Dust Storms Are Associated with an Increase in Outpatient Visits for Rheumatoid Arthritis

Abstract

Our study is the first to illuminate a previously underexplored dimension in the intricate interplay between environmental factors, specifically dust storms (DSs), and rheumatoid arthritis (RA) outcomes. An Auto Regressive Moving Average (ARMA) time-series estimation model was employed to analyze rheumatoid arthritis outpatient visits, dust storms, air pollution, and meteorology data in Taiwan from 2006 to 2012. The results show that females are three times more likely to experience RA-related issues and seek medical attention. Percentage analysis revealed a 10–15% increase in daily RA outpatient visits on post-event days 1 and 2 compared to non-DS days. However, the time-series estimation indicated a delayed and statistically significant (p < 0.05) increase in RA outpatient visits on post-event day 1 for males but not females. Additionally, a significant increase in RA outpatient visits (p < 0.05) was observed on post-event day 1 among individuals aged 61 and above. Environmental factors such as temperature and SO₂ showed strong significance across all genders and age groups (p < 0.001). The findings highlight distinct gender and age disparities regarding the impacts of DS on RA outpatient visits, emphasizing the heightened sensitivity of males to environmental pollutants and the vulnerability of the elderly population.

1. Introduction

Rheumatoid arthritis (RA) is a chronic autoimmune disorder in which the body’s immune system mistakenly attacks its own healthy tissues, leading to inflammation in the joints and surrounding tissues [1,2]. This condition, which is more prevalent in women than men, typically manifests in middle adulthood [3,4,5,6]. While the precise origins of RA remain elusive, current understanding attributes its onset to a combination of genetic and environmental factors [6,7,8,9].

Environmental contributors, such as cigarette smoke and ambient air pollution, have been significantly associated with RA [8,9,10,11]. Notably, smoking and certain inhaled pro-inflammatory agents have been implicated in promoting protein citrullination in the lungs, mediated by protein arginine deiminases (PADs) [12,13]. This enzymatic process converts positively charged protein arginine residues to neutral citrulline, a process known as citrullination. In individuals with a genetic predisposition, citrullination can trigger the production of anti-citrullinated peptide autoantibodies (ACPAs), which are strongly linked to the development of RA [9,14]. A specific study found that exposure to fine particulate matter (PM2.5) was independently associated with higher ACPA concentrations in elderly individuals with RA [14].

Ambient air pollution, which originates from sources such as transportation, industrial activities, agriculture, wildfires, and dust storms (DSs), plays a crucial role in this context. In Taiwan, during the winter and spring seasons, DSs originating in the deserts of Mongolia and China are carried by the wind to densely populated cities [15,16]. These DSs transport various particulates, particularly a higher proportion of coarse particles (PM10), along with a diverse array of elements, including natural crustal elements (Ca²⁺, Mg²⁺), microorganisms, and anthropogenic pollutants (Pb, As, NOx) [17,18].

In contrast to conventional urban air pollution, DSs are short-lived events, usually lasting only a few days. While past studies have established correlations between DSs and an increased risk of hospitalization due to cardiorespiratory diseases [19,20,21,22], the impact of DSs on RA has not been previously explored. Consequently, the objective of this study is to assess the relationships between DS events and RA. Utilizing a population-based national dataset, we analyze the association of DS events with the daily number of RA outpatient visits in Taiwan.

2. Materials and Methods

2.1. Data

The study incorporates data from four key areas: RA outpatient care, dust storms, air pollution, and meteorology. RA outpatient care data, covering the period from January 2006 to December 2012, were extracted from the National Health Insurance (NHI) research database, specifically focusing on cases with a primary diagnosis of RA (ICD-9-CM code 714).

Dust storm data were obtained from the air quality monitoring databases provided by the Taiwan Environmental Protection Agency (TEPA), collected from 55 air quality monitoring stations across Taiwan. TEPA defines a dust storm event as occurring when the concentration of total atmospheric PM exceeds 100 µg/m³, compared to the long-term average of 50 µg/m³.

