Mental health in patients with rheumatoid arthritis and axial spondyloarthritis: a cross-sectional, case–control tertiary centre study from Czechia Authors: Marketa Husakova, Veronika Balajkova, Karel Pavelka, Ladislav Senolt, Karolina Kuchaříková, Pavel Mohr, Jan Kotrs, Marta Olejárová Published: 2025 ABSTRACT Abstract AbstractObjectivesTo evaluate the mental health burden in rheumatoid arthritis (RA) and axial spondyloarthritis (axSpA) using diagnostic and self-reported tools and to examine its associations with current disease activity, patient-reported outcomes and barriers to appropriate care.DesignSingle-centre, cross-sectional, case–control study.SettingRheumatology centre of a tertiary care hospital, serving as a referral clinic with outpatient and inpatient care in Czech Republic.Participants233 patients with rheumatic diseases (113 RA, 120 axSpA) and 170 healthy controls (HC).Outcome measuresMental disorders (MD) were assessed through a structured psychiatric interview using the International Neuropsychiatric Interview (Mini International Neuropsychiatric Interview) administered by a trained professional and by self-reported questionnaires including the Beck Depression Inventory-II (BDI-II) and Beck Anxiety Inventory (BAI). Disease activity was evaluated with the Disease Activity Score-28 with C-reactive protein (DAS28-CRP) for RA and the Ankylosing Spondylitis Disease Activity Score with CRP (ASDAS-CRP) for axSpA, alongside patient-reported outcomes (PROs).ResultsAt least one MD was present in 24.8% of RA, 31.7% of axSpA and 7.0% of HC (p<0.001), driven mainly by major depression, dysthymia and suicidal ideation. Thresholds for depression and anxiety were significantly more frequent in RA and axSpA than HC (BDI-II: 18.6% vs 4.0% and 17.5% vs 1.7%; BAI: 38.9% vs 13.9% and 26.7% vs 11.0%; all p<0.001). Patients with MD had higher disease activity (RA: DAS28-CRP 3.12 vs 2.68, p<0.05; axSpA: ASDAS-CRP 2.39 vs 1.63, p<0.001). The presence of MD was associated with an increase of 0.40 in DAS28-CRP and 0.24 in ASDAS-CRP, although a 10-point increase in BDI-II was required for a similar association. Most patients with MD (75.0% RA, 89.6% axSpA) declined psychiatric counselling despite poorer concurrent outcomes in disease activity and PROs.ConclusionsMental disorders in RA and axSpA are closely associated with higher disease activity and unfavourable PROs, while access to and acceptance of psychiatric care remain markedly insufficient. Systematic integration of mental health assessment and management into rheumatology practice is strongly warranted.Keywords: Anxiety disorders, Depression & mood disorders, RHEUMATOLOGY STRENGTHS AND LIMITATIONS OF THIS STUDY The use of both structured psychiatric interviews (Mini International Neuropsychiatric Interview) and validated self-reported questionnaires (Beck Depression Inventory-II and Beck Anxiety Inventory) for robust assessment of mental disorders. The use of both structured psychiatric interviews (Mini International Neuropsychiatric Interview) and validated self-reported questionnaires (Beck Depression Inventory-II and Beck Anxiety Inventory) for robust assessment of mental disorders. Inclusion of a case–control design with both patients with rheumatoid arthritis (RA), axial spondyloarthritis (axSpA) and healthy controls. Inclusion of a case–control design with both patients with rheumatoid arthritis (RA), axial spondyloarthritis (axSpA) and healthy controls. Comprehensive evaluation of associations between mental disorders, disease activity (Disease Activity Score-28 with C-reactive protein (CRP) and Ankylosing Spondylitis Disease Activity Score with CRP) and patient-reported outcomes. Comprehensive evaluation of associations between mental disorders, disease activity (Disease Activity Score-28 with C-reactive protein (CRP) and Ankylosing Spondylitis Disease Activity Score with CRP) and patient-reported outcomes. Cross-sectional design prevents inference of causality. Cross-sectional design prevents inference of causality. Cohort characteristics may limit generalisability, as patients with very early or long-standing RA or axSpA were not included. Cohort characteristics may limit generalisability, as patients with very early or long-standing RA or axSpA were not included. Introduction Rheumatoid arthritis (RA) and axial spondyloarthritis (axSpA) impose a multidimensional burden, including impaired quality of life, reduced mobility, complications from anti-inflammatory therapy and frequent comorbidities—particularly mental health disorders (MD).1 2 Affective disorders (AD) are among the most common MD in RA3 and axSpA,4 though their reported prevalence varies depending on diagnostic methods, geographic context and patient characteristics. Psychiatric disorders in RA and axSpA are often identified through population-based registries using International Classification of Diseases (ICD) or Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnostic codes; however, such studies provide limited insight into disease-specific features or patient-reported outcomes (PROs) and include the full spectrum of patients with rheumatic diseases, from mild to severe cases. In contrast, self-administered questionnaires enable the assessment of depression and anxiety severity, as well as their associations with disease activity and PROs such as pain, fatigue, disability and quality of life. Nevertheless, the choice of tools and the cut-off points used to define depression or anxiety differ across studies, which may result in varying prevalence estimates.5 RA is a chronic autoimmune disease characterised by inflammation and subsequent damage to synovial joints, with associated systemic implications.6 A meta-analysis based on registry data reported pooled prevalence rates of major depressive disorder (MDD) at 16.8% (95% CI 10 to 24%) and dysthymia