Increased Risk of Psychiatric Disorders in Allergic Diseases: A Nationwide, Population-Based, Cohort Study

Background/objective Allergic diseases, such as bronchial asthma, allergic rhinitis, atopic dermatitis, and psychiatric disorders, are major health issues. There have been reports that allergic diseases were associated with depression or anxiety disorders. This study aimed to investigate the association between these allergic diseases and the risk of developing overall psychiatric disorders in patients from Taiwan. Methods This cohort study used the database of the Taiwan National Health Insurance Program. A total of 186,588 enrolled patients, with 46,647 study subjects who had suffered from allergic diseases, and 139,941 controls matched for sex and age, from the Longitudinal Health Insurance Dataset of 2000–2015, were selected from a sub-dataset of the National Health Insurance Research Database. Fine and Gray’s competing risk model analysis was used to explore the hazard ratio (HR), and 95% confidence interval, for the risk of allergic diseases being associated with the risk of developing psychiatric disorders during the 15 years of follow-up. Results Of the study subjects, 5,038 (10.8%) developed psychiatric disorders when compared to 9,376 (6.7%) in the control group, with significant difference (p < 0.001). Fine and Gray’s competing risk model analysis revealed that the adjusted HR was 1.659 (95% CI = 1.602–1.717, p < 0.001). In this study, we found that the groups of atopic dermatitis alone and the allergic rhinitis + atopic dermatitis were associated with a lower risk of psychiatric disorders, but all the other four groups, such as bronchial asthma alone, allergic rhinitis alone, bronchial asthma + allergic rhinitis, bronchial asthma + atopic dermatitis, and the combination of all these three allergic diseases, were associated with a higher risk of psychiatric disorders. Conclusion Allergic diseases are therefore associated with a 1.66-fold increased hazard of psychiatric disorders in Taiwan.


Data sources
The National Health Insurance (NHI) Program was launched in Taiwan in 1995, and as of June 2009, included contracts with 97% of the medical providers with approximately 23 million beneficiaries or more than 99% of the entire population (30,31). The National Health Insurance Research Database (NHIRD), which contains all the claims data of the beneficiaries, uses the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes to record diagnoses (32). The details of the program have been documented in previous studies (22,(33)(34)(35)(36)(37)(38)(39)(40)(41)(42).
A subset of the NHIRD, Longitudinal Health Insurance Database of a two million randomized sampled population in 2000-2015, was used to study the association between allergic diseases and the risk of psychiatric disorders. The present study used the NHIRD to identify patients with the diagnosis of allergic diseases, based on the ICD-9-CM codes, such as bronchial asthma (ICD-9-CM: 493.x), allergic rhinitis (ICD-9 CM code 477.X), and atopic dermatitis (ICD-9-CM code 691.X) during this period. The Institutional Review Board of the Tri-Service General Hospital approved this study and waived the need for individual written informed consent (IRB No. 2-105-05-040 and IRB No. 2-105-05-082).

study Design and sampled Participants
This study was of a population-based, matched-cohort design. Patients with newly diagnosed bronchial asthma, allergic rhinitis, or atopic dermatitis were selected from the Longitudinal Health Insurance Database from January 1, 2000, to December 31, 2015. The patients with asthma, allergic rhinitis, or atopic dermatitis before 2000 were excluded. This method could be viewed as a wash-out period to make sure all the allergic diseases were recent onset with references from other studies for the association between allergic diseases and psychiatric morbidity, using the NHIRD (6,7,11,15). This wash out period method could avoid the carry-over bias from exposure to the pre-existing allergic diseases. In addition, the patients diagnosed with dementia, depressive disorders, anxiety disorders, eating disorders, bipolar disorders, sleep disorders, and psychotic disorders, before 2000, or before their first visit for asthma, allergic rhinitis, or atopic dermatitis were also excluded. Each enrolled patient was required to have made at least three outpatient visits within the 1-year study period for allergic diseases according to these ICD-9-CM codes. A total of the patients who were enrolled, including 46,647 subjects with allergic diseases and 139,941 controls without allergic diseases, were matched for age, sex, and index date (Figure 1). in the politics, economy, and culture. Urbanization levels 3 and 4 were defined as a population between 149,999 and 499,999, and <149,999, respectively (43).

comorbidities
Comorbidities were assessed using the Charlson Comorbidity Index (CCI), which categorizes the comorbidities using the ICD-9-CM codes, scores each comorbidity category (44)(45)(46)(47)(48)(49), and combines all scores to calculate a single comorbidity score. A score of zero indicates that no comorbidities were found, and higher scores indicate higher comorbidity burdens (31).

