Skip to main content
Download PDF

Associations between ambient temperature and adult asthma hospitalizations in Beijing, China: a time-stratified case-crossover study

Respiratory Research volume 23, Article number: 38 (2022) Cite this article

Abstract

Background

Studies on the associations between ambient temperature and asthma hospitalizations are limited, and the results are controversial. We aimed to assess the short-term effects of ambient temperature on the risk of asthma hospitalizations and quantify the hospitalization burdens of asthma attributable to non-optimal temperature in adults in Beijing, China.

Methods

We collected daily asthma hospitalizations, meteorological factors and air quality data in Beijing from 2012 to 2015. We applied a time-stratified case-crossover design and fitted a distributed lag non-linear model with a conditional quasi-Poisson regression to explore the association between ambient temperature and adult asthma hospitalizations. The effect modifications of these associations by gender and age were assessed by stratified analyses. We also computed the attributable fractions and numbers with 95% empirical confidence intervals (eCI) of asthma hospitalizations due to extreme and moderate temperatures.

Results

From 2012 to 2015, we identified a total of 18,500 hospitalizations for asthma among adult residents in Beijing, China. Compared with the optimal temperature (22 °C), the cumulative relative risk (CRR) over lag 0–30 days was 2.32 with a 95% confidence interval (CI) of 1.57–3.42 for extreme cold corresponding to the 2.5th percentile (− 6.5 °C) of temperature distribution and 2.04 (95% CI 1.52–2.74) for extreme heat corresponding to the 97.5th percentile (29 °C) of temperature distribution. 29.1% (95% eCI 17.5–38.0%) of adult asthma hospitalizations was attributable to non-optimum temperatures. Moderate cold temperatures yielded most of the burdens, with an attributable fraction of 20.3% (95% eCI 9.1–28.7%). The temperature-related risks of asthma hospitalizations were more prominent in females and younger people (19–64 years old).

Conclusions

There was a U-shaped association between ambient temperature and the risk of adult asthma hospitalizations in Beijing, China. Females and younger patients were more vulnerable to the effects of non-optimum temperatures. Most of the burden was attributable to moderate cold. Our findings may uncover the potential impact of climate changes on asthma exacerbations.

Background

Asthma is among the most prevalent chronic airway diseases with significant public health consequences. According to the Global Burden of Disease, asthma affected about 358 million people in 2015, leading to an enormous disease burden worldwide [1]. In China, approximately 4.2% of the adult population is suffering from asthma, representing 45.7 million Chinese adults [2]. Asthma is characterized by reversible airflow obstruction with symptoms of coughing, wheezing, breathlessness and chest stuffiness [3]. As there is no definitive cure, the disease tends to recur and seriously affect patients’ quality of life. Considering the high prevalence of asthma and the difficulty in control and intervention, a comprehensive understanding of the risk factors of asthma exacerbations is imperative. Many environmental risk factors, such as air pollution, pollen, tobacco smoke and meteorological factors have been reported to contribute to asthma exacerbations, leading to hospitalization [4,5,6].

Among numerous meteorological factors, temperature has been the most studied [7]. Although some epidemiological studies have reported that both cold and heat exposures are associated with increased risks of asthma [8,9,10], the association between ambient temperature and asthma hospitalizations is still inconclusive. The majority of studies have only found significant cold effects on asthma hospital visits or admissions [11,12,13,14,15,16], while limited studies have provided evidence for increased risks of asthma caused by heat exposures [17,18,19]. With global climate changes, extreme weather events, especially heatwaves, are expected to increase in frequency and intensity [20]. The growing public concern about a warming climate urges health practitioners to further clarify the relationships between potential disease risks and heat exposures. Moreover, most previous studies only quantified the association between asthma and ambient temperature by ratio measures, such as relative risk or odds ratio. Yet, few studies have estimated the attributable burden as well [21, 22]. Compared with ratio measures, the attributable burden may provide more specific information on the actual influence of the exposures and benefits of prevention and intervention by calculating attributable risk measures, involving attributable fraction (AF) and attributable number (AN) [23].

The study aimed to explore the exposure–response relationship between ambient temperature and daily hospitalizations for asthma among adults in Beijing, China, and how the relationship varied within different sex and age groups. We also calculated the total hospitalization burden of asthma attributed to non-optimum temperature and the relative contributions separated into different temperature ranges, including extreme heat, moderate heat, moderate cold and extreme cold. Our findings may assist in developing suitable public health interventions and risk assessment methods for reducing asthma exacerbations resulting from abnormal ambient temperatures.

Methods

Data source

Our study area was Beijing, which is located in north China (39° 56′ N, 116° 20′ E). As the capital city of China, it covers an area of 16,410.54 km2, with more than 21 million residents in 2015. Beijing’s climate belongs to a temperate semi-humid continental monsoon climate characterized by hot, rainy summer and cold, dry winter. In 2015, a cross-sectional study conducted in Beijing among 26,166 Chinese adults aged 20 years or older showed that the prevalence of doctor-diagnosed asthma was 2.8% in males and 2.1% in females [24].

We obtained daily records of hospitalizations for asthma (emergency department or outpatient visits were not included) from January 1, 2012 to December 31, 2015 from the Beijing Public Health Information Center (http://www.phic.org.cn/). The center collected inpatient data from all the tertiary and secondary hospitals in Beijing by the electronic disease reporting system. Each record contained the hospital’s name, date of admission and discharge, patient’s gender, date of birth, residential address, and discharge diagnoses. The hospitalization data included in our study were identified by asthma as a primary discharge diagnosis according to the 10th version of the International Classification of Diseases (ICD-10, code J45-46). Further, individuals residing outside of Beijing or aged ≤ 18 years old were excluded from this study. The study was approved and exempt from the full ethical review by the Institutional Review Board of Peking Union Medical College Hospital because all the records included were anonymous.

