1. Introduction
Urticaria is a common health outcome. Around 15–20% of the population in the world suffers from urticaria [
1,
2]. Urticaria presents with a sudden appearance of erythematous patches with pruritus [
3]. The counts and sizes of patches (wheals) vary [
4]. Individual wheal disappear within 24 hours with no treatment [
2]. However, new wheals appear on a different part of the body in the case of recurrent episodes of urticaria. Acute spontaneous urticaria episodes lasts for up to six weeks [
4]. Acute spontaneous urticaria results from hypersensitivity towards the allergens resulting in the formation of Immunoglobulin E (IgE) antibodies that bind to receptors found on mast cells and basophiles. Upon re-exposure, the allergen is recognized by cell-bound specific IgE, leading to receptor crosslinking. This event leads to the release of many immune mediating enzymes, primarily histamine which causes increased capillary permeability that is presented in the form of swelling on the upper dermal layer of the skin [
3]. Common triggers for acute urticaria include viral infections of the upper respiratory track, allergens and psudoallergans, and drugs such as penicillin [
1]. Physical urticaria results from a response to physical stimuli. Physical urticaria is further classified into cold, heat, and dermographic and other physical stimuli-specific urticaria. Unlike other physical urticaria where the response to the stimuli is triggered by mast cells deregulation, cold urticaria response can also be triggered by the history of infections, autoimmune diseases and neoplasia [
1]. Heat urticaria is rare and is elicited by warm objects or air. Dermographic urticaria results in a swelling of skin along a line that is formed prior to appearance of the swelling (shearing of skin) [
5]. Among all physical urticaria subtypes, dermographic form is the most common and is seen mainly among children. Some special forms of urticaria such as cholinergic and contact urticaria are worth consideration, Cholinergic urticaria results from the increase in the body heat usually from the activity such as exercise (exercise induced urticaria). In contact urticaria, the wheals appear at the site of contact of triggers such as chemicals, cosmetics, food products, plants and drugs [
1].
Studies focusing on urticaria are limited [
6]. Women have a higher risk of the onset of urticaria than that among men [
7,
8,
9]. Urticaria is more common in younger adult age groups compared to older age groups [
3] However, the effect of a well-known determinant of health, socioeconomic status and factors such as access to care, occupation, and built environment on the likelihood of the occurrence of acute urticaria in particular, is not clear.
Knowledge of specific triggers or causes for the presentation of urticaria is much more developed. Potential triggers include environmental allergens such as pollen, spores originating from mold, dust mites, animal hair [
10], food products such as nuts, eggs, fish, seafood, mushrooms, and peas, among others [
3], infections particularly among children [
11], and physical factors such as cold and hot substance, light, pressure, and water [
12].
San Joaquin Valley, the southern part of the Central Valley of California, USA has one of the worst air quality outcomes in the United States [
13]. Agricultural infrastructure such as irrigation channels [
14], and animal feed and mobile sources contribute to the ozone-induced pollution in the Central Valley [
15], in addition to other sources. Studies have shown positive association between air pollution caused by ozone and hospital visits for the treatment of angioedema [
16], and the effect of air pollution caused by particulate matter (PM 2.5), ozone and nitrogen oxides on emergency department (ED) visits for the treatment of urticaria [
17]. Patients seeking immediate care can utilize ED services without pre-scheduling an appointment with a physician in the United States. Particulates of the smallest aerodynamic sizes (PM 2.5 and lesser) can have detrimental effects on the lungs and heart. The effect of air pollution and the other health outcomes triggered by hypersensitivity to allergens such as asthma particularly in the Central Valley have been addressed [
18,
19,
20]. The prevalence of acute urticaria was more than 50% among those who had allergic asthma or atopic dermatitis in one study [
21]. Efforts to understand acute urticaria health outcomes in the Central Valley are limited although the need for such efforts is warranted considering the air pollution and the possibility of a high threshold of environmental allergens originating from agriculture.
