Effect of Air Pollutants and Meteorological Factors on Daily Outpatient Visits of Allergic Rhinitis in Hohhot, China

Introduction

Air pollution is a major environmental risk factor affecting global health events. The human respiratory tract is directly connected to the outside world and is susceptible to air pollution and meteorological changes that can induce or exacerbate other systemic diseases. It has been shown that air pollution may lead to an increased incidence of respiratory diseases,^1,2 cardiovascular diseases^3,4 and population-based allergic diseases,⁵ and that there is a significant lag effect of air pollution on health. In recent years, with the industrialization of cities, westernized lifestyles, and drastic environmental changes, the incidence of AR has shown a significant increase, with prevalence rates as high as 50% of the global population.⁶ Previous studies have shown that the prevalence of allergic rhinitis varies regionally, and there is growing evidence that both environmental and genetic factors play important roles in the aetiology of allergic diseases.^7,8

China has a large population and even slight increases in pollutant concentrations can lead to significant health losses. Air pollutant concentrations are higher in the northern regions of China due to the dry climate and the burning of coal for heating in the winter, which can have more serious health effects.⁹ Hohhot is located on the northern border of China (Figure 1), with pocket-shaped landscapes with openings facing southwest, where air pollutants are blocked and retained to easily accumulate and aggregate, and diffusion conditions are poor, making it easy to form heavily polluted weather under unfavorable meteorological conditions. According to the composite index of ambient air quality in the Ecological Environment Bulletin of Inner Mongolia Autonomous Region in 2021, the air quality in Hohhot was below the average level of the autonomous region. In 2018, Wang Xueyan et al¹⁰ reported the current status of the prevalence of AR in six cities in Inner Mongolia region, and the prevalence rate of up to 32.4% has far exceeded the average level of the major cities in China in the previous statistics. A survey found that Risk factors related to AR include air pollutants, smoking history, antibiotic use, food allergy, drug allergy, asthma, family decoration in recent two years, etc. There are seasonal differences in the peak clinical symptoms of AR patients in Inner Mongolia, mainly from July to September. Artemisia annua, Chenopodium album and ragweed are the main pollen allergens in this area, and dust mite is the second largest allergen for AR patients in this area.¹¹ Therefore, this study investigated the short-term effects of air pollutants and meteorological factors on allergic rhinitis by analyzing the data of atmospheric pollutants, meteorology, and allergic rhinitis outpatient clinics in Hohhot over the past 3 years, with the aim of evaluating the health effects of air pollutants and meteorological factors and providing a scientific basis for future disease prevention and control.

Figure 1 Geographical location of Hohhot, Inner Mongolia.

Materials and Methods

Source of Information

Allergic rhinitis outpatient data were obtained from the case records of daily outpatient clinics in Inner Mongolia Autonomous Region People’s Hospital during the 3-year period from January 1, 2018, to December 31, 2020, including patient number, gender, age, International Classification of Diseases code (ICD-10), and date of diagnosis. Allergic rhinitis diseases were included according to ICD-10 (J30.401); non-Hohhot household cases were excluded according to home address; cases treated for trauma, surgery, and other reasons were excluded according to disease diagnosis; repeated outpatient cases were delimited by 28 d, and events occurring after more than 28 d were recorded as another event; case data with missing information were revalidated, and those that were still unavailable after revalidation were excluded medical records; clinical information that met the study criteria was extracted according to this criterion to create an allergic rhinitis database. The air pollutant data (PM_2.5, SO₂, etc.) and meteorological data (temperature, humidity, barometric pressure, etc.) for the same period were obtained from the China Air Quality Online Monitoring and Analyzing Platform and the China Meteorological Science Data Sharing Service Network.

Statistical Analysis

Descriptive Analysis and Spearman Correlation Analysis

The mean, standard deviation, median, quartile, minimum and maximum values of the pollutants, meteorological factors and the number of outpatient visits for allergic rhinitis were calculated using SPSS 22.0 statistical software to get a basic idea of the situation. Since the distribution of pollutant concentration and meteorological factor information was non-normal, the correlation between each pollutant and meteorological factor was analyzed by Spearman rank correlation, and the correlation coefficient (r) was used to express the degree of correlation between the variables, and significance was tested at the test level ɑ=0.05. When r>0, it means that there is a positive correlation between the variables; when r<0, it means that there is a negative correlation; the closer the absolute value of the correlation coefficient is to 1, the stronger the correlation is.

