Risk of Developing Depressive Disorders following Rheumatoid Arthritis: A Nationwide Population-Based Study

Background & Aims To evaluate the risk of depressive disorders among rheumatoid arthritis (RA) by using the Taiwan National Health Insurance Research Database (NHIRD). Methods We conducted a retrospective study of a matched cohort of 18 285 participants (3 657 RA patients and 14 628 control patients) who were selected from the NHIRD. Patients were observed for a maximum of 10 years to determine the rates of newly diagnosed depressive disorders, and Cox regression was used to identify the risk factors associated with depressive disorders in RA patients. Results During the 10-year follow-up period, 205 (11.2 per 1000 person-years) RA patients and 384 (5.1 per 1000 person-years) control patients were diagnosed with depressive disorders. In RA patients, most depressive disorders (n = 163, 80%) developed with five years of being diagnosed with RA. The incidence risk ratio of depressive disorders between RA patients and control patients was 2.20 (95% confidence interval [CI], 1.84–2.61, P<.001). After adjusting for age, sex, and comorbidities, RA patients were 2.06 times more likely to develop depressive disorders (95% CI, 1.73–2.44, P<.001) compared with the control patients. Hyperthyroidism (HR = 1.67) was an independent risk factor for depressive disorders in patients with RA. Conclusions The likelihood of developing depressive disorders is greater among RA patients than among patients without RA. Symptoms of depression should be sought in patients with RA.


Introduction
Rheumatoid arthritis (RA) is a systemic chronic autoimmune disorder affecting primarily cartilage and bone. In addition, several organs such as lung, vessels, and kidney may be involved. In general, RA affects 0.5-1% of the population [1]. Despite of substantial advances in medical treatment, chronic pain, fatigue, and functional disability causes much psychosocial burden for these patients with RA [2,3].
Therefore, interest in the psychiatric aspects of rheumatoid arthritis has grown. Several studies have reported a high depression prevalence rate in patients with RA [4][5][6][7][8][9][10][11][12], and a significant correlation was observed among depression and increased pain levels in patients with RA [13]. Comorbid depression has been also shown to independently increase disability and a low quality of life in patients with RA [14][15][16][17][18]. In addition, depression has been demonstrated to be an independent risk factor for non-suicide mortality in patients with RA [19]. Hence, the depression symptoms should be examined in patients with RA because depression is a modifiable illness that is amenable to treatment.
However, in previous studies focusing on depression among patients with RA, depression was often evaluated by using rating scales, such as the Beck Depression Inventory or the Epidemiological Studies Depression Scale, rather than diagnosis by a psychiatrist [5][6][7][8][9][10][11][12]. The real prevalence rate of clinical depressive disorders in patients with RA remains unknown. In addition, the design of former studies prevented the temporal causal relationship between RA and depressive disorders from being clarified because the designs of these studies were almost cross-sectional studies [4][5][6][7][8][9][10][11][12]. Furthermore, whether there is an independent risk factor for newly diagnosed depressive disorders in patients with RA has never been studied.
We performed a population-based retrospective cohort study using data derived from the National Health Insurance (NHI) system in Taiwan. The purpose of our study was to determine whether RA was associated with an increased risk of clinical depressive disorders. Independent risk factors for newly diagnosed depressive disorders among the patients with RA were also identified.

Data Sources
Instituted in 1995, the NHI program is a mandatory health insurance program that offers comprehensive medical care coverage, including outpatient, inpatient, emergency, and traditional Chinese medicine, to all residents of Taiwan; the coverage rate is as high as 99% [20]. The NHI research database (NHIRD) contains comprehensive information regarding clinical visits, including prescription details and diagnostic codes based on the International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM). The NHIRD is managed by the National Health Research Institutes (NHRI) and confidentiality is maintained according to the directives of the Bureau of the NHI. The data source for our study was the Longitudinal Health Insurance Database 2005 (LHID 2005), which is a dataset of NHIRD. Data for the LHID was collected by systematically and randomly sampling from the NHIRD; the database included the data of one million individuals. The NHRI of Taiwan reports that there were no significant differences in gender distribution, age distribution, or average insured payroll-related amount between the patients in the LHID and those in the original NHIRD [21]. Detailed information about the data source is provided on the NHRI website (http://nhird.nhri.org.tw). Comments, problems or requests about the data application could be sent to e-mail address at the NHRI (e-mail: nhird@nhri.org.tw).

Ethics Statement
The Institutional Review Board of the Taipei Veterans General Hospital approved this study (2013-03-035AC). Written consent from study patients was not obtained because the NHI dataset consists of de-identified secondary data for research purposes, and the Institutional Review Board of Taipei Veterans General Hospital issued a formal written waiver for the need for consent.