at 18.7% (95% CI −2–39%) among patients with RA.7 Data from the international cross-sectional study on comorbidities in RA indicated that depression was the most frequent comorbidity (15.0%, 95% CI 13.8 to 16.1%), though rates varied widely between countries. Other studies have shown MDD prevalence ranging from 2% to 33% across different populations.3 However, when using self-reported tools to identify depression, mild depressive symptoms were identified in up to 48% of patients with RA.7 In a Scottish early RA cohort, the prevalence of depression and anxiety was markedly higher than in healthy controls (HC; 12.2% and 19.0% vs 1.7%, respectively),8 and psychiatric comorbidity has been associated with worse disease outcomes, particularly in patients with depression.9 10 Recent data from the Rotterdam Early Arthritis CoHort further support a strong association between depression and disease activity: patients with depression had a higher number of tender joints, and after 2 years of follow-up, the likelihood of achieving remission was substantially lower than in those without depressive symptoms (OR 0.45, 95% CI: 0.25 to 0.80).11 AxSpA is characterised by inflammation of the axial skeleton, peripheral joints and entheses, leading to the development of bone bridges and subsequent limitations in mobility.12 Since 2009, axSpA has been classified into non-radiographic (nr-) and radiographic (r-) forms (formerly ankylosing spondylitis, AS).13 Both forms share similar disease activity and treatment response, but impaired mobility and elevated inflammatory markers are more likely to occur in r-axSpA.14 In 2008, increased rates of affective, anxiety and personality disorders in AS were reported in a Swedish registry study,15 and the increased prevalence of psychiatric comorbidity was recently identified to be higher in AS than the general population (13.8% vs 9.8 %, p<0.001), particularly with increased risk for MDD (OR 1.60 (95 % CI 1.43 to 1.79).16 In addition, MDD and anxiety disorders have been identified in 10.6% and 15.6% patients with axSpA, respectively.17 The meta-analysis of studies using self-reported questionnaires indicated a prevalence of mild depression in 38% (95% CI 30 to 45%) and moderate depression in 15% (95% CI 6 to 5%) of patients with axSpA. Importantly, patients with concomitant depressive symptoms had substantially worse disease activity and functional indices and achieved remission less frequently than those without depressive symptoms.5 Similar to RA, in axSpA registries, patients with concomitant depressive symptoms achieved lower rates of remission or lower disease activity compared with patients without depressive symptoms.18 Given these findings, our study aimed to assess current and past psychiatric comorbidities in RA and axSpA using both a structured diagnostic interview and validated self-reported questionnaires. In addition, we examined the impact of mental health burden on disease outcomes and explored whether patients with psychiatric symptoms sought appropriate care. Methods Study design and participants This cross-sectional, single-centre study was conducted at the outpatient department of the Institute of Rheumatology, Prague (Czech Republic), between June 2021 and December 2023. Patients with rheumatic diseases suffering from RA or axSpA and HC (without inflammatory disorders) were invited to participate. All participants provided written informed consent prior to inclusion, in accordance with the approval of the Ethics Committee of the Institute of Rheumatology, Prague (approval number 5704/2021). Following consent, participants were scheduled for structured interviews, self-reported questionnaires and sample collection. The study complied with the Declaration of Helsinki and followed Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines.19 Eligible participants were aged 18–65 and able to complete questionnaires independently. For RA and axSpA, a disease duration of 2–15 years was required to reduce bias from early disease phases.8 HCs were recruited through institutional advertisements, among accompanying persons, university students and staff. HCs were randomly frequency-matched to patient groups by age and sex. Exclusion criteria included conditions or medications that could affect mental health (see online supplemental table 1). A total sample size of 113 was calculated to achieve 90% statistical power, based on expected depression rates of 29% in RA,9 25.6% in axSpA20 and 5.5% for mood disorders in the general Czech population.21 Collected data included age, sex, family and housing status, body mass index, comorbidities (online supplemental table 2), pharmacotherapy, psychiatric/psychological treatment, addictions (smoking, alcohol, psychotropic substance use) and physical activity according to WHO weekly recommendations (online supplemental table 3). Measurements Patients characteristics. Patients with RA fulfilled either the 1987 ACR criteria22 or the 2010 RA Classification criteria.23 Collected variables included disease duration, presence of anti-citrullinated peptide antibodies and rheumatoid factor, extra-articular manifestations, joint deformities, current and past pharmacotherapy, radiographic features (including Steinbrocker stage) and disease activity assessed by the Disease Activity Score-28 with C-reactive protein (DAS28-CRP). All patients with axSpA met the ASAS classification criteria.13 Data included symptom onset (inflammatory back pain (IBP) or extra-musculoskeletal manifestations (EMMs)), disease duration, presence of IBP or EMMs, peripheral SpA signs, sacroiliac joint radiographic findings and treatment history. Disease activity and function were evaluated using Bath Ankylosing Spondylitis (BAS) indexes: Metrology Index (BASMI), Functional Index (BASFI), Disease Activity Index (BASDAI) and Ankylosing