We acquired daily meteorological data during the same period from the China Meteorological Data Sharing Service System (http://data.cma.cn/), including daily minimum, maximum, and mean temperature (°C), mean relative humidity (%), and mean wind speed (m/s). To adjust for the potential effects of air pollutants, daily air quality index (AQI) data were extracted from the China National Environmental Monitoring Centre (http://www.cnemc.cn/). The daily AQI value is a comprehensive indicator of air quality, determined by the maximum value of individual air quality indexes of six regularly monitored air pollutants (fine particulate matter, inhalable particulate matter, sulfur dioxide, nitrogen dioxide, ozone, and carbon monoxide).

Since influenza has been associated with asthma exacerbations [25], we also collected information on influenza epidemics defined as the proportion of isolates positive for influenza surpassed 30% of the maximum seasonal level (Influenza surveillance season was defined from the 27th week of the previous year to the 26th week of the following year) [26] from the Chinese National Influenza Center (http://www.chinaivdc.cn/cnic/).

Study design and data analysis

We used a time-stratified case-crossover design to estimate the association between ambient temperature and daily hospitalizations for asthma in adults. In contrast with the traditional case–control study, the case-crossover design has superiority in controlling the time-invariant confounders of individual characteristics because each person serves as his or her own control. For each admission, the case day was specified as the admission date. To minimize the bias of long-term trend, seasonality, and “weekday effects”, we matched 3 or 4 control days by the same day of the week earlier or later within the same month and year to the case days. A conditional quasi-Poisson regression model was applied to allow for over-dispersion in daily counts of asthma hospitalizations [27].

Previous studies have shown that the relationships between ambient temperature and asthma hospitalizations were non-linear with a lagged effect [8, 11]. Thus, a distributed lag non-linear model (DLNM) was adopted to assess the complicated non-linear and delayed temperature-health dependencies simultaneously [28] after adjusting for relative humidity, wind speed, AQI and influenza epidemics. We modeled the temperature exposure–response with a natural cubic spline with 4 degrees of freedom (df) and the lag-response with a natural cubic spline with 3 df in the log scale. We set a maximum lag of 30 days to explore the long delay of the effects because the effects of cold temperatures could last about 2–3 weeks with no substantial effects after more than 1 month [29], which was consistent with previous studies [8, 11, 30]. Although the effects of hot temperatures were more acute, the potential morbidity displacement by harvesting effects of high temperatures should be noticed [30]. Moreover, the possible delayed admission time should be considered because our study focused on hospitalizations for asthma rather than outpatient hospital visits or emergency department visits. The final model was as follows:

$$Log\left[ {E\left( {Y_{t} } \right)} \right] = \beta Temp_{t,l} + ns\left( {RH_{t} ,df = 4} \right) + ns\left( {WS_{t} ,df = 3} \right) + ns\left( {AQI_{t} ,df = 4} \right) + factor\left( {stratum} \right) + \gamma Influenza_{t} + \alpha$$

where t is the day of observation; E(Yt) denotes the expected daily counts of asthma hospitalizations on day t. Tempt,l is the cross-basis matrix to the temperature generated by DLNM. l represents the lag days. ns() is the natural cubic spline-smoothing function for non-linear variables, including RH, relative humidity; WS, wind speed, and AQI, air quality index. df means degrees of freedom. The optimal df selection for variable terms was based on minimizing the quasi-Poisson Akaike information criteria of the model. The degrees of freedom for relative humidity, wind speed, and AQI were set at 4, 3, 4, respectively. Stratum refers to the matched groups of case days and control days. Influenzat is a binary variable. If the day t is in the period of an influenza epidemic, the value of 1 otherwise 0. β and γ are the vectors of coefficients for corresponding variables. α is the model intercept.

We performed subgroup analyses stratified by gender and age (19–64 years old and ≥ 65 years old) to examine their potential modification effects. The minimum admission temperature (MAT) was determined by finding the temperature value corresponding to the lowest risk of total adult asthma hospitalizations from the cumulative exposure–response curve. We also took this value as the optimum temperature for all the subgroups. To quantitatively evaluate the effects of extreme heat and cold exposures, we calculated the relative risk (RR) and 95% confidence interval (CI) at the extreme values (the 97.5th percentile and the 2.5th percentile of daily mean temperature) relative to the MAT [31, 32]. The statistical methods of the study have been widely used in estimating the health impact of ambient temperature [33,34,35].

To quantify the AF and AN due to non-optimum temperatures under the framework of DLNM, we employed a backward perspective strategy that estimated the risk at time t as attributable to a series of past exposure events [23]. The formulas of backward AFx,t, and ANx,t at time t were as follows:

$$\mathrm{b}-{AF}_{x,t}=1-\mathrm{exp}\left(\sum_{l={l}_{0}}^{L}{\upbeta x}_{t-l}\text{,}l\right)$$
$$\mathrm{b}-{AN}_{x,t}=\mathrm{b}-{AF}_{x,t}\cdot {n}_{t}$$

where nt denotes the count of cases at time t. ANx,t and AFx,t implies the number of cases and the corresponding fraction at time t attributable to past exposures to x in the period t − l0,…, t − L. We further separated the temperature distribution into four components including the extreme cold and heat, and the moderate cold and heat [21, 31]. The extreme heat and cold were defined as temperatures higher than the 97.5th percentile and lower than the 2.5th percentile, respectively. The moderate heat and cold were defined as the ranges between the MAT and the 97.5th and 2.5th percentiles, respectively. Through Monte Carlo simulations, the estimation of 95% empirical confidence intervals (eCI) for AF and AN was determined as the related 2.5th and 97.5th percentiles of distributions by simulating 5000 random samples.

