Table 4

Summary of findings: association of variables facial dimensions, gender and ethnicity with RPE fit



Outcome
Studies assessing outcomeSummary of findings (n=number of studies)
Significant associationWeak association or mixed resultsNo association
Pass rates†26High (≥75%) overall user PR*: n=9.
Moderate (50%–74%) overall user PR: n=4.
Low (<50%) overall PR: n=4.
Variable PR between gender and/or ethnic groups: n=9.
Overall low or low-moderate PR in studies of non-white cohorts (n=12).

Association between facial dimensions (FD) and fit**25n=14.
FD association with fit: FW (n=7), FL (n=6), NRB (n=4), JW (n=4), LFL (n=3), NL (n=3), NP (n=3), NW (n=2), LW (n=2), BIOC (n=2), BECTO (n=2), BTMA (n=2).
Facial size and shape categories associated with FF (n=6).
Extremes in FW and FL or narrower faces associated with fit (n=2).
n=7.
Association of some facial dimensions (FW, LW, JW NW, NP, NL, NRB) but low correlation coefficient, poor predictors of fit, explain small proportion of variability, small absolute differences or effect size.
n=4.
No significant correlations.
No relationship between facial size categories and fit.
Association between gender and fit‡24n=12.
PR males>females (n=8).
PR females>males (n=2).
Gender differences in PR varied with model (n=1).
Association of facial dimensions and leak site attributed to gender.
Male gender is independent predictor for fit.
n=2.
PR males>females for at least 1 RPE model but comparable overall with inclusion of multiple models.
n=10.
No significant correlations. Comparable PR between genders.
Fit not predicted by gender.
Association between ethnicity and fit†5n=2.
Race specific models improve fit predictability.
Lower PR for Asians, highest for Caucasians.
n=2.
Lower PR for black females, No significant effect on FF.
Lower PR and FF for Asian females, race did not predict FF.
n=1.
Non-significant ethnicity-based variation in FF.
Association between mask factors and fit§20n=18.
Variability in FF based on brand.
Significant difference in FF/PR between brands (n=12).
Influence of RPE on fit within facial size categories (n=2) or shape (n=1).
RPE is determinant or predictor of fit (n=2).
n=1.
FF associated with number of sizes and models, not RPE design.
n=1.
Comparable PR between models.
  • *Overall user pass rates—percentage of participants successfully fit-testing on at least one RPE model.

  • †PR are reported as either (1) PR of users, as a percentage of participants who passed fit-testing on at least one respirator or (2) PR for RPE groups, as a percentage of participants who passed fit-testing for the respirator being tested.

  • ‡RPE fit as measured by respective studies, including fit/protection factor (FF), simulated workplace protection factor (SWPF), inward leakage (IL), fit-testing PR (PR).

  • §Mask factors are reported as any differences in FF or PR relating to mask factors such design, model, brand, shape or size.

  • BECTO, biectoorbitale breadth; BIOC, biocular breadth; BTMA, bitragion-menton arc; FL, face length; FW, face width; JW, jaw width; LFL, lower face length/menton-subnasale length; LW, lip width; NL, nose length/subnasale-nasion length; NRB, nasal root breadth; NW, nose width; PR, pass rates; RPE, respiratory protective equipment.