statistical analyses
All statistical analyses were performed using the SPSS for Windows, version 22.0 (IBM Corp., Armonk, NY, USA). χ 2 and t tests were used to evaluate the distributions of the categorical and continuous variables, respectively, with a Fisher's exact examination. Fine and Gray's competing risk analysis was used to determine the risk of psychiatric disorders since death can act as a competing risk factor for psychiatric disorders (40,50,51). The results were presented as hazard ratio (HR) with a 95% confidence interval (CI). Differences in the risk of psychiatric disorders between the study and control groups were estimated using the Kaplan-Meier method with the log-rank test. A twotailed p value < 0.05 was considered to indicate the statistical significance.

Baseline characteristics of the study Population
The baseline characteristics of the study population are depicted in Table 1. There were 46,647 subjects in the allergic diseases group and 139,941 non-allergic diseases in the control group, with a similar distribution of sex, age, marital status, education years, and monthly insured premiums. The mean CCR (SD) for the subjects was 1.32 (1.10) and 0.67 (1.68) for the allergic diseases and control group, respectively. The subjects had more medical visits in the spring and winter, with residence in the south, east, and offshore islands of Taiwan, living in levels 3 and   Of the study subjects, 5,038 (10.8%) developed psychiatric disorders when compared to 9,376 (6.7%) in the control group, in the 15-year follow-up. We have also examined the risk of psychiatric disorders associated with allergic diseases. After adjusting for age, sex, CCI scores, and all the covariates, the Fine and Gray's survival analysis revealed that the adjusted HR for psychiatric disorders was 1.659 for the subjects (95% CI = 1.602-1.717, p < 0.001) when compared with the control group. With reference to the female subjects, the male subjects were associated with a higher risk of psychiatric disorders. With reference to the younger group (aged < 20 years), older age groups (aged 20-49 years and aged ≥ 50 years) were associated with a higher risk of psychiatric disorders (p < 0.001) ( Table 2). Figure 2 shows the Kaplan-Meier analysis for the cumulative incidence of psychiatric disorders in the study subjects and the control groups (log-rank test, p < 0.001). Furthermore, each single increased score of CCI was associated with a 2% increased risk of psychiatric disorders, and study subjects who sought medical help in the summer (adjusted HR 0.914, p < 0.001), autumn (adjusted HR 0.748, p < 0.001), and winter (adjusted HR 0.921, p < 0.001) were associated with a lower risk of psychiatric disorders (data not shown). Study subjects who live in urbanization level 3 were associated with a lower risk of psychiatric disorders than those in level 4 (adjusted HR 0.817, p < 0.001). Study subjects who sought medical help in the medical centers (adjusted HR 0.504, p < 0.001) and regional hospitals (adjusted HR 0.635, p < 0.001) were associated with a lower risk of psychiatric disorders than those in the local hospitals (data not shown).
association Between Different allergic Diseases and risk of Psychiatric Disorders Table 3 reveals that the association between each group of allergic diseases and the individual psychiatric disorders using Fine and Gray's competing risk model. The subjects with atopic dermatitis were associated with a decreased risk of overall and individual psychiatric disorders.
In addition, the patients with allergic diseases who live in the middle (adjusted HR: 1.118, p < 0.001), southern (adjusted HR: 1.055, p = 0.003), and eastern Taiwan (adjusted HR: 1.596, p < 0.001) showed slightly higher risk of psychiatric disorders, with reference of those who live in northern Taiwan. The patients who live in area of urbanization level 3 (adjusted HR: 0.763, p < 0.001) showed lower risk of psychiatric disorders, with reference of those who live in the urbanization level 4. The patients who sought for medical help in the hospital centers (adjusted HR: 0.504, p < 0.001) and regional hospitals (adjusted HR: 0.635, p < 0.001) showed lower risk of psychiatric disorders with the reference of local hospitals (Data not shown). We divided the subjects with allergic diseases into six groups: bronchial asthma alone, allergic rhinitis alone, atopic dermatitis alone, bronchial asthma + allergic rhinitis, bronchial asthma + atopic dermatitis, allergic rhinitis + atopic dermatitis, and a combination of all these three allergic diseases. In this study, we found that the groups of atopic dermatitis alone and allergic rhinitis + atopic dermatitis were associated with a lower risk of psychiatric disorders, but the other four groups were associated with a higher risk of psychiatric disorders ( Table 5).