All statistical analyses were conducted using R software (version 3.6.1) with the packages of “gnm”, “dlnm” and “attrdl”. Statistical significance was regarded as a two-sided p < 0.05.

Sensitivity analysis

To test the robustness of our results, we conducted a series of sensitivity analyses: (1) changing df for relative humidity (3–5 df), wind speed (3–5 df) and AQI (3–5 df); (2) altering the df (3–5) of the natural cubic spline in the log scale for the lag-response space; (3) changing the maximum lag days to 14 and 21 in the lag parameter in DLNM; (4) applying different temperature metrics, including daily minimum temperature, maximum temperature and mean apparent temperature. The apparent temperature (AT) is an aggregative indicator to reflect the human thermal perception of ambient temperature. The indicator combines outdoor temperature with relative humidity and wind velocity. We calculated the daily mean AT using the following equations:

$${\text{AT}} = {\text{T}} + 0.{33} \times {\text{e}} - 0.{7}0 \times {\text{WS}} - {4}.00$$
(1)
$${\text{e}} = {\text{RH}}/{1}00 \times {6}.{1}0{5} \times {\text{exp}}\left[ {{17}.{27} \times {\text{T}}/\left( {{237}.{7} + {\text{T}}} \right)} \right]$$
(2)

where T denotes daily mean temperature (°C), WS refers to wind speed (m/s), e means water vapor pressure (hPa) which was calculated with Eq. (2); RH is relative humidity (%).

Results

Data description

Table 1 provides the descriptive statistics for daily adult asthma hospitalizations, meteorological variables and air quality from January 1, 2012 to December 31, 2015. During the study period, a total of 18,500 hospitalizations for asthma among adult residents were recorded in Beijing, a mean of 12.7 per day (range, 0–34). Of all these cases, there were more asthma hospitalizations for females (59.3%) than for males (40.7%). People aged 19–64 had more asthma hospitalizations than people aged 65 or above (59.9% vs. 40.1%). The median (minimum, maximum) values of daily mean temperature, minimum temperature, maximum temperature and mean apparent temperature were 14.0 °C (− 12.0 °C, 32.0 °C), 8.0 °C (− 18.0 °C, 27.0 °C), 21.0 °C (− 7.0 °C, 42.0 °C) and 6.2 °C (− 34.1 °C, 32.7 °C), respectively. The medians of relative humidity, wind speed and AQI were 54.0% (range, 8.0–98.0%), 8.0 m/s (range, 3.0–34.0 m/s) and 94.0 (range, 14.0–485.0), respectively. Figure 1 shows the time-series plots of daily meteorological variables, AQI and hospitalizations for asthma among adults in Beijing, 2012–2015.

Table 1 Descriptive statistics of daily adult asthma hospitalizations, meteorological variables and air quality in Beijing, China from 2012 to 2015
Full size table
Fig. 1
figure 1

Time-series plots for daily adult asthma hospitalizations, meteorological variables and air quality in Beijing, 2012–2015. (Tmean: mean temperature (°C), RH: relative humidity (%), WS: wind speed (m/s), AQI: air quality index)

Full size image

Associations between ambient temperature and adult asthma hospitalizations

Figure 2 shows the cumulative exposure–response curve exhibited a U-shaped relationship between daily mean temperature and total adult asthma hospitalizations across a 30-day lag period. For the whole study population, the MAT was 22 °C. Cumulative relative risk (CRR) of hospitalizations for asthma increased significantly below and above the MAT.

Fig. 2
figure 2

Lag-cumulative exposure–response associations between daily mean temperature and total adult asthma hospitalizations over lag 0–30 days in Beijing, 2012–2015

Full size image

Table 2 shows the cumulative effects of extreme temperatures on total adult asthma hospitalizations over different lag periods (0, 0–3, 0–7, 0–14, 0–21, and 0–30 days, i.e. the associations between single-day temperature exposures and the cumulative risks of asthma hospitalizations from the same day (lag 0) to the next few days). We found that both extreme cold (2.5th percentile vs. MAT) and heat (97.5th percentile vs. MAT) exposures were significantly associated with higher risks of asthma hospitalizations through lag 14 to 30 days, with the maximum CRR over lag 0–30 days (Extreme cold: CRR = 2.32, 95% CI 1.57, 3.42; Extreme heat: CRR = 2.04, 95% CI 1.52, 2.74). The CRRs associated with extreme cold were higher than those associated with extreme heat. Figure 3 shows the lag-response relationships between extreme temperatures and total adult asthma hospitalizations over lag 0–30 days. We observed that both the lagged effects of extreme cold and heat lasted for about 3 or 4 weeks. The extreme cold effects became significant on lag day 5 and lasted until lag day 22, whereas the extreme heat effects became significant on lag day 5 and persisted until lag day 25.