Air pollutant data include the average daily levels of SO₂ and O₃, measured by TEPA at the same 55 air quality monitoring stations. Meteorological data on daily mean ambient temperature were collected from 23 weather observatories managed by the Taiwan Central Weather Bureau, Ministry of Transportation and Communication.

2.2. Model

Our primary objective is to explore the relationship between RA outpatient care and the occurrence of dust storms. An Auto Regressive Moving Average (ARMA) time-series estimation model was employed using data from the NHI research database, covering the period from 2006 to 2012. The model is expressed as follows:

$${\mathrm{RA}}_t^{ } ={\mathsf{\alpha }}_{ }^{ }+{\mathsf{\beta }}_{ }^{ }{\mathrm{Trend}}_{ }^{ }+{{\displaystyle \sum }}_{\mathrm{j}=0}^4 {\mathsf{\Omega }}_{\mathrm{j}}^{ } +{{\displaystyle \sum }}_{\mathrm{j}=1}^{\mathrm{p}}{ \mathsf{\delta }}_{\mathrm{j}}^{ }{\mathrm{RA}}_{\mathrm{t}-\mathrm{j}}^{ }+ {\mathsf{\gamma }}_{ }^{ }{\mathrm{X}}_{\mathrm{t}}^{ }+{\mathsf{\epsilon }}_{\mathrm{t}}^{ }$$

where

• ${\mathrm{RA}}_{\mathrm{t}}^{ }$ is the daily number of RA outpatient visits.
• Trend is the time trend variable.
• ${\mathsf{\Omega }}_{\mathrm{j}}$ is a dummy variable for the jth day after a dust storm.
• ${\mathrm{RA}}_{\mathrm{t}-\mathrm{j}}$ denotes the number of daily RA outpatient visits on the jth day before day t.
• X_t represents controlled exogenous variables such as SO₂, and O_3.
• $\mathsf{\alpha },\mathsf{\beta },\mathsf{\delta }, \mathrm{and} \mathsf{\gamma } \mathrm{are}$ coefficient terms.
• ε_t is the error term.

Considering that RA symptoms may manifest not only on the day of the dust storm event but also in the subsequent days, we incorporated ${\mathsf{\Omega }}_{\mathrm{j}}$ into the model to capture dynamic effects. ${\mathsf{\Omega }}_0^{ }$ represents the day of the dust storm while ${\mathsf{\Omega }}_1^{ }{ \mathrm{to} \mathsf{\Omega }}_4^{ }$ represent the first to fourth day after the dust storm. The terms ${\mathrm{RA}}_{\mathrm{t}-\mathrm{j}}^{ }$ are autoregressive terms. The determination of optimal autoregressive lags relies on the Ljung–Box test and autocorrelation function (ACF). To address any potential serial correlations in residuals, we employed an MA (1,2) model that was selected based on the Akaike Information Criterion (AIC) and the Schwarz criterion value.

Furthermore, this study stratified the sample by gender and age to thoroughly examine differences in RA outpatient visits among subgroups. A two-sided p-value of <0.05 was considered statistically significant. Data analysis was conducted using STATA version 16.

3. Results

3.1. Descriptive Statistics

From 2006 to 2012, a total of 182,899 outpatient visits for RA were recorded in Taiwan. Gender analysis shows that 22.21% (40,626 visits) were for males, and 77.79% (142,273 visits) were for females, with daily counts per million of 15.89 and 55.64, respectively. Females are three times more likely to experience RA-related issues and seek medical attention, with an overall average daily count of 71.53 per million.

The population is divided into three age groups: 21–40, 41–60, and 61 and above, contributing to 10.87% (19,887 visits), 44.50% (81,388 visits), and 44.63% (81,624 visits) of RA outpatient visits, respectively. The average daily counts per million are 7.78, 31.83, and 31.92, respectively. RA outpatient visits are less common in those under 40 but notably increase thereafter. Individuals aged 60 and above account for 44.63% of visits, with an average daily count of 31.92, indicating an upward trend in RA issues with advancing age.

