A total of 601 patients were evaluated, including 321 patients (IMRD COVID+) and 280 controls (IMRD COVID-), with the majority being female with a similar median age. No differences in demographic data, including comorbidities, disease duration and IMRD, were observed between the two groups. A greater frequency of social isolation was observed in the control group (p = 0.001) and in the TNFi treatment group (p = 0.003). Table 1 summarizes the clinical and demographic data at baseline.

Among the patients with COVID-19 (n = 321), the most frequent symptoms were headache (60.6%), fever (55.6%), dysgeusia (54.3%), asthenia (53.7%), anosmia (52.5%) and cough (49.1%). Dyspnea was reported in 35.1% of patients. The medications used to treat COVID-19 were analgesics (53.7%), azithromycin (40.7%), oral corticosteroids (20.2%) and hydroxychloroquine (10.2%), but 16.1% did not use any medication.

Regarding COVID-19 outcomes, 77 (23.9%) patients sought hospital care, and 29 (9.0%) were hospitalized. Three of these patients were admitted to the intensive care unit and received mechanical ventilation. None of the patients died.

There was no significant difference between the cases and controls at the 6-month follow-up (V1, V2, and V3), regardless of the IMRD (Fig. 1).

Comparison of disease activity scores across the three study visits (v1, V2 and V3) in patients with systemic lupus erythematosus, measured by the SLEDAI (A); rheumatoid arthritis, measured by the CDAI (B); and axial spondylarthritis patients, measured by the BASDAI, comparing cases and controls matched for sex, age and epidemiological exposure. V1: visit 1 (inclusion); V2: visit 2; V3: visit 3. CDAI: Clinical Disease Activity Score; SLEDAI: Systemic Lupus Erythematosus Disease Activity Index; BASDAI: Bath Ankylosing Spondylitis Disease Activity Index. One-way ANOVA (95% CI), repeated measures, multiple comparison of the mean of each column with the mean of every other column

Although no difference was demonstrated in the mean activity scores, some patients self-reported worsening of disease activity after COVID-19 [30 with RA (32.2%), 32 with SLE (23.3%) and 2 with SpA (8.6%)]. Furthermore, we did not observe any significant difference in the mean activity score between the RA and SLE patients (Fig. 2).

Comparison of disease activity scores across the three study visits in patients with systemic lupus erythematosus (SLEDAI) (A) and rheumatoid arthritis (CDAI) (B) who self-reported clinical worsening after COVID-19. Visit 1 (inclusion); V2: visit 2; V3: visit 3. CDAI: Clinical Disease Activity Score; SLEDAI: Systemic Lupus Erythematosus Disease Activity Index. One-way ANOVA (95% CI), repeated measures, multiple comparison of the mean of each column with the mean of every other column

In RA patients who self-reported worsening of disease activity after COVID-19, we applied the flare definition, comparing the pre-COVID-19 CDAI score with the post-COVID-19 CDAI score (inclusion), visit 2 and visit 3, in a paired way. At inclusion, 12 patients (40.0%) had an increased CDAI ≥ 4.5 points compared to their pre-COVID status, but only 1 patient required a change in treatment. At visit 2, compared with visit 1, 5 new patients presented an increased CDAI ≥ 4.5 points, and 7 patients improved compared to the previous visit. At visit 3, 12 patients presented an increased CDAI, and of these, 8 new patients did not worsen at visit 2, and only one needed a change in treatment.

In addition, we compared the tender joint count (TJC), swollen joint count (SJC), patient global assessment (PGA) and medical global assessment (MGA) across the three visits, and no significant differences were observed. Eight patients in this group (26.6%) self-reported worsening of joint manifestations.

Comparing the disease activity in the group of patients with RA who self-reported worsening with that in the group of patients who did not, there was an association with post-COVID-19 MGA, assessed at baseline (ES = − 0.56; CI95% -1.01 to -0.11, p = 0.007), and with the CDAI and all its components at visit 3 (Table 2).

The size effect for MGA post-COVID-19 (V1) mean difference means was moderate (-0.56; CI95% -1.01 to -0.11), and for CDAI (-0.996; CI95% -1.45 to -0.51), TJC (-0.801; CI 95% -1.25 to -0.34), SJC (-0.818; CI-1.27 to -0.36), MGA (-1.003; CI -1.46 to -0.53) and PGA at V3 (-1.005; CI -1.45 to -0.53) was a large effect, demonstrating substantial differences between the groups.

Post-COVID-19 worsening in patients with RA was associated with diabetes mellitus (OR = 7.15; 95% CI 1.7–29.4, p = 0.005) and protection from the current use of TNF inhibitors before SARS-CoV-2 infection (OR = 0.51; 95% CI 0.2–0.9, p = 0.026).