Table 2 Lag-cumulative relative risks for total adult asthma hospitalizations associated with extreme cold exposure [2.5th percentile (− 6.5 °C) relative to MAT (22 °C)] and extreme heat exposure [97.5th percentile (29 °C) relative to MAT]
Full size table
Fig. 3
figure 3

Lag-response relationships between extreme cold exposure [2.5th percentile (− 6.5 °C) relative to MAT (22 °C)] and heat exposure [97.5th percentile (29 °C) relative to MAT] and total adult asthma hospitalizations in Beijing, 2012–2015. (MAT minimum admission temperature)

Full size image

Figure 4 shows the exposure–response plots for the subgroups over lag 0–30 days. The variation trends of CRR in all the subgroups followed a similar pattern as the total study population. Table 3 shows the lag-cumulative effects of extreme cold and heat exposures by subgroup analyses stratified by gender and age at different lag days. Among all the subgroups, the risks of extreme cold on adult asthma hospitalizations were higher than those of extreme heat. Significant associations between both the extreme cold and heat exposures and adult asthma hospitalizations in females appeared at lag times of 14–30 days. However, no significant association was detected in males. Female asthma patients seem to be more vulnerable to extreme temperatures than male patients. Regarding different age groups, the cumulative effects of extreme temperatures (both heat and cold) in the younger patients (19–64 years old) were significant at lag times of 14–30 days, while the cumulative effects were significant only at a lag of 30 days in the elderly (≥ 65 years old). The CRRs of extreme cold and heat exposures over lag 0–30 days were higher in the younger group (Extreme cold: CRR = 2.46, 95% CI 1.56, 3.89; Extreme heat: CRR = 2.39, 95% CI 1.71, 3.34) than the elderly (Extreme cold: CRR = 2.11, 95% CI 1.22, 3.64; Extreme heat: CRR = 1.56, 95% CI 1.00, 2.43). The results suggested that the younger patients were more likely to be exposed to higher risk in extreme temperatures.

Fig. 4
figure 4

Lag-cumulative exposure–response associations between daily mean temperature and adult asthma hospitalizations over lag 0–30 days stratified by gender and age in Beijing, 2012–2015

Full size image
Table 3 Lag-cumulative relative risks of adult asthma hospitalizations associated with extreme cold exposure [2.5th percentile (− 6.5 °C) relative to MAT (22 °C)] and heat exposure [97.5th percentile (29 °C) relative to MAT] stratified by gender and age
Full size table

Attributable burden of adult asthma hospitalizations due to ambient temperature

Table 4 shows the estimated AF and AN of adult asthma hospitalizations owing to both cold and heat exposures divided into contributions from moderate and extreme temperatures. The total fraction of hospitalization for asthma among adults caused by non-optimal temperatures over lag 0–30 days was 29.1% (95% eCI 17.5%, 38.0%), and most of them were related to the moderate cold (20.3%, 95% eCI 9.1%, 28.7%). Both extreme heat and cold exposures were responsible for a small portion, with the AF equal to 2.2% (95% eCI 1.3, 3.0) for extreme heat and 2.4% (95% eCI 1.4, 3.1) for extreme cold.

Table 4 Attributable fraction (%) and attributable number (95% eCI) of adult asthma hospitalizations due to different temperature ranges in the total population and subgroups stratified by gender and age
Full size table

The attributable risks varied among different subgroups. The overall estimated temperature-related burdens of asthma hospitalization were much higher in females (34.5%, 95% eCI 21.6, 44.2) than males (19.9%, 95% eCI 1.0, 33.5). The results were constant when further comparing the burdens resulting from different components of temperature between the two gender groups. Compared with the population aged ≥ 65 years old (26.0%, 95% eCI 7.7, 39.4), the total AF of non-optimum temperature was slightly higher in the population aged 19–64 years old (30.9%, 95% eCI 18.1, 40.7). The comparison between the two age groups was with little differences for the burdens attributable to moderate and extreme temperatures.

Sensitivity analyses

The risk estimates did not substantially differ after changing the dfs for relative humidity (3–5), wind speed (3–5), and AQI (3–5), altering the df (3–5) of the natural cubic spline in the log scale for the lag-response space, and changing the maximum lag days (14 or 21) in the model (see Additional file 1: Tables S1–S3 and Additional file 2: Fig. S1). Besides, our results were still robust by applying different temperature metrics, including daily minimum temperature, daily maximum temperature and daily mean apparent temperature (see Additional file 1: Table S4 and Additional file 2: Figs. S2–S4).

Discussion

Our results revealed that both low and high temperatures were significantly associated with increased risks of adult asthma hospitalizations. The associations were non-linear and followed U-shape curves in all the subgroups and the total population. Overall, a high fraction (29.1%) of adult asthma hospitalizations were attributable to non-optimum temperature over lag 0–30 days. Most of the hospitalization burden was attributable to moderate cold exposures. Females and younger patients were more susceptible to the short-term effects of extreme temperatures with greater burdens attributable to non-optimum temperatures. This is the first study estimating the attributable burden of adult asthma hospitalizations from ambient temperatures to the best of our knowledge.

