Hand disinfection at the right indications is generally considered to be one of the best way to prevent nosocomial infections. Despite decades of research and interventions, hand hygiene compliance is generally not at satisfactory levels worldwide. The best tool for understanding what influences healthcare behavior and the gold standard for investigating hand hygiene compliance is direct observation according to the WHO Five Moments (1, 2). However, these observations are not perfect, they only include fully observable healthcare actions, leaving an unknown degree of uncertainty about partially observable actions. In addition, Five Moments observations focus solely on hand disinfection, leaving out neighboring aspects of hand hygiene such as glove use. This limitation has been partially addressed, for example by the German “Aktion Saubere Hände” in 2019 by including the observation “gloves instead of hand disinfection” in their observation sheet template (12).
As healthcare behaviors changed drastically during the SARS-CoV-2 pandemic, we set out to develop a system that completely incorporates glove use into the Five Moments of Hand Hygiene, while also accounting for observation uncertainty. To this end, observations were made according to our extended Five Moments model, including the use of single-use gloves and a measure of uncertainty in three different ward types (ICU, IMC, and normal ward). Gloves used before patient contact or before an aseptic procedure could be new (i.e. clean), contaminated, disinfected, or unobservable (i.e. the location of the observer did not allow to see possible hand contact). The use of a new or disinfected glove after a hand disinfection did not impact the hand disinfection compliance rate for both before indications. In contrast, the use of a contaminated glove after a hand disinfection invalidated the hand disinfection and was counted as non-compliant.
Observing complex procedures involving several healthcare workers in modern wards with closed patient rooms can be very challenging for hygiene staff. The introduction of question marks in the observation sheet facilitates observation by allowing inconclusive partial observations to be accurately recorded. Thus, there is no pressure to decide for or against compliant behaviour if the hands could not be observed for the entire period, so observations can be expected to be more accurate. In addition, observation intervals with many question marks may indicate very difficult room configurations, hectic procedures, uncooperative staff or new but conscientious hygiene staff (Table 1). Without the option of a question mark, these observations would have been omitted entirely, potentially leading to a bias towards easily observable situations. Or, worse, classified as compliant or non-compliant based on intuition rather than objective observation. Here, additional fields for staff type and gender were relevant for targeted feedback and training but are also useful for keeping track of multiple people at the same time without compromising anonymity.
Direct observation data are often used as a measure of quality or to decide whether interventions are needed, and therefore need to be as reliable as possible. Relying on day-to-day experience is problematic because self-perception often differs drastically from reality. Operating room and endoscopy staff, for example, reported perceived HH rates of 73% and 95% respectively, but observations revealed compliance to be 11% (13). Our results show that especially the two before moments cannot be observed completely, leading to an uncertainty of 16% and 8%, respectively (Table 2, Fig. 2). This leaves a significant data gap for these moments, which are most critical for patient safety.
When used at the right time, gloves can contribute to patient safety, but in general glove use is not superior to hand hygiene. Frequent glove use has been associated with poor hand hygiene and inappropriate glove use can increase the risk of pathogen transmission via contaminated gloves (14–16). During the SARS-CoV-2 pandemic, our infection control nurses suspected that the extensive use of personal protective equipment (PPE) may have led to decreased hand hygiene compliance. While healthcare workers might see gloves as a convenient replacement for hand disinfection, the WHO guidelines state “in no way does glove use modify hand hygiene indications or replace hand hygiene” (1). The use of gloves actually complicates hand disinfection at the right indications (17). Sandbekken et al. observed a reduction of hand hygiene compliance of about 30% if gloves were worn in nursing home wards in Norway (18). Fuller et al. identified the use of gloves as one major risk factor for reduced hand hygiene compliance (14). Our results support this evaluation.
While glove use and compliance might differ for the individual moments (Table 3, Fig. 2) we have identified glove use as a general risk for hand disinfection compliance (Table 4). Table 3 shows that for aseptic procedures, the percentage of hand disinfection compliance while wearing gloves (21% of observations) is higher than the percentage of compliance without gloves (11%), which may lead to the erroneous conclusion that glove use increases HD compliance. But the corresponding odds ratio test (Table 4, line 2) shows that the opposite is true, due to the fact that observations without gloves are less frequent than those with gloves, and only a third of these observations were non-compliant.
Other authors found glove use to be associated with increased hand hygiene adherence before room entry and after room exit in a long-term care facility (19).
Some recent studies have also included the use of gloves in their compliance observations and reported similar results to ours. For example, Siebers et al. investigated compliance in an ICU and reported the highest compliance for after touching a patient (~ 50%) and the lowest compliance for before aseptic procedure (~ 20%), while our observations were 86% and 40%, respectively (4). They also reported that if glove use could replace hand disinfection, the overall compliance would increase from 33.9–72.1%, a finding that is similar to that of Haac et al. (4, 20).
Contaminated hands and gloves pose a threat to patient safety. Based on literature reports, glove contamination rates appear to depend on the overall frequency of hand hygiene, with gloves in combination with regular hand hygiene not leading to an increase in bacterial counts (21, 22). Yet, the longer healthcare workers go without hand hygiene, the higher the bacterial load on hands and gloves (21, 22)Because glove use is often associated with failure to change gloves during the treatment of the same patient, the risk of device-associated infections is increased (15, 16). As contaminated gloves are not covered by traditional direct observation, this risk is currently difficult to assess. Notably, in our observations, glove use after hand disinfection before patient contact was mostly correct, whereas about 20% of all before aseptic procedure observations had gloves contaminated between donning and indication (Table 2). This is the most important indication for the prevention of device-associated infections and requires a high level of compliance (3). Maybe, the time of glove donning was too early in the process, which could be addressed in training.
In the future combining these extended observations with investigations on infection rates will allow to more precisely estimate the risk caused by contaminated gloves, which could be mitigated by glove disinfection. Disinfecting gloves is more effective than disinfecting hands but it can decrease the protective capacity of the glove for the healthcare worker (23, 24).
Healthcare workers wear gloves primarily for self-protection and because they are perceived to save time compared to hand disinfection (6, 7). Although gloves are convenient in some respects, they are not without risk to the wearer. In general, wearing gloves increases the risk of adverse skin reactions, particularly from additives such as carbamates and thiurams (25). Dryness and rashes or itching are the main complaints about wearing gloves (25–27). While glove disinfection is great for reducing the risk of contamination for the patient, it can put the healthcare worker at risk. The majority of available disinfectants weaken gloves, causing them to tear more easily (23). As leaky gloves do not protect against contact with infectious materials, only gloves that have been shown to be disinfectant-compatible should be used. Another risk of frequent glove use is self-contamination due to poor doffing technique. In clinical practice, the correct method of doffing gloves is rarely used (28). In one study, for example, only 34% of healthcare workers used the CDC recommended method of doffing gloves (29). This risk is further increased if no hand disinfection is carried out after doffing the gloves.