Abstract
Introduction
The World Health Organization (WHO, 2011) identifies the transfer of healthcare associated infections (HAIs) by healthcare providers (HCPs) to be a major patient safety concern. This is because HAIs are often associated with prolonged length of hospital stay, long-term disability, microbial resistance, and increased cost and mortality (WHO, 2011). It has been estimated that more than 200,000 Canadians acquire HAIs each year, and that more than 8000 of these patients die as a result (Public Health Agency of Canada, 2013a; Zoutman et al., 2003). The annual healthcare costs for HAIs are significant. For example, Clostridium difficile infections cost C$46.1m and methicillin resistant Staphylococcus aureus infections C$36.3m per year. The risk of dying following an episode of C. difficile infection is estimated at 6% (Public Health Agency of Canada, 2013b).
Although many factors contribute to the development of HAIs, the consistent performance of hand hygiene (HH) prior to physical contact with patients has been reported to be the single most effective preventive strategy of these infections (Boyce and Pittet, 2002; Pittet et al., 2006). It has been reported that compliance with HH can be associated with at least a 20% reduction in the risk of developing HAIs (Harbarth et al., 2003). Despite the overwhelming evidence concerning the importance of HH and the negative consequences of HAIs, disappointing HH compliance rates continue to prevail. For instance, HH compliance among nurses has been reported to range between 33% and 53% (Erasmus et al., 2010; Mertz et al., 2011).
A plethora of research has explored the factors associated with HH compliance among HCPs. However, few studies have explored the rates and factors associated with HH compliance among undergraduate nursing students. For instance, Barrett and Randle (2008) explored nursing students’ self-perceived HH barriers, and reported that lack of compliance with HH was influenced by a perception of a lack of time, busyness, poor role modelling, the type of clinical procedure being completed, concerns over poor skin conditions, lack of knowledge, and the use of gloves. Furthermore, Cole (2009) reported that nursing students have a tendency to overestimate their HH knowledge and compliance due to their inability to give an objective account of their HH practices. This is concerning, given that lack of awareness and objectivity about one’s HH practices while in school may translate to poor HH compliance during post-graduation practice. The limited number of HH research studies among nursing students is also concerning, given their active involvement in direct patient care during their clinical practice experiences. Understanding nursing students’ HH practices and ensuring their compliance with HH guidelines prior to their entry into professional practice is paramount in light of the reported association between HH and HAIs (Van de Mortel et al., 2012). Therefore, the purposes of this study were to explore the self-perceived rates of HH compliance, and identify the predictors (i.e. facilitators and barriers) of HH compliance among undergraduate nursing students.
Methodology
Design
A descriptive, cross-sectional self-report survey was conducted on a convenience sample of 306 consenting participants to explore the self-perceived HH compliance among undergraduate nursing students. The sample was recruited in 2013 from a pool of 578 students enrolled in a Bachelor of Science in Nursing (BScN) program at a southwestern Ontario university, yielding a response rate of 52.9%. Students were eligible to participate in the study if they were registered in years 2, 3, or 4 of the program. Students enrolled in year 1 were excluded due to their limited clinical experience. On receipt of research ethics approval from the research ethics board of the home institution, the survey and a written description of the study were distributed to students during scheduled classes. Participation in the study was voluntary, and anonymity of responses was protected.
Questions for the study survey were specifically developed for the purpose of this research from existing HH literature, to elicit data pertaining to the participants’ demographic characteristics, self-perceived HH practices, perceptions of motivators and barriers to HH compliance. Other questions in the survey were informed by the theory of planned behaviour (TPB) (Ajzen, 1985, 1991). The TPB has been successfully used as a theoretical model for the identification of HCPs’ intentions to comply with HH guidelines (Pessoa-Silva et al., 2005; Whitby et al., 2006). The TPB explains how cognitive variables (attitude, subjective norms, perceived behaviour, behavioural control and intention) can predict behaviour, and offers theoretical insights for the study of HH behaviour (Ajzen, 1985).
The constructs of the TPB (attitude, subjective norms, perceived behaviour, behavioural control and intention) informed the formulation of several of the questions. That is, some questions elicited information that was attitude related, others were subjective norm related, perceived behaviour, behavioural control and intention related. Given the descriptive nature of the questions in our survey, each question was treated as an independent unit of analysis; hence it was not subjected to validity and reliability as this testing was beyond the scope of our study. Therefore, all items were treated as independent variables. Prior to administering the survey, it was piloted with 10 nursing students who provided feedback about clarity and understandability.
Self-perceived HH compliance was measured by questions that asked participants to indicate the percentage of time they performed HH before, after and both before and after having had direct patient contact during their clinical placement experiences. Proper HH compliance was defined by a participant indicating on the survey that they performed HH both before and after patient contact at least 90% of the time. Although the use of 90% as a cut-off criterion for compliance is higher than the 80% cut-off point used in existing literature (Budimir-Hussey et al., 2013; Sax et al., 2007), it is more lenient than a strict 100% compliance level, which has been suggested to be unrealistic (Jang et al., 2010; Voss and Widmer,1997).
