Fars province is located in the southwest of Iran, between longitude 27°31′N and 31° 42′N and latitude 50° 37′E and 55° 38′E. It covers an area of about 122,608 km² with a total population of 4.6 million and male to female ratio of 1: 1.03. It comprises 24 counties and 8,191 residential places including small and large villages, with Shiraz as the capital city (Figure 1).

Fars province has three distinct climatic regions with an average temperature of 16.8°C, ranging between 4.7°C and 29.2°C. Its mountainous area has moderately cold winters and mild summers in the north and northwest while the central regions have been identified with relatively rainy mild winters and hot dry summers. The third region located in the south and southeast has cold winters and hot summers. Due to the geo-climatic variation of the province, the region has different land covers from dense forests to bare plains and elevations of 115–3,115 m above sea level (Ghatee et al., 2020b).

This retrospective study collected the medical data of 349 patients hospitalized with acute asthma at the Nemazee hospital of Shiraz, Iran, during 2016–2019. Nemazee is the leading respiratory hospital in Fars province admitting people from all parts of the province with varied socioeconomic statuses. Patients whose spatial information was incomplete or related to provinces other than Fars were excluded from the study. Some patients were contacted randomly to ensure the spatial information of patients. This study was approved by the Ethics Committee of Shiraz University of Medical Sciences (IR.SUMS.MED.REC.1400.297).

The point residence of all patients was extracted from their medical records and recorded on the map based on the latitudes and longitudes of each point. The land cover vector layers and Digital Elevation Model (DEM) raster layer were acquired from the Department of Natural Resources in Fars province. The slope raster map was drawn through the spatial analyst tool, based on DEM, by computing the maximal rate of variation in value between each cell and its adjacent cells. The land cover layer showed spatial data on the different physical characteristics of the surface of the province.

The meteorological and rainfall data including temperature, maximum and minimum temperature, humidity, evaporation, number of rainy and frosty days, sunny hours, and rainfall from 18 synoptic meteorological and 86 rain-gauge stations were acquired from the Fars Province Weather Bureau. The mean values of each data related to the period of study were calculated. The generated variables were including Mean Annual Temperature (MAT), Minimum Mean Annual Temperature (MinMAT), Maximum Mean Annual Temperature (MaxMAT), Mean Annual Evaporation (MAE), Mean Annual Humidity (MAH), Mean Annual Rainfall (MAR), Mean Annual Frosty Days (MAFD), Mean Annual Rainy Days (MARD), and Mean Annual Sunny Hours (MASH). The annual iso-hydral, iso-humid, and frost days’ raster layers were generated using the Kriging interpolation method, and iso-thermal, iso-evaporation, and rainy days’ raster layers using the tension-based Spline interpolation model with a resolution grid of 2 × 2 km.

Geospatial and climatic data were analyzed using Arc GIS version 10.1. The provincial villages and cities’ point shape file layer was extracted with the raster layers. The identity tool was used to compute the geometric intersection of the layer obtained from the extraction of all raster layers with NTR hazard (polyline) and land cover (polygonal) vector layers to develop the final layer in which each point represented the properties of all the overlapped identity features from the above-mentioned raster and vector layers. The attribute of this layer was converted to an excel format for statistical analysis.

Statistical analysis was based on the presence or absence of patient in each point. The effect of geo-climatic factors on acute asthma was assessed after the spatial description of patients in Fars province. In this regard, residential points data including asthma reported and non-reported villages and cities were extracted from final province villages/cities point layer and comparison of environmental and climatic factors was done between asthma reported and non-reported points using univariate and multivariate logistic regression models. The statistical analyses were performed using SPSS version 21.

Adult cases with acute asthma were recruited from different areas of Fars province with varied climatic and geographical features. The climatic factors including MAT, MinMAT, MaxMAT, MAE, MAH, and MAR of 3 years ranged from 13.53°C to 27.38°C, 6.65°C–18.87°C, 20°C–31.79°C, 1659.5–3,308.2°mm, 31.66%–41.93%, and 133.57–607.58°mm, respectively, among the points with acute asthma in adults (Figures 2, 3). MARD, MAFD, and MASH were 16.77–60.04 days, 0.19–87.01 days, and 2482.51–3,602.52 h for these points, respectively (Figure 4). The range of DEM and slope varied from 349 to 2233 m and 0–28.37°, respectively (Figure 5).

Multivariate binary logistic regression analysis was performed to reveal the concomitant effects of significant variables detected by univariate analysis. Urban setting (p = 0.000, OR = 11.026, CI = 3.808–31.928) was the only significant factor detected by multivariate analysis while other variables lost their significance (Table 3).