Taiwan experienced 18 dust storm (DS) events over 109 days from 2006 to 2012. An analysis reveals a positive correlation between DS events and increased RA outpatient visits. On non-event days, the average daily count is 71.50, which is slightly higher than event days (66.92). Although counts on event days are lower, they consistently rise in the subsequent four days (78.39, 79.56, 72.83, and 68.22), reaching a peak on the first and second days after a DS event (

Table 1. Average daily RA outpatient visits according to gender and age group.

	Average Daily RA Outpatient Visits
	No Dust Storm	Day of Dust Storm	Post-Dust Day 1	Post-Dust Day 2	Post-Dust Day 3	Post-Dust Day 4	Observations (%)
Total	71.50	66.92	78.39	79.56	72.83	68.22	182,899 (100.00%)
Gender
Male	15.89	15.49	18.94	16.06	15.50	13.56	40,626 (22.21%)
Female	55.61	51.43	59.44	63.50	57.33	54.67	142,273 (77.79%)
Age
21–40	7.78	7.51	7.44	8.39	7.89	7.89	19,887 (10.87%)
41–60	31.82	30.30	34.06	33.72	33.61	30.83	81,388 (44.50%)
≥61	31.91	29.11	36.89	37.44	31.33	29.5	81,624 (44.63%)

).

3.2. Gender- and Age-Specific Patterns

Following a DS event, there is an immediate rise in RA outpatient visits for both males (18.94) and females (59.44), with visits peaking on the first day for males (18.94) and on the second day for females (63.50). Men might undertake higher levels of outdoor activity or experience occupational exposure during dust storms, leading to an earlier onset of symptoms and, consequently, earlier medical visits. In contrast, women may experience a slightly lower or delayed exposure, resulting in a peak in visits the following day. The impact of DSs on RA outpatient visits is most significant within the first two days after the event.

Age-related trends showed daily counts increasing from 7.78 to 8.39 and 7.89 on post-event days 2 and 3 for the 21–40 age group. Those aged between 41 and 60 experienced a surge from 31.82 to a peak of 34.06 on post-event day 1, remaining high at 33.72 on day 2. For individuals aged 61 and above, the most substantial rise occurred, with counts rising from 31.91 to 37.44 on post-event day 2. Variations in the temporal connection between DS outbreaks and RA outpatient visits are evident across different age groups.

3.3. Contrast Analysis

The percentage analysis shows a consistent pattern, with daily RA outpatient visits surging by 10–15% on post-event days 1 and 2 compared to non-DS days (100%) across both female and total populations (Figure 1). Males exhibit a similar trend, with visits peaking on post-event day 1. Among those aged 61 and above, a substantial disparity emerges, with daily RA outpatient visits skyrocketing by 15–20% on post-event days 1 and 2 compared to non-DS days (Figure 2). The 41–60 age group shows consistent increases of 5–10% on post-event days 1–3. In the 21–40 age group, the probability of occurrence is lower, peaking on post-event day 2. These insights highlight the temporal influence of DS events on RA outpatient visits, emphasizing demographic variations.

3.4. Regression Analysis

The ARMA time-series analysis in

Table 2. ARMA regression analysis for the relationship between DS and the number of daily RA outpatient visits according to gender.

	Number of Daily RA Outpatient Visits
	Total				Male				Female
Independent Variable	B	SE	p Value		B	SE	p Value		B	SE	p Value
Intercept	−31.9486	11.5864	0.0060	**	−9.2184	3.4727	0.0080	**	−26.3849	9.3630	0.0050	**
Time since dust storm
Day of dust storm	−8.0146	6.1632	0.1930		−0.8260	1.4500	0.5690		−7.4038	4.8788	0.1290
Post-dust day 1	9.9511	6.8482	0.1460		3.6872	1.7164	0.0320	*	5.9697	5.3157	0.2610
Post-dust day 2	7.8515	10.9950	0.4750		0.1625	3.3663	0.9620		7.7731	7.5873	0.3060
Post-dust day 3	−5.5288	7.1822	0.4410		−1.1538	2.0556	0.5750		−3.5710	5.6192	0.5250
Post-dust day 4	0.8573	8.6572	0.9210		−1.4477	2.3987	0.5460		1.6578	6.3635	0.7940
Temperature	0.8059	0.0916	0.0000	***	0.2207	0.0299	0.0000	***	0.6078	0.0753	0.0000	***
Humidity	0.1905	0.1089	0.0800		0.0400	0.0324	0.2180		0.1623	0.0881	0.0650
Rainfall	−1.2244	1.1345	0.2800		−0.1737	0.3570	0.6270		−1.0949	0.9115	0.2300
SO2	6.3050	0.6534	0.0000	***	1.7411	0.1952	0.0000	***	5.0236	0.5273	0.0000	***
O3	−0.0266	0.0621	0.6680		−0.0243	0.0187	0.1930		−0.0225	0.0505	0.6560
Trend	1.1431	0.0627	0.0000	***	0.2638	0.0201	0.0000	***	0.9023	0.0515	0.0000	***
MA1	0.1038	0.0168	0.0000	***	0.1153	0.0194	0.0000	***	0.1000	0.0175	0.0000	***
MA2	−0.4895	0.0178	0.0000	***	−0.2936	0.0192	0.0000	***	−0.4600	0.0182	0.0000	***

