Backgrounds

Previous studies linking biomass fuel use to hypertension have been inconsistent. We investigated the association between biomass fuel use and the risk of hypertension and blood pressure measures in older Chinese people.

Methods

The prospective cohort study of the Chinese Longitudinal Healthy Longevity Survey (CLHLS) included participants aged 65 years and older in 2011/2012 who were followed up until 2014 in 23 provinces in China. We explored the association between biomass fuel use and hypertension using the Cox proportional hazards model and examined the relationship of biomass fuel use with blood pressure measures using the generalized estimating equation. Additionally, we examined the effect of switching cooking fuels on hypertension during the follow-up.

Results

Among 3754 participants who were without hypertension at baseline, the mean age was 86 years old, and 47.5% of participants were men. Reported use of biomass fuel for cooking (50.2%) was associated with a higher risk of hypertension (incidence rate (IR) per 100 person-years: 13.15 versus 12.99, hazard ratio (HR) = 1.15, 95% confidence interval (CI) = 1.01–1.31). Biomass fuel use was related to systolic blood pressure (SBP) (β 1.10 mmHg, 95% CI: 0.48–1.72), diastolic blood pressure (DBP) (β 1.02 mmHg, 95% CI: 0.61–1.43) and mean arterial pressure (MAP) (β 1.03 mmHg, 95% CI: 0.63–1.43) elevation. Compared with persistent clean fuel users, participants who reported switching from clean to biomass fuels for cooking had a noticeably higher risk of hypertension (IR per 100 person-years: 14.27 versus 12.81, HR 1.49, 95% CI: 1.16–1.90) and higher SBP (3.71 mmHg), DBP (2.44 mmHg) and MAP (2.86 mmHg). Interaction and stratified analyses showed greater effect estimates of SBP and MAP in the oldest old people (≥85).

Conclusions

The use of biomass fuel for cooking was associated with greater hypertension risk, and the risk may be higher among those who switched from clean fuels to biomass fuels in the Chinese elderly population. Biomass fuel use was associated with a statistically significant but small absolute increase in blood pressure measures.