BJD E P I DE M I O L O G Y British Journal of Dermatology 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* S.K. Mahil iD ,1 M. Yates,2 S.M. Langan,1,3 Z.Z.N. Yiu iD ,4 T. Tsakok,1 N. Dand,5,6 K.J. Mason iD ,4,7 H. McAteer,8 F. Meynell,1 B. Coker,9 A. Vincent,9 D. Urmston,8 A. Vesty,8 J. Kelly,4 C. Lancelot,10 L. Moorhead,1 Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 H. Bachelez iD ,11,12 I.N. Bruce,13,14 F. Capon iD ,5,9 C.R. Contreras,15 A.P. Cope,2 C. De La Cruz,16 P. Di Meglio iD ,9,17 P. Gisondi iD ,18 K. Hyrich,13,14 D. Jullien,19,20 J. Lambert,21 H. Marzo-Ortega,22,23 I. McKinnes,24 L. Naldi,25 S. Norton,26 L. Puig iD ,27 R. Sengupta,28 P. Spuls,29 T Torres,30 R.B. Warren,4 H. Waweru,10 J. Weinman,31 C.M. Griffiths,4 J.N. Barker,9,17 M.A. Brown,2,9 J.B. Galloway,2 C.H. Smith1,9 and On behalf of the PsoProtect, CORE-UK study groups Linked Comment: G. Becher and A.D. Burden. Br J Dermatol 2021; 185:7–8. Summary Correspondence Background Registry data suggest that people with immune-mediated inflammatory Catherine Smith. diseases (IMIDs) receiving targeted systemic therapies have fewer adverse coron- Email: catherine.smith@kcl.ac.uk avirus disease 2019 (COVID-19) outcomes compared with patients receiving no systemic treatments. Accepted for publication 19 December 2020 Objectives We used international patient survey data to explore the hypothesis that greater risk-mitigating behaviour in those receiving targeted therapies may Funding sources account, at least in part, for this observation. See Appendix 2 for full details. Methods Online surveys were completed by individuals with psoriasis (globally) or rheumatic and musculoskeletal diseases (RMDs) (UK only) between 4 May Conflicts of interest and 7 September 2020. We used multiple logistic regression to assess the associa- See Appendix 3 for full details. tion between treatment type and risk-mitigating behaviour, adjusting for clinical S.K.M and M.Y. are joint first authors. and demographic characteristics. We characterized international variation in a mixed-effects model. See Appendix 1 for the full list of author Results Of 3720 participants (2869 psoriasis, 851 RMDs) from 74 countries, affiliations. 2262 (608%) reported the most stringent risk-mitigating behaviour (classified *Plain language summary available online here under the umbrella term ‘shielding’). A greater proportion of those receiv- ing targeted therapies (biologics and Janus Kinase inhibitors) reported shielding DOI 10.1111/bjd.19755 compared with those receiving no systemic therapy [adjusted odds ratio (OR) 163, 95% confidence interval (CI) 135–197]. The association between tar- geted therapy and shielding was preserved when standard systemic therapy was used as the reference group (OR 139, 95% CI 123–156). Shielding was asso- ciated with established risk factors for severe COVID-19 [male sex (OR 114, 95% CI 105–124), obesity (OR 137, 95% CI 123–154), comorbidity burden (OR 143, 95% CI 115–178)], a primary indication of RMDs (OR 137, 95% CI 127–148) and a positive anxiety or depression screen (OR 157, 95% CI 136–180). Modest differences in the proportion shielding were observed across nations. Conclusions Greater risk-mitigating behaviour among people with IMIDs receiving targeted therapies may contribute to the reported lower risk of adverse COVID- 19 outcomes. The behaviour variation across treatment groups, IMIDs and nations reinforces the need for clear evidence-based patient communication on risk-mitigation strategies and may help inform updated public health guidelines as the pandemic continues. 80 British Journal of Dermatology (2021) 185, pp80–90 © 2020 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. 81 What is already known about this topic? • At the beginning of the coronavirus disease 2019 (COVID-19) pandemic, patients with immune-mediated inflammatory diseases (IMIDs) on targeted systemic thera- pies were considered to be at higher risk of severe COVID-19. • Subsequent clinician-reported registry data suggest that targeted systemic therapy use is associated with fewer adverse COVID-19 outcomes compared with no sys- Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 temic therapy. What does this study add? • We characterize shielding behaviour in 3720 patients with IMIDs from a global self-report survey. • Use of targeted systemic therapy associates with increased shielding behaviour, compared with standard systemics or no therapy, as do demographic risk factors for severe COVID-19 including male sex and obesity. • Greater risk-mitigating behaviour among people with IMIDs receiving targeted thera- pies may contribute to the reported lower risk of adverse COVID-19 outcomes. • Behaviour variation across treatment groups reinforces the need for clear, evidence- based patient communication on risk-mitigation strategies. • These data may help to inform updated public health guidelines as the pandemic continues. The coronavirus disease 2019 (COVID-19) pandemic, caused by treatment or standard systemic agents.16–18 It remains unclear if the highly infectious SARS-CoV-2 virus, represents an unprece- this is attributable to therapeutic modulation of the host antiviral dented global health crisis.1,2 Death from COVID-19 is associated immune and inflammatory response (i.e. biological mechanisms) with male sex, older age, Asian/Black ethnicity and coexisting or enhanced shielding behaviour in patients receiving targeted conditions including cardiovascular disease and obesity.3,4 therapies (resulting in a lower infectious dose of SARS-CoV-2). Guided by international recommendations from the World There is an urgent need to address this knowledge gap because Health Organization (WHO), public health risk-mitigating mea- targeted and standard systemic therapies represent the mainstay sures such as social/physical distancing were introduced early in of treatment in moderate-to-severe IMIDs. the pandemic to limit community transmission of COVID-19.5–8 Rheumatic and musculoskeletal diseases (RMDs) and psoria- The WHO also recommended more stringent protection mea- sis are common IMIDs that are closely related: psoriasis is the sures to reduce exposure risk in groups at higher risk of severe commonest immune-mediated skin disease associated with COVID-19.9 This was referred to as ‘shielding’, and in the UK, inflammatory arthritis, and both conditions have a high preva- was incorporated into Government policy where individuals lence of multimorbidity and are effectively treated with tar- classed as clinically extremely vulnerable were advised to physi- geted and standard