The purpose of this study was to identify environmental and sociodemographic risk factors contributing to the visits for the treatment of acute urticaria at emergency departments (ED) of the medical facilities located in the southern Central Valley counties. The study was approved by the institution’s review board for the protection of human subjects.
2. Methods
This study was a retrospective cross-sectional analysis of emergency department visits for acute urticaria. Health data were collected from the Office of Statewide Planning and Development (OSHPD) in California [
22]. To maintain a license, hospitals are required to submit emergency department records to OSHPD. De-identified data are made available to the public and researchers within two years of the emergency department visit. Each emergency department record included information on the patient’s gender, age, race/ethnicity, expected source of payment, primary and secondary diagnoses, county, and zip code of residence. For this study, 2016 and 2017 data on emergency department visits were used of individuals residing within the eight San Joaquin Valley (southern area of the Central Valley) counties: Fresno; Kern; Kings; Madera; Merced; San Joaquin; Stanislaus; and Tulare. Zip code-level measures were collected from the US Census (using American Community Survey of 2015–2019) and the California Air Resources Board.
2.1. Outcome Assessment
The incidence rate of urticaria-related emergency department visits was the primary outcome of interest. According to the American Academy of Allergy Asthma and Immunology [
23], the diagnosis of urticaria is coded as following:
ICD-10 Code | Diagnosis |
---|
L50.0 | Allergic Urticaria |
L50.1 | Idiopathic Urticaria |
L50.2 | Cold and Heat Urticaria |
L50.3 | Dermatographic Urticaria |
L50.4 | Vibratory Urticaria |
L50.5 | Cholinergic Urticaria |
L50.6 | Contact Urticaria |
L50.8 | Chronic or Recurrent Urticaria |
L50.9 | Unspecified Urticaria |
In this study, these codes except L50.4 and L50.8 were used to extract urticaria-related ED events from OSHPD administered data. Allergic urticaria includes acute allergic reaction resulting in urticaria, allergic medicamentosa, and urticaria resulting from the use of a particular food or drug [
24]. Idiopathic urticaria is a diagnosis when urticaria erupts and the cause is unknown. Cold and heat urticaria are a result of cold or heat stimuli (physical urticaria), dermatographic urticaria includes autographism, dermatographia, and factitial urticaria. Cholinergic urticaria results from excess body heat, Contact urticaria results from contact with plants or other triggers. Unspecified urticaria is defined as urticaria resulting from exposure to food, drugs, infections, stress and insect bites [
24]. The aforementioned urticaria diagnosis codes exclude allergic contact dermatitis, antineurotic edema, giant urticaria, hereditary angioedema, Quincke’s edema, serum urticaria, solar urticaria, urticaria neonatorum, urticaria papulosa, and urticaria pigmentosa [
25].
In multivariate analysis (Poisson regression), the count of urticaria-related events per zip code was the primary outcome assessed. Zip codes vary in population size; therefore, the population at risk for contracting urticaria varies across zip codes. Population count estimates were used to adjust for the population at risk (offset) within each zip code. This method adjusts the model and allows for urticaria events to be treated as the numerator and the population at risk as the denominator so that regression coefficients can be interpreted as rate ratios. We used patient’s zip code of residence to merge hospital data with the US Census data.
2.2. Assessment of Covariates
We utilized additional data sources to identify the effect of covariates on ED visits for the management of acute urticaria. The percentage of agricultural workers and rurality of a zip code were the two primary covariates of interest. To estimate the percentage of agricultural workers in each zip code, we used 2011–2015 US Census estimates of agriculture, forestry, fishing, and hunting using the American Community Survey. In the 2012–2016 US Census data, population density is estimated by dividing the total population per zip code. Because the average value of population density per zip code was much higher (mean ≥ 19,000 residents per zip code), compared to the average values of other covariates (diesel particulates highest value less than 100 µg/m3), we used a natural log of the values of population density. High unevenness in the values of variables can affect the model fit and estimations. The percentage of agricultural workers and population density were treated continuously in analysis.