Note: MA, moving average; reference group: no dust storm. * p < 0.05; ** p < 0.01; *** p < 0.001.

provides valuable insights into the association between RA outpatient visits and DS events, considering a number of variables. Notably, a delayed and statistically significant (p < 0.05) increase in RA outpatient visits can be observed on post-event day 1 for the male subgroup. By contrast, no variations are found among the female subgroup or the total group on this day. Furthermore, temperature and SO₂ exhibit strong significance (p < 0.001) with positive coefficients in the estimations for all three groups, suggesting that higher temperatures and certain air pollutants are associated with an elevated likelihood of RA outpatient visits. Additionally, an ascending trend in RA outpatient visits is evident across Taiwan (p < 0.001). Remarkably, none of the five storm-related variables show significance among the female subgroup, indicating that DS does not significantly impact RA outpatient visits for females. For males, post-event day 1 exhibits significantly higher numbers of RA outpatient visits compared to non-DS days, underscoring a lag effect of one day after a DS event and revealing that males are at an increased risk of experiencing RA-related medical visits.

Table 3. ARMA regression analysis for the relationship between DS and the number of daily RA outpatient visits according to age group.

	Number of Daily RA Outpatient Visits
	21–40				41–60				≥61
Independent Variable	B	SE	p Value		B	SE	p Value		B	SE	p Value
Intercept	−2.0692	2.0735	0.3180		−9.6212	5.6262	0.0870		−27.0921	6.2422	0.0000	***
Time since dust storm
Day of dust storm	−0.3984	0.8270	0.6300		−3.4476	2.8730	0.2300		−4.0518	2.7657	0.1430
Post-dust day 1	−0.0417	1.1997	0.9720		3.4115	3.2733	0.2970		6.1803	3.0370	0.0420	*
Post-dust day 2	0.6337	1.3752	0.6450		1.8200	5.2086	0.7270		5.5197	4.4043	0.2100
Post-dust day 3	−0.0079	0.9761	0.9940		−0.9062	3.4209	0.7910		−2.8862	3.4281	0.4000
Post-dust day 4	0.3864	1.4128	0.7840		0.6251	3.8322	0.8700		−0.9059	3.7836	0.8110
Temperature	0.1155	0.0196	0.0000	***	0.4457	0.0455	0.0000	***	0.2815	0.0526	0.0000	***
Humidity	0.0181	0.0196	0.3550		0.0644	0.0527	0.2210		0.1328	0.0587	0.0240	*
Rainfall	−0.1750	0.2206	0.4280		−0.1802	0.4989	0.7180		−0.9077	0.6366	0.1540
SO2	0.7693	0.1139	0.0000	***	3.0136	0.3167	0.0000	***	3.3433	0.3479	0.0000	***
O3	−0.0178	0.0115	0.1210		−0.0130	0.0300	0.6640		−0.0292	0.0344	0.3970
Trend	0.0785	0.0128	0.0000	***	0.3527	0.0302	0.0000	***	0.7529	0.0371	0.0000	***
MA1	0.1095	0.0212	0.0000	***	0.0581	0.0187	0.0020	**	0.1692	0.0172	0.0000	***
MA2	−0.1196	0.0197	0.0000	***	−0.4175	0.0182	0.0000	***	−0.3877	0.0192	0.0000	***