systemic therapies. We focused on RMDs cally isolate at home and avoid face-to-face interactions.7 and psoriasis as representative IMIDs and used global self-re- Informed by pre-COVID-19 observational studies on drug-re- port survey data to explore the notion that individuals receiv- lated risks of serious infection,10–13 individuals with immune- ing different types of treatment exhibit distinct risk-mitigating mediated inflammatory diseases (IMIDs) receiving drugs that behaviours in the pandemic. affect the immune system were considered to be at higher risk of severe COVID-19.14,15 While limited evidence has been pub- lished to date on drug-specific COVID-19 risks in IMIDs, rheuma- Methods toid arthritis, systemic lupus erythematosus and psoriasis were Study design, participants collectively suggested as risk factors for death using UK primary care data linked to hospital records from 17 million adults.3 Glo- Two online, self-report surveys with aligned questions, per- bal clinician-reported registry data in rheumatic diseases, psoriasis mitting a combined analysis of data, were developed for peo- and inflammatory bowel disease have further suggested a differ- ple with psoriasis [Psoriasis Patient Registry for Outcomes, ential risk of severe COVID-19 associated with different treatment Therapy and Epidemiology of COVID-19 Infection Me (PsoPro- types. In particular, the use of targeted systemic therapies [biolog- tectMe): www.psoprotectme.org] and RMDs [COVID-19 ics and Janus Kinase (JAK) inhibitors] was associated with a Rheumatology Register (CORE-UK): https://www.redcap02. reduced risk of adverse COVID-19 outcomes, compared with no medstats.org.uk/redcap/surveys/?s=LCA3L4JHXW]. © 2020 The Authors. British Journal of Dermatology British Journal of Dermatology (2021) 185, pp80–90 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. 82 Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. PsoProtectMe (available in eight different languages) was pro- shielding behaviour in the pandemic. The key exposure mea- moted globally following its launch on 4 May 2020 and sure was IMID treatment type in the pandemic, comprising CORE-UK was subsequently launched on 12 June 2020 and three mutually exclusive categories (Table S2; see Supporting promoted in the UK. The surveys were disseminated via social Information): media, patient organizations (Table S1; see Supporting Infor- 1 Targeted therapy: biologics and JAK inhibitors [tumour mation) and clinical networks. The eligibility criterion was necrosis factor (TNF) inhibitors: abatacept, adalimumab, any person (all ages) with a clinician-confirmed diagnosis of certolizumab pegol, etanercept, golimumab, infliximab, psoriasis (PsoProtectMe) or RMD (CORE-UK), irrespective of rituximab; interleukin (IL)-17 inhibitors: brodalumab, Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 COVID-19 status (REC ref 20/YH/0135). Data were collected ixekizumab, secukinumab; IL-12/IL-23p40 or IL-23p19 and managed using REDCap electronic data capture tools inhibitors: guselkumab, risankizumab, tildrakizumab, licensed to King’s College London Division of Health and ustekinumab; IL-6 inhibitors: sarilumab, tocilizumab; JAK Social Care Research.19 inhibitors: baricitinib, tofacitinib]. 2 Standard systemic therapy: acitretin, apremilast, azathio- Variables prine, chloroquine, ciclosporin, dexamethasone, fumaric acid esters/dimethylfumarate, hydroxychloroquine, Minimum sufficient core sets of variables within the surveys leflunomide, methotrexate, mycophenolate mofetil, pred- were defined by our study group of clinicians, epidemiolo- nisolone, sulfasalazine, tacrolimus. gists, health data researchers and patient representatives. The 3 No systemic therapy. Patient Health Questionnaire-2 (PHQ-2) and Generalized Anx- iety Disorder 2-item (GAD-2) scales were used to screen for Patients on combination targeted and standard systemic depression and anxiety, respectively; scores of 3 or more were therapy were included in the targeted therapy group, and sur- positive. Adherence was assessed with a single-item question veys with missing treatment data were excluded. Apremilast which asked if the individual had stopped or delayed their was included in the standard systemic therapy group because medication in the pandemic. in clinician-reported registry analyses it was not grouped with Risk-mitigating behaviour was assessed with the following biologics (unlike JAK inhibitors).17,18 question: ‘Over the past 30 days, what methods have you been using to protect After excluding participants who self-reported current or yourself from COVID-19?’ Respondents could select any of the fol- prior confirmed/suspected COVID-19, associations with shield- lowing options: (1) Shielding (quarantine, strict distancing from ing status were assessed using: (i) a minimally adjusted logistic family members in the home); (2) Self-isolation (quarantine, regression model including age and sex covariates; and (ii) a staying home, avoiding others); (3) Social distancing (avoiding fully adjusted model including a consensus list of covariates crowds and large groups of people); (4) Using gloves and/or selected a priori as potentially influential on shielding behaviour masks during social interactions; (5) None. The most stringent on the basis of expert clinical opinion and existing evidence.20 risk-mitigating behaviour was classified under the umbrella term Treatment was included as a categorical variable in the fully ‘shielding’, encompassing (1) shielding and (2) self-isolation. adjusted model, with no systemic therapy as the reference ‘Shielding’ was considered as a binary variable; any respondent group. Country of residence was included as a cluster variable. who selected options (1) or (2) were coded as having shielded, A count of the number of comorbidities was generated. This and those selecting (3), (4) or (5) as having not shielded. was converted into a binary variable for analyses according to UK participants were also asked whether they were advised consensus clinical expert opinion of the study group: those with to shield using the following question: ‘Did you receive a letter or one or more comorbidity vs. those with no comorbidities. text from the NHS asking you to take additional protective measures including Two sensitivity analyses were performed on the fully to stay at home at all times and avoid all face-to-face contact for at least adjusted multivariable regression models: (i) multiple imputa- 12 weeks?’ tion using iterative chained equations with 20 sets of imputed data to account for missing covariate data; (ii) exclusion of respondents on no systemic therapy, with standard systemic Statistical methods therapy becoming the reference group. Adherence data was included as a covariate in this model. Data were extracted on 7 September 2020 and analysed using As the COVID-19 pandemic progressed in countries over Stata version 16. Continuous variables were reported using different time periods, we hypothesized that the impact of means (SD), and categorical/dichotomous variables as num- time on the relationship between treatment and shielding bers and percentages. To account for partially completed sur- behaviour would vary between countries. To explore this, veys, respondents who completed more than 50% of variables unadjusted estimates of shielding over time by treatment were included. Individuals completing CORE-UK and PsoPro- group were plotted for UK and non-UK survey respondents. tectMe were classified as having a primary diagnosis of RMD Based on these plots we re-ran the multivariable model with and psoriasis, respectively. UK respondents only including time as an interaction term We characterized the demographic, socioeconomic and dis- with treatment, and as a fixed covariate, with a comparison of ease-specific factors associated with the primary outcome of model fit. Time was converted to a binary variable, before or British Journal of Dermatology (2021) 185, pp80–90 © 2020 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. 83 after 30 June. Shielding in the UK appeared to decrease after descriptions see Table 1. Survey completion rates were high, this date, which also coincided with the reopening of hospi- with a median of 94% of covariates of interest completed (in- tality businesses (e.g. restaurants) across the UK. terquartile range of 89–95). A total of 650 surveys (17%) had To characterize international variations in shielding behaviour, 100% data completion. a mixed-effects logistic regression model was executed with Of the participants, 2299 (618%) participants were not country of residence as a random effect. The random effect cap- receiving a systemic agent for their psoriasis or RMD, 924 tures the difference between national and overall sample means, (248%) were receiving targeted therapies and 497 (134%) enabling estimation of case-mix adjusted rates. The national standard systemic agents. The three treatment groups had Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 effects on shielding were visualized using a caterpillar plot.21 similar baseline characteristics including age, ethnicity, mean number per household and comorbidities. Nonadherence was also similar: 90 of 495 (182%) patients in the standard sys- Results temic therapy group reported nonadherence against medical advice, compared with 138 of 923 (150%) receiving tar- Demographic, socioeconomic and clinical characteristics geted therapy. Of 257 (71%) participants who self-reported of study participants suspected or confirmed COVID-19, 80 (311%) were receiv- Self-reported data from 3720 individuals with a primary diagno- ing a targeted therapy, 46 (179%) a standard systemic agent sis of RMD (851, 229%) or psoriasis (2869, 783%, taking and 131 (510%) no systemic treatment. A lower proportion account of missing data) were available from 74 countries [in- of those with COVID-19 reported shielding (143 of 257, cluding UK (2578, 694%), Portugal (200, 54%), USA (165, 556%), compared with those without COVID-19 (2051 of 44%)]; for demographic, clinical and socioeconomic 3352, 612%). Table 1 Participant characteristics, by treatment No systemic Standard systemic Targeted Total therapy therapy therapy P-value Missing Total, n 3720 2299 497 924 Shielded, n (%)a 2262 (608) 1360 (592) 292 (588%) 610 (660%) < 0001 0 Advised to shield, n (%), UK only 1092 (424%) 523 (316%) 164 (514%) 405 (672%) < 0001 4 UK resident, n (%) 2578 (694%) 1656 (721%) 319 (646%) 603 (652%) < 0001 7 Male sex, n (%) 1174 (316%) 632 (275%) 158 (318%) 384 (416%) < 0001 2 Age in years, mean (SD) 492 (150) 493 (157) 490 (148) 490 (132) 084 25 Ethnicity, n (%) < 0001 0 White ethnicity 2990 (804%) 1856 (807%) 386 (777%) 748 (810%) Hispanic or Latino 158 (42%) 95 (41%) 29 (58%) 34 (37%) South Asian 127 (34%) 69 (30%) 23 (46%) 35 (38%) Japanese 90 (24%) 29 (13%) 18 (36%) 43 (47%) Black African/Caribbean/American 86 (23%) 68 (30%) 6 (12%) 12 (13%) Other 269 (72%) 182 (79%) 35 (70%) 52 (56%) Body mass index, mean (SD) 274 (61) 268 (58) 277 (60) 288 (65) < 0001 301 Alcohol > 14 units a week, n (%) 484 (136%) 291 (134%) 57 (120%) 136 (151%) 024 168 Current smoker, n (%) 460 (131%) 259 (121%) 69 (146%) 132 (148%) 0076 207 Full-time employment, n (%) 1664 (447%) 951 (414%) 223 (449%) 490 (530%) < 0001 0 Number in household, mean (SD) 27 (17) 27 (16) 29 (18) 27 (18) 019 9 Key worker, n (%) 985 (266%) 585 (256%) 146 (294%) 254 (276%) 016 15 Diagnosis, n (%) < 0001 54 Psoriasis 2869 (783%) 1539 (686%) 462 (930%) 868 (939%) Inflammatory arthritisb 529 (144%) 465 (207%) 23 (46%) 41 (44%) Connective tissue disease 127 (35%) 113 (50%) 9 (18%) 5 (05%) Axial spondyloarthritis 85 (23%) 75 (33%) 1 (02%) 9 (10%) Other 56 (15%) 53 (24%) 2 (04%) 1 (01%) COVID-19 diagnosis, n (%) 257 (71%) 131 (58%) 46 (97%) 80 (90%) < 0001 111 One or more comorbidity, n (%) 1651 (444%) 949 (413%) 224 (451%) 478 (517%) < 0001 0 Anxiety, n (%) 958 (272%) 605 (281%) 138 (292%) 215 (239%) 0032 197 Depression, n (%) 925 (262%) 593 (275%) 127 (269%) 205 (228%) 0024 196 a Shielded refers to participants who quarantined and self-isolated; binflammatory arthritis included any participant with a diagnosis of rheumatoid arthritis or psoriatic arthritis. © 2020 The Authors. British Journal of Dermatology British Journal of Dermatology (2021) 185, pp80–90 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. 84 Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 Figure 1 Age- and sex-adjusted associations with shielding. Each covariate was run as a predictor for shielding, adjusted for age and sex. Survey responders who were UK residents were asked if they had received an NHS letter advising them to shield, which was associated with shielding, odds ratio of 47 (95% confidence interval 39–56). Standard therapy and biologic therapy were both compared with no systemic therapy as a reference group. Obesity, defined as a body mass index greater than 30. RMD, rheumatic and musculoskeletal disease. 