Anti-tumor necrosis factor-α therapy is associated with less frequent mood and anxiety disorders in patients with rheumatoid arthritis


*Faruk Uguz, MD, Department of Psychiatry, Meram Faculty of Medicine, Selçuk University, Konya, Meram 42080, Turkey. Email:


Aims:  The purpose of the present study was to examine the current prevalence of mood and anxiety disorders, and factors related to mood and anxiety disorders in patients with rheumatoid arthritis (RA).

Method:  The study sample included 83 consecutive patients with RA who were admitted to a rheumatology outpatient clinic. Diagnoses of psychiatric disorders were determined using the Structured Clinical Interview for DSM-IV (SCID-I). To assess physical disability and disease activity, the Health Assessment Questionnaire and the Disease Activity Score, respectively, were used.

Results:  The prevalence of any mood or any anxiety disorder was 43.4%. The two most common psychiatric diagnoses were major depression (21.7%) and generalized anxiety disorder (16.9%). Mood and anxiety disorders were unrelated to sociodemographic features, disease-related factors, and medications for RA except anti-tumor necrosis factor-α (TNF-α). These disorders, however, were identified less frequently in patients with RA receiving anti-TNF-α drugs compared to patients who did not receive such medications.

Conclusion:  Patients with RA frequently have mood and anxiety disorders, and anti-TNF-α drugs may be useful for the mental status of these patients.

RHEUMATOID ARTHRITIS (RA) is a chronic progressive inflammatory systemic disease affecting primarily the joints,1,2 with a median prevalence rate of 0.33–1.07% according to geographic area.3 The disease is more frequent among women than men.3–5 Although RA may occur at any age, the majority of epidemiological data suggest that the disease commonly begins in the fifth decade of life.4,6 RA may exhibit systemic manifestations such as pulmonary, cardiovascular and nervous system involvement in addition to polyarthritis.7 The specific etiology is unknown, and it is a multifactorial disease, resulting from interactions among various factors (e.g. genetic, hormonal and infectious agents).4,6

In addition to physical disability, psychological distress and psychiatric disturbances appear to be frequent in patients with RA. It is suggested that subjects with RA have higher prevalence of depression and anxiety symptoms than healthy control subjects.8,9 The prevalence rate of depression in RA has been reported to vary in the range of 17.6–66.2% due to heterogenous assessment instruments (e.g. depressive symptom scales, clinical interviews, standardized research interviews).8,10–16 Although there are numerous studies on the prevalence of depression in patients with this medical illness, they mostly do not include the prevalence rate of major depressive disorder according to structured interviews based on DSM-IV criteria.17 Moreover, the data regarding prevalence of other DSM-IV mood disorders such as dysthymic disorder and bipolar disorder are limited in the literature.

Higher anxiety levels compared to healthy controls is another common psychiatric problem in patients with RA. Similar to depression, a varying prevalence rate has been reported in the wide range of 21–70% due to differences between study samples or instruments used.8,11,12,14 Also, some authors found that anxiety is more frequent than depression in patients with RA.14 Although anxiety symptoms have been widely examined, the number of reports on the prevalence of specific anxiety disorders is inadequate in this patient population.

Psychiatric comorbidity seems to be among the most important independent predictors of work disability in patients with RA.18 Depression has a negative impact on patients' physical function, their view of illness and satisfaction about current treatment of disease,19 and it increases the risk of mortality in these patients.10 In addition, studies have demonstrated that lifetime depressive symptoms are associated with an impairment in the quality of life of patients with RA.20,21 In contrast, psychological interventions such as cognitive behavioral therapy may be useful in the reduction of physical morbidity in addition to psychological morbidity in these patients.22

There were two aims for the present study: the first was to investigate the current prevalence of mood and anxiety disorders on the basis of a structured clinical interview; and second, to investigate the sociodemographic and clinical factors related to depressive and anxiety disorders in RA patients.


The study sample consisted of 83 consecutive patients with RA who were admitted to the Rheumatology Outpatient Clinic, Department of Physical Medicine and Rehabilitation, Meram Faculty of Medicine, Selcuk University in Konya, Turkey between March and September 2006. Illiteracy and cognitive incompetence, which make psychiatric interview difficult, concomitant severe medical illness (e.g. uncontrolled endocrine abnormalities, cardiovascular, neurological or pulmonary diseases), and usage of psychotropic medications within the last 4 weeks were exclusion criteria. The study was approved by the Ethics Committee of the Meram Medical School of Selçuk University.