We found that higher cumulative risks of adult asthma hospitalizations were related to both heat and cold exposures in Beijing, China. The extreme cold effect was higher than the extreme heat effect. Most previous studies only showed significant cold effects while no apparent association between heat exposures and asthma was detected. A similar study conducted in Shanghai, China reported that lower temperature (the 1st percentile of temperature relative to the median temperature) was associated with increased asthma hospitalizations with CRR = 1.79 (95% CI 1.18, 2.72) at lag 0–30 days [11]. Another study in Dongguan, China found that the CRR associated with extreme cold (the 5th percentile of temperature relative to the minimum morbidity temperature) for asthma outpatient visits was 1.04 (95% CI 1.00, 1.08) at lag 0–7 days [21]. On the other hand, a few studies have suggested positive associations between high temperatures and asthma hospitalizations. Lam et al. conducted a time-series study in Hong Kong and found that in hot seasons, the CRR associated with heat exposures (30 °C vs. 27 °C) was 1.19 (95% CI 1.06, 1.34) at lag 0–3 days [8]. The inconsistent results among various studies may be explained by differences in climate conditions, study designs and analytical approaches.

The intrinsic biological mechanisms are still unclear. The increased cold-related asthma hospitalizations may be associated with increasing bronchoconstriction [36], airway inflammation [37], mucus secretion [38] and decreasing effectiveness of immune responses [39] triggered by low temperatures directly. Moreover, cold temperatures can favor the transmission and survival of influenza viruses [40, 41], which may increase the risk of infection-related exacerbations of asthma. The possible mechanisms between heat exposures and asthma exacerbations point to bronchoconstriction mediated by the cholinergic reflex pathway [42] and airway inflammation aggravation through transient receptor potential channels [43]. Additionally, high temperatures play a key role in plants producing allergenic pollens with stronger allergenicity [4]. High concentrations of allergens, such as pollens and fungi in the air have been linked to the increased asthma severity in both children and adults [44]. On the other hand, people tend to spend more time indoors during extreme temperatures for obtaining better comfort, either using artificial heating during low ambient temperatures or using artificial cooling during high ambient temperatures. This extended stay indoors may increase the exposure to indoor molds, allergens or pollutants, which are known causes of asthma exacerbations [45,46,47].

Findings from our study showed that non-optimum temperatures were responsible for a substantial portion (29.1%) of adult asthma hospitalizations over lag 0–30 days. Most hospitalizations (20.3%) were attributed to the days with moderate cold temperatures. These results were consistent with several previous studies focusing on mortality [31, 35, 48]. However, Zhao et al. found that moderate heat exposure accounted for most of the morbidity burden of asthma in Dongguan, China [21]. The attributable burden for the temperature-asthma association may vary by distributions of days in different temperature ranges. Although extreme temperatures bring higher risk than moderate temperatures, the moderate cold days were the most in number in our study. Consequently, more attention should be paid to moderate cold when planning adaptation strategies and measures to reduce asthma hospitalizations.

Our subgroup analysis by gender showed that female patients with asthma were more vulnerable to ambient temperatures than male patients, which is similar to some epidemiological studies [12, 22]. The reasons for the discrepancy in the susceptibility of temperature-related asthma exacerbations by gender may point to bronchial hyperresponsiveness and estrogen in females [49]. As for subgroup analysis by age, we found that the younger population (19–64 years old) had higher risks of asthma hospitalizations attributable to ambient temperatures than the elderly, which was also shown in other prior studies [8, 14, 16]. Younger people were more vulnerable to temperature, possibly because they stay longer for work outdoors and are more likely to be exposed to abnormal temperatures.

There are some limitations to this study. Firstly, the temperature data were from fixed meteorological monitoring stations rather than individual exposure measurements, which may not reflect real exposures. Secondly, the data of some potential confounding factors, such as pollens, precipitation and thunderstorms, were unavailable for analysis. These factors are likely to impact the risk of asthma attacks and the number of hospitalizations [7]. Thirdly, since our study focused on the adult population in a single city, the extrapolation of our findings to other regions and the children population should be undertaken cautiously. Fourthly, mild asthma exacerbations treated in the outpatient setting or emergency department were not included in the study. We only focused on the association between ambient temperature and more severe asthma exacerbations requiring admission to a hospital ward, which may underestimate the effect of abnormal temperatures. Lastly, as an ecological study, the unit of analysis is a group of people instead of individuals [50]. Hence, the results should be interpreted with caution when applying to individuals. More comprehensive, individual-based epidemiological studies are needed in the future.

Conclusions

This study provides evidence of the non-linear associations between ambient temperature and adult asthma hospitalizations, and the corresponding disease burden that is mainly attributable to moderate cold in Beijing, China. The vulnerable populations including the youngers and females, need to strengthen their awareness of the adverse impacts of both extreme heat and cold exposures. Our findings may have significant implications that exposures to high and low temperatures should be considered as potentially preventable triggers of asthma hospitalizations. In the context of climate change, such evidence is crucial for planning proper health risk education to the public, tailoring effective intervention strategies and evaluating the overall burden of asthma hospitalizations associated with abnormal temperatures.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

AF:

Attributable number

AN:

Attributable fraction

AQI:

Air quality index

CI:

Confidence interval

CRR:

Cumulative relative risk

df:

Degrees of freedom

DLNM:

Distributed lag non-linear model

eCI:

Empirical confidence interval

ICD-10:

The 10th version of the International Classification of Diseases

MAT:

Minimum admission temperature

RH:

Relative humidity

RR:

Relative risk

WS:

Wind speed

References

  1. Soriano JB, Abajobir AA, Abate KH, Abera SF, Agrawal A, Ahmed MB, Aichour AN, Aichour I, Aichour MTE, Alam K, et al. Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Respir Med. 2017;5:691–706.

