Predicting respirator fitting using facial dimensions among public health workers in selected vector control units, Ministry of Health Malaysia

Abstract

A respirator is commonly used in the workplace to prevent harmful compounds from being inhaled. Many employers use respirators as their primary risk management tool for their employees. It has been demonstrated that an individual's facial dimensions and ethnicity can influence respirator fit. As most respirators on the market today are designed mainly based on western facial measurements, they may not be suitable for Malaysian facial dimensions. The aim of the study was to predict respirator fit based on the facial dimensions of public health workers involved in vector control unit. A questionnaire was used to gather worker background information. The Borg scale was used to assess the perceived comfort. Their 3D facial dimension was captured using a mobile face scanner app (Bellus 3D FaceApp) and measured using GOM Inspect Software. A respirator fit tester device (PortaCount 8040) was used to measure the quantitative respirator fit factor. A descriptive analysis was performed to describe the background information, perceived comfort, facial dimensions, and respirator fit factor. Multiple linear regression analysis was carried out to identify predictors of the respirator fit factor. The decision tree model was developed using RapidMiner to predict the respirator fit factor. The decision tree analysis acted as a complementary analysis to handle the methodological issues from the multiple linear regression. A total of 180 male subjects participated in this study, who were mainly Malay with a mean (standard deviation) age of 35.9 (8.1) years. After adjusting for confounders, the menton-sellion (p= 0.009, 95% CI= 0.008,0.052) and interpupillary distance (p= 0.033, 95% CI= 0.004.0.091) are significant facial dimension predictors for respirator fit. The respirator fit prediction model was established using menton-sellion, interpupillary distance, and the ethnicity, as the selected attributes give the lowest root mean square error of 0.97+/-0.205. In conclusion, menton-sellion and interpupillary distance are important facial dimensions to predict respirator fit, highlighting the need to consider ethnicity. This model could be very useful in the future for designing a suitable respirator for Malaysian workers.