The aim of this study was to evaluate the level of hand hygiene compliance among healthcare workers and visitors in the intensive care unit at the Onderstepoort Veterinary Academic Hospital.
Alcohol hand-rub was the prefered method of hand hygiene in this study. In contrast, Nakamura and colleagues  in a small animal private practice in the United States reported more (85%) handwashing with soap than alcohol-based hand rubs (11.6%) or chlorhexidine/betadine solution (3.8%). Alcohol-based rub is the prefered method as it is faster, more effective, and better tolerated by the skin . In addition, it has a broad spectrum of action and is able to kill most microorganisms in a hospital setting [7, 26–29]. A human study by Eggimann and Pittet  in Switzerland, reported a decrease in the prevalence of HAIs and multidrug-resistant Staphylococcus aureus (MRSA) after the introduction of alcohol-based sanitizers. Notwithstanding the effectiveness of alcohol-based sanitizers, mechanical hand washing with running water and antiseptic soap should also be used in veterinary Medicine as animals are more likely to be soiled with debris. Moreover, mechanical hand washing has been shown to be more effective against spore-forming bacteria and in the presence of organic matter [2, 26].
Most students in this study wore wristwatches and used personal cellphones during the care of patients. This is concerning as the wearing of jewellery, use of cell phones, and other personal equipment while treating a patient has been shown to increase risks of transmission of both HAIs and zoonotic infections [3, 9, 26]. Trick et al  showed a greater frequency of pathogens present on hands with jewellery compared to those without, regardless of the method of hand hygiene applied. Health care workers may also transport pathogens from healthcare facilities and introduce infections into their homes .
The majority of HCWs in this study did not perform hand hygiene before putting on or after removing gloves. The use of gloves is recommended in infection control but should not be a substitute for hand hygiene , which should therefore be performed prior to donning and after glove removal [9, 23, 26]. Hands could additionally be contaminated in the process of removing gloves, gloves could tear, or they may have microscopic defects resulting in contamination.
The overall hand hygiene compliance in the OVAH ICU was similar to the 20.6% reported by Shea and Shaw  in a small animal veterinary hospital in the United States of America. Similarly, low (30%- 40%) hand hygiene compliance in the human ICU have been reported in the Netherlands . In contrast, in a human study in the United Kingdom (UK), Randle et al  reported a higher (67.8%) hand hygiene compliance in the ICU. The overall low level of hand hygiene compliance in the current study could be attributed to time constraints between patients , a lack of perceived importance of hand hygiene , and low levels of hand hygiene compliance in the five moments observed in this study. The low compliance level in the current study is concerning as hand hygiene compliance has been shown to decrease cases of hospital-acquired infections and minimise exposure to zoonotic disease [13, 15]. Therefore, it is imperative that ongoing educational programs and surveillance be implemented for sustained changes in behaviour [3, 5, 12, 15, 17] and improvement in hand hygiene compliance [1, 15, 31].
Compliance based on the type of HCW and time of day
We observed lower compliance among doctors when compared to nurses and students with nurses having the highest level of compliance. Similarly, Salama and colleagues  in a human hospital in Kuwait reported lower compliance in doctors (38.4%) compared to nurses (50.0%). Our results and results of others suggest that nurses are more adherent to and knowledgeable about hand hygiene compared to other HCWs [5, 12, 31]. Furthermore, hand hygiene compliance was higher in the morning compared with that during other shifts. This could be because nurses worked mostly in the morning shift and had higher level of compliance compared to other healthcare workers. In addition, the higher number of HCWs present in the morning shift compared to other shifts could have reduced the ratio of patient to HCW allowing more time and better implementation of hand hygiene. In view of these findings we recommend the implementation of intervention strategies including education, feedback sessions, and monitoring of non-compliant doctors and students in order to improve compliance .
Type of moment
We observed low hand hygiene compliance after contact with the environmental formites such as medicine, drugs cabinets, drips, cages, cots, doors, telephones, bedding and leashes. In contrast, Randle Arthur, and Vaughan  in a human hospital in the United Kingdom reported a high (50%) compliance after contact with patient surroundings. The patient’s surroundings have been shown to harbour various microorganisms including those that cause hospital-acquired infections . Anastasiades et al  in South Africa isolated Staphylococcus aureus from the computer mouse and keyboards used in the human ICU. Similarly, and hygiene compliance in the current study was low after contact with surrounding and environmental fomites. In addition, studies have reported low hand hygiene compliance during activities that pose a low risk to HCWs when compared to patients, suggesting that HCWs perform hand hygiene for their own protection and not for that of patients’ [3, 5, 6]. This hypothesis is substantiated by the low level of compliance before an aseptic procedure (15.4%), before patient contact (18.8%) compared to high compliance after patient contact (32.2%), and after body fluid exposure (41.7%). This is most likely due to the hands of HCWs often being visibly soiled, sticky or gritty after these moments . Therefore, it is recommended that hand hygiene be performed anytime when in contact with patients and patients surrounding [2, 3, 8]. Moreover, interventions must be moment specific due to the variation in compliance based on the type of moment.
Based on the five moments of hand hygiene, nurses had the lowest compliance after body fluid contact, students had the lowest compliance before contact with a patient, and doctors had the lowest compliance after contact with patient surroundings and before patient contact. Compliance during the five moments in our study was significantly different between the three groups of HCWs. These findings suggest that the type of moment, as well as the type of HCW should be considered when developing hand hygiene intervention strategies.
To our knowledge, this is the first study in veterinary medicine in South Africa that used the ICAT tool to assess hand hygiene compliance. Therefore, the comparison to other studies is largely based on human studies which uses a similar system. The presence of the assessor in the ICU could have resulted in higher than normal levels of hand hygiene compliance. However, the assessor was discreet about the assessment questions and never communicated with the HCWs about the content of the assessment. The HCWs assessed in this study were not inclusive of other ICU workers such as kennel cleaners and cleaning staff members which are also capable of transmitting HAI pathogens. Nonetheless, the results in this study provide a useful indication of hand hygiene compliance level in the ICU at OVAH.