Categories
Nevin Manimala Statistics

Devices measuring transepidermal water loss: A systematic review of measurement properties

Skin Res Technol. 2022 Apr 12. doi: 10.1111/srt.13159. Online ahead of print.

ABSTRACT

OBJECTIVE: The objective of this review is to examine the reliability and measurement error of devices that measure transepidermal water loss (TEWL).

INTRODUCTION: TEWL is a physiological property of skin which increases when the epidermis is damaged. It is, therefore, a commonly utilised measure of skin barrier integrity. Devices measuring TEWL are available as open, semi-open or closed chamber. Studies of reliability examine the consistency of measurement, and/or responsiveness whereas measurement error scores in absolute terms the amount of error due to sources of variation.

INCLUSION CRITERIA: Studies examining the reliability and/or measurement error of TEWL measurement devices were included. Studies that only report on measurement of TEWL outcomes without examination of reliability and/or measurement error were excluded.

METHODS: The search strategy aimed to locate published and unpublished studies. Databases searched included PubMed, Embase, CINAHL and Web of Science, utilising identified keywords and limited to studies in English. Grey literature sources were searched to identify any unpublished documents. Study selection using the inclusion criteria was then assessed by two reviewers for methodological quality utilising the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) risk of bias tool to assess the reliability and measurement error of outcome measurement instruments.

RESULTS: A total of 22 devices were examined in the 38 included studies. The quality of study design was on average rated as ‘Adequate’ however reliability and measurement error statistical methods were on average rated as ‘Doubtful’.

DISCUSSION AND CONCLUSION: TEWL measurement devices were found to demonstrate good reliability and frequently correlated with other devices. However, measurement error was highly variable but improves under in vitro conditions. Future research should consider risk of bias factors when designing studies.

PMID:35411958 | DOI:10.1111/srt.13159

By Nevin Manimala

Portfolio Website for Nevin Manimala