Effectiveness and safety of lotion, cream, gel, and ointment emollients for childhood eczema: a pragmatic, randomised, phase 4, superiority trial.

Background: To our knowledge, there are no trials comparing emollients commonly used for childhood eczema. We aimed to compare the clinical effectiveness and safety of the four main emollient types: lotions, creams, gels, and ointments.

Methods: We did a pragmatic, individually randomised, parallel group, phase 4 superiority trial in 77 general practice surgeries in England. Children aged between 6 months and 12 years with eczema (Patient Orientated Eczema Measure [POEM] score >2) were randomly assigned (1:1:1:1; stratified by centre and minimised by baseline POEM score and age, using a web-based system) to lotions, creams, gels, or ointments. Clinicians and parents were unmasked. The initial emollient prescription was for 500 g or 500 mL, to be applied twice daily and as required. Subsequent prescriptions were determined by the family. The primary outcome was parent-reported eczema severity over 16 weeks (weekly POEM), with analysis as randomly assigned regardless of adherence, adjusting for baseline and stratification variables. Safety was assessed in all randomly assigned participants. This trial was registered with the ISRCTN registry, ISRCTN84540529.

Findings: Between Jan 19, 2018, and Oct 31, 2019, 12 417 children were assessed for eligibility, 550 of whom were randomly assigned to a treatment group (137 to lotion, 140 to cream, 135 to gel, and 138 to ointment). The numbers of participants who contributed at least two POEM scores and were included in the primary analysis were 131 in the lotion group, 137 in the cream group, 130 in the gel group, and 126 in the ointment group. Baseline median age was 4 years (IQR 2-8); 255 (46%) participants were girls, 295 (54%) were boys; 473 (86%) participants were White; and the mean POEM score was 9·3 (SD 5·5). There was no difference in eczema severity between emollient types over 16 weeks (global p value=0·77), with adjusted POEM pairwise differences of: cream versus lotion 0·42 (95% CI -0·48 to 1·32), gel versus lotion 0·17 (-0·75 to 1·09), ointment versus lotion -0·01 (-0·93 to 0·91), gel versus cream -0·25 (-1·15 to 0·65), ointment versus cream -0·43 (-1·34 to 0·48), and ointment versus gel -0·18 (-1·11 to 0·75). This result remained unchanged following multiple imputation, sensitivity, and subgroup analyses. The total number of adverse events did not significantly differ between the treatment groups (lotions 49 [36%], creams 54 [39%], gels 54 [40%], and ointments 48 [35%]; p=0·79), although stinging was less common with ointments (12 [9%] of 138 participants) than lotions (28 [20%] of 137), creams (24 [17%] of 140), or gels (25 [19%] of 135).

Interpretation: We found no difference in effectiveness between the four main types of emollients for childhood eczema. Users need to be able to choose from a range of emollients to find one that they are more likely to use effectively.

Funding: National Institute for Health and Care Research.