The diagnosis of RA was determined according to the Criteria of American Rheumatism Association.23 Mood and anxiety disorders were ascertained by means of the Structured Clinical Interview for DSM-IV (SCID-I).24 To determine physical disability and disease activity, we used the Health Assessment Questionnaire (HAQ)25 and the Disease Activity Score (DAS),26 respectively.

Initially, objectives and procedures of the study were explained to all participants, and written informed consent was obtained from them. After rheumatological assessment the patients were referred to the psychiatric outpatient clinic of the same hospital. Psychiatric interviews were carried out by a psychiatrist who was blinded to HAQ and DAS scores, and to pharmacological therapy for RA.

Data analysis was conducted using SPSS version 13.0 (SPSS, Chicago, IL, USA). For comparisons between the groups of patients with and without any depressive disorder or anxiety disorder, Mann–Whitney U-test was used for continuous variables, and χ2 test was used for categorical variables. When necessary, Fisher's exact test was used for categorical variables. Statistical significance was accepted as P < 0.05.


Demographic and clinical characteristics of the participants are presented in Table 1. Mean subject age was 49.9 ± 13.1 years (range, 19–76 years), and disease duration was 9.0 ± 8.3 years (range, 1–40 years). The patients were mostly female (89.2%), married (80.7%) and primary school graduates (89.2%). All except four were taking at least one medication for RA.

Table 1.  Subject characteristics
Parametern (%)
  1. TNF, tumor necrosis factor.

Age (years), mean ± SD49.91 ± 13.06
 Female74 (89.2)
 Male9 (10.8)
Educational level 
 Elementary school74 (89.2)
 Secondary or high school5 (6.0)
 University4 (4.8)
Marital status 
 Single3 (3.6)
 Married67 (80.7)
 Widowed, divorced or separated13 (15.7)
Employment status 
 Employed11 (13.3)
 Unemployed72 (86.7)
 Non-steroidal anti-inflammatory drugs77 (92.8)
 Corticosteroids50 (60.2)
 Metothrexate66 (79.5)
 Sulfosalazine38 (45.8)
 Chloroquine15 (18.1)
 TNF-α antagonists 
 Etanercept8 (9.6)
 Inflixumab8 (9.6)

Thirty-six subjects (43.4%) met the criteria for at least one mood or anxiety disorder according to SCID-I. The number of psychiatric disorders was one in 23 (27.7%), two in 10 (12.0%), and three in 3 (3.6%) of 83 patients with RA. The prevalence of any depressive disorder and any anxiety disorder was 34.9% (n = 29) and 22.9% (n = 19), respectively. Among the evaluated psychiatric disorders, major depression (n = 18, 21.7%) was the most prevalent. Generalized anxiety disorder (n = 14, 16.9%) and dysthymic disorder (n = 11, 13.3%) were other disorders frequently observed. None of the patients had a diagnosis of bipolar disorder. Prevalence rates for the remaining disorders are as follows: specific phobia, 6.0%; panic disorder, obsessive–compulsive disorder, agoraphobia, and posttraumatic stress disorder, 1.2% (Table 2).

Table 2.  Current prevalence of mood and anxiety disorders in RA patients (n = 83)
 n (%)
  1. RA, rheumatoid arthritis.

Major depression18 (21.8)
Dysthymic disorder11 (13.3)
Bipolar disorder– (–)
Generalized anxiety disorder14 (16.9)
Panic disorder1 (1.2)
Obsessive-compulsive disorder1 (1.2)
Specific phobia5 (6.0)
Social phobia1 (1.2)
Agoraphobia1 (1.2)
Post-traumatic stress disorder1 (1.2)
Any mood or anxiety disorder36 (43.4)
Any mood disorder29 (34.9)
Any anxiety disorder19 (22.9)

Statistical analysis indicated that any depressive disorder and anxiety disorders were unrelated to sex, age, marital status, educational level, employment status, disease duration, HAQ and DAS scores, or medications except anti-tumor necrosis factor-α (TNF-α) drugs. Patients with a depressive (3.4% vs 27.8%, Fisher exact test, P = 0.008) or an anxiety disorder (0% vs 25%, Fisher exact test, P = 0.017) were taking anti-TNF-α drugs significantly less frequently (Table 3). While among patients receiving anti-TNF-α drugs (n = 16) for RA, only one had a depressive disorder (6.3%) and none had an anxiety disorder; of RA patients not receiving these drugs, 28 (41.8%) had a depressive disorder and 19 (28.4%) had an anxiety disorder.