154) and a positive anxiety or depression screen (OR 157, Risk-mitigating behaviour differed by treatment type 95% CI 136–180) (Figure 2). In contrast, shielding was Overall, 2262 participants (608%) reported shielding. Of inversely associated with smoking (OR 073, 95% CI 063– 1632 participants self-reporting shielding in the UK, only 899 085), full-time employment (OR 066, 95% CI 049–088), (551%) reported being specifically advised to shield (no data at least four household members (OR 078, 95% CI 065– on shielding advice were available for non-UK participants). A 093) and key worker status (OR 055, 95% CI 044–068). greater proportion of those receiving targeted therapies (610 No association was found with age (OR 100, 95% CI 099– of 924, 660%) reported shielding compared with those 100) or white ethnicity (OR 085, 95% CI 062–117). receiving standard systemic agents (292 of 497, 588%) or no systemic therapy (1360 of 2299, 592%). Multivariable model sensitivity analyses After excluding those with self-reported COVID-19, we used logistic regression models to investigate the observed dif- To account for missing data (Table 1), the multivariable ferences in shielding by IMID treatment type. Compared with model was rerun following multiple imputation. The magni- the reference group of no systemic therapy, an age- and sex- tude and direction of associations did not change substantially adjusted model for shielding behaviour estimated an odds (Table S3; see Supporting Information). ratio (OR) of 100 [95% confidence interval (CI) 081–123] The model was also rerun excluding respondents on no sys- for those receiving standard systemic agents. In contrast, a sig- temic therapy, using standard systemic therapy as the refer- nificant association with shielding was observed for those ence group. The association between targeted therapy and receiving targeted therapies (OR 134, 95% CI 113–159) shielding was preserved (OR 139, 95% CI 123–156). Ther- (Figure 1). apy nonadherence was not associated with shielding A fully adjusted multivariable logistic regression analysis (Table S4; see Supporting Information). was performed with a categorical treatment exposure variable: The influence of time (survey completion date) on shield- (i) targeted therapy, (ii) standard systemic therapy, (iii) no ing behaviour was explored across treatment groups. Estimated systemic therapy. The no systemic therapy group was used as shielding behaviour generally decreased over time; however, the reference. Use of targeted therapy was associated with time had a differential impact in the UK compared with non- shielding compared with no systemic therapy (OR 163, 95% UK countries (Figures S1 and S2; see Supporting Information). CI 135–197). Standard systemic therapy (OR 117, 95% CI The multivariable model was therefore rerun with UK respon- 089–153) did not have a significant association with shield- dents only, first including time as a fixed covariate and sec- ing. There were associations with shielding for RMD (OR ondly as an interaction term with treatment. The association 137, 95% CI 127–148), male sex (OR 114, 95% CI 105– between targeted therapy and shielding was preserved, with a 124), comorbidity burden (defined as >1 comorbidity, OR better model fit for the interaction term (further details in 1.43, 95% CI 1.15-1.78), obesity (OR 137, 95% CI 123– Tables S5–S7; see Supporting Information). British Journal of Dermatology (2021) 185, pp80–90 © 2020 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. 85 Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 Figure 2 Fully adjusted model identifying associations with shielding. Covariates were determined a priori by an expert panel of collaborators. Country of residence was included as a cluster variable. Biologic therapy was compared with no systemic therapy as a reference group. Obesity, defined as a body mass index greater than 30. RMD, rheumatic and musculoskeletal disease. The multivariable model was also rerun limiting shielding time may help inform updated public health guidelines as the to only those who quarantined (i.e. option 1 of survey ques- pandemic continues. tion on risk-mitigating behaviour). The direction of the asso- Our dataset is based on a large sample of individuals self-re- ciations was preserved. porting RMD and psoriasis. Because there was no healthcare professional/record validation of survey responses, it is reas- suring that key risk factors for severe COVID-19 in the general There was modest variation in risk-mitigating behaviour population such as male sex and obesity were associated with across countries shielding. This is in keeping with public health messaging A greater proportion of participants in the UK reported shield- during the pandemic and indicates a representative and gener- ing compared with those elsewhere (633% vs. 550%). How- alizable sample. Shielding was recommended in groups of ever, UK participants were also less likely to receive a targeted individuals deemed at higher risk7,9 on the premise that this therapy (234% vs. 284%). A mixed-effects model further would reduce the risk of COVID-19 transmission. More showed modest variation around the sample mean in the pro- recently, evidence has emerged indicating shielding may also portion shielding in different countries, indicating broadly result in a less severe course of COVID-19 by reducing the similar risk-mitigating behaviours (Figure 3). Shielding was frequency and intensity of exposures to SARS-CoV-2, thus more prevalent in the UK, Canada and Argentina, but less lowering the infectious dose.22 There is a growing body of prevalent in Portugal and Japan. evidence indicating that SARS-CoV-2 viral load positively cor- relates with disease severity,23,24 and that in animal models, greater SARS-CoV-2 dose at exposure correlates with higher Discussion viral load and worse outcomes.25 We present global self-reported data on risk-mitigating beha- Notably, increasing age was not associated with shielding viours in 3720 individuals with inflammatory joint and skin behaviour in our dataset (despite an even spread of ages and disease across 74 countries. Established risk factors for severe 105% of our dataset being over age 70 years). This finding is COVID-19 outcome including male sex, obesity and comor- in keeping with a recent international study of 8317 individu- bidity burden were associated with stringent risk-mitigating als from the general population showing that age did not pre- behaviour (classified here under the umbrella term ‘shielding’, dict whether individuals took health precautions (mask encompassing self-reported shielding, quarantine, staying wearing, social distancing, handwashing, staying home).20 home or distancing within the home). Notably, use of tar- Instead, beliefs that taking health precautions are effective and geted therapies (biologics and JAK inhibitors) was associated a concern for one’s own health were important predictors. with shielding in comparison with no systemic therapy or Consistent with this, we identified an association between standard systemic therapy. Although the differences in shield- shielding and anxiety/depression. A larger proportion of par- ing behaviours across treatment groups in UK respondents ticipants also reported shielding compared with those advised were preserved when time was used as an interaction term, to shield, which may reflect the elevated rates of self-reported the observed decline in estimated shielding behaviour over anxiety. Anxiety and depression have also been reported in © 2020 The Authors. British Journal of Dermatology British Journal of Dermatology (2021) 185, pp80–90 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. 