Spondylitis Disease Activity Score with CRP (ASDAS-CRP).24 Blood samples for CRP were obtained from patients with RA and those with axSpA within 3 days of questionnaire completion. Self-reported questionnaires for all participantsAll participants completed questionnaires covering demographic data, occupation, the EQ-5D for quality of life and the Health Assessment Questionnaire (HAQ).25 26 Global health was rated by both patients and physicians using a 0–100 visual analogue scale. Mental health assessment. Mental disorders were assessed using the Czech-validated version of the Mini International Neuropsychiatric Interview (M.I.N.I.), a structured diagnostic interview aligned with DSM-IV and ICD-10 criteria.27 The M.I.N.I. facilitates the diagnostic process in both psychiatric and primary care settings.28 Briefly, M.I.N.I. is organised into lettered modules (commonly labelled A–P), each corresponding to a specific diagnostic category. Each module begins with screening items that reflect the principal diagnostic criteria for the disorder. Positive screening responses are followed by clarifying questions specific to that condition; all items are answered dichotomously (‘yes’/‘no’), and one or more diagnostic outcome fields at the end of each module indicate whether the criteria for disease have been met. Interviewers are required to read all questions verbatim, with bracketed phrases provided to help explain symptoms to the participant when necessary. In our study, we used the following modules: A, current or lifetime MDD; B, dysthymia; C, suicidality; D, (hypo)manic episodes (HME); E, panic disorder (PD); L, psychotic disorders; and O, generalised anxiety disorder (GAD). Interviews were conducted by three trained clinicians (MH, MO, VB). Depressive and anxiety symptoms were evaluated with Czech-validated versions of the Beck Depression Inventory-II (BDI-II) and Beck Anxiety Inventory (BAI).29,32 Each consists of 21 items rated on a 0–3 scale. Scores indicate the following categories: (1) BDI-II: no depression (0–13), mild (14–19), moderate (20–28) and severe (29–63) and (2) BAI: no anxiety (0–9), mild (10–18), moderate (19–29) and severe (30–63).32 33 Psychiatric examination procedure. All participants were offered psychiatric counselling. Responses were categorised as follows: ‘not needed’ (for those without any positive finding in the M.I.N.I.) or, among those with at least one identified mental disorder, as ‘already receiving psychiatric care’, ‘positive response’, ‘declined offer’ or ‘unwilling to respond’. Statistical analysis All statistical analyses were performed on complete case data. Controls were used for descriptive and case–control comparisons only; all case-only models were fit in the patient cohort. Continuous variables were summarised as means with 95% CIs or medians with IQRs, depending on distribution. For univariate subgroup analyses, the t-test was used for normally distributed variables, while the Wilcoxon exact test or Mann–Whitney test was applied for non-parametric data. Correlation analyses were conducted where appropriate. A p value <0.05 was considered statistically significant. To identify factors associated with a positive M.I.N.I. outcome, a multiple logistic regression model was used where any positive M.I.N.I. category result was treated as one while none as 0.34 A backwards stepwise regression approach (with pseudo R² re-evaluation at each step) was chosen for its interpretability and suitability in exploratory settings, helping to select a subset of independent variables that best explained the presence of a mental disorder. Final model exploratory value was assessed using McFadden’s pseudo R² (R=0.177), indicating modest explanatory power, and Cohen’s kappa coefficient (κ=0.335), reflecting fair agreement beyond chance. To explore feature contributions towards CRP values (DAS 28 for RA and ASDAS for axSpA), we have employed linear regression modelling. Both models were constructed using backwards stepwise selection (with R² re-evaluation at each step). Mild heteroskedasticity was detected in both; thus, we have refitted them with weights computed as the inverse of the squared residuals as a commonly used heuristic (WLS). Analyses were conducted using Python libraries: statsmodels (v0.14.1) and scikit-learn (v1.3.2).35 36 Coefficient visualisation was done using forestplot (v0.3.3) and data processing with pandas (v2.1.4) and numpy (v1.26.3).37 Patient and public involvement statement Patients were first involved during the preparatory stage of the study, when three patients with axSpA and three healthy volunteers piloted the questionnaires on a voluntary basis to assess the time burden of completion. Patients and the public were not involved in developing the research questions, outcome measures or study design. During recruitment, the Czech patient organisation REVMA-LIGA was informed about the study and invited to distribute the advertisement among its members. Plans for dissemination include sharing the main study results with REVMA-LIGA and KLUB BECHTĚREVIKŮ, which support patients with RA and axSpA, respectively, once the primary findings are published. Results Demographic data Tables1 2 present the demographic and clinical characteristics of patients with RA and axSpA, as well as HC who completed all study procedures. In total, 233 patients with rheumatic diseases—113 with RA and 120 with axSpA—out of 248 meeting the inclusion criteria, completed all study assessments. 170 individuals were recruited as HC, from which age- and sex-frequency-matched counterparts were randomly selected for the axSpA (n=118) and RA (n=101) groups, without applying a fixed matching ratio. 