Table 3.  Subject characteristics vs presence of depressive or anxiety disorder (n, %)
 Presence of depressive disorderPresence of anxiety disorder
n = 29
n = 54
n = 19
n = 64
  • Fisher exact test;

  • χ2 test;

  • §

    Mann–Whitney U-test.

  • DAS, Disease Activity Score; HAQ, Health Assessment Questionnaire; SD, standard deviation; TNF-α, tumor necrosis factor-α.

 Female27 (93.1)47 (87.0)0.48419 (100)55 (85.9)0.110
Marital status  0.310  0.907
 Single0 (0)3 (5.6) 1 (5.3)2 (3.1) 
 Married23 (79.3)44 (81.5) 15 (78.9)52 (81.3) 
 Widowed, divorced or separated6 (20.7)7 (13.0) 3 (15.8)10 (15.6) 
Educational level  0.233  0.983
 Elementary school28 (96.6)46 (85.2) 17 (89.5)57 (89.1) 
 Secondary or high school1 (3.4)4 (7.4) 1 (5.3)4 (6.3) 
 University0 (0)4 (7.4) 1 (5.3)3 (4.7) 
Employment status  0.087  0.443
 Unemployed28 (96.6)44 (81.5)  18 (94.7)54 (84.4)
 Non-steroidal anti-inflammatory drugs27 (93.1)50 (92.6)1.00017 (89.5)60 (93.8)0.616
 Corticosteroids16 (55.2)34 (63.0)0.63912 (63.2)38 (59.4)1.000
 Methotrexate22 (75.9)44 (81.5)0.57716 (84.2)50 (78.1)0.750
 Sulfasalazine14 (48.3)24 (44.4)0.81911 (57.9)27 (42.2)0.296
 Chloroquine4 (13.8)11 (20.4)0.5596 (31.6)9 (14.1)0.097
 TNF-α antagonists1 (3.4)15 (27.8)0.0080 (0)16 (25.0)0.017
Disease duration (years), mean ± SD7.58 ± 7.649.74 ± 8.550.192§8.42 ± 9.529.15 ± 7.920.528§
Age (years), mean ± SD48.79 ± 12.9950.51 ± 13.170.727§52.68 ± 14.6449.09 ± 12.550.105§
HAQ1.48 ± 0.841.56 ± 0.840.702§1.53 ± 0.931.54 ± 0.810.896§
DAS4.80 ± 1.725.25 ± 1.390.382§4.90 ± 1.775.15 ± 1.440.696§


This study presents three main results: (i) approximately half of the patients with RA (43.4%) had a depressive or an anxiety disorder during psychiatric assessment; (ii) the two most common specific psychiatric disorders were major depression (21.7%) and generalized anxiety disorder (16.9%) in patients with RA; and (iii) perhaps, more importantly, the patients with RA receiving anti-TNF-α therapy had less frequent prevalence of any mood or anxiety disorders.

In the present study we found the current prevalence of any depressive and any anxiety disorder to be 34.9% and 22.9%, respectively. These rates are higher than those (3.9–6.5% for any mood disorder, 6.8%–9.7% for any anxiety disorder) estimated in the general population.27–29 In patients with RA, depressive disorders have a prevalence of approximately 20–40% according to most of studies based on clinical interviews,15,16,30,31 which is consistent with the present results, although some authors reported a considerably higher frequency of these disorders.14

