Polderman 2003.

Methods	Parallel‐group, randomised controlled trial This study was carried out in a secondary care setting at a single dermatology clinic in the Netherlands
Participants	28 participants with dyshidrotic hand eczema, with duration of 4 months to 34 years Dropouts: 3 Inclusion criteria of the trial Dyshidrotic hand eczema Exclusion criteria of the trial Younger than 18 years old Use of systemic immunosuppressive or immunomodulating medication in the last 2 months Pregnancy History of UV sensitivity or skin malignancy Study population Gender: not stated Age: not stated
Interventions	Intervention • UVA‐1 irradiation 40 J/cm² on the hands in 15/15 participants 5 times weekly for 3 weeks Control intervention • Placebo (simulated blue light) in 10/13 participants Emollients probably were allowed in both groups Participants were followed until 6 weeks after the end of treatment Duration 9 weeks (3 weeks active treatment, 6 weeks follow‐up)
Outcomes	Primary outcomes of the trial Observer‐rated severity by the dyshidrotic eczema area and severity index (DASI; based on sum score for severity 1 = mild, 2 = moderate, 3 = severe for, respectively, vesicles, erythema, desquamation, itch, multiplied by score for affected area); time point unclear Other outcomes VAS for itch (probably participant‐rated) at the end of each week and 3 and 6 weeks after treatment Observer‐rated reduction in severity for separate items of DASI at the end of each week and 3 and 6 weeks after treatment Adverse events
Notes	Primary outcome probably at week 3 (i.e. at end of treatment). Analysis based on intention‐to‐treat principle. There was a follow‐up 6 weeks after treatment, but only summary data were given for the treatment group Study authors were contacted on 28 February 2014 but remained not responsive Declarations of interest: not stated Funding: not stated Sample size rationale: not stated
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "they were randomly assigned to either UVA‐1 (n=15) or placebo treatment (n=13) by an independent investigator using a lottery system" Comment: reference to a lottery system, which is an adequate way to prevent selection bias
Allocation concealment (selection bias)	Low risk	Quote: "they were randomly assigned to either UVA‐1 (n=15) or placebo treatment (n=13) by an independent investigator using a lottery system" Comment: participants were assigned to different study arms by an independent investigator using a lottery system
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "placebo treatment comprised of TL tubes, emitting visible light, covered with a blue plastic plate to mimic the blue UVA‐1 light" Comment: participants: placebo design to mimic the intervention; participants wore protective eyewear and protection on their forearms during both interventions. For participants, the placebo is probably indistinguishable from the actual treatment; however personnel who delivered the treatment were not blinded. Given that the staff had to administer the treatment, and we could not think of a better way to blind participants and staff, this was considered as low risk
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "a blinded investigator was responsible for the evaluation of the parameters" Comment: observer‐blinded; this was another person, then the one who assigned participants to treatment arms
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "analysis was performed according to the intention‐to‐treat principle" Comment: intention‐to‐treat analyses
Selective reporting (reporting bias)	Unclear risk	No trial registration found. No major differences between Methods and Results sections for results during the treatment phase, except that results 6 weeks after treatment are very scarce. Study authors remarked that for ethical reasons, a lot of participants were prescribed topical corticosteroids; therefore these results are less reliable and probably are poorly reported
Other bias	Unclear risk	Baseline comparisons: no baseline comparisons regarding group differences (randomisation check) Diagnostic certainty: yes The study was completed