15 patients were excluded due to incomplete questionnaire responses, and two withdrew consent during questionnaire completion. Table 1Demographic data of patients with rheumatoid arthritis, axial spondyloarthritis and healthy controlsFeatureRAHCRA vs HCaxSpAHCAxSpA vs HCRA vs AxSpAn=113n=101n=120n=118Age, median (IQR)51.00 (46.00–58.00)51.00 (45.00–58.00)0.38042.50 (34.00–49.00)44.50 (30.25–52.75)0.782<0.001Sex, female/male, %82.30/17.7080.20/19.800.69638.14/62.5038.14/61.860.920<0.001BMI, <30/>30, %75.22/24.7880.20/19.800.38481.67/18.3378.81/21.190.5820.611Marital status Single/divorced/widowed, %77.6874.260.56271.4364.410.2490.277 Married/partner, %22.3225.740.56228.5735.590.2490.277Number of adults in household >1/=1, %84.07/15.9376.24/26.760.15580.00/20.0079.66/20.340.9480.420City population <20 000, % (n)55.7531.68<0.000143.3036.440.0760.090 <100 000, % (n)29.2058.42<0.000120.0012.710.0760.090 >100 000, % (n)15.049.90<0.000136.6750.850.0760.090Work ability Full time/part time, %66.9788.12<0.0185.9382.200.149<0.001 Sick leave/unemployed/dispension, %17.434.95<0.019.175.080.149<0.001 Student/parental-leave/pension, %15.606.93<0.015.0012.710.149<0.001Comorbidities Without/up to two selected dis./ >two selected dis.33.93/48.21/17.8638.61/48.51/12.870.30750.00/44.17/5.8353.39/38.98/7.630.843<0.01Quality of life and function HAQ, median (IQR)0.38 (0.00–1.38)0.00 (0.00–0.12)<0.00010.25 (0.00–0.75)0.00 (0.00–0.00)<0.0001<0.01 5D-EQ, median (IQR)0.73 (0.59–0.80)1.00 (0.80–1.00)<0.00010.76 (0.66–0.81)1.00 (0.89–1.00)<0.00010.845 GHA: patients, median (IQR)28.00 (17.00–40.00)6.00 (2.00–16.00)<0.000130.00 (13.00–50.00)4.00 (1.00–10.00)<0.00010.214 GHA: physician, median (IQR)25.00 (14.00–38.00)5.00 (1.00–10.00)<0.000126.00 (14.00–44.50)3.00 (0.00–8.75)<0.00010.308Prevalent psychiatric comorbidity Without/present, %84.96/15.0497.03/2.970.07795.00/5.0096.61/3.390.2370.146Physical activity according to WHO recommendation Yes/no, %54.87/45.1355.45/44.550.93369.17/30.8366.10/33.900.615<0.05Alcohol consumption (WHO limits) Lower/equivalent/higher, %88.50/10.62/0.8880.20/15.84/3.960.06477.50/20.83/1.6733.90/79.66/15.250.848<0.05Smoking Never/in past/current, %58.41/24.77/16.8262.38/26.73/10.890.78063.33/10.83/25.8762.71/21.18/16.140.072<0.05Psychoactive drug abuse Never/in the past/current, %86.73/0.88/0.8890.10/9.10/0.000.57678.33/20.0/1.6772.88/25.43/1.690.8480.414axSpA, axial spondyloarthritis; BMI, body mass index; 5DEQ, five-dimensional quality of life; GHA, global health assessment; HAQ, Health Assessment Questionnaire; HC, healthy controls; IQR, interquartile range; RA, rheumatoid arthritis. Demographic data of patients with rheumatoid arthritis, axial spondyloarthritis and healthy controls axSpA, axial spondyloarthritis; BMI, body mass index; 5DEQ, five-dimensional quality of life; GHA, global health assessment; HAQ, Health Assessment Questionnaire; HC, healthy controls; IQR, interquartile range; RA, rheumatoid arthritis. Table 2Disease characteristics of rheumatoid arthritis and axial spondyloarthritis RA (n=113)axSpA (n=120)Disease duration Since the first symptom, median (IQR)7.98 (4.22–13.70)12.22 (6.99–18.43) Since diagnosis, median (IQR)6.16 (2.98–12.44)6.89 (3.26–9.61)First symptoms of axSpA IBP, %NA72.50 Arthritis, %NA14.17 Uveitis, %NA10.00 Idiopathic bowel disease, %NA1.67 Dactylitis, %NA0.83 Enthesitis, %NA0.83Radiographic sacroiliitisNA64.46Immunological parameters Rheumatoid factors positivity, %71.68NA ACPA positivity, %69.03NA Anti-MCV positivity, %32.14NA HLA-B27 positivityNA90.00Disease activity CRP mg/l, median (IQR)2.40 (0.93–5.60)2.96 (1.52–7.35) DAS28-CRP, median (IQR)2.82 (2.09–3.44)NA ASDAS-CRP, median (IQR)NA1.86 (1.28–2.79) BASDAI, median (IQR)NA2.80 (1.40–4.40) BASFI, median (IQR)NA1.30 (0.20–3.12) BASMI, median (IQR)NA2.60 (2.00–3.40)Current therapy Without bDMARDs or tsDMARDs, %57.5277.50 bDMARDs or tsDMARDs, %42.4822.50 Without corticotherapy, %77.880.00 Current corticotherapy, %22.120.00ACPA, autoantibodies against citrullinated peptide; anti-MCV, autoantibodies against citrullinated vimentin; ASDAS-CRP, Ankylosing spondylitis disease activity score with CRP; axSpA, axial spondyloarthritis; BASDAI, Bath AS disease activity index; BASFI, Bath AS functional index; BASMI, Bath AS metrology index; b/ts DMARDs, b(biological), ts(targeted synthetic)DMARDs- disease modifying drugs; CRP, C-reactive peptide; DAS28-CRP, Disease activity score with CRP; IBP, inflammatory back pain; RA, rheumatoid arthritis. Disease characteristics of rheumatoid arthritis and axial spondyloarthritis ACPA, autoantibodies against citrullinated peptide; anti-MCV, autoantibodies against citrullinated vimentin; ASDAS-CRP, Ankylosing spondylitis disease activity score with CRP; axSpA, axial spondyloarthritis; BASDAI, Bath AS disease activity index; BASFI, Bath AS functional index; BASMI, Bath AS metrology index; b/ts DMARDs, b(biological), ts(targeted synthetic)DMARDs- disease modifying drugs; CRP, C-reactive peptide; DAS28-CRP, Disease activity score with CRP; IBP, inflammatory back pain; RA, rheumatoid arthritis. Mental distress in patients with RA and axSpA: evidence from structured interviews and self-reported questionnaires for depression and anxiety The mental health burden in RA and axSpA was confirmed by both the M.I.N.I. interview and self-reported questionnaires. According to the M.I.N.I., 24.78% of patients with RA and 31.67% those with axSpA met the criteria for at least one MD, significantly more than their respective control groups (table 3). While patients with RA and those with axSpA showed similar psychiatric profiles, both groups differed significantly from controls in rates of MDD, dysthymia and suicidal ideation; patients with axSpA also showed significantly more GAD (table 3). Table 3Variability of mental disorder, Beck Depression Inventory and Beck Anxiety Inventory in patients with rheumatoid arthritis, axial spondyloarthritis and healthy controls RAHCaxSpAHCRA vs HCAxSpA vs HCRA vs AxSpAn=113n=101n=120n=118M.I.N.I. Without any