Compared with depressive disorders, the available data regarding prevalence of anxiety disorders are limited. El-Miedany and El-Rasheed reported a 70% prevalence of anxiety in an interview-based study.14 Recently, similar to the present observations, Arnold et al. reported a 25% lifetime prevalence of any anxiety disorder according to DSM-IV criteria17 using a structured clinical interview.30 In the present subjects the most common anxiety disorder was generalized anxiety disorder, with a prevalence rate of 16.9%. Arnold et al. found a prevalence rate of generalized anxiety disorder of 7.5% in patients with RA, but the most common anxiety disorder was specific phobia in their sample.30 Contrary to the prevalence rate for any anxiety disorder, there are differences between the current study and that of Arnold et al.30 with regard to frequencies of other specific anxiety disorders in addition to generalized anxiety disorder. While the present rate of panic disorder (1.2% vs 7.5%), post-traumatic stress disorder (1.2% vs 5.0%), social phobia (1.2% vs 5.0%), and specific phobia (6.0% vs 12.5%) were lower, the rate of obsessive–compulsive disorder (1.2% vs 0%) and agoraphobia (1.2% vs 0%) was slightly higher compared to that in the Arnold et al. study.30 These discrepancies could be due to differences between studied samples and evaluated time period (current vs lifetime).

In the present sample any depressive disorder was found to be unrelated to demographic characteristics, disease duration, disease activity and physical disability. Similarly, Dickens et al. found no association between a diagnosis of depression and demographic and disease-related factors such as disease duration, disease activity and functional impairment.15 Results of the present study are consistent with those of a number of previous studies in terms of association of depression with age, sex, disease duration,10,13,16 and education level.16 In contrast, some authors reported that depressed patients are less likely to be married,10,16,32 have a lower education level,10,32 and a longer disease duration.32 In addition, a number of studies have reported an association between depression and physical disability.10,13,16,32 These heterogenous results may be due to methodological variations including study sample, interview or assessment instruments, and diagnostic criteria.

Similar to the results for any depressive disorder, we found no association between any anxiety disorder and demographic and clinical factors. To date there has been no published study reporting factors related to any anxiety disorder according to DSM-IV criteria17 in the literature, although some authors noted a connection of anxiety with poor physical function12 and shorter disease duration.8,14

TNF-α is a dominant pro-inflammatory cytokine target for therapy of RA. Recently, anti-TNF-α drugs such as infliximab and etanercept have been approved for treatment for RA.33 Perhaps the most important finding of the current study is that the patients with RA taking anti-TNF-α drugs had a significantly lower frequency of any depressive and anxiety disorders compared to patients not receiving these drugs. This finding suggests the existence of a biological connection between depressive and anxiety disorders and RA, when considered with the other results of the present study: These disorders are not associated with demographic and clinical characteristics of patients with RA.

TNF-α is one of the most important cytokines in the pathogenesis of RA, and patients with RA have an elevated serum concentrations of this cytokine that is correlated with disease activity.2,34 In addition, TNF-α seems to be associated with psychiatric disturbances, particularly depression in RA. Tuglu et al. reported that compared to controls, major depressed patients had significantly higher serum TNF-α levels, which decreased to levels comparable with those of the controls after antidepressant treatment.35 O'Brien et al. found that currently depressed and selective serotonin re-uptake inhibitor (SSRI)-resistant patients had higher TNF-α levels compared to previously SSRI-resistant currently euthymic subjects and healthy subjects, and healthy subjects and the euthymic group who had previously not responded to SSRI had similar TNF-α levels.36 Increased cytokine concentrations such as that for TNF-α also appears to be associated with the occurrence of depressive symptoms during interferon-α treatment.37 Additionally, some authors reported augmented TNF-α release in anxious women38 and a positive correlation between levels of TNF-α and degree of anxiety symptoms.39 Pro-inflammatory cytokines may affect central monoamine turnover.40 Moreover, some authors hypothesized that decreases in brain-derived TNF-α due to antidepressant treatment regulate intracellular signaling pathways of α2-adrenergic receptor.41 Consequently, anti-TNF-α drugs may reduce anxiety and depressive symptoms by affecting elevated TNF-α-induced changes in neurotransmitter release or functions.

The main advantage of the present study was that the physician conducting the psychiatric interviews was blinded to patient treatments. The main limitations were the cross-sectional design and relatively small sample size. In addition, we did not examine personal or family histories of mental disorders.

In conclusion, the present results suggest that patients with RA who were admitted to a rheumatology outpatient clinic have frequent psychiatric disturbances, and anti-TNF-α therapies may positively affect these patients' mental status. Given that suppression of pro-inflammatory cytokines does not occur in depressed patients who do not respond to SSRI,36 additional anti-TNF-α drugs may be effective in the reduction of depressive symptoms in psychiatric patients with treatment-resistant depression. But prospective controlled studies should be performed for association between anti-TNF-α drugs and psychiatric disorders.