risk of Psychiatric Disorders stratified by covariates
We analyzed the data by the stratification of the factors such as sex, age, marital status, education levels, urbanization level, geographic areas of residence, seasons of medical visits, monthly insured premiums, and levels of care from medical services providers. We found that the subjects, who were either male or female, married or not married, and in different educational years, urbanization levels, residence areas, seasons of visits, insured premiums, and levels of care, were associated with an increased risk of psychiatric disorders. In the different age groups, subjects aged < 20 years were associated with a lower risk of psychiatric disorder, while subjects in other age groups (≥20 years) were associated with a higher risk ( Table 6).

hr analysis of Psychiatric Disorders in Patients With Different Medications for the Treatment of asthma
The different medications of treatment durations more than 30 days for the treatment of bronchial asthma were grouped into four classes by the number of subjects who had used these medications: oral prednisolone, inhaled steroids, beta-agonist, aminophylline, leukotriene receptor antagonists, and anti-IgE   antibody (omalizumab). Oral prednisolone usage was associated with a lower risk of developing psychiatric disorders in bronchial asthma patients. Besides, usage of beta-agonist, aminophylline, and leukotriene receptor antagonists was also associated with a lower risk of developing psychiatric disorders while adjusted with all the covariates and comorbidities, either with or without prednisolone usage ( Table 7).

DiscUssiOn
In this study, we examined the association between the overall allergic diseases and the risk of psychiatric disorders. After adjusting the covariates, the adjusted HR was 1.659 for the subjects (95% CI = 1.602-1.717, p < 0.001) when compared with the control group. In other words, the adult patients with allergic diseases had a 1.66-fold increased risk of developing psychiatric disorders. The Kaplan-Meier analysis revealed that the study subjects had a significantly higher 15-year psychiatric disorders-free survival rate than the controls. Bronchial asthma and allergic rhinitis were associated with the risk of overall psychiatric disorders, but atopic dermatitis was associated with a lower risk of psychiatric disorders. Bronchial asthma was associated with the risk of individual psychiatric disorders, such as dementia, anxiety disorders, depressive disorders, eating disorders, bipolar disorders, sleep disorders, and psychotic disorders, but allergic rhinitis was associated with a lower risk of dementia and an increased risk of other psychiatric disorders. However, the increasing number of allergic diseases was associated with an increased risk of psychiatric disorders. In comparison to the previous studies for individual allergic diseases in several psychiatric disorders, such as anxiety disorders, depressive disorders, bipolar disorders, schizophrenia, and ADHD (4, 6-8, 10-15, 43), this is the first study on the topic of the association of allergic diseases and the broader spectrum of psychiatric disorders. Atopic dermatitis seems to be an exception in these allergic diseases, by being associated with a lower risk of psychiatric disorders, even though previous reports found that atopic dermatitis are associated with ADHD and autism spectrum disorders (52)(53)(54)(55). Further study is needed to clarify the association between atopic dermatitis and the risk of psychiatric disorders.
The results of this study showed the association between overall allergic diseases and the risk of psychiatric disorders, which echoed the findings of other studies on the association between the overall allergic or atopic diseases and the risk of psychiatric   disorders: for example, children with eczema, asthma, or hay fever had more emotional, conduct, and hyperactivity problems in a cross-sectional study in Denmark (56), and atopic diseases might be associated with an elevated risk of developing depression in a birth cohort study in Finland (57). Several studies have reported the associations between bronchial asthma or other atopic diseases and psychiatric disorders, such as affective disorders, anxiety disorders, schizophrenia, substance-related disorders, autismspectrum disorders, or ADHDs (9,(58)(59)(60)(61)(62). Some researchers have also reported the associations between atopic dermatitis and attention-deficit/hyperactivity disorder, or speech disorders in childhood (63), and the association between allergic rhinitis and the risk of suicides and depression (64). Furthermore, parental exposure to occupational asthmagens might also be associated with children autism-spectrum disorders (65). Previous studies have also reported the associations between other immune or inflammation-related diseases, such as periodontitis (66), glutenrelated illnesses (67), amyotrophiclateral sclerosis (68), psoriasis (69), food or other allergies (70), and idiopathic environmental intolerance (71) and the risk of psychiatric disorders. Conversely, psychiatric disorders might also affect the immune systems (72). In this study, the overall allergic diseases were associated with affective disorders, anxiety disorders, psychotic disorders, and dementia, respectively. The interplay between allergic diseases and psychiatric disorders, therefore, warrant further study.
The underlying mechanism of the association between allergic diseases and psychiatric disorders remains unclear. The "orchestration" of the pro-inflammatory cytokines plays an important role in the pathogenesis of allergy-related diseases, such as asthma (73). Reports have shown that cytokines such as interleukin (IL)-6, tumor necrosis factor-alpha (TNF-α), IL-10, and monocyte chemoattractant protein-1/CCL2 might be well associated with depressive, bipolar, or anxiety disorders (74,75). Reports also show that dementia is related to peripheral pro-inflammatory factors released by periodontal inflammatory mediators such as C-reactive protein, IL-6, haptoglobin, TNF-α, and fibrinogen (76)(77)(78). Furthermore, some studies have also found that inflammation play an important role in eating disorders (79,80) and sleep disorders (81)(82)(83).
Additionally, we have included several psychosocial factors in the analysis, such as marital status, educational level, monthly insured premiums, urbanization level, and residence. In comparison to the allergy-disease subjects who sought medical help in the local hospitals, study subjects who received medical help from medical centers or regional hospitals were associated with a lower risk of psychiatric disorders. These findings suggest that disadvantageous socioeconomic factors might well contribute to the risk of psychiatric disorders.
We found that the study subjects of those aged < 20 years had a lower risk in developing psychiatric disorders. We therefore hypothesize that the reason for a lower risk is that the maximal follow-up time is 15 years, which might not have, as yet, reached the age of onset of the most major psychiatric disorders for the patients of allergic diseases (84). Further study is needed to clarify the association between age and the risk of psychiatric disorders in the patients with allergic diseases.
Furthermore, we found that oral prednisolone usage was associated with a lower risk of developing psychiatric disorders in bronchial asthma patients. Moreover, the usage of beta-agonist, aminophylline, and leukotriene receptor antagonists was also associated with a lower risk of developing psychiatric disorders, with or without prednisolone usage. The mechanisms and effects of these medications on the risk of psychiatric disorders in bronchial asthma patients merit further study.

