Is Cold Apparent Temperature Associated With the Hospitalizations for Osteoporotic Fractures in the Central Areas of Wuhan? A Time-Series Study

Abstract

Osteoporosis is alarming problem due to aggravation of global aging, especially in China. Osteoporotic fracture (OF) is one of the most severe consequents of osteoporosis. Many previous studies found that environmental factors had adverse effects on human health. Cold temperature was associated with OF and bone metabolism in prior observational and experimental researches. However, few studies had been conducted on the acute effect of low temperature and OF. Data on daily meteorological factors and hospitalizations for OF were collected from Wuhan, China, between January 1, 2017 to December 24, 2019. Apparent temperature (AT), comprehensively considered a variety of environmental factors, was calculated by ambient temperature, relative humidity and wind speed. A generalized linear regression model combined with distributed lag non-linear regression model (DLNM) with quasi-Poisson link was used to explore the association between AT and the number of hospitalizations for OF. Subgroup analyses stratified by gender, age and the history of fracture were applied for detecting susceptible people. The exposure-response curve of AT and OF were generally U-shaped with lowest point at 25.8°C. The significant relationship of AT-OF existed only in cold effect (−2.0 vs. 25.8°C) while not in warm effect (37.0 vs. 25.8°C). Statistically significant risks of OF for cold effects were only found in females [RR = 1.12 (95%CI: 1.02, 1.24) at lag 2 day], aged <75 years old [RR = 1.18 (95%CI: 1.04, 1.33) and 1.17 (95%CI: 1.04, 1.33) at lag 2 and 3 days, respectively] and people with history of fracture [RR = 1.39 (95%CI: 1.02, 1.90) and 1.27 (95%CI: 1.05, 1.53) at lag 1 and 2 days, respectively]. The significant associations of AT on OF were only found in cold effect. The females, people aged <75 years and people with history of fracture possibly appeared to be more vulnerable. Public health departments should pay attention to the negative effect of cold AT and take measures in time.

Seasonally Dependent Change of the Number of Fractures after 50 Years of Age in Poland—Analysis of Combined Health Care and Climate Datasets

“…Several studies evaluated seasonality and the effect of weather conditions on the incidence of fractures across the countries and regions [ 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 31 , 32 , 33 , 34 , 35 ]. Practically all studies show an increased incidence of fractures in the elderly in the colder season, although monthly distribution may be different worldwide.…”

“…From the other side, in the southern hemisphere, more fractures occur in June, July, and August, even though these temperatures in winter months are higher in comparison with the northern hemisphere [ 35 ]. Many studies also show a direct relationship between fracture incidence and temperature, not only with season [ 22 , 32 ].…”

Effects and Interaction of Meteorological Factors on Pulmonary Tuberculosis in Urumqi, China, 2013–2019

“…Second, distributed lag non-linear model (DLNM) was built to estimate the lag and non-linearity of meteorological factors on the risk of PTB. Exposure-response dimension fitting adopted cubic polynomial function without setting nodes; Exposure-lag dimension fitting was modeled by natural cubic splines with 3 degrees of freedom having an internal knot at equally spaced value ( 19 ); The degree of freedom of spline function of meteorological factors was set to 3; Since the median incubation period of PTB was 7 weeks ( 20 , 21 ), while avoiding visit delay and making the model extract more complete information. Thus, the maximum latency was determined as 12 weeks.…”

Extreme temperature exposure and urolithiasis: A time series analysis in Ganzhou, China

“…According to previous studies (8,18) and the minimum principle of the quasi-Akaike information criterion (q-AIC), the lag duration was determined to be 10 days, and r and h were included in the model with a natural cubic spline function (df = 3), and the df of time was selected to be 7/year. Taking the daily mean temperature of 10 • C as the reference temperature, the predicted value range was 0∼32 • C with an interval of 0.1 • C. Warm and cold effects on urolithiasis hospitalizations were estimated at the ambient temperature values of P99 (30.4 • C) and P1 (2.9 • C) (19). To identify vulnerable people, we further conducted subgroup analyses by gender (male and female) and age groups (0-20, 21-40, 41-60, and >60 years old).…”