Among the patients with SLE who self-reported worsening of disease after COVID-19, 10 reported the appearance of new clinical manifestations. However, no significant differences were observed regarding sex, age, comorbidities, or concomitant medications. Worsening was only associated with anosmia, a symptom of COVID-19 (OR = 2.43; 95% CI = 1.07–5.52, p = 0.03). The self-reported clinical manifestations of the SLE patients who were worse after COVID-19 (n = 32) at visit 3 were skin rash (5 patients), arthritis (3 patients), proteinuria (2 patients) and interstitial lung disease (1 patient).

By comparing the pre- and post-COVID-19 SLEDAI scores according to the flare definition, we observed that 6 patients (18.0%) had increased SLEDAI scores ≥ 4.0 points, but only 2 patients required a change in treatment. At visit 2, compared with visit 1, most patients showed improvement in the SLEDAI score, and only 3 patients experienced improvement in the SLEDAI score at visit 3. However, 3 new patients had increased SLEDAI scores compared with those at visit 2.

Among SpA patients, only 2 patients self-reported worsening after COVID-19, and one of them experienced progressive worsening at all three visits, based on the BASDAI.

There was no statistically significant association between COVID-19 outcomes (hospital care, hospitalization, and ICU admission) and post-COVID-19 disease activity at visit 3.

After a 6-month follow-up, a statistically significant difference was observed in the FACIT and DASS-21 scores in the three domains between the cases and controls (Fig. 3).

Comparisons among the FACIT (A), DASS-21D (B), DASS-21 A (C) and DASS-21 S (D) median (IQR) scores in patients and controls at visit 3. IQR: Interquartile range. FACIT-F: Functional Assessment of Chronic Illness Therapy–Fatigue; DASS-21D: Depression Anxiety and Stress Scale–Dominium Depression; DASS-21 A: Depression Anxiety and Stress Scale–Dominium Anxiety; DASS-21 S: Depression Anxiety and Stress Scale–Dominium Stress. Mann‒Whitney test; CI = 95%

The median FACIT score was 37 (27–44) in cases and 40 (32–47) in controls (p = 0.0003); the median DASS-21 D in cases was 3 (0–8) and 2 (0–5) in controls (p = 0.0005); the median DASS-21 A in cases was 4 (1–9) and 3 (1–6) in controls (p < 0.001); and the median DASS-21 S in cases was 6 (2–12) and 5 (2–9) in controls (p < 0.018). No difference was observed regarding the FACIT-F and DASS-21 scores when comparing patients with RA, SLE and SpA in each case group (Fig. 4).

Distribution and comparison of FACIT (A), DASS21D (B), DASS21A (C) and DASS21S (D) median scores after a 6-month follow-up in patients with rheumatoid arthritis (RA), systemic lupus erythematosus (SLE) and axial spondyloarthritis (axSpA) only in the patient group (IMRD COVID-19 +). FACIT-F: Functional Assessment of Chronic Illness Therapy – Fatigue; DASS-21D: Depression Anxiety and Stress Scale Domain Depression Scale; DASS-21 A: Depression Anxiety and Stress Scale Domain Anxiety Scale; DASS-21 S: Depression Anxiety and Stress Scale Domain Stress Scale; PsoA: psoriatic arthritis; RA: rheumatoid arthritis; axSpA: axial spondyloarthritis; SLE: systemic lupus erythematosus. Independent-samples Kruskal‒Wallis test, 95% CI

Univariate linear models revealed no associations between FACIT or DASS-21 and comorbidities, IMRD endpoints (type and specific therapy), or outcomes related to COVID-19, such as symptoms and treatment. Considering the FACIT-F and DASS-21 scores, there was no significant correlation between disease activity in RA patients and that in SLE patients in the case group (Fig. 5).

Correlation between disease activity in patients with systemic lupus erythematosus, as measured by the SLEDAI, and rheumatoid arthritis, as measured by the CDAI, only in the case group (IMRD COVID-19 +), with FACIT-F (A and E) and DASS 21-D (B and F), DASS 21-A (C and G) and DASS 21-S (D and H) scores

Specifically, when we analyzed the group of patients who experienced worsening of disease activity after COVID-19, we observed that RA patients also experienced a decrease in FACIT-F (31.5 vs. 36.4; p = 0.047) and DASS-21-S (19.0 vs. 13.0; p = 0.031) scores compared to those with no worsening activity. Additionally, the means were considered moderate for FACIT-F (0.47; from 0.03 to 0.91) and DASS 21-S (-0.44; from − 0.88 to -0.007) (Table 3). In SLE patients, no significant difference was observed.

There was no statistically significant association between COVID-19 outcomes (hospital care, hospitalization, or ICU admission) and FACIT-F or DASS-21 scores at visit 3.