Data analysis
Data analysis was performed using the IBM statistical package for the social sciences (SPSS 21.0). Basic descriptive statistics were used to describe the demographic characteristics of the sample and the self-reported frequency of HH before and after patient contact. Preliminary bivariate chi-square and t-test comparisons of the independent variables were performed to compare participants who identified themselves as HH compliant versus participants who identified themselves as non-compliant. Given the exploratory nature of this study, a forward stepwise binary logistic regression approach was performed to determine the independent predictors of self-perceived HH compliance. All variables having a P value of 0.25 or less in the bivariate analysis were included in the logistic regression iteration process to avoid deletion of potentially significant predictors from the final multivariate regression model (Hosmer and Lemeshow, 2000). A 95% confidence interval (CI) was used as the criterion to determine statistical significance.
Results
Female participants comprised 83% (n = 253) of the sample. The mean age of participants was 23.2 years (SD ± 4.47; range 19–48). Participants’ mean number of clinical placements was 5.73 (SD ± 2.06; range 2–11). Overall, 74.8% (n = 229) of participants were determined to have met the criteria of being compliant, indicating that their compliance was 90% or greater both before and after having had direct patient contact.
Chi-square comparisons of self-perceived HH compliance (≥90%) and non-compliance (≤89%) and beliefs.
Indicates P ≤ 0.25 and inclusion in multivariate analysis.
HH: hand hygiene; HCP: healthcare provider; HAI: healthcare associated infection.
Stepwise logistic regression depicting the independent predictors of HH compliance.
B: unstandardised coefficient; SE: standard error; OR: odds ratio; CI: confidence interval; P value: probability of accepting the null hypothesis at an alpha of 0.05; HH: hand hygiene.
The Cox and Snell R2 and Nagelkerke R2 indicate that the seven independent predictors in the model explain a modest 20.5–30.3% of the variance in HH compliance in this sample.
Discussion
The majority of participants (74.8%) in this study indicated their HH compliance was 90% or greater in the moments both before and after having had direct physical patient contact. The levels of self-reported HH compliance in this study were surprisingly high when compared to the existing literature of observations of nurses’ HH compliance levels, which range between 33% and 53% (Erasmus et al., 2010; Mertz et al., 2011). The finding of high levels of HH compliance in this study were very similar to those reported by Cole (2009), who reported that nursing students self-perceive surprisingly high levels of HH compliance. While it is possible that nursing students may have higher HH compliance than nurses, it may also suggest that participants had difficulty making objective self-assessments about their HH practices, or have poor insight into their actual HH behaviour. It is therefore important that both students and nurse educators be knowledgeable about the often present discrepancy between self-reported HH compliance rates and actual observed HH compliance rates.
The regression results suggested that seven variables were significant predictors of participants’ HH compliance: the decision to perform HH as a result of concerns about reprimand/discipline if HH guidelines are not followed, and the motivation to perform HH in order to protect the patient; participants’ numbers of clinical placements and the perception that the clinical nursing instructor consistently performed HH were also significant predictors of HH compliance; furthermore, the variables busyness, forgetfulness and the perception that the use of alcohol hand rubs damages the skin were all independent predictors of HH compliance in this study.
Participants who were motivated to perform HH by concerns about being reprimanded or disciplined if they did not follow HH guidelines were 4.3 times more likely to comply with HH guidelines than participants who did not share these concerns. This suggests motivation to perform HH was driven by the perception that HH behaviour was being observed and assessed by referent individuals (e.g. clinical nursing instructor, unit staff members) during their clinical placements, and that if they failed to comply with HH guidelines they would be disciplined. Unfortunately, how participants defined the experience of discipline or reprimand if they failed to perform HH during their clinical experiences was not adequately explored in this study. Therefore, future research in this population is necessary to explore the students’ perceptions of what constitutes reprimand/discipline and how this impacts HH practices.
In this study, participants who were motivated to perform HH by the belief that HH protects the patient from infection were 2.4 times more likely to comply with HH guidelines than those who did not hold this belief. When reported HH compliance rates before and after physical contact were compared, 80% of participants perceived they were HH compliant before having physical contact with their patients, and 95% of participants indicated that they were HH compliant after having contact with their patients. The finding of greater HH compliance after patient contact seems to contradict participants’ reported motivation for HH compliance being protection of the patient from infection. This suggests that participants’ motivations to perform HH were similar to findings for postgraduate HCPs – with the motivation to perform HH being based on an individual’s evaluation of risk to personal safety, rather than on patient protection (Erasmus et al., 2010; Jang et al., 2010; Korniewicz and El-Masri, 2010). Interestingly, the variable protection of self from infection was not statistically significant in the final logistic regression model, suggesting that the observed frequencies were a mere function of chance.
A concerning finding suggested decreased HH compliance correlating with greater numbers of clinical experiences. The unadjusted and adjusted results suggested there was a significant difference in HH compliance across different levels (years) of study (P = 0.012). The percentage of compliant participants in the second (79.7%) and third (78.7%) years of the program was greater than that of the fourth year (61.6%) of study. The regression findings suggested that participants’ numbers of clinical placements, which increase as the students advance in level, could be used as an independent predictor of HH compliance. Specifically, participants who had a greater number of clinical experiences were 18.5% less likely to be compliant with HH guidelines (OR 0.815; 95% CI 0.702–0.947). One possible explanation for this finding could be that greater HH compliance was the result of having recently received HH education, but participants’ HH compliance decreased with the passage of time and lack of reinforcement of HH education. Including HH compliance learning activities in each level of the nursing program could make students more aware of their HH behaviour.