Study Population
Using data extracted from the LHID 2005, we conducted a retrospective cohort study of patients aged 20 years and older who were newly diagnosed with RA by a rheumatologist between January 1, 2000 and December 31, 2008. RA was defined as ICD-9-CM code: 714. We excluded patients who were diagnosed with depressive disorders (ICD-9-CM code: 296.2X-296.3X, 300.4, and 311.X) before enrollment. Insurance premiums, calculated according to the beneficiary's total income, were used to estimate monthly income. Monthly income was grouped into low income (monthly income ,20,000 New Taiwan Dollar [NTD]), median income (20,000 NTD#monthly income ,40,000 NTD), and high income (monthly income $40,000 NTD). Urbanization was divided into three groups: urban, suburban, and rural. Urbanization and monthly income levels were used to represent socioeconomic status. For each cirrhotic patient included in the final cohort, 4 age-, sex-, and enrolment-date-matched control patients who were not diagnosed with rheumatoid arthritis or depressive disorder were randomly selected from the LHID 2005. All RA and control patients were observed until diagnosed with a depressive disorder by a psychiatrist, death, withdrawal from the NHI system, or December 31, 2009. The primary clinical outcome assessed was psychiatrist-diagnosed depressive disorder.

Statistical Analyses
The incidence of newly diagnosed depressive disorders in the RA and control patients, stratified by gender and age (equal or older than 60 years old or younger than 60 years old) was calculated. Independent t-tests and chi-squared tests were used to examine the differences in the demographic characteristics between the RA and control patients.
A Cox proportional-hazards regression model was used to identify variables that predicted depressive disorder in the RA and control patients, and in the RA patients only. Control variables, such as age; sex; common comorbidities, including hypertension, diabetes mellitus, dyslipidemia, coronary artery disease, congestive heart failure, cirrhosis, hyperthyroidism, hypothyroidism, cerebrovascular disease, and malignancy; urbanization; and monthly income were included as covariates in the univariate model. Factors that demonstrated a moderately significant statistical relationship in the univariate analysis (P,.1) were entered by forward selection in a multivariate Cox proportional-hazards regression model [22].
The Perl programming language (version 5.12.2) was used to extract and compute data. The Microsoft SQL Server 2005 (Microsoft Corp., Redmond, WA, USA) was used for data linkage, processing, and control sampling. IBM SPSS (version 19.0 for Windows; IBM Corp., New York, NY, USA) and SAS statistical software (version 9.2; SAS Institute Inc., Cary, NC, USA) were used to perform all statistical analyses. The comparisons resulting in a P-value of less than .05 were considered to indicate a statistically significant relationship.

Participant Selection (Figure 1)
The sample comprised 3657 RA patients and 14 628 control patients without depression, among whom 72.8% were female. The median age at enrollment was 51 years (interquartile range [IQR], 41-61 years), and the median follow-up period was 4.72 years (IQR, 2.59-7.42 years) for RA patients and 4.92 years (IQR, 2.69.21-7.55 years) for control patients. The comorbidities, including hypertension, diabetes mellitus, dyslipidemia, congestive heart failure, hyperthyroidism, hypothyroidism, and cerebrovascular disease were more common in RA patients than control patients. The comparisons of the demographic and clinical variables between the RA and control patients are presented in Table 1.

Incidence Rate of Depressive Disorders
During the study period, 205 (11.2 per 1000 person-years) RA patients and 384 (5.1 per 1000 person-years) control patients were diagnosed with depressive disorders. The incidence risk ratio (IRR) of depressive disorders between RA patients and control patients was 2.20 (95% CI, 1.84-2.61, P,.001). When stratified by gender and age, the IRR of depressive disorders was still higher in the RA patients than in the control patients. The results are shown in Table 2. In RA patients, most depressive disorders (n = 163, 80%) developed with five years of being diagnosed with RA.

Rheumatoid arthritis on Risks of Clinical Depression
After adjusting for age, sex, comorbidities, urbanization, and monthly income, the hazard ratio (HR) for developing depressive disorders during the follow-up period was 2.06 times (95% CI, 1.73-2.44, P,.001) greater for the RA patients than for the control patients (Table 3 and figure 2).