    Article  Google Scholar 

  2. Huang K, Yang T, Xu J, Yang L, Zhao J, Zhang X, Bai C, Kang J, Ran P, Shen H, et al. Prevalence, risk factors, and management of asthma in China: a national cross-sectional study. Lancet. 2019;394:407–18.

    Article  PubMed  Google Scholar 

  3. Romaszko-Wojtowicz A, Cymes I, Draganska E, Doboszynska A, Romaszko J, Glinska-Lewczuk K. Relationship between biometeorological factors and the number of hospitalizations due to asthma. Sci Rep. 2020;10:9593.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. D’Amato G, Cecchi L. Effects of climate change on environmental factors in respiratory allergic diseases. Clin Exp Allergy. 2008;38:1264–74.

    Article  CAS  PubMed  Google Scholar 

  5. Gilmour MI, Jaakkola MS, London SJ, Nel AE, Rogers CA. How exposure to environmental tobacco smoke, outdoor air pollutants, and increased pollen burdens influences the incidence of asthma. Environ Health Perspect. 2006;114:627–33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Diette GB, McCormack MC, Hansel NN, Breysse PN, Matsui EC. Environmental issues in managing asthma. Respir Care. 2008;53:602–15 (discussion 616-607).

    PubMed  Google Scholar 

  7. Bodaghkhani E, Mahdavian M, MacLellan C, Farrell A, Asghari S. Effects of meteorological factors on hospitalizations in adult patients with asthma: a systematic review. Can Respir J. 2019;2019:3435103.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Lam HC, Li AM, Chan EY, Goggins WB 3rd. The short-term association between asthma hospitalisations, ambient temperature, other meteorological factors and air pollutants in Hong Kong: a time-series study. Thorax. 2016;71:1097–109.

    Article  PubMed  Google Scholar 

  9. Xu Z, Huang C, Hu W, Turner LR, Su H, Tong S. Extreme temperatures and emergency department admissions for childhood asthma in Brisbane Australia. Occup Environ Med. 2013;70:730–5.

    Article  PubMed  Google Scholar 

  10. Hu Y, Cheng J, Jiang F, Liu S, Li S, Tan J, Yin Y, Tong S. Season-stratified effects of meteorological factors on childhood asthma in Shanghai, China. Environ Res. 2020;191:110115.

    Article  CAS  PubMed  Google Scholar 

  11. Zhang Y, Peng L, Kan H, Xu J, Chen R, Liu Y, Wang W. Effects of meteorological factors on daily hospital admissions for asthma in adults: a time-series analysis. PLoS ONE. 2014;9:e102475.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  12. Son JY, Bell ML, Lee JT. The impact of heat, cold, and heat waves on hospital admissions in eight cities in Korea. Int J Biometeorol. 2014;58:1893–903.

    Article  PubMed  Google Scholar 

  13. Abe T, Tokuda Y, Ohde S, Ishimatsu S, Nakamura T, Birrer RB. The relationship of short-term air pollution and weather to ED visits for asthma in Japan. Am J Emerg Med. 2009;27:153–9.

    Article  PubMed  Google Scholar 

  14. Wang YC, Lin YK. Temperature effects on outpatient visits of respiratory diseases, asthma, and chronic airway obstruction in Taiwan. Int J Biometeorol. 2015;59:815–25.

    Article  PubMed  Google Scholar 

  15. Cong X, Xu X, Zhang Y, Wang Q, Xu L, Huo X. Temperature drop and the risk of asthma: a systematic review and meta-analysis. Environ Sci Pollut Res Int. 2017;24:22535–46.

    Article  PubMed  Google Scholar 

  16. Kabir AF, Ng CFS, Yasumoto S, Hayashi T, Watanabe C. Effect of ambient temperature on daily nebulized asthma hospital visits in a tropical city of Dhaka, Bangladesh. Int J Environ Res Public Health. 2021;18:890.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Soneja S, Jiang C, Fisher J, Upperman CR, Mitchell C, Sapkota A. Exposure to extreme heat and precipitation events associated with increased risk of hospitalization for asthma in Maryland, U.S.A. Environ Health. 2016;15:57.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  18. Lin S, Luo M, Walker RJ, Liu X, Hwang SA, Chinery R. Extreme high temperatures and hospital admissions for respiratory and cardiovascular diseases. Epidemiology. 2009;20:738–46.

    Article  PubMed  Google Scholar 

  19. Wang J, Zhang Y, Li B, Zhao Z, Huang C, Zhang X, Deng Q, Lu C, Qian H, Yang X, et al. Asthma and allergic rhinitis among young parents in China in relation to outdoor air pollution, climate and home environment. Sci Total Environ. 2021;751:141734.

    Article  CAS  PubMed  Google Scholar 

  20. D’Amato G, Holgate ST, Pawankar R, Ledford DK, Cecchi L, Al-Ahmad M, Al-Enezi F, Al-Muhsen S, Ansotegui I, Baena-Cagnani CE, et al. Meteorological conditions, climate change, new emerging factors, and asthma and related allergic disorders. A statement of the World Allergy Organization. World Allergy Organ J. 2015;8:25.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Zhao Y, Huang Z, Wang S, Hu J, Xiao J, Li X, Liu T, Zeng W, Guo L, Du Q, Ma W. Morbidity burden of respiratory diseases attributable to ambient temperature: a case study in a subtropical city in China. Environ Health. 2019;18:89.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Pan R, Gao J, Wang X, Bai L, Wei Q, Yi W, Xu Z, Duan J, Cheng Q, Zhang Y, Su H. Impacts of exposure to humidex on the risk of childhood asthma hospitalizations in Hefei, China: effect modification by gender and age. Sci Total Environ. 2019;691:296–305.