86 Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 Figure 3 Caterpillar plot of observed and estimated risk-mitigating behaviour, by nation. The grey markers are the observed national proportions of survey respondents who shielded. The blue markers are the predicted random national effect on shielding from a mixed-effects model, with 95% confidence intervals in red. The black horizontal line represents the overall mean. previous observational studies, underscoring the mental health the COVID-19 Global Rheumatology Alliance registry,17 bio- burden of the pandemic (which may at least in part be due to logic/targeted synthetic systemic drug use was associated with the impact of social isolation).26–28 While this finding sug- lower odds of being hospitalized compared with patients gests accurate and representative data capture, more data are receiving no systemic therapies. This effect was largely driven required on the severity and temporality of anxiety and by TNF inhibitors as most patients on biologics were receiving depression. this subgroup. A decreased risk of hospitalization or death was Our study indicates a greater likelihood of shielding overall also associated with TNF inhibitor biologics compared with in individuals with a primary diagnosis of RMD compared no treatment among 525 patients with inflammatory bowel with psoriasis; however, the reasons underlying this are not disease and COVID-19 reported to SECURE-IBD (Surveillance clear. It may be attributable to differences in illness percep- Epidemiology of Coronavirus Under Research Exclusion for tion,29 use of treatments and prevalence of comorbidities. Inflammatory Bowel Disease).16 In contrast, the standard sys- IMID-specific COVID-19 risks are unknown, and neither RMD temics sulfasalazine or 5-aminosalicylate were associated with nor psoriasis were included in WHO and national public a higher risk of hospitalization or death. Our previous study health shielding recommendations per se.7,9 The reasons of 374 patients with psoriasis and COVID-19 reported to the underlying differences in shielding behaviours between treat- PsoProtect registry further suggested an association between ment groups, including patient perceptions of COVID-19 risk, biologics (pooled data on TNF, IL-17 and IL-23 inhibitors) also warrant further study. Although there is a paucity of data and reduced risk of hospitalization, compared with standard on treatment-related beliefs in psoriasis, recent single-centre systemic therapies.18 Although exploration of possible biologi- cross-sectional patient survey data in inflammatory bowel dis- cal mechanisms underlying these associations is warranted ease indicate patients perceive biologics to be riskier than (e.g. cytokine-targeted biologics may attenuate a severe sys- other therapies.29 These perceptions may influence shielding temic inflammatory response to COVID-1930), our current behaviours and are important to address during patient–clini- study highlights shielding behaviour as an important unmea- cian interactions. sured potential mediator in these datasets. The differences in Our global data on shielding behaviour builds on the find- shielding behaviours across treatments supports the notion that ings from a recent single-centre audit of 1693 UK patients greater protective shielding behaviour (resulting in a lower with rheumatic diseases.26 Of these, 46% self-reported shield- infectious dose of SARS-CoV-2) in those receiving targeted ing; however, shielding among different treatment groups was therapies may account, at least in part, for the observed associ- not explored. In line with our findings, the audit found that a ations. Thus, conclusions from clinician-reported registry data lower proportion of individuals with COVID-19 shielded about medication-related COVID-19 risk should be interpreted (39%) compared with those without COVID-19 (47%). Our in this context, and further research efforts are required to study also complements emerging findings from international quantify potential mediation through shielding. clinician-reported registries, indicating differences in COVID- A greater proportion of participants from the UK reported 19 outcomes between different treatment types. Among 600 shielding compared with those elsewhere, which may reflect patients with rheumatic diseases and COVID-19 reported to cross-national differences in public health messaging. British Journal of Dermatology (2021) 185, pp80–90 © 2020 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. 87 However, these data should be interpreted with caution as our Masanori Okuse, Ms Dagmara Samselska, Ms Isabella Tosi and dataset is dominated by UK participants. Due to limited cap- Ms Ya-Hsin Wang. We are also incredibly thankful to Engine ture of socioeconomic data, we were unable to fully adjust for Group UK for their generous creative input and website this confounder in the analysis; however, we did identify that expertise. household density and full-time employment were inversely associated with shielding. Both shielding behaviour and clini- cal decision-making around systemic therapies globally (in- References cluding access to medications) may be affected by 1 Wu Z, McGoogan JM. Characteristics of and important lessons Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 socioeconomic variables such as income and education,31,32 from the coronavirus disease 2019 (COVID-19) outbreak in China: which may in turn influence outcome of COVID-19.4 Linkage summary of a report of 72 314 cases from the Chinese Center for between health, social, behavioural and employment data Disease Control and Prevention. JAMA 2020; 323:1239–42. should thus be prioritized in future research. 