MARD was a significant factor related to acute asthma in adults in the current study. The rainy days create humid and wet conditions resulting in homes being damp. Such conditions favor the growth of allergens like dust mites, mold, and fungi. Evidence supports that exposure to such indoor allergens is causally related to asthma incidence (Ghosh et al., 2018; Campo et al., 2019). Home dampness and molds are essential elements influenced by humidity levels and are associated with asthma symptoms (Williamson et al., 1997; Cox-Ganser, 2015). In addition, on rainy days, people prefer to stay at home, increasing exposure to these allergens. Similar to our study, Akpinar-Elci. et al. showed an increase in asthma visits during rainy months, while an inverse relationship was found by Wen-Chao Ho et al. (Ho et al., 2007; Akpinar-Elci et al., 2015). They explained that rainy days might interact with particulate matter (PM10), creating a washout effect (Ho et al., 2007; Akpinar-Elci et al., 2015).

A negative association between high temperature and asthma hospitalization was detected in our study. These results are consistent with some research in the field of asthma. A study by Son JY et al. in eight major cities in Korea found a significant association between asthma hospitalization and low temperature (Son et al., 2014). Similar results for childhood asthma were reported by Guo et al. in Shanghai (Guo et al., 2012). An increase in respiratory bacterial and viral infections associated with infiltration of inflammatory factors and mucus secretion has often followed a cold temperature, which may explain the effect of low temperature on asthma exacerbation (Li et al., 2011). However, some studies have demonstrated increased asthma hospitalization in high temperatures (Lin et al., 2009; Yamazaki et al., 2013). These different patterns of association between temperature and asthma geographically may be related to the effects of temperature on some asthma risk factors that differ in prevalence and seasonality by region (Buckley and Richardson, 2012). The impact of temperature on asthma exacerbation may also be related to differences in the levels of exposure, public health interventions, health and social care services, and susceptibility of subpopulations (Michelozzi et al., 2006; Anderson and Bell, 2011).

Evaporation was another significant factor negatively affecting asthma occurrence in this study. To our knowledge, this was the first evaluation of this factor in the literature. The rate of water evaporation depends on climatic factors such as temperature, solar radiation, wind, and relative humidity. Sunny, hot, dry, and windy conditions produce higher evaporation rates. This is consistent with our data that showed low acute asthma incidence in warm and low humidity conditions associated with increased evaporation. However, we failed to find a statistically significant value for humidity.

The urban setting was the most critical factor affecting asthma in this study. Previous asthma studies reported that urban lifestyle could affect individual health through industrial emissions, occupation, and daily use of products (Khan et al., 2010). Global urbanization is associated with increased environmental pollution caused by the combustion of fossil fuels, biomass, emissions from agriculture, and windborne mineral dust and organic matter. Higher PM2.5 and PM10 levels are associated with increased use of asthma medication, GP consultations, and hospitalization by activating cytokine pathways and inflammation in the airway (Nel et al., 1998; Nel, 2005; Brandt et al., 2015; Clifford et al., 2017). In addition to air pollution, the socioeconomic difference between urban and rural environments could affect the quality and access to healthcare and, finally asthma hospitalization. Urban locations generally tend to have a prototype environment that predisposes individuals to asthma, but it can be decreased if socioeconomic issues are addressed. Similar to our results, Solé et al. (2007) reported higher asthma prevalence among adolescents living in urban areas compared to the rural ones in Brazil. We previously found increased asthma hospitalization occurrences in urban regions of Fars province compared with rural areas (Kanannejad et al., 2022). The farming environment of rural areas is one explanation for lower asthma prevalence in such places (Strachan, 1989). Consumption of unpasteurized cow milk, exposure to farm animals, and fodder in rural areas may have a protective effect on asthma (Pechlivanis and von Mutius, 2020). Increased exposure to farm animals results in higher microbial exposures through which the protective effect might be mediated. Also, the consumption of unpasteurized milk can modulate cytokine production patterns toward the Th-1 (T helper-1) response, which could be responsible for the observed protective effect. One study reported that increased endotoxin exposure at birth was associated with a lower risk of allergic sensitization and eczema (Simpson et al., 2006). On the other hand, some studies have shown higher asthma occurrences in rural areas due to low-quality healthcare services, lower socioeconomic statuses, and cultural barriers to access to medicine compared with urban ones (Pesek et al., 2010; Roy et al., 2010).

The slope was another significant environmental factor affecting acute asthma occurrence in Fars province, where more asthma hospitalization was detected in areas with a lower pitch. Our previous study on childhood populations showed the powerful effect of slope on asthma hospitalization (Kanannejad et al., 2022). It may be related to higher population density in regions with a downward slope that usually is connected with increased air pollutants, nanoparticles, and transmissible infectious viral agents (Bröms et al., 2009). This study is the first of its kind which reflects slope effect on adult asthma, worldwide.

There are some limitations in this study. Poor adherence to prescribed medication is a cause of asthma exacerbation, which was not considered in this study. Therefore, the applied models do not predict factors responsible for individual-level hospitalization. In addition, daily mean concentrations of air pollutants as another leading cause of asthma attacks should be considered in future studies.