Nonadherence to systemic immune-modifying therapy in people with psoriasis during the COVID-19 pandemic: findings from a global cross-sectional survey

Background Nonadherence

Treatment adherence refers to the extent to which a person's behaviour in taking medication corresponds with agreed recommendations from a healthcare provider. 1 This is in terms of dosage, timing and frequency of drug administration, and is based on available clinical evidence demonstrating optimal outcomes with minimal side-effects. 1 Nonadherence can be described as either unintentional or intentional. 2 Unintentional nonadherence occurs when an individual wants to follow their prescribed treatment regimen but is unable to do so due to lack of capacity or specific resources (e.g. forgetfulness or cost of medications). 2 In contrast, intentional nonadherence occurs when an individual decides not to follow recommendations. 2 Intentional nonadherence is a complex behaviour influenced by several factors, including perceptual factors that influence their motivation to start and continue treatment (e.g. beliefs and preferences) and mental health. [3][4][5] Before the COVID-19 pandemic, an association between mental health conditions such as anxiety or depression and nonadherence was found in a range of longterm conditions, 3,5 including immune-mediated inflammatory diseases such as psoriasis; 6 however, it is unclear whether these associations have changed during the pandemic.
The mainstay of treatment for moderate-to-severe psoriasis is systemic immune-modifying therapy, either standard systemic or targeted biologic therapies. 7 These agents have been the subject of considerable attention during the COVID-19 pandemic due to their mechanism of action (i.e. immune modulation) and established association with infections. 8,9 At the beginning of the COVID-19 pandemic, risk mitigation strategies for individuals receiving immune-modifying therapies focused on behaviours to limit exposure to the virus (e.g. shielding) rather than discontinuing treatment, given the importance of ensuring ongoing control of psoriasis, 10 including avoiding demand on healthcare services due to disease flares. Data from registry-and population-based COVID-19 studies 9,11-16 have supported this approach, with evidence suggesting that risks of adverse outcomes in this group are mainly driven by the same factors as in the general population (e.g. age, sex, ethnicity and comorbid burden).
Emerging research indicates that the COVID-19 pandemic may have negatively impacted adherence to systemic immune-modifying therapies in people with psoriasis, [17][18][19][20][21][22][23][24] with nonadherence ranging between 1.6% 20 and 68.5% 21 and considerable variation reported across systemic immune-modifying therapy types. 21 Global mental health has also been adversely affected in the COVID-19 pandemic. [25][26][27][28] Understanding the extent of nonadherence to different systemic immune-modifying therapies, and the potential factors contributing to it, is the first necessary step in helping patients to gain more benefit from their What is already known about this topic?
• Before the COVID-19 pandemic, mental health disorders including anxiety and depression were shown to be associated with nonadherence to immune-modifying therapy in individuals with immune-mediated diseases. • The COVID-19 pandemic has led to a rise in the global prevalence of mental health disorders in the general population.
• The extent of nonadherence to immune-modifying therapy in individuals with psoriasis and its association with mental health during the COVID-19 pandemic remains poorly understood.

What does this study add?
• One-quarter of individuals with psoriasis who completed PsoProtectMe reported nonadherence to systemic immune-modifying therapy during the pandemic, primarily due to concerns around their immunity. • After adjusting for confounders (age, sex, ethnicity, comorbidity) and country of residence, the association between a positive screen for anxiety and nonadherence to immune-modifying therapy in individuals with psoriasis was attenuated and no longer significant. • Future research in larger populations should further explore pandemic-specific drivers of treatment nonadherence. • Clear communication of the reassuring findings from population-based research regarding immune-modifying therapy-associated adverse COVID-19 risks to individuals with psoriasis is essential, to optimize adherence and disease outcomes. Downloaded from https://academic.oup.com/bjd/advance-article/doi/10.1093/bjd/ljac144/6957428 by guest on 03 March 2023 medication(s). Furthermore, the COVID-19 pandemic provides an important opportunity to better understand the potential complex relationship between mental health and treatment nonadherence.
To address this, we analysed self-report data from the PsoProtectMe international survey, which explores the global impact of the COVID-19 pandemic on individuals with psoriasis and includes validated screens for anxiety (Generalized Anxiety Disorder-2 29 ) and depression (Patient Health Questionnaire-2 30 ). Our specific objectives were: 1To characterize the extent of, and reasons underlying, self-reported nonadherence to systemic immune-modifying therapy in people with psoriasis during the COVID-19 pandemic.
2To assess whether an association exists between a positive screen for anxiety or depression and self-reported nonadherence to systemic immune-modifying therapy in people with psoriasis during the COVID-19 pandemic.