Generalized Additive Model

Time series analysis (TSA) has been widely used in environmental epidemiology and is the most commonly used method to study the acute effects of air pollution and meteorological factors on human beings.¹² Generalized Additive Models (GAM) are the most widely used time series models, which can simultaneously assess the effects of both linear and nonlinear relationships with the dependent variable. Environmental factors that have linear and nonlinear relationships with the dependent variable can be assessed simultaneously, while correcting for a variety of confounders.¹³ For the total population, the number of allergic rhinitis day visits is a small probability event, which can be considered to approximate a Poisson distribution. Therefore, in this study, a time series analysis of pollutants, meteorological factors, and the number of allergic rhinitis clinic visits was performed using a generalized summation model based on the Poisson distribution. Natural cubic spline (ns) was used to control for the long-term trend by treating the date, and the day of the week (dow) effect was controlled by a dummy variable. As the relationship between meteorological factors and outpatient attendance was also nonlinear, it was also controlled for with ns. The degrees of freedom of the model parameters were selected by using Akaike’s information criterion (AIC) and referring to related literature, and the smaller the value of AIC, the better the fit of the model.

The basic model is expressed as:

In our model, Y_t is the actual number of AR visits on day t and ε_t is the expected number of AR visits on day t; β₀ is the residual; β_xt is a regression coefficient representing the coefficient of linear effect on the number of visits; xt is the pollutant concentration on day t; s is a spline smoothing function for the time series at t; q is a nonparametric spline smoothing function used to control for the meteorological variables (including mean temperature, relative humidity, and barometric pressure); df is the degrees of freedom of the natural spline function; and dow is a dummy variable to control for weekly effects. We consider 6–10 df as calendar time and 3–7 df as environmental factors.

Relative Risk Calculation

At the same time, the relative risk (RR) of allergic rhinitis outpatient visits due to each 10 μg/m³ increase in pollutant concentration was calculated with the formula: RR = EXP (β × 10). Since the health effects of air pollution have a certain lag, this study analyzed the single-day lag effect of air pollutants from 0 to 7 days (lag0-lag7) and the cumulative lag effect from 2–8 days (lag01-lag07). lag0 is the value of air pollutant concentration on the day of exposure, lag1 is the air pollutant concentration of the day before the exposure, and so on up to lag7. Cumulative lag is the moving average of pollutant concentrations on the day of exposure and the days before exposure, eg, lag02 represents the three-day average of air pollutant concentrations for lag0, lag1, and lag2.

Sensitivity Analysis

In addition, sensitivity analyses were conducted for the health effects of pollutant exposure to evaluate the stability of the model results by constructing a two-pollutant model in which the pollutant concentration corresponding to the optimal lag (the lag with the largest effect in a single-day lag) was used as the study variable and other pollutants with the same lag were introduced. R4.2.3 software mgcv, gam package was used for the analysis with a test level of 0.05.

Interaction

On the basis of the above studies, the effects of the combined effects of mean temperature, relative humidity and pollutants on the number of deaths from related diseases were analyzed, and three-dimensional spatial maps of the effects of the combined effects of mean temperature, relative humidity and pollutants on the number of outpatient visits for phase allergic rhinitis were fitted to give a visual picture of the effects of the combined effects of the two on the number of outpatient visits by means of the described spatial distribution characteristics.

Results

Description of the Data

Table 1 shows the basic statistics of daily outpatient visits, air pollutants and meteorological factors in Hohhot AR from 2018 to 2020. The daily average concentrations of PM_2.5, PM₁₀, SO₂, NO₂, CO, and O₃ during the study period were 38.1 μg/m³, 83.3 μg/m³, 36.1 μg/m³, 15.1 μg/m³, 14.2 mg/m³, and 99.95 μg/m³, respectively. According to the national “Ambient Air Quality Standard (GB3095-2012)”, the limit of the secondary concentration of the daily average concentrations of PM_2.5 and PM₁₀ in Hohhot during the study period exceeded the national standard values (15 μg/m³, 45 μg/m³), and the average concentrations of SO₂, NO₂, CO, and O₃ did not exceed the national secondary limit values. The average temperature in Hohhot during the three-year period was 7.4 °C, with a daily minimum of −21.36 °C and a daily maximum of 27.99 °C, and the relative humidity during the study period was 46.81%, with a minimum of 9.25% and a maximum of 95%. The AR clinic had a minimum of 0 and a maximum of 173 daily outpatient visits during the three-year period.

