Influence of the COVID-19 Pandemic on the Treatment of immune mediated Rheumatic and Musculoskeletal Diseases during the first pandemic period: A Systematic Review

1. Introduction

In 2019 the novel coronavirus SARS-CoV-2 was identified in China and spread rapidly over countries, causing a worldwide pandemic (1–5). Initially regarded as primarily affecting the lungs, corona virus disease 19 (COVID-19) is now known to be a multisystemic infectious disease that affects different organ systems (6). The course of COVID-19 ranges from mild, to severe and critical cases (depending on risk factors) often requiring intensive care (7).

Patients with immune-mediated rheumatic and musculoskeletal diseases (imRMD) are at higher risk of infections especially due to the use of immunosuppressive medication. The pandemic has therefore raised concerns amongst rheumatologists, especially regarding immunocompromised patients. Data from 2021 show that the risk for infection with SARS-CoV-2 is not increased (8) or is only slightly (9) elevated in patients with imRMDs compared to the general population, but, if infected, the risks for hospitalization or for a severe disease course are increased (7). Most conventional synthetic (csDMARD), biological (bDMARD) and targeted synthetic disease modifying antirheumatic drugs (tsDMARDs) do not seem to increase the risk of infection with SARS-CoV-2 or the risk of poor outcomes of COVID-19, exceptions being glucocorticoids > 10 mg/day, rituximab, mycophenolate mofetil (MMF) and, potentially, Janus kinase inhibitors (JAKi) (9).

The disease course of patients with imRMDs and SARS-CoV-2 infection has been studied widely, but systematic reviews describing the influence of the pandemic on the treatment of imRMDs are lacking.

The aim of this systematic review is to describe the influence of the COVID-19 pandemic on the management of imRMDs on adherence and changes in medications, on the access to rheumatological care and medications and on the use of other health care delivery forms.

2. Materials and Methods

This review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (10) . A comprehensive search was carried out in PubMed, Cochrane and Embase databases regarding publications from 01.12.2019 to 31.10. 2021. We used specific MeSH headings and additional keywords to identify studies (see search strategy in Appendix 1).

We selected articles in English or German including adult patients with imRMDs that evaluated the influence of the COVID-19 pandemic on the general management of imRMDs (Influence on adherence or changes in medications), on the access to rheumatological care and medications, and on the use of other health care delivery forms.

There were no restrictions regarding to study-design except for systematic reviews and case reports that were excluded as well as articles on the disease outcomes in case of SARS-CoV-2 infection. Studies found were screened independently by two reviewers (MS, SBa) for inclusion. In the first phase the studies were screened for title and abstract, followed by full-text screening and data extraction. In case of disagreement, consensus was reached after discussion between the two raters. Quality rating was performed according to the Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence (11). Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia) (12) was used as the literature management program and Zotero (Corporation for Digital Scholarship, Fairfax, USA) as reference management software (13). Because no randomized controlled trials were published and data were very heterogenous no meta-analysis was performed. Out of the included studies two clusters of “influences of COVID-19 pandemic” were formed and analyzed further regarding: (i) medical management of imRMDs; (ii) health care in general regarding imRMDs.

3. Results

Study Selection

The search strategy identified 5969 potentially relevant studies. Based on title and abstract and after removal of duplicates, 155 studies were assessed in full-text screening. A final total of 35 studies with data from 182’746 patients and from 2018 rheumatologists were included in the systematic review. Figure 1 shows the study flow.

Study characteristics and levels of evidence

The majority of included studies were surveys or questionnaires (14–47), which are evidence grade IV according to the Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence (11). One was a cohort study, graded level III (48). Seventeen studies were from Europe (15,16,18,19,21,22,24,28–31,37,39,41,42,46,47), three from Africa (14,17,34) , ten from North America (23,25–27,33,35,36,38,45,48,49) and five from Asia (20,32,40,43,44). Details of the included studies are shown in Table 1.

• Outcomes
• 1) Influence on the medical management of imRMDs

Non-availability of rheumatic medication was a prevalent issue, important examples being hydroxychloroquine (HCQ) and tocilizumab. Shortages or difficulties in the availability of HCQ was an issue in 16–69% of patients (14,15,17,20,23,30,47), with the highest level reported from India (20). Shortages of tocilizumab was reported in 14% of patients (45,47).

Non-adherence to prescribed drugs was another issue. Non-adherence was defined in the different studies very heterogeneously as changing of medication, adaptation of dose or interval without professional health advice or as stopping of medication or irregular intake of medication without professional health advice. The overall non-adherence rate among all included studies was 4–46% of patients (16,26,29–32,43). High rates of non-adherence were reported by four studies (20,21,24,33), with the highest level (46% of patients) reported from India (20). The second highest level of non-adherence (almost 40% of patients) was reported from North America (33). A Swiss study (16) comparing medication adherence of patients with different imRMDs before and during the pandemic found only slight adherence reductions. A significant increase in non-adherence was only seen in patients with axial spondyloarthritis (axSpA) (13% medication non-adherence in pre-COVID-19 period versus 20% during the first wave, P = 0.003). The lowest level of medication non-adherence was reported from Denmark (21). In this study, compliance with medication was compared between the start of the first lockdown to three months later, when society was gradually reopened. Low levels of non-adherence were reported (4-6% at the start of the lockdown versus 2-4% three months later).

