Introduction

Depression is extremely common in patients with rheumatoid arthritis (RA), with an incidence 1.7-fold greater than in individuals without RA [1]. The rate of major depression diagnosed in clinical interviews of RA patients is 16.8%, and the rate of depression in RA patients screened with the Hospital Anxiety and Depression Scale (HADS) has been reported to be between 14.8% and 48% [2].

Hospitalization, loss of workforce, healthcare costs, and mortality are higher in patients with depression and RA compared to those with RA only [3, 4]. Anxiety in RA patients has been reported at the rate of 25.1%. In 16.3% of RA patients, anxiety and depression are seen together [5].

Patients with depression and anxiety symptoms show a reduced response to disease-modifying anti-rheumatic drugs (DMARDs) and glucocorticosteroids, and there is worsening of physical function and disease activity. Depression symptoms may be a prognostic marker of poor rheumatological out-comes [6, 7].

The presence of depression at the start of treatment has been shown to reduce the response to treatment by 30% [8]. Depression symptoms have also been determined to reduce remission rates [9, 10].

Material and methods

Participants

This study included patients aged 18–85 years, who presented at our Rheumatology Polyclinic and were diagnosed with RA according to the ACR 2010 diagnostic criteria [11]. Patients were excluded from the study if they had psychosis, were pregnant or breastfeeding, were receiving psychiatric treatment, or had a diagnosis of malignancy.

A record was made for each patient of age, sex, marital status, employment status, body mass index (BMI), disease duration, comorbidities, drugs used in treatment, the Disease Activity Score in 28 joints based on erythrocyte sedimentation rate (DAS28–ESR), the Health Assessment Questionnaire (HAQ) score and the Hospital Anxiety and Depression Scale (HADS) score.

The HADS is used to determine the risk of anxiety and depression in a patient, the level, and severity of change. There are two subscales that separately evaluate anxiety and depression. In the Turkish version of the scale the cut-off points have been determined as 10 for the anxiety subscale, and 7 for the depression subscale [12].

The study protocol was approved by the Ethics Committee of Akdeniz University Faculty of Medicine (07.02.2018 decision number: 150).

Statistical analysis

Descriptive information was presented as means, standard deviations, frequencies, and ranges as appropriate. The χ2 test for categorical data, Student’s t-test or the Mann-Whitney U test for continuous data was performed for comparisons between demographic and clinical characteristics of patients with or without anxiety and depression.

Multivariate logistic regression analyses were performed to evaluate the relationship of demographic and clinical factors with anxiety or depression in patients with rheumatoid arthritis.

The associations between independent variables and anxiety/depression were expressed as the odds ratio (OR) with 95% confidence interval (CI). A p-value less than 0.05 was considered statistically significant. All analyses were done using IBM SPSS Statistics 20.

Results

One hundred and eighty-two patients with RA were included in this study. The mean age of the patients was 52.9 ±12.0 years (range 21–81 years), and the mean disease duration was 11.7 ±8.9 years (range 1–50 years). Seventy-five and three tenths percent of the patients were female. The mean DAS28 score was 2.8 ±1.1 and the mean HAQ score was 1.08 ±1.2.

Table I summarizes the descriptive statistics for age, BMI, disease duration, DAS28, HAQ, and HADS scores, sex, marital and working status, education level, comorbidities and drugs used for RA (conventional synthetic disease-modifying antirheumatic drugs [csDMARDs] and biological disease-modifying antirheumatic drugs [bDMARDs]).

Table I

Demographic and clinical characteristics

VariableMean ±SDRange
Age [years]52.9 ±12.021 – 81
BMI [kg/m²]29.1 ±4.918.1–48.7
Disease duration [years]11.7 ±8.91 – 50
DAS282.8 ±1.10.9–6.6
HAQ1.08 ±1.20 – 3
HADS – anxiety7.5 ±4.30 – 17
HADS – depression7.2 ±4.00 – 19
Number (%)
 Anxiety +46 (25.3)
 Depression +92 (50.3)
Sex
 Women137 (75.3)
 Men45 (24.7)
Marital status
 Married149 (81.9)
 Single33 (18.1)
Working status
 Retired47 (25.8)
 Housewife102 (56)
 White-collar worker19 (10.4)
 Blue-collar worker14 (7.7)
Education
 Uneducated39 (21.4)
 Primary school92 (50.5)
 High school30 (16.5)
 University21 (11.5)
Comorbidities
 None116 (63.7)
 Present66 (36.3)
Drugs
 csDMARD127 (69.8)
 bDMARD55 (30.2)

[i] BMI – body mass index, bDMARD – biologic disease-modifying antirheumatic drug, csDMARD – conventional synthetic diseasemodifying antirheumatic drug, DAS28 – Disease Activity Score with 28-joint counts, HADS – Hospital Anxiety Depression Scale, HAQ – Health Assessment Questionnaire.