Note: MA, moving average; reference group: no dust storm. * p < 0.05; ** p < 0.01; *** p < 0.001.

presents an age-based analysis of the ARMA results, providing insights into the relationship between DS and RA outpatient visits across different age groups. Among individuals aged 61 and above, a notable trend emerges. On post-event day 1, there is a significant surge in RA outpatient visits compared to non-DS days (p < 0.05), intriguingly suggesting a delayed impact of DSs on RA outpatient visits, specifically for the elderly population. Surprisingly, no discernible association between RA outpatient visits and DS is observed within the age groups 21–40 and 41–60. Additionally, mirroring the findings within gender subgroups, environmental factors such as temperature and SO₂ exhibit robust significance across all age groups (p < 0.001). These consistent findings underscore the influence of environmental factors on RA outpatient visits, highlighting a potential area for further investigation.

The dummy variable of the time trend shows a strongly significant and positive coefficient in all estimations (p < 0.001). This suggests there is a positive association between the time trend and RA outpatient visits, indicating that RA incidences have been increasing over time.

4. Discussion

The literature has consistently highlighted the adverse health effects of dust storms, linking them with various diseases, including respiratory [12,22], cardiac [17], and metabolic disorders [18]. The particulate matter in DSs, especially coarse particles (PM10), poses a significant threat to the respiratory system, leading to conditions such as asthma, bronchitis, silicosis, and pneumonia. This sets the stage for understanding the potential impacts of dust storms on rheumatoid arthritis, a chronic autoimmune disorder primarily affecting joints and the surrounding tissues.

Rheumatoid arthritis is a long-term autoimmune condition in which the immune system erroneously attacks the body’s own healthy tissues, resulting in inflammation mainly in the joints and nearby tissues. [1,2]. Although the precise cause of rheumatoid arthritis remains under investigation, it is widely accepted that a complex interplay of genetic and environmental factors contributes to its development [6,7,8,9]. Notably, environmental elements such as cigarette smoke and exposure to ambient air pollution have been significantly linked to an increased risk of developing RA [8,9,10,11]. The exact mechanism remains unclear, but inhaling air pollutants, such as particulate matter (PM), likely results in elevated levels of free radicals within the lungs, triggering oxidative stress, and increasing the production of antibodies and pro-inflammatory cytokines like IL-1 and IL-6 [23,24,25,26]. These cytokines contribute to respiratory diseases and can enter the bloodstream, leading to systemic inflammation. One possible mechanism is that these inflammatory factors travel through the bloodstream to the joints, causing synovial inflammation and hyaline cartilage degradation [23]. This systemic inflammation may exacerbate autoimmune conditions like RA. Dust storms contribute to this environmental concern, as they contain various types of particulates, including an elevated proportion of coarse particles (PM10) [17,18]. While considerable research has focused on the relationship between RA, smoking, and genetic factors, the influence of environmental factors like air pollution and dust storms on RA development remains underexplored [27].

This study aims to address this research gap by investigating the potential connection between DS and RA outpatient visits. Utilizing a rigorous ARMA time-series regression model, we analyzed the 2006–2012 NHI database with DS data while controlling for air quality and ambient temperature data. Our findings reveal a notable positive correlation between dust storm occurrences and the utilization of RA outpatient services. Notably, males exhibited a significant increase in RA outpatient visits one day after DS onset, indicating a swift response to environmental pollutants. By contrast, this effect was not observed in females, highlighting potential gender-specific vulnerabilities. Despite a higher overall probability of RA incidence in females, our study underscores the heightened sensitivity of males to environmental pollutant risks.

An age-based analysis further nuances this relationship, showing that individuals over 61 years old exhibit a significant increase in RA outpatient visits on the first day following a DS. Despite their ability to stay indoors during DS events, the more fragile immune systems of the elderly may contribute to their increased vulnerability. This age-specific variation raises questions about the differential vulnerability and responsiveness of different age cohorts to DS-triggered RA incidents, necessitating further exploration and targeted research.