    Article  CAS  PubMed  Google Scholar 

  23. Gasparrini A, Leone M. Attributable risk from distributed lag models. BMC Med Res Methodol. 2014;14:55.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Zhang H, Dong L, Kang YK, Lu Y, Wei HH, Huang J, Wang X, Huang K. Epidemiology of chronic airway disease: results from a cross-sectional survey in Beijing, China. J Thorac Dis. 2018;10:6168–75.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Gerke AK, Yang M, Tang F, Foster ED, Cavanaugh JE, Polgreen PM. Association of hospitalizations for asthma with seasonal and pandemic influenza. Respirology. 2014;19:116–21.

    Article  PubMed  Google Scholar 

  26. Cowling BJ, Wong IO, Ho LM, Riley S, Leung GM. Methods for monitoring influenza surveillance data. Int J Epidemiol. 2006;35:1314–21.

    Article  PubMed  Google Scholar 

  27. Armstrong BG, Gasparrini A, Tobias A. Conditional Poisson models: a flexible alternative to conditional logistic case cross-over analysis. BMC Med Res Methodol. 2014;14:122.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Gasparrini A. Modeling exposure-lag-response associations with distributed lag non-linear models. Stat Med. 2014;33:881–99.

    Article  PubMed  Google Scholar 

  29. Ye X, Wolff R, Yu W, Vaneckova P, Pan X, Tong S. Ambient temperature and morbidity: a review of epidemiological evidence. Environ Health Perspect. 2012;120:19–28.

    Article  PubMed  Google Scholar 

  30. Feng F, Ma Y, Zhang Y, Shen J, Wang H, Cheng B, Jiao H. Effects of extreme temperature on respiratory diseases in Lanzhou, a temperate climate city of China. Environ Sci Pollut Res Int. 2021;28:49278–88.

    Article  PubMed  Google Scholar 

  31. Gasparrini A, Guo Y, Hashizume M, Lavigne E, Zanobetti A, Schwartz J, Tobias A, Tong S, Rocklöv J, Forsberg B, et al. Mortality risk attributable to high and low ambient temperature: a multicountry observational study. Lancet. 2015;386:369–75.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Bai L, Li Q, Wang J, Lavigne E, Gasparrini A, Copes R, Yagouti A, Burnett RT, Goldberg MS, Cakmak S, Chen H. Increased coronary heart disease and stroke hospitalisations from ambient temperatures in Ontario. Heart. 2018;104:673–9.

    Article  CAS  PubMed  Google Scholar 

  33. Guo Y, Barnett AG, Pan X, Yu W, Tong S. The impact of temperature on mortality in Tianjin, China: a case-crossover design with a distributed lag nonlinear model. Environ Health Perspect. 2011;119:1719–25.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Chen K, Breitner S, Wolf K, Hampel R, Meisinger C, Heier M, von Scheidt W, Kuch B, Peters A, Schneider A, Group KS. Temporal variations in the triggering of myocardial infarction by air temperature in Augsburg, Germany, 1987–2014. Eur Heart J. 2019;40:1600–8.

    Article  CAS  PubMed  Google Scholar 

  35. Fu SH, Gasparrini A, Rodriguez PS, Jha P. Mortality attributable to hot and cold ambient temperatures in India: a nationally representative case-crossover study. PLoS Med. 2018;15:e1002619.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Koskela HO. Cold air-provoked respiratory symptoms: the mechanisms and management. Int J Circumpolar Health. 2007;66:91–100.

    Article  PubMed  Google Scholar 

  37. Liu H, Hua L, Liu Q, Pan J, Bao Y. Cold stimuli facilitate inflammatory responses through transient receptor potential melastatin 8 (TRPM8) in primary airway epithelial cells of asthmatic mice. Inflammation. 2018;41:1266–75.

    Article  CAS  PubMed  Google Scholar 

  38. Juan Y, Haiqiao W, Xie W, Huaping H, Zhong H, Xiangdong Z, Kolosov VP, Perelman JM. Cold-inducible RNA-binding protein mediates airway inflammation and mucus hypersecretion through a post-transcriptional regulatory mechanism under cold stress. Int J Biochem Cell Biol. 2016;78:335–48.

    Article  PubMed  CAS  Google Scholar 

  39. Shephard RJ, Shek PN. Cold exposure and immune function. Can J Physiol Pharmacol. 1998;76:828–36.

    Article  CAS  PubMed  Google Scholar 

  40. Lowen AC, Mubareka S, Steel J, Palese P. Influenza virus transmission is dependent on relative humidity and temperature. PLoS Pathog. 2007;3:1470–6.

    Article  CAS  PubMed  Google Scholar 

  41. Tang JW. The effect of environmental parameters on the survival of airborne infectious agents. J R Soc Interface. 2009;6(Suppl 6):S737-746.

    PubMed  PubMed Central  Google Scholar 

  42. Hayes D Jr, Collins PB, Khosravi M, Lin RL, Lee LY. Bronchoconstriction triggered by breathing hot humid air in patients with asthma: role of cholinergic reflex. Am J Respir Crit Care Med. 2012;185:1190–6.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Deng L, Ma P, Wu Y, Ma Y, Yang X, Li Y, Deng Q. High and low temperatures aggravate airway inflammation of asthma: evidence in a mouse model. Environ Pollut. 2020;256:113433.

