Fig 1.
Flow chart of experimental design.
Panel A: Field sampling of ABHS from bulk refillable dispensers. Panel B: Evaluation of rate of ethanol evaporation of dispenser and ABHS under accelerated stability conditions.
Table 1.
Visual and quality characteristics of ABHS samples obtained from bulk refillable dispensers in community settings.
Table 2.
Results of FDA impurity analysis of ABHS samples from bulk refillable dispensers.
Table 3.
Results of alcohol concentration analysis of ABHS samples from bulk refillable dispensers.
Fig 2.
Ethanol concentration in ABHS samples sourced from single school district.
Ethanol concentration of ABHS samples from single school district in South Carolina. Panel A: Distribution of ethanol concentrations (% v/v) measured from 39 individual dispensers. Black dashed line indicates 60% ethanol, the minimum concentration of ethanol required for the ABHS product to be compliant with FDA OTC Drug regulations [23]. Panel B: Average ethanol concentration is associated with status of refill cap placement. A total of 39 ABHS samples from a single school district was analyzed for ethanol concentration. Upon inspection of photos, many dispensers had refill caps removed (n = 19), some had caps installed (n = 7), and some were unable to determine based on photographs (n = 13). When grouped by status of cap placement, a significant difference in average ethanol concentrations (% v/v) was observed. Refill cap status (with or without) was significantly associated with rate of ethanol loss (P < .05) based on one-way ANOVA with cap status as treatment and average ethanol concentration as independent variable. Letters above each bar graph indicate statistically different mean ethanol percentages using Tukey’s HSD post-hoc test for separation of means.
Fig 3.
Observed ethanol loss in ABHS stored in bulk refillable dispensers (pooled across all ABHS types) under accelerated stability conditions (40 ± 2°C, RH of 75 ± 5%).
Rate of monthly ethanol loss of ABHS stored in various bulk refillable dispensers held in accelerated stability conditions. For both panels, ethanol content is pooled across all ABHS types (e.g., foam, gel, liquid/WHO), since ABHS type was not found to significantly impact the rate of ethanol evaporation (data shown in S1 Fig; P > 0.05). Controls were two sanitary sealed 1200mL refill bottles of ABHS products from GOJO Inc. (details in Materials and Methods). Panel A: Observed ethanol loss over 6 months of storage time for ABHS stored in bulk refillable dispensers. Samples were stored at 40 ± 2°C, RH of 75 ± 5%, and aliquots for analysis were measured after 0, 1, 2, 3, and 6 months of storage. Bolded shapes on the graph represent mean ethanol percent (v/v) values for duplicate aliquots at each indicated time point. Dashed lines represent a linear fit trendline based on all data points. Dispenser types appear on the legend, and details of each dispenser type appear in S1 File in the supplemental materials. Panel B: Monthly loss of ethanol represented as the inverse slope of a linear trendline of ethanol content as a function of time when dispenser × ABHS combinations were stored for 6 months at accelerated conditions (from Panel A). Error bars represent standard error of the mean. Dispenser type was significantly associated with rate of ethanol loss (P < .05) based on one-way ANOVA with dispenser type as treatment and average monthly ethanol loss as independent variable. Letters above each bar graph indicate statistically different mean ethanol percentages using Tukey’s HSD post-hoc test for separation of means.