This paper pools four diaper trials in 1601 infants to test how strongly skin wetness and skin pH track with diaper dermatitis. Wetness was measured as transepidermal water loss and pH with a skin electrode, alongside nurse grades of rash. In the combined data, diapered skin was wetter and had a higher pH than adjacent undiapered skin, and grades were higher when wetness or pH was in the elevated group. Wetness showed the stronger association at each grading period. No heavy-metal concentration is reported. The pH and wetness results are skin-physiology findings and are not entered in the ledger.
Key numbers
This source does not report a heavy-metal concentration, a metal release value, or a metal limit of its own.
Methods (brief)
Four trials, each about 16 weeks, compared coded paper diapers and cloth diapers. Skin in the diaper area was graded every two weeks. Wetness used evaporimeter readings at the suprapubic area and buttock just after diaper removal, minus a thigh control. pH was read on diapered suprapubic skin and on the thigh, avoiding dermatitis. Normal versus elevated groups were defined as pH below 5.9 versus above 6.1 and transepidermal water loss below 9.0 versus above 11.0, then analyzed by two-way ANOVA.
Evidence fitness
A statistical association, in diapered infants, between higher skin wetness, higher skin pH, and higher diaper-dermatitis grades.
They do not support Any heavy-metal concentration in diapers, diaper components, wipes, or creams; A causal proof that wetness or pH produces the rash. The analysis is an association across grouped clinical observations; Named-product performance. Test diapers are coded by letter, and about 30% of observations were dropped to create a gap between normal and elevated groups.
Limitations
About 30% of the base was excluded by the gap between normal and elevated cut points. Grading-period rows are paired observations, not 1601 independent infants at every visit. Diaper products are coded, not identified. The text does not name the country of enrollment. The second author’s surname is printed in the text layer as MiUigan, read here as Milligan.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.