2 Onder G, Rezza G, Brusaferro S. Case-fatality rate and characteris- Collecting data via an online survey may have limited partici- tics of patients dying in relation to COVID-19 in Italy. JAMA 2020; pation to more tech-literate individuals and those more con- 323:1775–6. nected to media. The study sample was mostly female (as 3 Williamson EJ, Walker AJ, Bhaskaran K et al. Factors associated with COVID-19-related death using OpenSAFELY. Nature 2020; expected in survey-based studies), of white ethnicity, and their 584:430–6. https://doi.org/10.1038/s41586-020-2521-4. diagnoses were self-reported, which further limits the general- 4 Mathur R, Rentsch CT, Morton C et al. Ethnic differences in izability of the results. Ascertainment bias may overestimate the COVID-19 infection, hospitalisation, and mortality: an OpenSA- overall proportion shielding, because those more concerned FELY analysis of 17 million adults in England. medRXiv 2020; about COVID-19 risk may be more likely to participate. Our https://doi.org/10.1101/2020.09.22.20198754 (preprint). sample, in which a greater proportion reported receiving tar- 5 World Health Organization. Tracking public health and social mea- geted therapies compared with standard systemic agents, may sures. Available at: https://www.who.int/emergencies/diseases/ novel-coronavirus-2019/phsm (last accessed 15 Jan 2021). not be representative of patients receiving systemic therapies 6 Centers for Disease Control and Prevention CDC. Coronavirus Dis- more broadly. A disparate group of medications is also classified ease 2019 (COVID-19). CDC. People at increased risk: and other together as standard systemic agents. Potential selection bias people who need to take extra precautions. 2020 Available at: may be addressed through systematic recruitment of partici- https://www.cdc.gov/coronavirus/2019-ncov/need-extra-preca pants enrolled in pharmacovigilance registries. Future linkage to utions/index.html (last accessed 15 Jan 2021). registry and healthcare records may also validate self-reported 7 GOV.UK. Guidance on shielding and protecting people who are demographic and clinical characteristics. clinically extremely vulnerable from COVID-19. Available at: https://www.gov.uk/government/publications/guidance-on-shie Shielding of at-risk individuals remains a global public lding-and-protecting-extremely-vulnerable-persons-from-covid-19/ health priority. Our study indicates that use of targeted thera- guidance-on-shielding-and-protecting-extremely-vulnerable-persons- pies is associated with shielding in individuals with RMD and from-covid-19 (last accessed 15 Jan 2021). psoriasis, compared with no systemic treatment or standard 8 World Health Organization. Information note on COVID-19 and systemic agents. This may contribute to the reported lower NCDs. Available at: https://www.who.int/publications/m/item/ risk of adverse COVID-19 outcomes associated with targeted covid-19-and-ncds (last accessed 15 Jan 2021). therapies reported by IMID registries. The observed differences 9 World Health Organization. Overview of public health and social measures in the context of COVID-19: interim guidance. 18 May in shielding across treatment groups, IMIDs, nations and time 2020. Available at: https://apps.who.int/iris/bitstream/handle/ may inform future updates of public health recommendations 10665/332115/WHO-2019-nCoV-PHSM_Overview-2020.1-eng.pdf? for COVID-19 risk-mitigating behaviours. Capture and consid- sequence=1&isAllowed=y. eration of risk-mitigating behaviour is important in future 10 Kalb RE, Fiorentino DF, Lebwohl MG et al. Risk of serious infec- studies of COVID-19 risk across people with IMIDs and on tion with biologic and systemic treatment of psoriasis: results different types of systemic treatments. from the Psoriasis Longitudinal Assessment and Registry (PSO- LAR). JAMA Dermatol 2015; 151:961–9. 11 Ramiro S, Sepriano A, Chatzidionysiou K et al. Safety of synthetic Acknowledgments and biological DMARDs: a systematic literature review informing the 2016 update of the EULAR recommendations for management We are very grateful to the patients who have contributed to of rheumatoid arthritis. Ann Rheum Dis 2017; 76:1101–36. PsoProtectMe and CORE-UK. We would like to acknowledge 12 Rutherford AI, Subesinghe S, Hyrich KL, Galloway JB. Serious the professional and patient organizations who supported or infection across biologic-treated patients with rheumatoid arthritis: promoted PsoProtectMe and CORE-UK (see Table S1). We are results from the British Society for Rheumatology Biologics Regis- grateful for the input of Prof. Lars Iversen, Prof. Nick Rey- ter for Rheumatoid Arthritis. Ann Rheum Dis 2018; 77:905–10. 13 Galloway JB, Hyrich KL, Mercer LK et al; BSRBR Control Centre nolds, Prof. Joel Gelfand, Ms Christine Janus and Ms Melissa Consortium; British Society for Rheumatology Biologics Register. Sweeney. We would like to acknowledge the following indi- Anti-TNF therapy is associated with an increased risk of serious viduals for help with translating the PsoProtectMe survey: Dr infections in patients with rheumatoid arthritis especially in the Haleema Alfailakawi, Dr Wisam Alwan, Dr Rosa Andres Ejar- first 6 months of treatment: updated results from the British Soci- que, Dr Ines Barbosa, Ms Carmen Bugarin Diz, Ms Katarzyna ety for Rheumatology Biologics Register with special emphasis on Grys, Dr Mahira Hamdy El Sayed, Mr Tran Hong Truong, Mr risks in the elderly. Rheumatology (Oxford) 2011; 50:124–31. © 2020 The Authors. British Journal of Dermatology British Journal of Dermatology (2021) 185, pp80–90 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. 88 Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. 14 Price E, MacPhie E, Kay L et al. Identifying rheumatic disease patients at high risk and requiring shielding during the COVID-19 Appendix 1 pandemic. Clin Med (Lond) 2020; 20:256–61. 