Design, study participants and recruitment strategy
A cross-sectional survey design was employed. The inclusion criterion was any individual reporting a clinician-confirmed diagnosis of psoriasis. There was no age limit (questions could be completed by the individual or on behalf of someone else). Following its launch on 4 May 2020, PsoProtectMe (https:// psoprotectme.org/; available in nine different languages) was promoted globally via social media, patient organizations and clinical networks (REC reference 20/YH/0135). Data were collected and managed using REDCap (Research Electronic Data Capture) tools licensed to King's College London Division of Health and Social Care Research. 31

Materials
We defined a minimum sufficient core set of variables within PsoProtectMe with our study group of clinicians, researchers and patient representatives. 10,32 Treatment adherence was determined by a question which asked whether the participant had stopped or delayed their systemic immune-modifying medication during the COVID-19 pandemic, and the underlying reason(s) for doing so (see https://psoprotectme. org/ for full questionnaire). All immune-modifying treatments are listed in Table S1; see Supporting Information. If a respondent was on co-therapy (i.e. prescribed more than one standard systemic and/or targeted therapy), their adherence was assessed for each medication separately. If a respondent who was receiving co-therapy was nonadherent to one or more of their medications, they were classified as nonadherent overall. We used the Generalized Anxiety Disorder 2-item (GAD-2) 29 and the Patient Health Questionnaire-2 (PHQ-2) 30 to screen for anxiety and depression, respectively; scores of ≥ 3 were considered positive.

Statistical analysis
Data were analysed using SPSS (version 27.0 for Windows). Due to the continuously evolving nature of the COVID-19 pandemic, we specifically assessed the extent, reasons for, and potential determinants of systemic immune-modifying therapy nonadherence during the first year of the pandemic (to 16 March 2021), when most global populations were being affected by some form of containment measure.
We used descriptive statistics to summarize the findings; we reported continuous variables using medians and interquartile ranges (IQRs), and categorical/dichotomous variables as numbers and percentages. We used binomial logistic regression to assess associations between positive screens for anxiety or depression and nonadherence to systemic immune-modifying therapy. These models were then adjusted for potential confounders (age, sex, ethnicity and medical comorbidity), as well as country of residence. All underlying assumptions for all statistical tests were met, unless otherwise stated. A P-value of ≤ 0.05 was used to demonstrate statistical significance.

Results
Clinical and sociodemographic characteristics of the study population  Table 1. Table S1 provides a breakdown of the standard systemic and targeted immune-modifying therapies used by respondents.
Reasons for nonadherence to systemic immune-modifying therapy are presented in Table 2. Concerns about their own immunity ('I didn't want it to affect my immunity') was the most common reason for discontinuation (48.2%), followed by advice from their doctor ['My doctor told me to stop it' (30.7%)], concern about complications (19.1%) and having no medicine supply (15.2%).  After adjustment for potential confounders (age, sex, ethnicity, medical comorbidity) and country of residence, although the direction of the effect remained, the associations were attenuated and anxiety was no longer significantly associated with nonadherence to overall systemic (OR 1.20, 95% CI 0.92-1.56) or targeted (OR 1.33, 95% CI 0.94-1.89) immune-modifying therapy. After removing respondents whose reason for nonadherence was solely their doctor's advice or having no supply, a positive screen for anxiety was also associated with nonadherence to systemic immune-modifying therapy in the unadjusted model (OR 1.34, 95% CI 1.02-1.76), and this association was similarly attenuated and no longer significant after adjustment for confounders (OR 1.20, 95% CI 0.89-1.60).