According to the Turkish validation scores of HADS with a cut-off value of 10 for anxiety and cut-off value of 7 for depression, anxiety was detected in 25.3% of the patients, and depression was detected in 50.3% of the patients.

Tables II and III present the comparisons of sex, marital and working status, education level, comorbidities and drugs used between the patients with or without anxiety and depression. There was a statistically significant difference in working status between patients with or without anxiety and in sex education level and working status between patients with or without depression (p < 0.05).

Table II

The χ2 test results of demographic characteristics between patients with and without anxiety

Sexp = 0.182
WomenMenTotal
Anxiety –n (%)99 (72.3)37 (82.2)136 (74.7)
Anxiety +n (%)38 (27.7)8 (17.8)46 (25.3)
Totaln (%)137 (100)45 (100)182 (100)
Marital statusp = 0.239
MarriedSingleTotal
Anxiety –n (%)114 (76.5)22 (66.7)136 (74.7)
Anxiety +n (%)35 (23.5)11 (33.3)46 (25.3)
Totaln (%)149 (100)33 (100)182 (100)
Educationp = 0.108
UneducatedPrimaryHigh schoolUniversityTotal
Anxiety –n (%)26 (66.7)68 (73.9)22 (73.3)20 (95.2)136 (74.7)
Anxiety +n (%)13 (33.3)24 (26.1)8 (26.7)1 (4.8)46 (25.3)
Totaln (%)39 (100)92 (100)30 (100)21 (100)182 (100)
Working statusp = 0.019
RetiredHousewifeWhite-collarBlue-collarTotal
Anxiety –n (%)42 (89.4)71 (69.6)15 (78.9)8 (57.1)136 (74.7)
Anxiety +n (%)5 (10.6)31 (30.4)4 (21.1)6 (42.9)46 (25.3)
Totaln (%)47 (100)102 (100)19 (100)14 (100)182 (100)
Comorbiditiesp = 0.809
NonePresentTotal
Anxiety –n (%)86 (74.1)50 (75.8)136 (74.7)
Anxiety +n (%)30 (25.9)16 (24.2)46 (25.3)
Totaln (%)116 (100)66 (100)182 (100)
Drugsp = 0.281
DMARDBiologicalTotal
Anxiety –n (%)92 (72.4)44 (80)136 (74.7)
Anxiety +n (%)35 (27.6)11 (20)46 (25.3)
Totaln (%)127 (100)55 (100)182 (100)
Table III

The χ2 test results of demographic characteristics between patients with and without depression

Sexp = 0.008
WomenMenTotal
Depression –n (%)60 (43.8)30 (66.7)90 (49.5)
Depression +n (%)77 (56.2)15 (33.3)92 (50.5)
Totaln (%)137 (100)45 (100)182 (100)
Marital statusp = 0.902
MarriedSingleTotal
Depression –n (%)74 (49.7)16 (48.5)90 (49.5)
Depression +n (%)75 (50.3)17 (51.5)92 (50.5)
Totaln (%)149 (100)33 (100)182 (100)
Educationp = 0.002
UneducatedPrimaryHigh schoolUniversityTotal
Depression –n (%)13 (33.3)41 (44.6)20 (66.7)16 (76.2)90 (49.5)
Depression +n (%)26 (66.7)51 (55.4)10 (33.3)5 (23.8)92 (50.5)
Totaln (%)39 (100)92 8100)30 (100)21 (100)182 (100)
Working statusp = 0.006
RetiredHousewifeWhite-collarBlue-collarTotal
Depression –n (%)32 (68.1)40 (39.2)12 (63.2)6 (42.9)90 (49.5)
Depression +n (%)15 (31.9)62 (60.8)7 (36.8)8 (57.1)92 (50.5)
Totaln (%)47 (100)102 (100)19 (100)14 (100)182 (100)
Comorbiditiesp = 0.082
NonePresentTotal
Depression –n (%)63 (54.3)27 (40.9)90 (49.5)
Depression +n (%)53 (45.7)39 (59.1)92 (50.5)
Totaln (%)116 (100)66 (100)182 (100)
Drugsp = 0.949
DMARDBiologicalTotal
Depression –n (%)63 (49.6)27 (49.1)90 (49.5)
Depression +n (%)64 (50.4)28 (50.9)92 (50.5)
Totaln (%)127 (100)55 (100)182 (100)

