Adverse skin reactions to personal protective equipment during COVID‐19 pandemic in Italian health care workers

Abstract To avoid exposure to SARS‐COV‐2, healthcare professionals must use personal protective equipment (PPE). Their use has been related to a series of adverse effects; the most frequent adverse events were headache, dyspnoea, and pressure injuries. Skin adverse effects are very common, including contact dermatitis, itching, erythema, and acneiform eruptions. The objective of this study is to evaluate the skin problems caused by personal protection equipment (PPE) in health care workers (HCWs) and to individuate eventual risk factors. From May to June 2020 a retrospective observational multi‐centric study conducted by an online survey sent by email, involving 10 hospital centers, was performed. We considered as independent variables gender and age, occupational group and sector, time of utilization, type and material of PPE. We tested 3 types of PPE: gloves, bonnet, and mask for different time of utilization (<1, 1–3, 3–6, >6 h). We performed a multiple logistic regression model to correlate them with skin adverse events occurrence. Among all the 1184 participants, 292 workers reported a dermatological pathology: 45 (15.41%) had psoriasis, 54 (18.49%) eczema, 38 (13.01%) acne, 48 (16.44%) seborrheic dermatitis, and 107 (36.64%) other. In our sample previous inflammatory dermatological conditions, female sex, prolonged use of PPE were significant risk factors for developing skin related adverse events considering all the PPE considered. The use of PPE is still mandatory in the hospital setting and skin adverse reactions still represent a global problem. Although data from Europe are limited, our study highlighted the importance of the problem of PPE skin reactions in a large sample of Italian healthcare professionals.

The aim of this study was to evaluate the skin problems caused by personal protection equipment (PPE) in health care workers (HCWs) suggesting preventive measures to avoid the risk to develop skin diseases.

| MATERIALS AND METHODS
This work was designed a retrospective observational multi-centric study conducted by an online survey sent by email, involving 10 hospi- We performed descriptive statistics to describe sociodemographic aspects of participants and study variables characteristics, which were presented through theoretical score ranges, arithmetic means, standard deviations. Pearson bivariate correlations were performed to check multi-collinearity and to give some preliminary information into relationships between dermatological disease and the use of PPE. P values were considered significant if they were <0.05.
In a second stage the significant predictors of the first stage were entered together in a final multiple logistic regression models. Data were stratified by gender, age, occupational group and sector, time of PPE utilization, type and material of PPE. Crude odds ratios (ORs) and adjusted ORs for all the other entered variables, along with 95% confidence intervals, were calculated. To analyze the collected data, we used the STATA 16 statistical package.

| RESULTS
The sample included 1184 participants; skewness and kurtosis were used to investigate the distribution of the collected data and Shapiro- We observed in 25 (2.11%) workers a loss of occupational days due to dermatological illness; in 56 times workers asked for occupational physician surveillance; in 30 cases, HCWs were removed from their workplaces and treated with specialist dermatological prescriptions.
We evaluated 3 types of PPE: gloves (Table 1), bonnet ( Table 2) and mask ( We first analyzed the adverse effects on the hands; we found a statistically significant correlation (p < 0.05) between adverse effects and dermatological pathologies (atopic dermatitis [AD], SD, acne and psoriasis), age, sex, and time of utilization (Table 5). There was no correlation with the type of gloves (p = 0.28).
The regression analysis showed an OR of 1.56 for women, an higher OR for young adults (the highest risk has been observed between 21 and 30 years) and an increased risk with time of utilization of the gloves; the job and the working sector did not have some role. The model of regression had p < 0.001, R2 of 0.06, sensitivity of 71.07% and specificity of 53.55%; the model describes well the set of observations (goodness of fit p = 0.41).
Concerning the adverse effects linked to bonnet use, we found a statistically significant (p < 0.05) correlation between adverse effects and dermatological pathologies, age, sex, and time of utilization; there was no correlation with age (p = 0.12) and type of PPE The regression analysis showed an OR of 1.87 for women, an higher OR for young adults (the highest risk has been observed between 21 and 30 years) and an increased risk with time of utilization of the gloves; the job and the working sector did not have some

| DISCUSSION
There is growing evidence that COVID-19 is associated with a variety Regarding PPEs related acne, which may be considered a subtype of acne mechanica, it occurs more frequently in subjects with a previous history of acne vulgaris. In this case, the endogenous skin characteristic that may explain this increased prevalence is hyperseborrhea, that leads to the development of an acne prone skin.
The high temperature of the face covered by the mask induces an increased sebum excretion rate by 10% for each 1 C rise. In those subjects with a baseline overproduction of sebum this promotes the occurrence of acneic lesions especially in concomitance with surgical mask use. 15 Finally, psoriatic patients are more likely to develop inflammatory lesions compared with the unaffected subjects as a result of maskrelated Koebner phenomenon. 16 All the PPEs adverse effects have shown to be time-dependent.
The prolonged use of these devices promotes the development of a warm, moist, occlusive environment, a well described risk factor for the occurrence of various inflammatory lesions.
Last, our study has put in evidence that women more likely than men report the occurrence of skin problems caused by all the examined PPEs. Possible reasons for these gender disparity can be individuated in the use of cosmetics, that can be a triggering factor