MD, % (n)75.22 (85)93.07 (94)68.33 (82)92.37 (109)<0.001<0.0001ns At least one MD, %24.786.9331.677.63 MDD, %7.080.0012.500.00<0.01<0.0001ns Dysthymia, % (n)*6.67 (7/105)1.89 (2/101)9.52 (10/105)1.64 (2/118)0.098<0.05ns Suicidality, %7.080.008.331.69<0.01<0.05ns HME, %0.880.000.000.00nsnsns PD, %8.853.968.333.99nsnsns Psychotic disorders, %0.000.000.000.00NANANA GAD, %6.190.999.240.85ns<0.01nsDepression (BDI-II) Without ([0–13), %81.4296.0482.5098.31<0.001<0.001ns Mild (14–19), %10.623.9612.501.69 Moderate (20–28), %6.190.004.170.00 Severe (29-63), %1.770.000.830.00Anxiety (BAI) Without ([0–9), %61.0686.1473.3388.98<0.0001<0.0010.059 Mild (10–18), %25.6612.8720.0011.02 Moderate (19–29), %11.500.994.170.00 Severe (30–63), %1.770.002.500.00*Only people negative for MDD were interviewed for dysthymia.axSpA, axial spondyloarthritis; BAI, Beck Anxiety Inventory; BDI-II, Beck Depression Inventory; GAD, generalised anxiety disorder; HC, healthy controls; HME, (hypo)manic episode; MD, mental health disorders; MDD, major depressive disorder; M.I.N.I, Mini International Neuropsychiatric Interview; NA, not analysed; ns, not significant; PD, panic disorder; RA, rheumatoid arthritis. Variability of mental disorder, Beck Depression Inventory and Beck Anxiety Inventory in patients with rheumatoid arthritis, axial spondyloarthritis and healthy controls Only people negative for MDD were interviewed for dysthymia. axSpA, axial spondyloarthritis; BAI, Beck Anxiety Inventory; BDI-II, Beck Depression Inventory; GAD, generalised anxiety disorder; HC, healthy controls; HME, (hypo)manic episode; MD, mental health disorders; MDD, major depressive disorder; M.I.N.I, Mini International Neuropsychiatric Interview; NA, not analysed; ns, not significant; PD, panic disorder; RA, rheumatoid arthritis. Mood disorders (MDD, dysthymia and HME) were present in 14.16% of patients with RA and 20.83% of those with axSpA, compared with 1.89% (p<0.01) and 1.64% (p<0.0001) in HC, respectively. Anxiety disorders (GAD+PD) were found in 14.16% (RA) and 15.00% (axSpA), vs 0.98% (p<0.05) and 1.68% (p<0.01) in their control groups. Self-reported outcomes mirrored diagnostic findings. Thresholds for depression were reached by 18.58% (RA) and 17.50% (axSpA), vs 3.96% and 1.69% of controls, respectively. Anxiety was reported by 38.93% (RA) and 26.67% (axSpA), compared with 13.86% and 11.02% of controls (all p<0.001). Patients with RA and axSpA were also more likely to experience moderate to severe symptoms of both depression and anxiety (table 3). Impact of mental burden on disease outcomes in RA and axSpA Patients with axSpA and at least one MD had higher disease activity, poorer functional scores and reduced quality of life compared with those without (table 4). Similarly, in RA, MD was associated with increased DAS28-CRP scores, particularly due to more tender joints. However, CRP levels, autoantibody status and radiographic findings did not differ significantly between M.I.N.I.-positive and M.I.N.I.-negative groups for either disease (table 4, (online supplemental table 4). Multivariable logistic regression models identified key factors associated with MD (figure 1a, b). In RA, alcohol consumption above WHO-recommended limits increased the risk of MD nearly fivefold. Additional contributors included higher anxiety and disease activity. In axSpA, anxiety was the strongest predictor of MD, while ASDAS-CRP and BASFI were not significant. In both models, cohabitation with at least one other adult reduced the likelihood of MD. Table 4Differences in clinical variables in patients with rheumatoid arthritis or axial spondyloarthritis suffering from mental health disorder RAaxSpAAny MD (M.I.N.I.)Without MD (M.I.N.I.)Any MD (M.I.N.I.)Without MD (M.I.N.I.)(n=28)(n=85)(n=38)(n=82)Age, median (IQR)53.00 (45.25–57.00)51.00 (46.00–58.00)ns43.50 (34.50–47.75)42.00 (34.00–49.00)nsSex: female/male, %82.14/17.8682.53/17.65ns44.74/55.2634.15/65.85nsBMI: <30/>30, %71.43/28.5776.47/23.53ns84.21/15.7980.49/19.51nsDisease activity CRP mg/l, median (IQR)2.45 (1.16–7.46)2.35 (0.92–5.00)ns3.50 (1.30–9.82)2.77 (1.53–6.61)ns DAS28-CRP, median (IQR)3.12 (2.62–3.90)2.68 (2.02–3.38)<0.05NANANA TJC, median (IQR)4.00 (1.75–6.25)2.00 (0.00–3.00)<0.05NANANA SJC, median (IQR)1.00 (0.00–2.75)0.00 (0.00–2.00)nsNANANA ASDAS-CRP, median (IQR)NANANA2.39 (1.78–3.39)1.63 (1.05–2.45)<0.001 BASDAI, median (IQR)NANANA4.05 (3.00–6.17)2.05 (1.10–3.90)<0.0001 BASFI, median (IQR)NANANA2.90 (1.42–4.78)0.60 (0.10–2.00)<0.0001 BASMI, median (IQR)NANANA2.70 (2.05–3.40)2.40 (2.00–3.15)nsQuality of life and function HAQ, median (IQR)0.75 (0.38–1.62)0.38 (0.00–1.12)ns0.62 (0.25–1.12)0.12 (0.00–0.62)<0.001 5D-EQ, median (IQR)0.71 (0.52–0.80)0.73 (0.59–0.85)0.0940.62 (0.52–0.72)0.80 (0.73–1.00)<0.0001 GHA: patients, median (IQR)30.50 (24.75–49.00)23.00 (13.00–39.00)0.07250.00 (31.75–68.00)21.00 (10.00–40.00)<0.0001 GHA: physician, median (IQR)30.00 (24.75–44.25)21.00 (10.00–37.00)<0.0544.00 (26.50–62.75)19.50 (11.00–32.00)<0.0001ASDAS-CRP, Ankylosing spondylitis disease activity score with CRP; axSpA, axial spondyloarthritis; BASDAI, Bath AS disease activity index; BASFI, Bath AS functional index; BASMI, Bath AS metrology index; BMI, body mass index; DAS28-CRP, Disease activity score with C-reactive protein; 5DEQ, five-dimensional quality of life; GHA, global health assessment; HAQ, Health Assessment Questionnaire; MD, mental health disorders; M.I.N.I, Mini international neuropsychiatric interview; NA, not analysed; RA, rheumatoid arthritis; SJC, swollen joint count; TJC, tender joint count. Differences in clinical variables in patients with rheumatoid arthritis or axial spondyloarthritis suffering from mental health disorder ASDAS-CRP, Ankylosing spondylitis disease activity score with CRP; axSpA, axial spondyloarthritis; BASDAI, Bath AS disease activity index; BASFI, Bath AS