Discussion
The key findings of our study are that (1) the incidence rate of depressive disorders in patients with RA was 11.2 per 1000 personyear; (2) the risk of clinical depressive disorders was higher when patients with RA (HR = 2.06) and most depressive disorders (n = 163, 80%) developed within five years of being diagnosed with RA; (3) hyperthyroidism (HR = 1.67) might be an independent risk factor for depressive disorders among patients with RA.
The strengths of this study are the large sample size, the long follow-up period, and the diagnosis of clinical depressive disorders and rheumatoid arthritis by specialists. In addition, our study design involved an unbiased participant selection process. Because participation in the NHI is mandatory and all residents of Taiwan can access low-cost health care, referral bias is low and follow-up compliance is high.
Consistent with the results of previous studies [4][5][6][7][8][9][10][11], we found that the risk of depression in RA patients was higher than that in the non-RA control patients. Chronic inflammation is one of the possible causes. RA is a chronic inflammatory disease, and accumulating evidence implicates inflammation as a critical mediator in the pathophysiology of depressive disorders [23][24][25][26]. In the Cardiovascular Health Study of 4 268 patients, a significant association was noted between high levels of C-reactive protein and depression symptoms, even after adjusting for confounding factors [27]. Administration of cytokines including tumor necrosis factor-a in animal and human studies has been shown to induce depressive symptoms, including sleep disturbance, anorexia, loss of interest and mood disturbances [28]. Potential involved mechanisms include direct effects of pro-inflammatory cytokines on monoamine, dysregulation of the hypothalamic-pituitary-adrenal axis, impaired neuroplasticity and structural and functional brain changes [29].
The symptoms of RA such as pain and fatigue may be another possible cause for higher risk of depression in RA patients than that in the non-RA control patients. Symptoms of RA such as pain and fatigue may impair patients' function and quality of life and influence patients' perception of health status, and make the patients more prone to develop depressive symptoms [12,30]. RA patients often suffered pain and pain intensity was more clearly related to disability in RA patients [31]. Several studies have shown that chronic pain was an important risk factor for depression and suicidal behavior [32][33][34]. Previous study has shown that the degree of depression in RA patients varies in proportion to the level of pain [31].
Furthermore, physical comorbidities might be another possible cause for a higher prevalence of depressive disorder in RA patients. Our findings are consistent with the results of previous studies that found that a higher percentage of many physical comorbidities in RA patients, such as hypertension [35], diabetes mellitus [36], congestive heart failure [37], hypothyroidism [38], and cerebrovascular disease [39]. All of above diseases had been proposed to be related with increased the risk of depression [40][41][42][43]. Finally, long-term drug use in RA patients and inevitable side effects of various medications might also impaired quality of life in RA patients and increased the risk of depression in RA patients [44,45]. Previous study showed that disease-modifying drugs used for RA patients had been related to higher rates of depression, and suicidal ideation among RA patients, serving as an alert to the importance of considering also this factors in therapeutic decisions [46]. According to previous studies, conservative estimates of the prevalence of depressive disorder among patients with RA ranged between 13 and 20% [13,45,47]. This prevalence rate was much higher than the rate calculated in this study. The possible reason for these two discrete findings might be partly due to difference in the methods of assessing depression [13]. In the most previous studies regarding RA and depression [5][6][7][8][9][10][11][12], rating scales, such as the Beck depression inventory, rather than diagnosis by a psychiatrist, were used for evaluation of depression in those studies. People who were recorded as depressed on rating scales may not meet the diagnostic criteria for clinical depressive disorders. However, depressive disorders in RA patients may be underdiagnosed in our study because symptoms of depressive disorders may mimic symptoms of RA, such as easily fatigue, sleep disturbance, and loss of interests with multiple somatic complaints. Physicians may have viewed RA patients' depressive symptoms as physical discomforts caused by RA and overlooked these symptoms, rather than viewing these symptoms as warning signs of depressive disorder.
There has been some discussion in the literature regarding the factors that influence the development of depression in patients with RA [48][49][50]. However, whether there is certain physical comorbidity as an independent risk factor for new-onset depressive disorders following a RA diagnosis has never been studied. In our study, we found that hyperthyroidism was an independent risk factor for depressive disorders in patients with RA. Studies have shown that mood disturbance, especially depressive disorders, often accompanies hyperthyroidism patients [51][52][53][54]. Alterations in components of the hypothalamic-pituitary-thyroid axis have been documented for the development of depressive disorder among patients with hyperthyroidism [53]. In addition, functional impairment caused by hyperthyroidism might play a vital role in causing psychological distresses including depressive disorders among patients with hyperthyroidism [55].  Our findings have certain limitations. Information regarding the family history of psychiatric disorders [56,57], lifestyle factors [58], self-perceived health, disability [59], and environmental factors (life stresses, such as recent major life event, domestic violence, childhood trauma, and social support) [60][61][62][63] are not included in the NHIRD, and all of which might be associated with the risk of development of depressive disorders. In addition, the severity of RA could not be ascertained. Whether the severity of RA influences the risk of developing depressive disorders warrants further study, despite previous studies indicating that disease activity of RA alone does not equate with depressive symptoms in RA patients reliably [47]. Furthermore, physicians may view RA patients' depressive symptoms as physical discomforts caused by RA and overlook these symptoms. Therefore, in our study, the incidence of clinical depressive disorders may be underestimated.
This population-based retrospective cohort study demonstrated that the risk of developing depressive disorders is higher among RA patients than those without RA, and that RA is an independent risk factor for depressive disorders. Hyperthyroidism is an independent risk factor for developing depressive disorders among patients with RA. Because depression was related to impairing the quality of life of RA patients, and depressive disorder is modifiable illness that is amenable to treatment, symptoms of depression should be sought in patients with RA. Future population-based prospective studies are needed to further investigate the association between RA and the risk of developing depressive disorders.