    Article  CAS  PubMed  Google Scholar 

  44. Dharmage SC, Perret JL, Custovic A. Epidemiology of asthma in children and adults. Front Pediatr. 2019;7:246.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Segura-Medina P, Vargas MH, Aguilar-Romero JM, Arreola-Ramírez JL, Miguel-Reyes JL, Salas-Hernández J. Mold burden in house dust and its relationship with asthma control. Respir Med. 2019;150:74–80.

    Article  PubMed  Google Scholar 

  46. Paterson CA, Sharpe RA, Taylor T, Morrissey K. Indoor PM2.5, VOCs and asthma outcomes: a systematic review in adults and their home environments. Environ Res. 2021;202:111631.

    Article  CAS  PubMed  Google Scholar 

  47. Kanchongkittiphon W, Mendell MJ, Gaffin JM, Wang G, Phipatanakul W. Indoor environmental exposures and exacerbation of asthma: an update to the 2000 review by the Institute of Medicine. Environ Health Perspect. 2015;123:6–20.

    Article  PubMed  Google Scholar 

  48. Chen R, Yin P, Wang L, Liu C, Niu Y, Wang W, Jiang Y, Liu Y, Liu J, Qi J, et al. Association between ambient temperature and mortality risk and burden: time series study in 272 main Chinese cities. BMJ. 2018;363:k4306.

    Article  PubMed  PubMed Central  Google Scholar 

  49. McCallister JW, Mastronarde JG. Sex differences in asthma. J Asthma. 2009;45:853–61.

    Article  Google Scholar 

  50. Sedgwick P. Ecological studies: advantages and disadvantages. BMJ. 2014;348:g2979.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We especially thank the Beijing Public Health Information Center for providing the data.

Funding

This work was supported by Chinese Academy of Medical Sciences (CAMS) Initiative for Innovative Medicine [Grant number 2017-I2M-2-001]; National Natural Science Foundation [Grant number 91643208]; and National Key Research and Development Plan [Grant number 2017YFC0211703 and 2015CB553402].

Author information

Authors and Affiliations

  1. Department of Medicine, Peking Union Medical College Hospital, Peking Union Medical College & Chinese Academy of Medical Sciences, No.1 Shuaifuyuan Wangfujing Dongcheng District,, Beijing, 100730, China

    Yuxiong Chen, Dehui Kong, Jia Fu, Yongqiao Zhang, Yakun Zhao, Yanbo Liu, Zhen’ge Chang, Yijie Liu, Xiaole Liu, Kaifeng Xu & Zhongjie Fan

  2. Department of Biochemistry, State Key Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, 100005, China

    Chengyu Jiang

Authors
  1. Yuxiong Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Dehui Kong
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jia Fu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yongqiao Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yakun Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yanbo Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Zhen’ge Chang
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Yijie Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Xiaole Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Kaifeng Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Chengyu Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Zhongjie Fan
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

ZF obtained the original data and funding. ZF and YC designed the study. DK, JF, YoZ and ZC preprocessed the data. YaL, YiL and YaZ analyzed the data. YC drafted the initial manuscript. XL, KX, CJ and ZF reviewed and edited the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Zhongjie Fan.

Ethics declarations

Ethics approval and consent to participate

The study was approved and exempt from the full ethical review by the Institutional Review Board of Peking Union Medical College Hospital because all the records included were anonymous (S-K1097).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

Lag-cumulative relative risks for total adult asthma hospitalizations associated with extreme heat exposure [97.5th percentile (29 °C) relative to MAT (22 °C)] and cold exposure [2.5th percentile (− 6 °C) relative to MAT] over lag 0–30 days using different degrees of freedom of confounders. Table S2. Lag-cumulative relative risks for adult asthma hospitalizations associated with extreme heat exposure [97.5th percentile (29 °C) relative to MAT (22 °C)] and cold exposure [2.5th percentile (− 6.5 °C) relative to MAT] over lag 0–30 days using different degrees of freedom of natural cubic spline in the log scale for the lag-response space. Table S3. Lag-cumulative relative risks for adult asthma hospitalizations associated with extreme cold exposure (2.5th percentile relative to MAT) and heat exposure (97.5th percentile relative to MAT) using different maximum lag days in the distributed lag non-linear model. Table S4. Lag-cumulative relative risks for adult asthma hospitalizations associated with extreme cold exposure (2.5th percentile relative to MAT) and heat exposure (97.5th percentile relative to MAT) over lag 0–30 days employing different temperature metrics.

Additional file 2: Figure S1.

Lag-cumulative exposure–response associations between daily mean temperature and total adult asthma hospitalizations in Beijing, 2012–2015 using different maximum lag days in the distributed lag non-linear model. Figure S2. Lag-cumulative exposure–response associations between daily minimum temperature and total adult asthma hospitalizations in Beijing, 2012–2015. Figure S3. Lag-cumulative exposure–response associations between daily maximum temperature and total adult asthma hospitalizations in Beijing, 2012–2015. Figure S4. Lag-cumulative exposure–response associations between daily mean apparent temperature and total adult asthma hospitalizations in Beijing, 2012–2015.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chen, Y., Kong, D., Fu, J. et al. Associations between ambient temperature and adult asthma hospitalizations in Beijing, China: a time-stratified case-crossover study. Respir Res 23, 38 (2022). https://doi.org/10.1186/s12931-022-01960-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12931-022-01960-8

Keywords

Respiratory Research

ISSN: 1465-993X

Contact us