Since the COVID-19 outbreak, public health authorities have looked for the best evidence on infection risk and prognosis to guide their choices. The interpretation of the observational studies that variably reported increased infection rates and a poor prognosis in patients with systemic rheumatic diseases (SRD) has been limited by factors such as the selection of patients in the care of tertiary referral centres, the small available sample sizes for the less prevalent diseases, the description of SRDs as a single broad category, and the neglected influence of comorbidities [1].
The use of big healthcare data has become essential in gathering crucial information for a reliable identification of high-risk groups. The Chronic Related Group (CReG) system is an experimental approach of classification of chronic patients created to predict the medical resources needed to ensure their care. This system automatically assigns a diagnosis to a subject according to medical administrative records over a pre-set period. Specifically, CReG system relies on the registration and integration of disease-specific codes used to determine the share of healthcare costs, hospital discharge diagnoses codes and access to the prescription of drugs or therapeutic procedures uniquely associated with a specific condition [2], [3].
In this analysis, we compared incidences and 30-day outcomes of 40,490 SRD patients (Table 1 ) to 471,6119 subjects dwelling in the Lazio Region, the second most populated region of Italy that includes the Rome metropolitan area. SRDs and comorbidity diagnoses were derived from the CReG classification while data on COVID-19 infection from a regional digital network. The risk was expressed as incidence rate ratio adjusted for demographics and comorbidities. We focused on the period from the 20th of February 2020 to the 31st of December 2020 to selectively assess a cohort of unvaccinated patients.
Table 1.
Prevalence of SRDs in Lazio and demographics of the affected patients.
| SRD diagnosis | Total patients, n | Prevalence, % | Overlapping SRDs, n (%) | Female, n (%) | Age 18–35, n (%) | Age 36–59, n (%) | Age 60–79, n (%) | Age ≥ 80, n (%) |
|---|---|---|---|---|---|---|---|---|
| RA | 14838 | 0.31 | 1463 (9.86) | 11581 (79.0) | 1005 (0.10) | 4755 (0.23) | 7243 (0.58) | 1835 (0.46) |
| PsA | 11273 | 0.24 | 867 (7.72%) | 6862 (61.0) | 611 (0.06) | 5192 (0.25) | 4915 (0.40) | 519 (0.13) |
| axSpA | 2422 | 0.05 | 305 (12.59) | 1118 (46.1) | 255 (0.03) | 1276 (0.06) | 763 (0.06) | 128 (0.03) |
| SLE | 3936 | 0.08 | 682 (17.33) | 3388 (86.0) | 412 (0.04) | 2261 (0.11) | 1141 (0.09) | 122 (0.03) |
| pSS | 3393 | 0.07 | 989 (29.15) | 3208 (94.5) | 114 (0.01) | 1324 (0.06) | 1628 (0.13) | 327 (0.08) |
| SSc | 2256 | 0.05 | 522 (23.14) | 2056 (91.1) | 109 (0.01) | 836 (0.04) | 1107 (0.09) | 204 (0.05) |
| MSD | 952 | 0.02 | 296 (31.09) | 797 (83.7) | 83 (0.01) | 418 (0.02) | 405 (0.03) | 46 (0.01) |
| UCTD | 2977 | 0.06 | 695(23.25) | 2745 (92.2) | 386 (0.04) | 1673 (0.08) | 853 (0.07) | 65 (0.02) |
| SV | 1560 | 0.03 | 118 (7.56) | 978 (62.6) | 206 (0.02) | 560 (0.03) | 610 (0.05) | 184 (0.05) |
SRD: systemic rheumatic disease; SD: standard deviation; RA: rheumatoid arthritis; PsA: psoriatic arthritis; axSpA: axial spondylarthritis; SLE: systemic lupus erythematosus; pSS: primary Sjögren syndrome; SSc: systemic sclerosis; MSD: myositis-spectrum disorders; UCTD: undifferentiated connective tissue disease; SV: systemic vasculitis.
Table 2 reports peculiar patterns in terms of incidence, hospitalisation, intensive care unit (ICU) admission and death for the different SRDs. COVID-19 risk was increased in patients with Psoriatic Arthritis and Undifferentiated Connective Tissue Disease, possibly as the result of reduced adherence to protecting behaviours. These conditions are indeed less frequently treated with immunosuppressants or associated with visceral involvement, circumstances that have been reported to lead to highly perceived individual risk [4], [5], [6]. The hospitalization risk was higher in patients with Axial Spondylarthritis, Systemic Erythematosus Lupus (SLE), Systemic Vasculitis, intensive care unit (ICU) admission risk was higher in Systemic Erythematosus Lupus and primary Sjögren's syndrome patients, while increased mortality was reported in patients with Rheumatoid Arthritis, SLE, primary Sjögren Syndrome, and Scleroderma. It can be argued that the patients who are more likely to present pulmo-renal complications are more susceptible to worse outcomes. The high prevalence of lung fibrosis and the specific vasculopathy could explain the especially high mortality in scleroderma patients.
Table 2.