Moreover, while the influence of DS events on RA outpatient visits varies across age groups, this study also underscores the pervasive impact of environmental factors. In this context, temperature and SO₂ levels emerge as robust variables, demonstrating consistent and strong statistical significance (p < 0.001) across all age groups. This universal significance highlights the critical role of these environmental factors in exacerbating RA conditions, transcending age boundaries. A review study on weather effects in RA found a positive correlation between outdoor temperature and RA in six studies conducted in coastal regions with consistently high relative humidity, where temperature largely controls absolute humidity. If air humidity impacts the body, it likely does so by affecting the microclimate close to the skin. This positive correlation between microclimate vapor pressure, outdoor temperature, and RA could explain the link between temperature and RA [28]. The impact of SO₂ can be attributed to the free reactive oxygen species released by inhaled SO₂ in the respiratory tract, which can activate nuclear factor kappa B (NF-κB), a key regulator of pro-inflammatory cytokine production in RA patients [29]. These findings not only shed light on the complex interplay between DS events, age, and RA outpatient visits but also emphasize the crucial role of environmental variables, calling for a holistic understanding of the multifaceted factors influencing RA outcomes. Further in-depth research and analysis in this field could potentially unravel more intricate connections, paving the way for targeted interventions and enhanced patient care strategies.

Overall, the time trend variable is positively associated with RA outpatient visits, indicating that RA incidences have been increasing over time. Therefore, preventive measures against RA in relation to DSs are to avoid exposure to dust by staying indoors during DS events, keeping windows closed, and using air purifiers. If staying indoors is impossible, wearing a mask should be a necessary precaution. People must continually pay close attention to air quality forecasts. Emphasizing preventive measures is particularly crucial for vulnerable populations, such as males and the elderly. Health authorities and local clinics can play a proactive role in disseminating information and advocating preventive measures to mitigate the impacts of DSs on RA incidences.

The strength of this manuscript lies in its exploration of a novel aspect by investigating the potential connection between dust storms and RA outcomes, thereby contributing to the existing literature discussing environmental influences on autoimmune disorders. However, there are limitations. First, data limitations are as follows: the reliance on retrospective data, particularly from the NHI research database, may pose challenges in terms of potential biases, missing variables, or the inability to establish a causal link due to the observational nature of the study. Second, potential confounders include the following: despite statistical adjustments, unmeasured confounding variables or factors not accounted for in the model could influence the observed associations. Third, the data are specific to Taiwan and may not be applicable to regions with different cultural, demographic, and environmental standards, or healthcare systems.

5. Conclusions

Our study is the first to illuminate a previously underexplored dimension in the intricate interplay between environmental factors, specifically dust storms, and rheumatoid arthritis outcomes. By analyzing Taiwan’s National Health Insurance claim data alongside dust storms, air pollution, and meteorology data, we identified a significant association between dust storms and an increased risk of RA outpatient visits. The results indicate that females are three times more likely to experience RA-related issues and seek medical attention compared to men. Percentage analysis revealed a 10–15% surge in daily RA outpatient visits on the first and second days following a dust storm compared to non-dust storm days. However, regression analysis showed that males, particularly those above 61, are more vulnerable to the effects of dust storms, with a significant spike in RA outpatient visits observed on the first day after a dust storm. These findings suggest a gender- and age-specific sensitivity to environmental pollutants, highlighting the need for targeted preventive strategies. Moreover, the consistent significance of environmental factors such as temperature and SO₂ levels across all age groups underscores the broader influence of these variables on RA conditions. This study contributes to the growing body of work in the literature on the environmental determinants of autoimmune disorders, particularly RA, by focusing on the previously underexplored area of dust storms.

In summary, this research highlights the importance of mitigating exposure to dust storms, especially for vulnerable populations such as males and the elderly, to prevent RA exacerbations. Public health measures, including staying indoors during dust storms, using air purifiers, and wearing masks, are essential preventive strategies. Further research is needed to explore the complex relationships between environmental factors and autoimmune disorders, aiming to improve patient care and develop targeted interventions.