The drugs mostly changed or stopped were bDMARDs and JAKi (28,32). Low-dose prednisolone and csDMARDs were the least likely medications to be stopped. Longer disease duration of the underlying rheumatic disease and higher disease activity were significantly associated with medication discontinuation (19). Disease flares were described in high proportions of the patients (63-74 %), who had stopped their DMARDS (15,18).

Regarding different rheumatologic diseases, a study from India (43) found that 43% of patients with inflammatory arthritis, 31% with systemic lupus erythematodes (SLE), and 13% with inflammatory myositis and scleroderma (P < 0.05) stopped their treatment. Further detailed information regarding non-adherence to, or non-availability of medication is shown in Table 2.

Many studies reported reasons for changes in medication-taking. The following factors were significantly associated with change in at least one medication due to patients’ fear of COVID-19 (21): male sex (odds ratio (OR) 1.51, 95% confidence interval (95% CI) 1.21–1.89), age >80 years compared to <39 years (OR 0.11, 95% CI 0.006–0.52), lower education (OR 0.56, 95% CI 0.45–0.69), being employed (OR 1.52, 95% CI 1.16–1.99), and use of bDMARDs (OR 1.86, 95% CI 1.02–3.81). Further, there was a direct correlation of incidence of SARS-CoV-2 infections in the general population and medication non-adherence: the higher the incidence of COVID-19, the lower the medication adherence (24).

Lastly, there was also an influence of the pandemic on the treating physician regarding medication. Rheumatologists reduced the dose of steroids in 23–36% of patients (14,47) and, in 17% of patients, steroids were stopped completely (14). In contrast, csDMARDs were stopped only rarely (in 2 % of patients) (14), whereas bDMARDs were stopped more frequently (in 33% of patients) (15). In a few cases drug application intervals of bDMARDs were extended (28). Moreover rheumatologists were hesitant to start a bDMARD in 75% (47) or a tsDMARD in 14% (14) of cases.

• 2) Influence on healthcare in general

Singh et al. (45) reported an increase in alternative types of visits to the rheumatologist related to COVID-19 compared to the pre-COVID-19 era, such as telephone visits (plus 53%), video-based Veterans Affairs Video Connect (VVC) visits (plus 44%) and clinical video tele-health (CVT) visits with a facilitator (plus 29%). Bos et al. (46) reported telephone visits to be the most commonly used form of remote consultation, with 80% of rheumatologists using exclusively telephone consultations. In-person visits were conducted only in special circumstances, such as for joint aspiration (46).

From the patients’ perspective high levels of unwillingness to health care visits were reported (21–86%) (14,15,27,30,36). The highest levels with 86% of patients unwilling to attend the hospital were reported from Turkey (30). Inability to communicate with or to see the rheumatologist was also frequently reported by 7% (21) to 39% (20) of patients (17,20,21,25,30).

George et al. (27) and Banerjee et al. (36) reported avoidance of laboratory testing in 42% and 47% of patients, respectively. Patients with imRMDs were significantly less likely to avoid in-person visits (OR 0.79 (95% CI 0.70–0.89)) or laboratory tests compared to patients with non-autoimmune RMDs (35% versus 39%, OR 0.84 (95% CI 0.73–0.96) (26,48). Other factors associated with avoidance of in-person visits and laboratory testing were older age, low socioeconomic status, living in urban areas or in countries with higher COVID-19 activity and regarding medication receiving a bDMARD or JAKi (48).

4. Discussion

This systematic review showed that COVID-19 influenced health care behavior in patients, as well as in rheumatologists and other doctors. In many cases patients or doctors discontinued established medication. Further, the pandemic resulted in a collapse of supply chains, causing non-availability of medication, especially in the case of HCQ and tocilizumab. Healthcare appointments took place less frequently than usual and telehealth emerged as a solution, with remote consultations with physicians or with newly established telerehabilitation services.

Medication non-adherence was a common problem among patients. A possible explanation could be the low availability of remote consultations at the beginning of the pandemic, resulting in feelings of insecurity with patients stopping their medication as a self-management strategy. The classes of medication that were discontinued most frequently were bDMARDs and JAKi (26,30), possibly because these immunosuppressive medications are considered the most dangerous regarding infections.

Between different imRMDs, relevant differences in non-adherence to medication have been reported. Low numbers of non-adherence were reported in patients with vasculitis (37,38). Patients with vasculitis are usually aware of the disease course with serious relapses in the absence of maintenance therapy, which results in adherence to treatment (38). Another factor increasing medical compliance is that parenteral treatments are often only possible in the hospital setting and are therefore not postponed by patients.