Table 1 Characteristics of Daily Number for Allergic Rhinitis (AR) Outpatients, Air Pollutants, and Meteorological Factors

Time series plots of atmospheric pollutants, meteorological factors and daily AR visits showed similar trends for PM_2.5, PM₁₀, SO₂, NO₂ and CO, all showing the highest levels of pollution in the winter and lighter levels in the summer (Figure 2). The pollution cycle of O₃ is significantly different from that of the other atmospheric pollutants, with the highest levels of pollution in the summer and lighter levels in the winter. The number of allergic rhinitis visits in July and August was higher than that of other pollutants. high in July and August, with a clear long-term and seasonal trend.

Figure 2 Time series diagram of atmospheric pollutants, meteorological factors and the number of AR outpatient visits. Notes: Pression: atmospheric pressure, Humidity: Atmospheric relative humidity, Temperature: atmospheric temperature, AR: Number of outpatients with allergic rhinitis.

Relationship Between Air Pollution, Environmental Factors and AR Outpatients

Spearman rank correlation was used to analyze the correlation between air pollution and meteorological factors, and the results are shown in Figure 3. There was a negative correlation between O₃ and other pollutants, and significant positive correlations were found among the remaining pollutants. There were significant negative correlations between temperature and PM_2.5 (r= −0.39, P<0.01), PM₁₀ (r= −0.26, P<0.01), SO₂ (r= −0.55, P<0.01), and CO (r= −0.16, P<0.01), while there was a positive correlation between temperature and O₃ (r=0.86, P<0.01). Relative humidity showed negative correlations with PM₁₀ (r= −0.16, P<0.05), SO₂ (r= −0.17, P<0.01) and O₃ (r= −0.04, P=0.21), and negative correlations with PM_2.5 (r=0.08, P<0.01), NO₂ (r=0.004, P=0.90) and CO (r=0.02, P=0.37) showed positive correlations.

Figure 3 Spearman Correlation Coefficients Between Air Pollutants and Meteorological Variables. Notes: Pression: atmospheric pressure, Humidity: Atmospheric relative humidity, Temperature: atmospheric temperature, AR: Number of outpatients with allergic rhinitis.

Short-Term Effects of Air Pollutants on Allergic Rhinitis

After adjusting for long-term trends in time, temperature, relative humidity, and weekly effects to explore the effects between air pollutants and daily outpatient visits for allergic rhinitis at different lag times are shown in Figure 4. The single-day effects of PM_2.5 and PM₁₀ both reached a maximum at exposure lag1, at which time the RR and 95% CI of the effects of pollutant concentrations on AR outpatient visits for every 10 μg/m³ of increase in pollutant concentration were 1.008 (1.001–1.016) and 1.002 (1.000–1.005), respectively. The effects of SO₂ were maximized at lag5 1.072 (1.033–1.113), NO₂ at lag2 1.020 (1007–1.034), CO at lag6 1.033 (1.014–1.052), and O₃ reaches a maximum of 0.987 (0.980–0.994) at lag4. In the cumulative lag model, the effects of atmospheric pollutants PM_2.5, PM₁₀, SO₂, NO₂, and CO on AR all showed an upward trend with the accumulation of time. The cumulative effects of PM_2.5, PM₁₀, SO₂, and CO all reached a maximum at lag07, when the RR and 95% CI of the effect of every 10 μg/m³ increase in pollutant concentration on the number of outpatient visits for population AR were 1.036 (1.014–1.052), 1.987 (0.9804–0.9936), respectively. were 1.036 (1.02–1.051), 1.011 (1.006–1.017), 1.199 (1.081–1.33), and 1.031 (1.004–1.058), respectively, and the cumulative effect of NO₂ was maximized at lag06 1.055 (1.024–1.087), and the cumulative effect of O₃ changed with time, first increasing and then decreasing. changes, increasing and then decreasing, with a maximum effect of 0.9802 (0.968–0.993) at lag05, and all the cumulative effect of air pollutants is stronger than the single-day effect.

Figure 4 Correlation between air pollutant concentration and daily average outpatient amount of AR in different delay periods. Notes: Pression: atmospheric pressure, Humidity: Atmospheric relative humidity, Temperature: atmospheric temperature, AR: Number of outpatients with allergic rhinitis.

Two-Pollutant Model

To further evaluate the robustness of the model, a dual-pollutant model was developed based on the strongest lag effect of the single-pollutant model. Considering the strong covariance between PM_2.5 and PM₁₀, PM₁₀ was included in PM_2.5, so PM₁₀ was excluded from the dual-pollutant model for PM_2.5, but PM _(10–2.5) was included in the model. Specific results for AR relative risks and 95% CI based on the dual-pollutant model are given in Table 2. When PM _(10–2.5), SO₂, NO₂, CO, and O₃ were included, there was no significant change in the relative risk values for PM_2.5. Similarly, after introducing other pollutants into the single pollutant model for the gaseous pollutants SO₂, NO₂, CO, and O₃, the results show that the change in RR values remains insignificant, suggesting that the model developed is robust.