In blue-collar workers presence of anxiety was significantly higher than in retired patients. The depression rate was significantly higher in women compared to men, in uneducated patients compared to patients who are high school or university graduates and in housewives than retired patients. Biological disease-modifying antirheumatic drug and csDMARD users were compared; no statistically significant difference was found in terms of anxiety and depression (Tables II and III).

Tables IV and V present the comparisons of age, disease duration, BMI, DAS28 and HAQ scores. Both patients with anxiety and depression have significantly higher scores of DAS28 and HAQ compared to patients without anxiety and depression (p < 0.05), indicating higher disease activity and worse functional status.

Table IV

Comparison of demographic and clinical characteristics between patients with and without anxiety

VariableAnxiety –Anxiety +p-value
Age53.3 ±12.651.8 ±10.10.452
Body mass index29.0 ±4.629.3 ±5.60.744
Disease duration12.3 ±9.010.1 ±8.50.093
DAS282.6 ±1.03.3 ±1.20.002
HAQ0.9 ±1.11.7 ±1.4< 0.001

[i] DAS28 – Disease Activity Score with 28-joint counts, HAQ – Health Assessment Questionnaire.

[ii] Statistical significance was calculated using the Mann-Whitney U test.

Table V

Comparison of demographic and clinical characteristics between patients with and without depression

VariableDepression –Depression +p-value
Age51.6 ±12.154.2 ±11.90.152
Body mass index28.6 ±4.829.6 ±4.90.194
Disease duration11.5 ±9.111.9 ±8.80.735
DAS282.5 ±0.93.1 ±1.20.001
HAQ0.6 ±0.751.6 ±1.47< 0.001

[i] DAS28 – Disease Activity Score with 28-joint counts, HAQ – Health Assessment Questionnaire.

[ii] Statistical significance was calculated using the Mann-Whitney U test.

The multivariate logistic regression analysis using anxiety and depression as dependent variables is presented in Table VI. The Health Assessment Questionnaire score was significantly associated with both presence of anxiety (OR = 1.04, 95% CI: 1.00–1.09, p = 0.041) and depression (OR = 1.10, 95% CI: 1.04–1.16, p = 0.000). Being a university graduate was significantly associated with decreased risk of depression (OR = 0.16, 95% CI: 0.02–0.94, p = 0.043).

Table VI

Multivariate analysis for demographic and clinical factors associated with anxiety and depression in patients with rheumatoid arthritis

VariableAnxietyDepression
FactorsRR (95% Cl)p-valueRR (95% Cl)p-value
Age0.99 (0.95–1.03)0.9911.02 (0.97–1.06)0.347
Body mass index1.02 (0.94–1.10)0.5621.05 (0.98–1.13)0.148
Disease duration0.96 (0.91–1.01)0.2000.99 (0.95–1.04)0.854
DAS281.37 (0.94–1.99)0.0951.31 (0.92–1.87)0.124
HAQ1.04 (1.00–1.09)0.0411.10 (1.04–1.16)0.000
Sex2.12 (0.49–9.07)0.3083.15 (0.94–10.61)0.063
Marital status1.31 (0.47–3.58)0.5990.52 (0.19–1.43)0.209
Education
 Uneducated0.2750.063
 Primary school1.17 (0.45–3.04)0.7351.27 (0.49–3.26)0.612
 High school1.05 (0.27–4.09)0.9340.55 (0.14–2.12)0.393
 University0.11 (0.00–1.37)0.0870.16 (0.02–0.94)0.043
Working status
 Retired0.5920.251
 Housewife1.27 (0.30–5.38)0.7411.04 (0.29–3.68)0.946
 White-collar worker2.00 (0.34–11.76)0.4423.48 (0.74–16.39)0.113
 Blue-collar worker3.25 (0.57–18.41)0.1813.62 (0.66–19.68)0.136
Comorbidities1.09 (0.47–2.50)0.8321.70 (0.79–3.64)0.171
Drugs1.77 (0.73–4.28)0.1991.40 (0.64–3.05)0.390

[i] DAS28 – Disease Activity Score with 28-joint counts, HAQ – Health Assessment Questionnaire.

