely between trials (I2 92% and 78%) and only two African studies were included, so the size of benefit is uncertain.
Using health assessments of 263,374 Swedish adults aged 20-69 for the Åstrand test and 95,043 for the Ekblom-Bak test, the authors built age- and sex-specific percentile reference values for estimated VO2max. These come from two specific cycle tests, each with its own bias, so they are only meaningful for comparing a result from the same test.
Across 18 studies in youth and 24 in adults, trials gave mixed results on whether sitting time changes cardiorespiratory fitness, but pooled adult trials of interventions to cut sitting (with or without more activity) showed a mean difference of 3.16 mL/kg/min in peak oxygen consumption (95% CI 1.76-4.57). Certainty was very low, and many interventions also added activity.
Among 15,056 adults at a Chinese health-check centre, the "high stable" triglyceride-glucose (TyG) index trajectory had an adjusted hazard ratio of 1.69 (95% CI 1.40-2.02) for hypertension versus the "low increasing" group. This is a retrospective study of people attending a health-check centre, so it shows association only.
Among 8513 US adults with hypertension, those in the highest triglyceride-glucose (TyG) index quartile had higher odds of uncontrolled hypertension than the lowest quartile (OR 1.49, 95% CI 1.21-1.89). The data are a cross-sectional survey, so they show association only and cannot show which came first.
Pooling 270 randomised trials with 15,827 participants, aerobic, resistance, combined, interval and isometric training all lowered resting blood pressure compared with non-intervention controls (P<0.001 for each), with the largest fall for isometric training: 8.24 mm Hg systolic and 4.00 mm Hg diastolic. The ranking between modes is partly indirect and the trials differ, so it is not a personal prescription.
In 7 cohort studies of 19 548 healthy adults followed for a median 5.3 years, average systolic blood pressure was 1.25 mm Hg higher at 12 g of daily alcohol and 4.90 mm Hg higher at 48 g than with no drinking, with no threshold apparent. With only 7 observational studies, the link is not proof that alcohol itself raised pressure.
Over a mean 28.8 years, atrial fibrillation was diagnosed in 10 of 228 elite Swedish female endurance athletes (4.4%) and 23 of 1368 matched women (1.7%), a hazard ratio of 2.56 (95% CI 1.22-5.37) that rose to 3.67 (1.71-7.87) after adjusting for hypertension. Only 33 cases occurred in total, so the estimates are imprecise.
In 290,913 UK Biobank adults, more alcohol was linked to higher systolic and diastolic blood pressure and to hypertension in both men and women (p<0.001). Current smoking went with lower readings overall, but among current smokers more cigarettes per day went with higher systolic pressure, so the smoking picture is mixed and should not be read as causal.