For other covariates, additional measures were retrieved from the CalEnvironScreen v1.0 tool, Office of Environmental Health Hazard Assessment, Sacramento, USA which aggregated data to the zip code level from various data sources including the US Census and the California Air Resources Board, among others [
26]. Administered by the California Office of Health Hazard Assessment, CalEnviroScreen v1.0 is an environmental and social justice screening tool used in California to allocate cap-in-trade funds to the most disadvantaged communities. A variety of socioeconomic and environmental pollutant measures are used within the tool. The measures used in our study included the percentage of the population living below two times the federal poverty level (percentage of poverty), the percentage of the population that is non-white or Hispanic/Latino (percentage of non-white), the percentage of the population younger than 10 years of age and older than 65, the percentage of the population over the age of 16 that is unemployed and eligible for the labor force (percentage of unemployed), the percentage of the population older than 25 years of age with less than a high school education (percentage of less than high school education), and diesel exhaust particulate matter (Diesel PM). The diesel PM was measured as the spatial distribution of gridded diesel PM from on road and non-road sources for a 2012 summer day in July (kg/day). The California Air Resources Board [
27] administers diesel PM emissions data and shares it with the CalEnvironScreen system. We found that percentage of poverty and percentage of non-white had a strong positive multicolinearity effect (Pearson’s correlation value 0.744,
p-value < 0.05). Therefore, we combined the two variables and used the dichotomous composite in the final multivariate model. This process is referred to as forming a composite to avoid the multicolinearity effect [
28]. Multicollinearity can also affect the model estimations. All other variables were treated continuously in the analyses unless stated otherwise and cases with missing data were eliminated from the analysis.
2.3. Statistical Analysis
To fit the discrete nature of the outcome variable, Poisson-based regression was used. Pearson’s
r and unadjusted Poisson models were conducted to test bivariate associations between each of the predictor variables considered in the analysis and the outcome variable. If the predictor variable was not previously explored in the literature or was not significant, it was not included in the final adjusted model. White’s test of heteroscedasticity demonstrated that an ordinary least squares model was a poor fit for these data (
p value of <0.001) because of a violation of the assumption of homogeneity of error variance. Therefore, ordinary least squares regression was not used in the final model although it was used in a fully adjusted model to investigate collinearity across predictor variables. Statistical analysis was conducted in the R data analysis system [
29].
3. Results
There were 14,417 acute urticaria-related emergency department visits from 2016 to 2017 in the San Joaquin Valley composing 0.4% of all ED visits (
n = 3,237,113). As reflected in
Table 1, unspecific urticaria-related ED events had highest attribution towards the distribution of urticaria codes with 73% of all urticaria diagnoses, followed by allergic urticaria-related ED events with 26% of the total. All other urticaria-related events were rare (less than 1%). These results highlight that the dependent variable mainly contained acute spontaneous urticaria. Less than 1% (
n = 99) of missing data were removed and we successfully merged 14,417 cases with zip codes, and these were used for the analysis.
As shown in
Table 2, a greater percentage of women (56.2%) utilized acute urticaria-related ED services compared to those utilized by men (43.8%). Children younger than five years of age (24.3%) and children aged five to nine (15.1%) had the highest percentage of ED utilization compared to that utilized by any other age group. The mean age of the population was 20.8 years with a standard deviation of 19.3 years. Among racial/ethnic groups, Hispanic/Latino was the most common group for the utilization of ER for acute urticaria treatment followed by white (23.6%) and black/African Americans (5.3%) people. For insurance use type, Medicaid covered the largest percentage of ED visits (69.7%) followed by private insurance (19.5%).
