Discussion

The bacterial burden of clothing depends on a variety of behavioral and organizational factors including personal hygiene, laundering habits, and exposure to environmental contaminants. To limit the variability of laundering, storage, and transport of staff uniforms, many hospitals issue uniforms as an infection prevention strategy. Few studies have compared the bioburden of common hospital attire of healthcare workers to date.Reference Mitchell, Spencer and Edmiston ¹⁶ In this study, we aimed to describe differences in bacterial abundance across military-approved uniforms and hospital-provided scrubs in an inpatient population of military nursing staff. We report a significantly higher bacterial burden on military-approved versus hospital-provided scrubs across all time points.

Scrubs in our study demonstrated a time-dependent accumulation of bacteria (except on back), implying acquisition from the hospital environment (eg, from patients or a contaminated hand). Our findings support smaller studies that have reported that sterile uniforms can accrue large numbers of bacteria over the course of a single shift.Reference Sanon and Watkins ¹⁷ However, military-approved uniforms did not demonstrate this pattern. Possible reasons for this may include the significantly different rates of contamination upon arrival or the inherent differences in the microbial colonization of different fabric compositions (a phenomenon which has been previously reported).Reference Takashima, Shirai, Sageshima, Ikeda, Okamoto and Dohi ¹⁸ Additionally, repeat sampling of uniform areas throughout the day may have diluted bacterial accumulation at 4 and 8 hours of wear, which may have removed residing bacteria that would otherwise be present in subsequent samples. Other studies have circumvented repeat sampling by rotating among a pool of participants throughout different days and time points.Reference Burden, Cervantes, Weed, Keniston, Price and Albert ¹⁹ On the other hand, sampling subjects throughout the day allowed us to capture a more robust sample, simultaneously controlling for behavioral and textile variables that can occur in a rotating paradigm. Finally, due to the demonstrated rate of textile reinoculation that can occur in just 3 hours,Reference Burden, Cervantes, Weed, Keniston, Price and Albert ¹⁹ along with the apparent resistance of hospital scrubs to repeat sampling error, it is likely that a more complex phenomenon was at work here. Future studies focused explicitly on elucidating inherent microbial properties of different textiles should certainly control for real-world variables such as laundering, storage, transport, undershirts, etc.

The waist, sleeve cuff, and abdomen regions of both uniform types demonstrated the highest amount of bacterial accumulation. The prevalence of bacteria in these regions is likely due to either direct contact with contaminated surfaces (including contaminated gloves) or self-inoculation from contact with inadequately cleansed hands.Reference Munoz-Price, Arheart and Mills ²⁰ Indeed, environmental contamination of the immediate patient environment has been identified as an independent risk factor for contamination of healthcare workers’ gloves and gowns.Reference Morgan, Rogawski and Thom ²¹ These findings highlight the need for frequent hand hygiene during episodes of care and improved hospital surface decontamination, which is often suboptimal and a contributor to nosocomial infections.Reference Peretz, Koiefman, Dinisman, Brodsky and Labay ^{22 –} Reference Gardner, Muller and Prior ²⁴

The number of bacteria recovered from wrists showed no differences across uniform type nor meaningful alteration over time, though we did find that the left wrist harbors higher levels of bacteria in general. We propose that this may be attributed to the tendency of participants to wash their dominant hand more thoroughly (majority were right handed) or the prevalence of watches, bracelets, and other accessories worn on the left wrist. These findings support the need for “bare below the elbows” policies in use at other facilities.Reference Bearman, Bryant and Leekha ²⁵ Moreover, these findings suggest that bacterial contamination of hospital textiles bear unique properties compared to dermal surfaces.

Our findings, combined with studies demonstrating that sterilized uniforms can become reinoculated by 50% (of the 8 hours total bacteria) in as little as 3 hours,Reference Burden, Cervantes, Weed, Keniston, Price and Albert ²⁶ highlight the need for the daily washing of uniforms. This is particularly important given that 30%–42% of the hospital staff do not wear a fresh uniform daily.Reference Wiener-Well, Galuty, Rudensky, Schlesinger, Attias and Yinnon ^{9 ,} Reference Callaghan ²⁷ While more focused studies are needed to understand the inherent microbial attraction and retention of various textiles in a more controlled paradigm, this study provided the first, real-world snapshot of the bacterial abundance of 2 distinct uniform types in a military hospital.

Future studies should delve deeper into the underlying factors that influence bacterial accumulation (hand hygiene, laundering, patient acuity, etc) because they may identify other avenues for infection prevention. In this study, we observed a pattern of elevated bacterial colonization on military uniforms previously worn (ie, not recently laundered), though this trend was not statistically significant (Table 3). In addition to the limitations of self-reported data, lack of direct observation impeded our ability to track potential environmental exposures that may have occurred prior to the participant’s arrival to work. Another important limitation of this study was that vertical transfer of pathogenic bacteria was not specifically assessed. While no study to date has specifically demonstrated staff to patient transmission through the uniform, a number of hospital textiles have been cited as possible sources of patient infections.Reference Fijan and Å ostar Turk ²⁸ Moreover, it has been proposed that the impact of healthcare textiles may be underestimated due to lack of point-source investigations of textiles during outbreaks or infection.Reference Mitchell, Spencer and Edmiston ¹⁶ On the other hand, the bactericidal effectiveness of industrial and residential laundering has been previously demonstrated,Reference Lakdawala, Pham, Shah and Holton ^{29 –} Reference Riley, Williams, Owen, Shen, Davies and Laird ³¹ though it is not a fool-proof solution.Reference Mitchell, Spencer and Edmiston ¹⁶

Table 3

Colony-Forming Units (CFU) on Arrival Based on Days Worn Since Last Wash

Note. IQR, interquartile range.

In conclusion, we found that bacteria are present at much higher levels on military-approved uniforms than on hospital-provided scrubs at our institution. While additional work is needed to demonstrate a direct association between staff uniforms and nosocomial infections,Reference Weber, Khan, Fader and Weber ¹⁵ staff uniforms may contribute to the transmission of pathogens between patients.Reference Mitchell, Spencer and Edmiston ¹⁶ Hospital uniform policies could be expanded to require on-site outfit changes to minimize environmental contamination or, at minimum, daily laundering. The accumulation of bacteria observed on hospital-provided scrubs additionally warrants a closer look at the microbial environment within the hospital. In the future, integrated infection prevention approaches should consider behavioral and engineering factors such as hand hygiene, uniform policy, and environmental sanitation. Policy makers should additionally bear in mind that the high financial burden of antimicrobial resistance may be offset by critical changes to uniform policies, such as the expansion of hospital-provided scrubs to all personnel who provide direct patient care. We hope that the findings from this study can be used to educate staff about the role that hospital attire plays in the risk of cross contamination to improve adherence to hospital uniform washing guidelines recommended by the Centers for Disease Control and Prevention.