Discussion

In this study, depression was determined in 50.3% of the patients and anxiety in 25.3%. The DAS28–ESR and HAQ scores were determined to be significantly higher in patients with depression and anxiety compared to those without. Depression was determined at a higher rate in females, those with a low level of education, and housewives. A university level of education was found to reduce the risk of depression. Anxiety was determined at higher rates in blue-collar workers.

In a systematic review of 21 studies that included 4,447 RA patients, the prevalence of depression was found to be 48% [13]. Altan et al. [14] reported a 54% rate of depression and 38% of anxiety in patients with RA, using cut-off points of 10 for anxiety and 7 for depression in the current study, according to HADS. Isık et al. [15] used HADS-A and HADS-D scales and found 41.5% depression and 13.4% anxiety in patients with RA, and these rates were found to be significantly higher than in the control group.

Different rates have been reported in various studies on the subject of the frequency of depression and anxiety, and these differences have been attributed to the design of the study, the scales used, and a possible relationship with geography and socioeconomic status [13].

In a study of German and Brazilian patients with RA, depression was determined at a higher rate in the Brazilians. High disease activity has been found to be associated with depression [16].

In an Italian cohort of 490 patients, depression was determined in 14.3%. There was determined to be a significant increase in the risk of depression with male sex, a high HAQ score, patient global evaluation and the use of antidepressants [17].

It is accepted that there is a two-way relationship between RA and depression [18, 19]. Depression is seen more in RA patients, and there has been found to be an increased risk of RA development in individuals with depression. There are increased proinflammatory cytokines in depression similar to in RA, and these cytokines are reduced with antidepressant treatment [20].

In patients with major depressive disorder, the risk of developing RA is increased by 38% compared to the normal population and the risk of RA development has been reported to be reduced in those using antidepressants compared to non-users [21]. Some anticytokine treatments used in RA have been determined to have positive effects on depression [22].

In a study by Ng et al. [23], depression and anxiety were found to be strongly correlated with DAS28–ESR. Depression has been determined to be significantly low in patients using etanercept. It has been reported that female sex and disease activity are strong predictors for depression.

In a study of 18,421 RA patients taking biological treatment, Matcham et al. [8] determined that the response to treatment in 1 year decreased by 20–40% when depression was present at the start of biological treatment.

Fragoulis et al. [24] examined 848 early RA patients and reported the prevalence of anxiety and depression to be 19.0% and 12.2%, respectively. A relationship was found between depression and anxiety, disease activity and a poor functional result. A relationship was also found between CRP levels and depression but not with anxiety.

In a meta-analysis, Zhang et al. [25] determined higher disease activity and lower quality of life in RA patients with depression compared to those without depression, but no difference was determined with respect to pain and functional disability. A low socioeconomic status, female sex, young age, and functional limitations have been reported to be factors associated with depression in RA patients [26]. Depression is generally associated with the severe form of RA [27].

Watad et al. [28] determined a higher rate of anxiety in RA patients than in a control group. Female sex, young age, smoking, and low socioeconomic status were reported to be independent factors associated with anxiety. In another study, low socioeconomic status and high DAS28 scores were determined to be associated with anxiety [24].

In our study, when bDMARD and csDMARD users were compared, no statistically significant difference was found in terms of anxiety and depression. In our study, we found a significant difference between the patients’ DAS28 and HAQ scores and the presence of anxiety and depression.

Similarly, in a study, bDMARDs and cDMARDs were not found to be superior to each other in terms of depression [29]. There is a need for further studies specifically investigating the effects of bDMARDs on depression and anxiety.

Depression was determined to be related to the global health score in a study of 464 RA patients, and anxiety was determined to be related to being married and functional disability [30]. In another study, it was reported that the presence of anxiety and depression in patients with RA can cause suicide and diminished quality of life and can worsen the prognosis of RA [31].

Study limitations

The study had a relatively low number of patients, a cross-sectional design, was conducted in a single center, and it lacked a control group.

In addition, the socioeconomic status of the patients was not examined. As the study included patients who were being followed up in a tertiary level center, the results may not be generalizable to all RA patients.

Conclusions

Depression and anxiety are observed extremely frequently in RA patients. It should be taken into consideration that this may affect the remission response of patients, the prognosis, and even mortality, and there should be collaboration with the psychiatry department in these cases.