Discussion
The extent of nonadherence to systemic immune-modifying therapy within our global sample of individuals with psoriasis during the first year of the COVID-19 pandemic is substantial (25.3%). Furthermore, these data indicate that although a positive screen for anxiety was associated with nonadherence in crude analyses, the association was attenuated after adjustment for confounders.
The proportion of individuals with psoriasis reporting nonadherence to systemic immune-modifying therapy is higher than indicated by some studies conducted before the pandemic. 33,34 For example, in studies by Chan et al. 33 and Thorneloe et al., 34 14.2% and 16.4% of study participants were nonadherent, respectively. Immediately following the worldwide spread of COVID-19, concern emerged regarding the safety of immune-modifying medications, specifically in relation to potential increased susceptibility to and/or severity of COVID-19. Biologic drugs targeting tumour necrosis factors interleukin (IL)-12/23, IL-23 and IL-17 can increase individuals' risk of contracting common and opportunistic infections. 8,9 However, importantly, there is now a growing reassuring evidence base supporting the continuation of systemic immune-modifying therapies during the pandemic. Registry data suggest that biologics do not increase susceptibility to and/or severity of COVID-19. 8,12,20,23,24,[35][36][37][38] For example, our registry-based studies 12,13 and others 11,14,15 indicated that biologic use was associated with lower risk of COVID-19-related hospitalization compared with nonbiologic systemic therapies, albeit with potentially confounding shielding behaviour differences across treatment groups. 10 Larger-scale population-based research suggests no differences in COVID-19-related death in adults prescribed targeted immune-modifying therapy compared with those on standard systemic therapy. 16 These data have helped to inform international clinical guidance supporting treatment adherence during the pandemic. 11 Our findings (both unadjusted and adjusted logistic regression models) indicate that depression was not significantly associated with nonadherence to systemic immune-modifying therapy during the pandemic. In contrast, prior research conducted before the pandemic suggested an association between depression and treatment nonadherence across a range of chronic conditions, 3,5 including immune-mediated inflammatory diseases. 6 Pre-pandemic research in psoriasis also indicated that medication nonadherence was linked to feeling anxious about potential drug adverse effects. 39 Indeed, in our unadjusted logistic regression model, we identified an association between anxiety and nonadherence to systemic immune-modifying therapy, more specifically targeted therapy. As targeted therapies are a newer treatment compared with standard systemic therapies, 40 patients may be more concerned about the potential adverse effects of this therapy both during and beyond the COVID-19 pandemic. Because good clinician communication is positively correlated with patient treatment adherence, 41 it is important for clinicians to clearly communicate the findings from population-based research regarding immune-modifying therapy-associated risks to patients, both during and beyond the COVID-19 pandemic.
However, after adjustment for potential confounders, while the direction of the effect remained, the association between anxiety and nonadherence to targeted immune-modifying therapy was attenuated and no longer significant, although wide confidence intervals reflect a lack of precision of the estimate, and replication in larger populations would be of value. The COVID-19 pandemic is a unique period, with major impacts on global mental health in the general population. [25][26][27][28] There may be different drivers of treatment nonadherence during this period compared with pre-pandemic, such as concerns surrounding own immunity, as suggested in our study and others. [18][19][20][22][23][24] As the COVID-19 pandemic evolves, it is important to understand and address pandemic-specific drivers of treatment nonadherence in psoriasis, to improve medication behaviours and subsequent patient outcomes.
A strength of our study is that we have expanded a limited evidence base on the association between immune-modifying treatment nonadherence and mental health during the COVID-19 pandemic in individuals with psoriasis using our global dataset. However, this study has several limitations; for example, the cross-sectional design precludes any inferences about causality or directionality of associations. The statistical tests employed may have been underpowered, increasing the probability of type II errors, as reflected by the width of the confidence intervals reducing the certainty of the estimates, and do not take into account the duration of nonadherence. Self-report assessments have been shown to underestimate true levels of medication nonadherence 42 due to reasons such as social desirability bias. Individuals nonadherent to immune-modifying treatment or with low computer literacy may have been disinclined to participate in this online study, potentially introducing ascertainment bias and under-estimating some of our findings. Our study population included self-selecting responders to PsoProtectMe and was dominated by females of white ethnicity residing in the UK, therefore limiting the generalizability of our results. To validate self-reported sociodemographic and clinical characteristics, future research should consider linkage to registry and healthcare records.
In conclusion, we have identified a substantial prevalence of nonadherence to immune-modifying treatment (roughly 25%) during the COVID-19 pandemic in our international sample of individuals with psoriasis, primarily due to concerns surrounding respondents' immunity. As global mental health has been adversely affected in the pandemic, 25-28 our finding that depression and anxiety were not strongly related to treatment nonadherence is reassuring. Future research should further explore pandemic-specific drivers of treatment nonadherence in larger populations. Communication of the reassuring findings from population-based research regarding immune-modifying therapy-associated adverse COVID-19 risks to individuals with psoriasis is essential, to optimize adherence and disease outcomes.
Sanofi, outside the submitted work. P.S. has done consultancies in the past for Sanofi 111017 and AbbVie 041217 (unpaid); has received departmental independent research grants for TREAT NL registry from pharma companies since December 2019; is involved in performing clinical trials with many pharmaceutical industries that manufacture drugs used for the treatment of diseases such as psoriasis and atopic dermatitis, for which financial compensation is paid to the department/hospital; and is chief investigator