functional index; BASMI, Bath AS metrology index; BMI, body mass index; DAS28-CRP, Disease activity score with C-reactive protein; 5DEQ, five-dimensional quality of life; GHA, global health assessment; HAQ, Health Assessment Questionnaire; MD, mental health disorders; M.I.N.I, Mini international neuropsychiatric interview; NA, not analysed; RA, rheumatoid arthritis; SJC, swollen joint count; TJC, tender joint count. (a) Multivariable regression model to develop MD for RA. (b) Multivariable regression model to develop MD for axSpA. ASDAS–CRP, Ankylosing Spondylitis Disease Activity Score with C-reactive Protein; BAI, Beck Anxiety Inventory; BASFI, Bath Ankylosing Spondylitis Functional Index; DAS28-CRP, Disease Activity Score with C-reactive Protein; HAQ, Health Assessment Questionnaire. RA patients with depression or anxiety were older than those without these symptoms (54.00 (IQR 51.00–61.00) vs 50.50 (IQR 44.00–57.25) and 53.50 (IQR 48.75–60.25) vs 50.00 (IQR 44.00–56.00), respectively; both p<0.05; online supplemental table 5), but did not differ in disease duration. RA patients with depression or anxiety had poorer HAQ scores (1.75 (IQR 0.75–2.62) and 1.12 (IQR 0.50–1.88), respectively), lower 5D-EQ scores (0.52 (IQR 0.52–0.72) and 0.59 (IQR 0.52–0.72), respectively) and higher disease activity (DAS28-CRP 3.29 (IQR 2.76–4.97) and 3.14 (IQR 2.09–3.90), respectively) compared with those without mood disorders (HAQ: 0.38 (IQR 0.00–0.88) and 0.25 (IQR 0.00–0.62), respectively; both p<0.0001; 5D-EQ: 0.73 (IQR 0.69–0.85), p<0.01 and 0.59 (IQR 0.52–0.72), p<0.0001; DAS28-CRP: 2.68 (IQR 2.09–3.30) and 2.59 (IQR 2.12–3.18), respectively; both p<0.05; see online supplemental table 5). Furthermore, a higher proportion of RA patients without depression or anxiety were free of corticosteroids compared with those with depression or anxiety (82.60% vs 57.10% and 82.60% vs 70.40%, respectively; both p<0.05; see online supplemental table 5). In axSpA, poor outcomes in all PROs were found in people with depression or anxiety (see online supplemental table 5). Depressive and anxious axSpA exhibited higher disease activity (ASDAS-CRP 3.31 (IQR 2.47–3.90) and 2.67 (IQR 2.07–3.43), BASDAI 5.40 (IQR 3.90–6.70) and 4.95 (IQR 3.20–6.33), respectively) than those without ((ASDAS-CRP 1.75 (IQR 1.14–2.41) and 1.61 (IQR 1.03–2.43), BASDAI 2.40 (IQR 1.20–3.90) and 2.05 (IQR 1.10–3.73)), all p<0.0001 (see online supplemental table 5). Furthermore, patients with depressive axSpA had significantly higher serum CRP levels (9.30 (IQR 2.90–22.70)) compared with non-depressive patients (2.61 [IQR 1.40–5.83]), p<0.05 (see online supplemental table 5). Contribution of mental health burden on disease activity Significant positive correlation was found among all mental health parameters (presence of any MD, BDI-II and BAI) and disease activity indexes in both RA and axSpA, though stronger in axSpA. In RA, DAS28-CRP correlated with the presence of MD (r=0.20, p<0.05), BDI-II (r=0.32, p<0.001) and BAI (r=0.29, p<0.01). In axSpA, ASDAS-CRP showed stronger correlations: r=0.35 (MD), r=0.52 (BDI-II) and r=0.50 (BAI); all p<0.0001 (online supplemental figure 1A,B). To further explore these associations, linear regression models were used. Due to the strong correlation between BDI-II and BAI (r=0.69, p<0.0001), only BDI-II was included in the models. In RA, the presence of any MD increased DAS28-CRP by 0.4 points, an effect mirrored by a 10-point increase in BDI-II (table 5). In axSpA, after adjusting for age and sex, ASDAS-CRP increased by 0.24 points with MD presence and by up to 0.7 points with a 10-point increase in BDI-II (table 5). Table 5Linear regression model of the relationship between mental burden and disease activity in rheumatoid arthritis and axial spondyloarthritisCoef.SETP0.025–0.975RA: DAS 28 Sex -male0.00970.0380.2530.801−0.066–0.086 Age0.00460.0022.1610.0330.000–0.009 BDI-II0.04230.00130.959<0.0010.040–0.045 Presence of any MD (M.I.N.I.)0.37020.02415.198<0.0010.322–0.418 Intercept2.30420.10222.653<0.0012.103–2.506axSpA - ASDAS-CRP Sex, male0.18230.0444.157<0.0010.095 - 0–269 Age0.02100.0038.171<0.0010.016–0.026 BDI-II0.07340.00612.816<0.0010.062–0.085 Presence of any MD (M.I.N.I.)0.24110.0514.729<0.0010.140–0.342 Intercept0.47950.1174.113<0.0010.249–0.710ASDAS-CRP, Ankylosing spondylitis disease activity score with C-reactive protein; axSpA, axial spondyloarthritis; BDI-II, Beck Depression Inventory; Coef, coefficient; DAS28, Disease activity score; MD, mental health disorders; M.I.N.I, Mini International Neuropsychiatric Interview; RA, rheumatoid arthritis; SE, standard error. Linear regression model of the relationship between mental burden and disease activity in rheumatoid arthritis and axial spondyloarthritis ASDAS-CRP, Ankylosing spondylitis disease activity score with C-reactive protein; axSpA, axial spondyloarthritis; BDI-II, Beck Depression Inventory; Coef, coefficient; DAS28, Disease activity score; MD, mental health disorders; M.I.N.I, Mini International Neuropsychiatric Interview; RA, rheumatoid arthritis; SE, standard error. Psychological care for people with rheumatic diseases and mental health burden Psychiatric counselling was offered to all 72 individuals with a diagnosed MD, yet most patients with RA or axSpA—as well as HC—declined the offer (table 6). Only a minority had previously sought mental healthcare. In RA, those who declined psychiatric care had higher DAS28-CRP scores (3.12 (IQR 2.52–3.74) vs 2.68 (IQR 2.02–3.38), p=0.080) and were more likely to meet the threshold for anxiety (BA I≥10: 55.00% vs 30.60%, p=0.060), compared with individuals without MD. Table 6Reflection of psychiatric counselling in people with detected mental health disorderPresence of any MD (M.I.N.I.)RAaxSpAControlsn=28n=38n=13Psychiatric counselling and care already provided n (%)5 (17.85)3 (7.89)3 (23.07)Psychiatric counselling offeredAccepted n (%)2 (7.14)1 (2.63)1 (7.70)Declined n (%)*17 (60.71)30 (78.95)9 (69.23)Without any response to psychiatric counselling n (%)4 (14.30)4 (10.53)0Controls - presence of any MD of all 170 healthy individuals recruited.