15 British Association of Dermatologists. Dermatology advice regard- ing medication acting on the immune system: adults, paediatrics Author affiliations and young people. Available at: https://www.bad.org.uk/shared/ 1 St John’s Institute of Dermatology, 9NIHR Biomedical get-file.ashx?itemtype=document&id=6674 (last accessed 15 Jan 2021). Research Centre; Guy’s and St Thomas’ NHS Foundation Trust 16 Brenner EJ, Ungaro RC, Gearry RB et al. Corticosteroids, but not and King’s College London, London, UK 2 Centre for Rheumatic Diseases, 5Department of Medical Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 TNF antagonists, are associated with adverse COVID-19 outcomes in patients with inflammatory bowel diseases: results from an and Molecular Genetics, School of Basic and Medical Bio- international registry. Gastroenterology 2020; 159:481–91.e3. sciences, Faculty of Life Sciences and Medicine; 26Psychology 17 Gianfrancesco M, Hyrich KL, Al-Adely S et al; COVID-19 Global Department, Institute of Psychiatry, Psychology and Neuro- Rheumatology Alliance. Characteristics associated with hospitalisa- science; 31School of Cancer and Pharmaceutical Sciences; tion for COVID-19 in people with rheumatic disease: data from the COVID-19 Global Rheumatology Alliance physician-reported King’s College London, London, UK 3 registry. Ann Rheum Dis 2020; 79:859–66. Faculty of Epidemiology, and Population Health, London 18 Mahil SK, Dand N, Mason KJ et al; PsoProtect study group. Factors School of Hygiene and Tropical Medicine, London, UK 4 associated with adverse COVID-19 outcomes in patients with pso- Dermatology Centre, Salford Royal NHS Foundation Trust, riasis – insights from a global registry-based study. J Allergy Clin The University of Manchester; 13Kellegren Centre for Rheuma- Immunol 2020; 147:60–71. tology, Manchester University NHS Foundation Trust; Manch- 19 Harris PA, Taylor R, Thielke R et al. Research electronic data cap- ester Academic Health Science Centre, NIHR Manchester ture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomedical Research Centre, Manchester, UK 6 Biomed Inform 2009; 42:377–81. Health Data Research UK, London, UK 7 20 Clark C, Davila A, Regis M, Kraus S. Predictors of COVID-19 vol- School of Medicine, Keele University, Keele, UK 8 untary compliance behaviors: an international investigation. Glob The Psoriasis Association, Northampton, UK Transit 2020; 2:76–82. 10 International Federation of Psoriasis Associations 21 Harrison XA, Donaldson L, Correa-Cano MA et al. A brief introduc- 11 Department of Dermatology, AP-HP H^ opital Saint-Louis, tion to mixed effects modelling and multi-model inference in Paris, France ecology. PeerJ 2018; 6:e4794. 12 22 Guallar MP, Meiri~ no R, Donat-Vargas C et al. Inoculum at the time INSERM U1163, Imagine Institute for Human Genetic of SARS-CoV-2 exposure and risk of disease severity. Int J Infect Dis Diseases, Universite de Paris, Paris, France 14 2020; 97:290–2. Versus Arthritis Centre for Genetics and Genomics, 23 Pujadas E, Chaudhry F, McBride R et al. SARS-CoV-2 viral load pre- Manchester Academic Health Science Centre, The University of dicts COVID-19 mortality. Lancet Respir Med 2020; 8:e70. Manchester, Manchester, UK 24 Zheng S, Fan J, Yu F et al. Viral load dynamics and disease severity 15 Catedra de Dermatologia, Hospital de Clinicas, Facultad de in patients infected with SARS-CoV-2 in Zhejiang province, China, Ciencias Medicas, Universidad Nacional de Asuncion, Paraguay January–March 2020: retrospective cohort study. BMJ 2020; 369: 16 m1443. Clinica Dermacross, Santiago, Chile 17 25 Imai M, Iwatsuki-Horimoto K, Hatta M et al. Syrian hamsters as a St John’s Institute of Dermatology, School of Basic & small animal model for SARS-CoV-2 infection and countermeasure Medical Biosciences, Faculty of Life Sciences & Medicine, development. Proc Natl Acad Sci U S A 2020; 117:16587–95. King’s College London, London, UK 26 Cleaton N, Raizada S, Barkham N et al. COVID-19 prevalence and 18 Section of Dermatology and Venereology, University of the impact on quality of life from stringent social distancing in a Verona, Verona, Italy single large UK rheumatology centre. Ann Rheum Dis 2020; https:// 19 Department of Dermatology, Edouard Herriot Hospital, doi.org/10.1136/annrheumdis-2020-218236 27 Jia R, Ayling K, Chalder T et al. Mental health in the UK during Hospices Civils de Lyon, University of Lyon, Lyon, France 20 the COVID-19 pandemic: cross-sectional analyses from a commu- Groupe de recherche sur le psoriasis (GrPso) de la Societe nity cohort study. BMJ Open 2020; 10:e040620. francßaise de dermatologie, Paris, France 21 28 Pfefferbaum B, North CS. Mental health and the Covid-19 pan- Department of Dermatology, Ghent University, Ghent, demic. N Engl J Med 2020; 383:510–12. Belgium 29 Koliani-Pace JL, Haron AM, Zisman-Ilani Y et al. Patients’ perceive 22 NIHR Leeds Biomedical Research Centre, Leeds Teaching biologics to be riskier and more dreadful than other IBD medica- Hospitals Trust, UK tions. Inflamm Bowel Dis 2020; 26:141–6. 23 30 Robinson PC, Richards D, Tanner HL, Feldmann M. Accumulating Leeds Institute of Rheumatic and Musculoskeletal Medi- evidence suggests anti-TNF therapy needs to be given trial priority cine, University of Leeds, Leeds, UK 24 in COVID-19 treatment. The Lancet Rheumatol 2020; 2:e653–5. Institute of Infection, Immunity and Inflammation, 31 Wang Z, Tang K. Combating COVID-19: Health equity matters. University of Glasgow, Glasgow UK. Nat Med 2020; 26:458. 25 Centro Studi GISED, Bergamo, Italy 32 Scala E, Megna M, Amerio P et al. Patients’ demographic and 27 Department of Dermatology, Hospital de la Santa Creu i socioeconomic characteristics influence the therapeutic decision- Sant Pau, Universitat Aut onoma de Barcelona, Barcelona, Cat- making process in psoriasis. PLoS One 2020; 15:e0237267. alonia, Spain British Journal of Dermatology (2021) 185, pp80–90 © 2020 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. 89 28 Royal United Hospitals Bath NHS Trust, Bath, UK from AbbVie, Almirall, Biocad, Boehringer-Ingelheim, Janssen, 29 Department of Dermatology, Amsterdam Public Health/ Kyowa Kirin, Leo Pharma, Novartis, Pfizer and UCB, outside Infection and Immunology, Amsterdam University Medical the submitted work. I.N.B. reports grants from the National Centers, Location AMC, Amsterdam, the Netherlands Institute for Health Research during the conduct of the study; 30 Department of Dermatology, Centro Hospitalar do Porto, grants from Genzyme Sanofi, grants and personal fees from Portugal GSK and UCB, personal fees from Astra Zeneca, Aurinia, Eli Lilly, ILTOO and Merck Serono, outside the submitted work. F.C. reports consultancy fees from AnaptysBio and grants from Appendix 2 Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 Boheringer-Ingelheim, outside the submitted work. A.P.C. reports grants from Eli Lilly, outside the submitted work. P.D.M. reports grants and personal fees from UCB, and per- Funding sources sonal fees from Janssen and Novartis, outside the submitted We acknowledge financial support from the Department of work. P.G. reports personal fees from AbbVie, Amgen, Eli Lilly, Health via the National Institute for Health Research (NIHR) Janssen, Novartis, Pierre Fabre, Sandoz and UCB, outside the Biomedical Research Centre based at Guy’s and St Thomas’ submitted work. K.H. reports grants from Bristol Myers Squibb NHS Foundation Trust and King’s College London, the NIHR and Pfizer, and personal fees from AbbVie, outside the submit- Manchester Biomedical Research Centre and the Psoriasis Asso- ted work. D.J. reports personal fees from Amgen, and personal ciation. The views expressed are those of the author(s) and fees and nonfinancial support from AbbVie, Celgene, Janssen- not necessarily those of the NHS, the NIHR, or the Depart- Cilag, Leo-Pharma, Lilly, MEDAC and Novartis, outside the ment of Health and Social Care. S.K.M. is funded by a Medical submitted work. H.M.