Table 3 illustrates the acute urticaria-related ED visits per 1000 in the population by demographic characteristic. The incidence rate was computed as the total of the number of cases divided by the number of individuals at risk (
n = estimated 9,226,932), multiplied by 1000. The number of people at risk being the entire population of the selected counties. Overall, the San Joaquin Valley had an incidence rate of 1.56 per 1000 persons from 2016–2017. Women had a higher rate (1.9 per 1000 persons) of acute urticaria-related ED visits compared to those among men (1.5 per 1000 persons). Children less than five years of age had the highest rate of ED visits (5.5 per 1000 persons) followed by those among youth with age less than nineteen years (3.4 per 1000 persons). Black/African American (2.0 per 1000 persons) and Hispanic/Latino (2.2 per 1000 persons) residents had nearly twice the rate of ED visits in comparison to the rate of ED visits among white residents (1.2 per 1000 persons).
Table 4 illustrates descriptive statistics for the measures included in analyses as well as Pearson’s r for each measure with the outcome of interest. There were 201 zip codes included in the analysis. All measures were treated continuously. In the analysis of bivariate correlations with the rate of urticaria-related ED visits, we found that the percentage of agricultural workers and population density were positively correlated with the rate of urticaria ED visits (Pearson’s r = 0.234,
p < 0.05 and Pearson’s r = 0.249,
p < 0.05, respectively). Poverty and non-white concentrations were positively associated with the rate of urticaria ED visits (Pearson’s r = 0.413,
p < 0.001 and Pearson’s r = 0.393,
p < 0.001, respectively). Diesel exhaust PM was positively associated with urticaria ED visits (Pearson’s r = 0.166,
p < 0.01).
In
Table 5, the results of a Poisson-based multivariate analysis of the rate of urticaria-related ED visits are shown. Adjusting for covariates in the model, we found that population density was positively associated with ED visits (RR = 2.817, CI = 2.752, 2.883,
p < 0.05) where one percent increase in population density resulted in 2.81 times the risk of acquiring acute urticaria We found that the percentage of agricultural workers was positively associated with ED visits (RR = 1.490, CI = 1.284, 1.728,
p < 0.05) where a one percent increase in the percentage of agricultural workers was associated with a 1.49 times increase in the rate of urticaria ED visits. Diesel exhaust PM was significant and positively associated with urticaria-related ED visits (RR = 1.007; CI = 1.005, 1.008,
p < 0.05) where a one unit increase in diesel PM was associated with a 6% increase in urticaria-related ED visits. The composite of the percentage of those living in low poverty areas and areas with a high percentage of the white population was negatively associated with urticaria-related ED visits (RR = 0.628, CI = 0.600, 0.657,
p < 0.05). In other words, zip codes (
n = 61) with low levels of poverty and low levels of non-white individuals have about a 37% lower risk of acute urticaria-related ED visits compared to all other zip codes included in the analysis (
n = 140).
As shown in
Figure 1, acute urticaria rates were higher in zip codes with higher levels of poverty.
As shown in
Figure 2, acute urticaria rates were higher in zip codes with higher percentage of non-whites.
5. Conclusions
In this population-based study based in the Central Valley of California, we found that high and medium percentage of poverty and high and medium percentage of non-white population, greater percentage of agricultural workers, young age (less than 18), and higher concentration of diesel exhaust were associated with a high risk of acute urticaria. In particular, the magnitude of the risk of acute urticaria was highest among zip codes with high population density, high percentage of agricultural workers and those containing high and medium percentage of individuals living in poverty and had high and medium percentage of racial minorities. Furthermore, it is possible that most agricultural workers lived in zip codes with high and medium percentage of poverty and the greater to medium percentage of racial minorities. With the unique situation of farming practices (production of diverse commodities) and workforce (substantial amount of migrant and minority workers) in California compared to the other regions in the United States, it is important to understand the needs of farm workers as they may have these environmental and social challenges that could put them at risk for acute urticaria. Further studies should explore more about agricultural worker living conditions, access to care and economic situation, and should also emphasize designing intervention to reduce acute urticaria outcomes targeted for other communities such as those living in urban and suburban neighborhoods with high social and economic inequalities.