*Reasons were as follows: postpone, problems in past and not necessary to solve it currently, belief that professional help is not needed.axSpA, axial spondyloarthritis; MD, mental health disorders; M.I.N.I, Mini International Neuropsychiatric Interview; RA, rheumatoid arthritis. Reflection of psychiatric counselling in people with detected mental health disorder Controls - presence of any MD of all 170 healthy individuals recruited. Reasons were as follows: postpone, problems in past and not necessary to solve it currently, belief that professional help is not needed. axSpA, axial spondyloarthritis; MD, mental health disorders; M.I.N.I, Mini International Neuropsychiatric Interview; RA, rheumatoid arthritis. In axSpA, individuals who declined care had significantly worse clinical outcomes than those without MD. They showed higher disease activity (ASDAS-CRP>2.1: 60.60% vs 31.70%, p<0.001; BASDAI>4.0: 54.50% vs 24.40%, p<0.01), worse function (BASFI: 3.10 [IQR 1.50–5.00] vs 0.60 [IQR 0.10–2.00], p<0.001) and lower quality of life (EQ-5D: 0.62 [IQR 0.52–0.72] vs 0.80 [IQR 0.73–1.00], p<0.0001). They were also less likely to be free of depression or anxiety symptoms (BDI-II<14: 54.50% vs 95.10%, p<0.001; BAI<10: 30.30% vs 90.25%, p<0.0001). Discussion This study is the first to integrate both self-reported and structured interview-based assessments to evaluate the impact of mental health burden on disease outcomes in RA and axSpA. We found that the presence of any MD, as identified by the M.I.N.I., was significantly associated with increased disease activity in both conditions. Despite these negative clinical and PROs, the uptake of follow-up psychiatric care was remarkably low, underscoring the need for improved strategies to engage both patients and healthcare providers in addressing mental health as part of comprehensive rheumatologic care. Mental disorders and psychological distress in relation to disease outcomes In our cohort of individuals with established RA and axSpA, one in four patients with RA and nearly one in three patients with axSpA met the criteria for a mental disorder. Severe conditions such as MDD and suicidal ideation were observed in 7.0% of patients with RA and in 12.5% and 8.0% of patients with axSpA, respectively. Suicidal thoughts are known to be more prevalent in individuals with chronic conditions, with arthritis increasing risk by 1.71-fold.38 However, data on suicide attempts in patients with RA remain scarce, and no increased risk has been established for axSpA specifically.39 Marrie et al suggested an increased risk of suicide in patients with RA, particularly those with concomitant depression (HR 15.2, 95% CI 9.22 to 25.1) or anxiety (HR 8.38, 95% CI 5.15 to 13.6).40 Prior studies have reported a wide range in MDD prevalence: up to 33% in RA3 7 8 and approximately 26% in axSpA.2 16 20 41 42 Our findings align with those from population-based studies using the M.I.N.I., such as the German NAKO study (MDD in 9.5% of RA, 8.5% of AS),43 and the Dutch Lifelines cohort, which reported affective or anxiety disorders in ~12% of patients with RA.44 Conversely, Khan et al found MDD in 41% of patients with RA using the M.I.N.I., though their sample was skewed towards those with higher disease activity.45 The strong association between disease activity and depressive symptoms—especially the subjective components of DAS28—has been well established.7 10 46 Similarly, depressed or anxious patients with axSpA consistently report poorer outcomes, including an ~0.9-point increase in BASDAI compared with non-depressed patients.47 In our study, both patients with RA and those with axSpA with current or past psychiatric diagnoses (via M.I.N.I.) had higher disease activity and worse PROs. Interestingly, a 10-point increase in BDI-II was needed to produce a similar effect, suggesting that even mild depression or anxiety may contribute to poorer disease control. This highlights the clinical relevance of both diagnostic and symptom-level assessments. These findings support the need for routine screening for psychiatric comorbidities in rheumatology, including intensity-based tools for depression and anxiety. Further research is needed to explore whether tailored mental health interventions could improve disease outcomes. Recent data show lower remission rates in depressed patients with RA10 and higher anxiety—but not depression—among those with difficult-to-treat RA,48 reinforcing the complex relationship between mental health and rheumatic disease control. Addressing mental health needs in RA and axSpA The poorer disease activity and PROs observed in patients with RA and axSpA with mental health burdens emphasise the need for more integrated care. Although managing comorbidities is a well-established part of treatment in these diseases,1 2 6 49 mental healthcare remains insufficient. In our cohort, few patients reported current or prior psychiatric care, and even fewer accepted follow-up counselling despite clear psychological needs. In inflammatory disorders such as RA and axSpA, the consequences of shared biological mechanisms with mental illnesses, psychological adaptive processes, psychosocial factors or even general attitudes and barriers to seeking psychiatric care may contribute to unmet psychological needs. Several factors may contribute to this gap. The biological overlap between inflammation and mental illness is increasingly recognised. Shared immune mechanisms—including dysregulated activity of macrophages, T cells and innate lymphoid cells—alongside pro-inflammatory cytokines (eg, IL-1, IL-6, IL-17, TNF-α)—drive both joint pathology and systemic symptoms such as pain and fatigue. These cytokines may also influence brain function, either by crossing the blood–brain barrier or signalling via the neural axis, thereby disrupting mood-regulating circuits.50 Beurel et al recently reviewed the interplay between the immune system and neuropsychiatric symptoms, highlighting how depression-related symptoms, such as sadness and suicidal ideation, closely linked to serotonin pathway dysregulation, might be influenced by interferon (IFN)-α and interleukins (IL)−2, IL-6 and IL-8.51 Additionally, IL-1β, IL-17A and TNF-α may affect serotonin and norepinephrine circuits, potentially triggering anxiety.51 In RA, the beneficial effects of targeted anti-inflammatory drugs, including TNF and IL-6 inhibitors, on mental health have been suggested.6 8 This raises the hypothesis that shared improvements in physical and mental well-being among individuals undergoing anti-inflammatory therapy might lead to an underestimation of psychological needs. Consequently, physical symptoms might not always be correctly distinguished from mental health challenges. Moreover, the severity of depressive symptoms in chronic diseases, such as RA, may particularly reflect psychosocial factors and individual adaptive strategies.52 Coping mechanisms in RA and axSpA must address daily stressors such as pain, reduced functionality and medical procedures. Unfortunately, mental disorders and the decision to seek psychiatric care often face negative social perceptions, including embarrassment or fear of judgement. Insufficient knowledge about mental health issues has also been identified as a significant barrier to seeking psychiatric care.53 Beyond these general barriers, challenges specific to rheumatic diseases may further complicate access to mental health support. In younger patients and their families, additional concerns include the risk of misattributed physical symptoms, rising healthcare costs and the use of psychiatric medications.54 Altogether, these findings call for routine mental health screening, destigmatisation efforts and improved communication between rheumatology and mental health services to ensure that patients receive comprehensive, individualised care. Study limitations and future directions The outcomes of our study should be considered in the context of several limitations. First, its cross-sectional design prevents the assessment of causal relationships or changes over time. While psychiatric counselling was offered to individuals with a positive M.I.N.I. screen, follow-up data on whether participants sought care are unavailable. Thus, we could not assess the progression of psychological burden or its long-term impact on disease outcomes, particularly in those without appropriate mental health support. Second, our cohort did not include individuals with very early or long-standing RA or axSpA and was not designed to estimate epidemiological prevalence. Instead, the focus was on understanding the association between mental health burden and disease outcomes. Still, our findings align with prevalence data from comparable studies. Most participants had relatively low disease activity, yet still reported substantial psychological distress. This suggests that residual inflammation may contribute to mental health symptoms, though the link between immune activation and affective disorders is likely complex and not solely mechanistic. Future research should investigate how immune dysregulation affects neuroplasticity and the biological pathways underlying depression and anxiety in RA and axSpA. Finally, the lack of a structured scale to assess reasons for refusing psychiatric care limits insight into patient attitudes. Further studies should explore psychological adaptation, stigma and barriers to care, ideally using more nuanced tools to understand help-seeking behaviour and tailor interventions. Conclusion Mental health burden, identified via M.I.N.I., was present in nearly 25% of patients with RA and 31% those with axSpA, even in the context of low to moderate disease activity. Rates of depressive symptoms and psychiatric disorders were comparable between the two conditions and markedly higher than in individuals without inflammatory disease. These findings highlight that psychological distress is common in RA and axSpA and may negatively influence disease outcomes. Yet, most affected patients did not receive adequate mental health support. Addressing psychological needs through proactive identification and reducing barriers to care may improve overall outcomes and quality of life in this population. Supplementary material Footnotes Funding: This work was supported by the Czech Health Research Council (NU21-09-00297) and by the project (Ministry of Health, Czech Republic) for conceptual development of research organisation 00023728 (Institute of Rheumatology). Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-103686). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study involves human participants and was approved by the Ethics Committee of the Institute of Rheumatology, Prague: No 5704/2021. Participants gave informed consent to participate in the study before taking part. Data availability free text: Due to the sensitive nature of the questions asked in this study, survey respondents were assured that raw data would remain confidential and would not be shared. The data used in this study is confidential. Patient and public involvement: Patients and/or the public were involved in the design, conduct, reporting or dissemination plans of this research. Refer to the Methods section for further details. Data availability statement Data are available upon reasonable request. References Supplementary Materials Data Availability Statement Data are available upon reasonable request.