Infection rates and thirty-day hospitalisation, ICU admission, and death rates in SRD patients with COVID-19.
| Infection |
Hospitalization |
ICU admission |
Death |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Diagnosis | n | Tested patients, n (%) | Cases, n | Adjusted IRR (95% CI) | Cases, n | Adjusted IRR (95% CI) | Cases, n | Adjusted IRR (95% CI) | Cases, n | Adjusted IRR (95% CI) |
| RA | 13375 | 3218 (24.06) |
453 | 1.09 (0.99–1.19) |
94 | 1.18 (0.96–1.45) |
16 | 1.47 (0.90–2.41) |
30 | 1.50a (1.04–2.17) |
| PsA | 10370 | 2747 (26.49) |
394 | 1.21c (1.10–1.33) |
60 | 1.15 (0.89–1.48) |
12 | 1.54 (0.87–2.72) |
9 | 0.89 (0.46–1.73) |
| axSpA | 2117 | 575 (27.16) |
84 | 1.21 (0.98–1.50) |
15 | 1.89a (1.14–3.14) |
1 | 0.82 (0.12–5.82) |
2 | 1.02 (0.25–4.12) |
| SLE | 3254 | 908 (27.90) |
128 | 1.14 (0.96–1.36) |
24 | 2.16c (1.45–3.22) |
5 | 3.67b (1.52–8.83) |
5 | 2.67a (1.10–6.44) |
| pSS | 2404 | 640 (26.62) |
84 | 1.12 (0.90–1.38) |
17 | 1.58 (0.98–2.54) |
5 | 4.13b (1.71–9.96) |
6 | 2.51a (1.12–5.62) |
| SSc | 1734 | 528 (30.45) |
46 | 0.84 (0.63–1.12) |
10 | 1.23 (0.66–2.31) |
0 | – | 6 | 4.60c (2.06–10.29) |
| MSD | 656 | 170 (25.91) |
22 | 1.03 (0.68–1.56) |
4 | 1.78 (0.67–4.74) |
0 | – | 0 | – |
| UCTD | 2282 | 726 (31.81) |
95 | 1.26a (1.03–1.54) |
10 | 1.45 (0.78–2.70) |
0 | – | 0 | – |
| SV | 1442 | 409 (28.35) |
53 | 0.99 (0.75–1.30) |
14 | 1.81a (1.07–3.06) |
2 | 1.33 (0.33–5.38) |
4 | 2.31 (0.86–6.18) |
| Overlapping SRDs | 2856 | 836 (29.27) |
100 | 1.06 (0.87–1.28) |
20 | 1.79b (1.16–2.78) |
6 | 3.96c (1.78–8.84) |
5 | 2.45a (1.02–5.91) |
| No SRDs | 4675629 | 1054387 (22.55) |
149092 | 1.00 (0.99–1.00) |
17517 | 1.00 (0.98–1.02) |
2570 | 0.99 (0.93–1.05) | 3821 | 0.99 (0.92–1.08) |
| Lazio region | 4716119 | 1065144 (22.59) | 150551 | – | 17785 | – | 2617 | – | 3888 | |
SRD: systemic rheumatic disease; SD: standard deviation; RA: rheumatoid arthritis, PsA: psoriatic arthritis; axSpA: axial spondylarthritis; SLE: systemic lupus erythematosus; ICU: intensive care unit; pSS: primary Sjögren syndrome; SSc: systemic sclerosis; MSD: myositis-spectrum disorders; UCTD: undifferentiated connective tissue disease; SV: systemic vasculitis.
P < 0.05.
P < 0.01.
P < 0.001.
In conclusion, we showed how the CReG system classification allows the identification of high-risk SRD patients on a large scale and highlights the heterogeneity in their clinical behaviours. This methodology could be fruitfully extended to the assessment of other potential SRD-related complications such as cancer and cardiovascular events.
Funding
No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to write this article.
Disclosure of interest
The authors declare that they have no competing interest.
References
- 1.Grainger R., Kim A.H.J., Conway R., et al. COVID-19 in people with rheumatic diseases: risks, outcomes, treatment considerations. Nat Rev Rheumatol. 2022;18:191–204. doi: 10.1038/s41584-022-00755-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sorlini M., Perego L., Silva S., et al. The chronic related groups (CReG) model for ensuring continuity of care for chronically ill patients: pilot experience of the planning stage in Bergamo (Italy) Ig Sanita Pubbl. 2012;68:841–861. [PubMed] [Google Scholar]
- 3.Rea F., Corrao G., Ludergnani M., et al. A new population-based risk stratification tool was developed and validated for predicting mortality, hospital admissions, and health care costs. J Clin Epidemiol. 2019;116:62–71. doi: 10.1016/j.jclinepi.2019.08.009. [DOI] [PubMed] [Google Scholar]
- 4.Kipps S., Paul A., Vasireddy S. Incidence of COVID-19 in patients with rheumatic disease: is prior health education more important than shielding advice during the pandemic? Clin Rheumatol. 2021;40:1575–1579. doi: 10.1007/s10067-020-05494-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cook C., Cox H., Fu X., et al. Perceived risk and associated shielding behaviors in patients with rheumatoid arthritis during the Coronavirus 2019 pandemic. ACR Open Rheumatol. 2021;3:834–841. doi: 10.1002/acr2.11340.0. [published online ahead of print, 2021 Sep 8] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mahil S.K., Yates M., Langan S.M., et al. Risk-mitigating behaviours in people with inflammatory skin and joint disease during the COVID-19 pandemic differ by treatment type: a cross-sectional patient survey. Br J Dermatol. 2021;185:80–90. doi: 10.1111/bjd.19755. [DOI] [PMC free article] [PubMed] [Google Scholar]