Andrea Bauer, Henriette Rönsch, Peter Elsner, Daan Dittmar, Cathy Bennett, Marie‐Louise A Schuttelaar, Judit Lukács, Swen Malte John, Hywel C Williams

Occupational irritant hand dermatitis (OIHD) causes significant functional impairment, disruption of work, and discomfort in the working population. Different preventive measures such as protective gloves, barrier creams and moisturisers can be used, but it is not clear how effective these are. This is an update of a Cochrane review which was previously published in 2010. To assess the effects of primary preventive interventions and strategies (physical and behavioural) for preventing OIHD in healthy people (who have no hand dermatitis) who work in occupations where the skin is at risk of damage due to contact with water, detergents, chemicals or other irritants, or from wearing gloves. We updated our searches of the following databases to January 2018: the Cochrane Skin Specialised Register, CENTRAL, MEDLlNE, and Embase. We also searched five trials registers and checked the bibliographies of included studies for further references to relevant trials. We handsearched two sets of conference proceedings. We included parallel and cross-over randomised controlled trials (RCTs) which examined the effectiveness of barrier creams, moisturisers, gloves, or educational interventions compared to no intervention for the primary prevention of OIHD under field conditions. We used the standard methodological procedures expected by Cochrane. The primary outcomes were signs and symptoms of OIHD developed during the trials, and the frequency of treatment discontinuation due to adverse effects. We included nine RCTs involving 2888 participants without occupational irritant hand dermatitis (OIHD) at baseline. Six studies, including 1533 participants, investigated the effects of barrier creams, moisturisers, or both. Three studies, including 1355 participants, assessed the effectiveness of skin protection education on the prevention of OIHD. No studies were eligible that investigated the effects of protective gloves. Among each type of intervention, there was heterogeneity concerning the criteria for assessing signs and symptoms of OIHD, the products, and the occupations. Selection bias, performance bias, and reporting bias were generally unclear across all studies. The risk of detection bias was low in five studies and high in one study. The risk of other biases was low in four studies and high in two studies.The eligible trials involved a variety of participants, including: metal workers exposed to cutting fluids, dye and print factory workers, gut cleaners in swine slaughterhouses, cleaners and kitchen workers, nurse apprentices, hospital employees handling irritants, and hairdressing apprentices. All studies were undertaken at the respective work places. Study duration ranged from four weeks to three years. The participants' ages ranged from 16 to 67 years.Meta-analyses for barrier creams, moisturisers, a combination of both barrier creams and moisturisers, or skin protection education showed imprecise effects favouring the intervention. Twenty-nine per cent of participants who applied barrier creams developed signs of OIHD, compared to 33% of the controls, so the risk may be slightly reduced with this measure (risk ratio (RR) 0.87, 95% confidence interval (CI) 0.72 to 1.06; 999 participants; 4 studies; low-quality evidence). However, this risk reduction may not be clinically important. There may be a clinically important protective effect with the use of moisturisers: in the intervention groups, 13% of participants developed symptoms of OIHD compared to 19% of the controls (RR 0.71, 95% CI 0.46 to 1.09; 507 participants; 3 studies; low-quality evidence). Likewise, there may be a clinically important protective effect from using a combination of barrier creams and moisturisers: 8% of participants in the intervention group developed signs of OIHD, compared to 13% of the controls (RR 0.68, 95% CI 0.33 to 1.42; 474 participants; 2 studies; low-quality evidence). We are uncertain whether skin protection education reduces the risk of developing signs of OIHD (RR 0.76, 95% CI 0.54 to 1.08; 1355 participants; 3 studies; very low-quality evidence). Twenty-one per cent of participants who received skin protection education developed signs of OIHD, compared to 28% of the controls.None of the studies addressed the frequency of treatment discontinuation due to adverse effects of the products directly. However, in three studies of barrier creams, the reasons for withdrawal from the studies were unrelated to adverse effects. Likewise, in one study of moisturisers plus barrier creams, and in one study of skin protection education, reasons for dropout were unrelated to adverse effects. The remaining studies (one to two in each comparison) reported dropouts without stating how many of them may have been due to adverse reactions to the interventions. We judged the quality of this evidence as moderate, due to the indirectness of the results. The investigated interventions to prevent OIHD probably cause few or no serious adverse effects. Moisturisers used alone or in combination with barrier creams may result in a clinically important protective effect, either in the long- or short-term, for the primary prevention of OIHD. Barrier creams alone may have slight protective effect, but this does not appear to be clinically important. The results for all of these comparisons were imprecise, and the low quality of the evidence means that our confidence in the effect estimates is limited. For skin protection education, the results varied substantially across the trials, the effect was imprecise, and the pooled risk reduction was not large enough to be clinically important. The very low quality of the evidence means that we are unsure as to whether skin protection education reduces the risk of developing OIHD. The interventions probably cause few or no serious adverse effects.We conclude that at present there is insufficient evidence to confidently assess the effectiveness of interventions used in the primary prevention of OIHD. This does not necessarily mean that current measures are ineffective. Even though the update of this review included larger studies of reasonable quality, there is still a need for trials which apply standardised measures for the detection of OIHD in order to determine the effectiveness of the different prevention strategies.