The nonlinear association between temperature anomaly and serum lipid profiles: A nationwide longitudinal panel study

Abstract

Background

Temperature anomaly (TA) caused by global climate change has been a risk factor for various health outcomes, but few studies reported its effects on serum lipid profiles. Therefore, we conducted a large population-based longitudinal panel study to investigate the association between TA and serum lipid profiles.

Methods

Individuals with at least two lipid tests and completing questionnaire surveys were included. Monthly TA were calculated based on an assimilated dataset of in situ observations, satellite remote sensing measurements and weather research forecast simulations. The TA values were categorized as three types: TA > 0 (sudden warming), TA < 0 (sudden cooling) and TA = 0 (no significant change). A fixed effect model was used to quantify the associations between lipid profiles and TA.

Results

A total of 295,233 individuals (689,621 visits) were included. Each increment in TA was significantly associated with increased risk of hyperbetalipoproteinemia [relative risk (RR) = 1.048, 95 % confidence interval (CI): 1.028–1.069], hypoalphalipoproteinemia (RR = 1.191, 95 %CI: 1.138–1.247), and hypercholesterolemia (RR = 1.056, 95 %CI: 1.033–1.080). Their corresponding attributable fractions were 4.62 % (95 %CI: 2.72 %–6.48 %), 16.06 % (95 %CI: 12.12 %–19.82 %) and 5.30 % (95 %CI: 3.19 %–7.37 %). The exposure-response relationships between TA and the risk of four types of dyslipidemia were estimated as J-shaped nonlinear curves centered at the TA value of zero. Sudden warming appeared to have a more pronounced effect on the risk of hyperbetalipoproteinemia, hypoalphalipoproteinemia and hypercholesterolemia than sudden cooling, whereas the opposite trend is observed for hypertriglyceridemia. All these associations were modified by climate zones and individual characteristics.

Conclusion

Temperature anomaly associated with increased risk of dyslipidemia, with varying associations across different climate zones and population subgroups. These results provide insights for the development of policies aiming at the prevention of dyslipidemia.