limitations
The present study has several limitations that warrant consideration. First, similar to previous studies using the NHIRD on allergic diseases as aforementioned, not all the data were recorded in the NHIRD, and we were unable to evaluate the severity, weakness severity, laboratory parameters, or lung function examinations in asthma patients. Second, other factors, such as genetic, psychosocial, and environmental factors, were not included in the dataset. Third, for patients who have to pay for their own for overthe-counter drugs for the allergic diseases, their self-medications would not have been included in the NHIRD. There is no study about the rates of self-medications for these allergic diseases. However, due to the high coverage of medical providers (97%) and beneficiaries (more than 99%) of the NHI system in Taiwan, most of the people would ask for help from the NHI-contracted hospital or clinics for their allergic diseases.

strength of This study
One of the primary strengths of this study is the use of ICD-9 codes, and a number of studies have demonstrated the accuracy and validity of several diagnoses in the NHIRD, including DM (85,86), cancer (87)(88)(89), myocardial infarction (85,90,91), and central nervous system diseases, such as Tourette syndrome (92), and stroke (85,(93)(94)(95), outcomes (89), mortality (85,96), or comorbidity (89,96). In a wide spectrum of conditions, some studies have also demonstrated the concordance between Taiwan's National Health Survey and the NHIRD on various diagnoses (97), medication usage (97), and health system utilizations (97,98). Correspondingly, the long-term observation period from 2000 to 2015 allowed for more credibility, when compared with other similar studies, to propose physical mechanisms and plausible hypotheses. Finally, and most importantly, we have tried to explain the mutual biological and psychological mechanism between the allergic diseases and the psychiatric disorders.

cOnclUsiOn
We have evaluated the risk of psychiatric disorders in association with the allergic diseases such as asthma, allergic rhinitis, and atopic dermatitis in Taiwan's population, using the representative population-based data. We have demonstrated that the patients with allergic diseases were at a significantly higher risk of psychiatric disorders than the control group. Further studies are therefore needed for patients with allergic diseases not only to prevent its clinical exacerbation but also to decrease the possibility of developing psychiatric disorders. We sincerely hope that this study will provide the necessary information for an earlier intervention for patients with allergic diseases.

FUnDing
This study was supported in part by the Tri-Service General Hospital Research Foundation (TSGH-C106-002, TSGH-C106-106, and TSGH-C107-004). These funding agencies did not influence the study design, data collection and analysis, decision to publish, or preparation of the manuscript.