Table 2 The RR of the Effects of Air Pollutants on Outpatients with Allergic Rhinitis (AR) in Two-Pollutant Model

Interaction of Temperature and Pollutants

From the results of Spearman correlation analysis, temperature was positively correlated with AR and negatively correlated with PM_2.5, PM₁₀, SO₂ and NO₂. Therefore, we used the concentration of atmospheric pollutants with the strongest effect to study the interaction between atmospheric pollutants and temperature. Figure 5 shows the effect of the interaction of atmospheric pollutants and air temperature on the clinic visits of patients with allergic rhinitis. The results show that PM_2.5 and PM₁₀ peaks are similar, with peaks occurring at high-temperature-high pollutant concentration levels and low-temperature-high pollutant concentration levels. The interactions of SO₂, CO, and air temperature on AR outpatient visits are unimodal, with all peaking at high-temperature-high pollutant levels. The peaks for NO₂ occur at low-temperature-high pollutant levels. O₃ showed high portal volumes except for high temperature-low pollutant concentration levels.

Figure 5 The combined effect of air pollutants and temperature on the number of outpatients with AR. Notes: Pression: atmospheric pressure, Humidity: Atmospheric relative humidity, Temperature: atmospheric temperature, AR: Number of outpatients with allergic rhinitis.

Interaction Between Relative Humidity and Pollutants

Correlation analysis showed that outpatient visits for allergic rhinitis were positively correlated with relative humidity and negatively correlated with PM₁₀, SO₂, and NO₂. Figure 6 shows the effect of the interaction of atmospheric pollutants and relative humidity on outpatient visits for AR patients. At low relative humidity and high pollutant concentrations, outpatient visits were highest for PM_2.5 and CO. PM₁₀ and SO₂ showed higher AR visits at high humidity-high pollutant concentrations. the highest peaks for O₃ occurred at low pollutant concentrations-low humidity. and NO₂ showed relatively stable outpatient visits, except for low values at low relative humidity and high pollutant concentrations.

Figure 6 The combined effect of air pollutants and relative humidity on the number of outpatients with AR. Notes: Pression: atmospheric pressure, Humidity: Atmospheric relative humidity, Temperature: atmospheric temperature, AR: Number of outpatients with allergic rhinitis.

Discussion

Hohhot is a typical northern city, belonging to a semi-arid region, with a mid-temperate continental monsoon climate, typical of the Mongolian Plateau continental climate, with obvious climatic changes in the four seasons, and a large temperature difference between the year and the day.¹⁴ Meteorological factors and pollutants show cyclical fluctuations. O₃, RH, and temperature are higher in the summer and lower in the winter, and air pollutants (PM_2.5, PM₁₀, SO₂, NO₂, and CO) are characterized by “higher in the winter” as a result of winter heating by the residents of Hohhot, PM₁₀, SO₂, NO₂ and CO) are characterized by “higher in winter and lower in summer”.

Effects of Air Pollutants on Daily AR Visits

In recent years, the prevalence of AR has increased rapidly, especially in the western and northern regions.^10,15 According to domestic reports, the total prevalence rate of AR in the western China was as high as 34.31% in 2005.¹⁶ In 2009, it was found that the total average prevalence rate of AR in the northern China was 9.3%.¹⁷ Meta-analysis on the relationship between allergic rhinitis and environmental air pollutants in China showed that PM_2.5, PM₁₀, NO₂ and SO₂ all increased the risk of allergic rhinitis.¹⁸ We carried out a study on the influence of ambient air pollutants on the number of allergic rhinitis outpatients in Hohhot, it also found that elevated short-term exposure levels of PM_2.5, PM₁₀, SO₂, NO₂, and CO had a significant effect on the increase in the number of AR outpatient visits in Hohhot. A domestic study¹⁹ showed that each 10 μg/m³ increase in air pollutant concentration (1 mg/m³ for CO) caused an increase in AR clinic visits in Xinxiang of 1.036 (1.020–1.051), 1.011 (1.006–1.017), 1.199 (1.081–1.33), 1.055 (1.024–1.087), and 1.033 (1.014–1.052). Studies in Beijing and Chongqing have obtained similar results to ours.^20,21 Studies in Europe and Korea have similarly shown that PM_2.5, PM₁₀, SO₂, NO₂, and CO exposures increase the incidence of allergic rhinitis.^22,23 However, Hajat et al found that the correlation between PM_2.5, PM₁₀, and AR was not significant in London.²⁴ This may be due to the differences in risk factors for exposures between the cities, including environmental factors (geographic location, climate type, and air quality), populations (children, adults, and elderly), and allergen exposure (timing, dose, and quantity). In-depth studies should be conducted to clarify the relationship between the specific chemical composition of particulate matter (eg, ions, dust, black carbon, heavy metals, volatile organic compounds) and the source of AR. The effect of O₃ on AR is still controversial. The results of this study show that short-term exposure to O₃ may be related to the reduction of outpatient visits to allergic rhinitis, which is consistent with the results of the studies conducted in London,²⁴ and in Xinxiang¹⁹ in China. This may be related to the lower O₃ concentration in Hohhot, where Bocci et al reported that the effect of ozone on specific diseases is closely related to the exposure measure, and that low-dose exposure to O₃ activates the antioxidant system of the blood.²⁵ On the other hand, it may also be related to the lifestyles of the residents, where crowds of people go out in hot and highly radiated weather to reduce their exposure to O₃.