The sudden discontinuation of anti-rheumatic therapy is a relevant issue, because it can lead to disease flares. A large proportion of patients with different imRMDs reported a flare after modifying their treatment (15,18). This supports the recommendation of not stopping treatment during the pandemic in situations other than suspected or confirmed SARS-CoV2-infection, because resulting disease flares and higher requirements for glucocorticoids could increase the risk of SARS-CoV-2 infection (36).

The pandemic, and mostly the fear of infection with SARS-CoV2, had a severe influence on the medication behavior of rheumatologists. A large proportion of rheumatologists reduced the dose or frequency of steroids (14,47,51), many changed DMARDs (14,22) or stopped them (14,15) and there was hesitancy to start new DMARDs (47,51).

N. Rebić et al. conducted a systematic review about the adherence to medication in patients with imRMDs (51) . They described non-adherence rates of 6.5- 34.2% and discontinuation rates of 2- 31.4% which are similar rates compared to the overall non-adherence rate of 4- 46% in our systematic review. They found slightly higher numbers of physicians who reduced the dose of steroids (23- 56 % v.s. 23- 36% in our review) and they also reported of a reluctance to start bDMARDS or tsDMARDS.

Different non-compliance rates to medical visits were reported between the different studies (22,30). Patients with autoimmune RMDs were significantly less likely to avoid in-person visits and laboratory tests compared to patients with non-autoimmune rheumatic diseases (26,48). These results may be explained with the fact that patients with imRMDs needed close monitoring because of their disease as well as their immunosuppressive treatment and that the fear of an infection with SARS-CoV2 was a more dominant factor determining behavior. Interestingly a study from North America (48) reported a normalization of the rates of follow-up visits a few months after the start of the pandemic, suggesting a rapid adaptation of patients and doctors to the pandemic circumstances.

The COVID-19 pandemic posed many challenges, but it also opened new opportunities for the development of health care systems. Because of the environmental risk factors for acquiring a SARS-CoV2-infection before vaccines existed, practical steps to reduce the infection risk were introduced, including social distancing, hand hygiene and use of face masks (9). As a consequence of social distancing patient consultations were performed remotely whenever possible, leading to an increase in telehealth care. Prior to 2019 telehealth care was very rare or non-existent, but its use grew rapidly during the COVID-19 pandemic. Nevertheless, there were huge differences between different countries in the implementation of telehealth. Data from the USA and Australia showed an increase in telehealth, whereas data from India reported that only a small proportion of patients were aware that telehealth existed, and even fewer used it. It is likely that many patients with imRMDs, especially those from non-urban parts of emerging countries, had no access to telehealth care during the COVID-19 pandemic. A major goal of telehealth was to try to avoid disruption of health care and to prevent patients from stopping their medication. In addition, telephone and video-based consultations were preferred in “stable patients” with known disease course and without the need for changing immunosuppressive medication (45).

Strengths and Limitations

This study has several strengths. Although most studies were of low evidence grade (III or IV), data from 182’746 patients and 2018 interviewed rheumatologists were included. Together with the fact that studies from four different continents and different countries were included, these large numbers paint a global picture of the influence of the pandemic on imRMDs. The influence covers a broad spectrum of issues occurring during the COVID-19 pandemic, including compliance with medication, shortage of certain medications and problems with delivery of health care. In addition, new aspects are described, such as telemedicine and telerehabilitation services, which were set up as substitutes for former in-person routine care.

The study also has limitations. First, regarding the heterogeneity of the reported outcomes and the study designs, it was not possible to carry out a meta-analysis and to evaluate the results statistically using odds ratios. Unclear and inhomogeneous reporting and extensive variation in research methods between studies creates barriers for comparison and generalization to specific patient populations. Furthermore, definitions of non-adherence with medication varied between studies, making it challenging to synthesize findings meaningfully. Most of the included studies were surveys, which leads to some typical limitations regarding the study design. Surveys may lead to inclusion biases, as patients who are more interested or worried about COVID-19 are generally more willing to participate. Responses are self-reported and cannot be verified. Survivorship bias is also probable, as very sick or deceased patients cannot participate. Further, patients with a relatively higher socioeconomic status have a greater online presence and affinity to online surveys and are therefore probably overrepresented. As many of the results were published only as case reports and congress abstracts those results were not included in the present study. Important information may therefore have been missed. Finally, this review shows only results regarding the influence on the treatment of imRMDs of the first wave of the COVID-19 pandemic lasting February 2020 to August 2020.

5. Conclusions

The COVID-19 pandemic influenced the management of patients with imRMDs, especially during first wave when no vaccine was available. The influence of the pandemic was diverse regarding adherence to medication, shortage of some medications, adherence to doctor visits or laboratory testing, and regarding governmental interventions. To preserve adherence to health care the COVID-19 pandemic was a starting point for new health care systems. Virtual clinics and remote and telehealth consultations were implemented. These new forms of health care delivery should be spread and implemented worldwide to routine clinical practice to be ready for future pandemics.