-O. reports grants and personal fees from Research Council (MRC) Clinical Academic Research Partner- Janssen and Novartis; and personal fees from AbbVie, Celgene, ship award (MR/T02383X/1). S.M.L. is supported by a Well- Eli Lilly, Pfizer, Takeda and UCB, outside the submitted work. come Senior Research Fellowship in Clinical Science (205039/ I.McI. reports grants from AbbVie, Astra Zeneca, Boerhinger, Z/16/Z) and by Health Data Research UK (grant no. LOND1). Bristol Myers Squibb, Celgene, Eli Lilly, Janssen, Novartis, N.D. is also funded by Health Data Research UK (grant no. Roche and UCB, outside the submitted work. L.P. reports MR/S003126/1), which is funded by the UK MRC, Engineer- grants and personal fees from AbbVie, Almirall, Amgen, Boeh- ing and Physical Sciences Research Council, Economic and ringer Ingelheim, Janssen, Lilly, Novartis and UCB; and per- Social Research Council, Department of Health and Social Care sonal fees from Bristol Myers Squibb, Fresenius-Kabi, Mylan, (England), Chief Scientist Office of the Scottish Government Pfizer, Samsung-Bioepis, Sandoz and Sanofi, outside the sub- Health and Social Care Directorates, Health and Social Care mitted work. R.S. reports speaker fees, consultancy and/or Research and Development Division (Welsh Government), grants from AbbVie, Biogen, Celgene, MSD, Novartis and UCB. Public Health Agency (Northern Ireland), British Heart Foun- T.T. reports grants and personal fees from AbbVie, Almirall, dation and Wellcome Trust. Z.Z.N.Y. is funded by a NIHR Amgen, Arena Pharmaceuticals, Biocad, Biogen, Boehringer Academic Clinical Lectureship through the University of Ingelheim, Bristol-Myers Squibb, Celgene, Eli Lilly, Janssen, Manchester. C.E.M.G. is a NIHR Emeritus Senior Investigator Leo Pharma, MSD, Novartis, Pfizer, Samsung-Bioepis and San- and is funded in part by the MRC (MR/101 1808/1). doz, during the conduct of the study. R.B.W. reports grants C.E.M.G. and R.B.W. are in part supported by the NIHR and personal fees from AbbVie, Almirall, Amgen, Celgene, Eli Manchester Biomedical Research Centre. Lilly, Janssen, Leo, Novartis, Pfizer and UCB, and personal fees from Arena, Avillion, Boehringer Ingelheim, Bristol Myers Squibb and Sanofi, outside the submitted work. H.W. is on the Appendix 3 Board of the International Federation of Psoriasis Associations who have received grants from AbbVie, Almirall, Amgen, Conflict of interest statements Boehringer Ingelheim, Bristol Meyers Squibb, Celgene, Eli Lilly, Janssen, Leo Pharma, Novartis, Pfizer, Sun Pharma and UCB, S.K.M. reports departmental income from AbbVie, Celgene, Eli outside the submitted work. J.W. has presented talks for Lilly, Janssen-Cilag, Novartis, Sanofi and UCB, outside the sub- Abbott, AbbVie, Bayer, Boehringer Ingelheim, Chiesi, Roche mitted work. K.J.M. reports personal fees from Leo Pharma and and Merck. C.E.M.G. reports grants and personal fees from Novartis, outside the submitted work. H.McA. reports grants AbbVie, Janssen, Lilly, Novartis and UCB Pharma, and grants from AbbVie, Almirral, Amgen, Celgene, Dermal Laboratories, from Almirall, Amgen, BMS, Leo and Pfizer, outside the sub- Eli Lilly, Janssen, Leo Pharma, T and R Derma, and UCB, out- mitted work. J.N.B. reports grants and personal fees from Abb- side the submitted work. F.M. has nothing to disclose. D.U. Vie, Lilly, J & J and Novartis, during the conduct of the study. reports grants from AbbVie, Almirral, Amgen, Celgene, Dermal J.B.G. reports personal fees from AbbVie, Janssen, Novartis, Pfi- Laboratories, Eli Lilly, Janssen, Leo Pharma, T & R Derma and zer, Sanofi and UCB, and grants from Eli Lilly, outside the sub- UCB, outside the submitted work. L.M. reports personal fees mitted work. C.H.S. reports grants from AbbVie, Novartis, from AbbVie, Celgene, Janssen, Leo Pharma, Novartis and Pfizer and Sanofi, outside the submitted work. No other con- UCB, outside the submitted work. H.B. reports personal fees flicts of interest have been reported. © 2020 The Authors. British Journal of Dermatology British Journal of Dermatology (2021) 185, pp80–90 published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists. 90 Shielding behaviour and inflammatory skin and joint disease, MAHIL et al. Table S5 UK-only analysis: multivariable logistic regression Supporting Information model characterizing the association between therapy and Additional Supporting Information may be found in the online shielding behaviour (reference group is no systemic therapy). version of this article at the publisher’s website: Table S6 UK-only analysis: multivariable logistic regression model characterizing the association between therapy and Table S1 Organizations that supported or promoted PsoPro- shielding behaviour (reference group is no systemic therapy), tectMe and/or CORE-UK. with time of survey completion included as a fixed covariate. Table S2 Therapy breakdown. Table S7 UK-only analysis: multivariable logistic regression Downloaded from https://academic.oup.com/bjd/article/185/1/80/6600367 by Fundacio Biblioteca Josep Laporte user on 20 February 2023 Table S3 Imputed multivariable logistic regression model model characterizing the association between therapy and characterizing the association between therapy and shielding shielding behaviour (reference group is no systemic therapy), behaviour (reference group is no systemic therapy). with time of survey completion included as an interaction Table S4 Multivariable logistic regression model character- term with treatment. izing the association between biologic therapy and shielding Figure S1 Estimated shielding over time, UK respondents behaviour with standard systemic therapy as the comparator only. group. Figure S2 Estimated shielding over time, non-UK respon- dents only. British Journal of Dermatology (2021) 185, pp80–90 © 2020 The Authors. British Journal of Dermatology published by John Wiley & Sons Ltd on behalf of British Association of Dermatologists.