Interaction of Air Pollutants with Temperature and Humidity

Air pollution and air temperature are not isolated in the environment. In this study, it was found that the increase in PM₁₀, NO₂, and O₃ concentrations under low temperatures led to varying increases in outpatient visits for AR. Relevant national studies have pointed out that physiological defenses of the upper respiratory tract decrease under low temperatures. Viral infections inhibit the activity of ciliated cells and phagocytosis of alveolar macrophages, which in turn causes the invasion of bacteria, including Streptococcus pneumoniae and Klebsiella pneumoniae.²⁶ Meanwhile, persistent low temperatures may cause the combustion of petroleum products to release more toxic byproducts, such as NO_2.²⁷ These factors contribute to the frequency of respiratory illnesses under low-temperature conditions. In this study, it was also found that PM_2.5, SO₂, and CO lead to the highest number of AR clinic visits at high temperatures, and Ying Zhang et al found that increased PM_2.5 concentrations at high temperatures increased the risk of respiratory illnesses.²⁸ This may be due to the fact that sustained high temperatures promote photochemical reactions in the atmosphere, which leads to an increased production of atmospheric pollutants, such as PM_2.5. Temperatures may indirectly influence respiratory disease predisposing factors such as viral and bacterial activity and outdoor survival time, leading to an increase in allergic rhinitis sufferers.²⁹

Numerous studies have shown a significant correlation between humidity and respiratory diseases.^30,31 The present study demonstrated that the effects of PM_2.5, CO, and O₃ were more pronounced at low humidity, and that droplet and direct contact transmission were the main modes of transmission of respiratory diseases, and that a low-temperature, dry environment provided suitable conditions for droplet-borne diseases. Studies have shown that exposure to low relative humidity leads to dehydration, decreased elasticity, and weakened cilia motility at all levels of the tracheal and bronchial mucosa, making it difficult to remove dust, bacteria, and other adherents from the airway surface. Dust and bacteria lead to the release of various inflammatory factors, increasing airway reactivity and leading to an increased risk of respiratory disease. And PM₁₀, NO₂, and SO₂ have more significant effects on allergic rhinitis at high humidity, which is consistent with the findings of Wang et al.³² The effect of relative humidity on the variation of air pollution needs to be further investigated.

Our results will help to understand the effects of air pollution and meteorological factors on AR. In addition, as the average concentration of atmospheric pollutants and the incidence of allergies continue to increase globally, our findings will provide a basis for future research on the effects of increasing concentrations of atmospheric pollutants on AR and may also help to develop relevant health protection strategies. However, there are some limitations in our research. First of all, ecological bias is inevitable Research shows that keeping pets and home decoration are risk factors for allergic rhinitis,³³ and regular indoor ventilation is related to the alleviation of rhinitis.³⁴ In this study, the monitoring data of outdoor fixed locations are used as the proxy data for personal exposure to pollutants. Because individuals spend most of their time indoors, it may lead to measurement errors.³⁵ Although the magnitude of the error is difficult to quantify, it usually results in a decrease in the risk estimate.³⁶ This study collected data from only one hospital, and therefore the generalizability of the findings may be limited.

Conclusion

In summary, short-term exposure to PM_2.5, PM₁₀, SO₂, NO₂, and CO in Hohhot had a significant effect on the increase in the number of allergy outpatient visits, and O₃ exposure had a significant effect on the decrease in the number of AR outpatient visits. The public should pay more attention to air pollution, and it is recommended that governmental departments rationally allocate health resources and carry out relevant health education to reduce exposure to air pollutants and decrease their health hazards to residents.