Interestingly, if participants perceived that their clinical nursing instructor consistently performed HH when necessary, they were 2.2 times more likely to be compliant with HH guidelines than those who did not hold this perception. Similarly, Snow et al. (2006) reported that mentors’ HH practices were the strongest predictor of students’ HH practices. It has been reported that nursing student HH compliance has been both positively and negatively influenced by other HCPs’ HH behaviour (Barrett and Randle, 2008; Gould and Drey, 2013; Lusardi, 2007). This finding calls attention to the fact of the strong influence that role modelling proper HH behaviour has on students’ HH compliance, and suggests that nursing instructors’ HH practices have a significant impact on nursing students’ HH compliance.
Participants in this study were 77% less likely to comply with HH guidelines if they perceived themselves to be busy when performing patient care than those who did not hold this belief. Similar findings of nursing students perceiving HH compliance as poor when they were busy have been reported (Barrett and Randle, 2008; Lusardi, 2007). Education provided to nursing students needs to reinforce that when busy there is an increased risk of non-compliance, and also to encourage students to identify strategies that increase personal and team compliance (e.g. reminding busy colleagues if they forget to perform HH). It is also important to emphasise to nursing students and nurses in general that failure to comply with HH because of busyness contradicts HH guidelines and can result in the transmission of HAIs.
Participants were 64% less likely to comply with HH guidelines if they held the perception that forgetfulness was a barrier to their HH compliance than those who did not hold this perception. Similar findings among HCPs have been reported (Budimir-Hussey et al., 2013; Pittet, 2000). Forgetfulness cannot be accepted as an excuse for lack of compliance with HH, given the serious patient safety consequences of poor compliance.
Participants were 84% less likely to comply with HH guidelines if they perceived that using alcohol hand rub for HH was damaging to their skin than those who did not hold this belief. The mistaken perception that the use of alcohol hand rubs causes damaged skin conditions such as dryness and irritation on hands has frequently been reported in the literature as a barrier to compliance. However, alcohol-based hand rubs are actually among the best tolerated HH agents, due to the addition of emollients (Chamorey et al., 2011; Kampf and Loffler, 2007), and therefore should not be a concern with regard to potentially causing skin damage. It is important to ensure that both nursing faculty and nursing students understand the relationship between HH agents and skin damage.
Concepts from the TPB were used to inform the questions developed for this study. The TPB proposes that attitudes, subjective norms and perceived behavioural control are predictive factors of behavioural intention, and intention is considered to be the immediate antecedent of the behaviour (Ajzen, 1991). This proposition is supported by the findings from this study. Specifically, participants were motivated to perform HH because of the attitude or perception that HH protects patients from infection. An individual’s attitude towards the behaviour is the degree to which performance of the behaviour is positively or negatively valued. The TPB further proposes that the subjective norm is the perception of social pressure to engage or not engage in behaviour and that an individual’s perception about the behaviour is influenced by the judgement of significant others (Ajzen, 1991). In this study, a participant’s perception of being observed impacted HH compliance based on an assessment of risk of reprimand/discipline. Additionally, the perception that a clinical nursing instructor consistently performed HH added support for the notion that social pressures influence HH behaviour. The TPB defines perceived behavioural control as one’s perception of the ability to perform behaviour (Ajzen, 1991). In this study, the perception of busyness, of forgetfulness and that use of alcohol hand rub was damaging to skin can be considered factors that could impact a participant’s perception of having control over their performance of HH.
Despite the fact that our survey questions were theoretically driven by literature review and the TPB, the independent predictors in this study explained only 20.5–30.3% of the total variance of undergraduate nursing student HH compliance. This finding highlights the need for future research given that a sizeable percentage of the variance in the outcome remains unexplained. The high sensitivity result of this model (93.9%) suggests it was a good model for classifying those who comply with HH guidelines. However, the model’s lower specificity result (32.5%) suggests that it is less accurate in classifying non-compliants. The overall precision of the model’s ability to classify a student correctly as compliant or non-compliant was a modest 78.4%, which indicates that future work is still needed to understand better the predictors of HH compliance among nursing students.
This study was not without limitations. Given the subjective self-report nature of this study, there was a possibility of a recall, social desirability bias and/or a self-selection bias whereby participants may have reported higher levels of HH compliance than reality despite the anonymous format of the questionnaire. Participants may have consciously or unconsciously taken into consideration the fact that the author of this study is also a nursing instructor at the University of Windsor, and may have indicated what they believed to be the most acceptable answer on the questionnaire. To minimise the risk of bias, the author was not present in the classroom while the questionnaire was administered or completed.
Another important limitation of this study was that it did not measure participants’ actual observed rates of HH compliance during clinical experiences. Also, not adequately explored was how participants conceptually defined busyness, forgetfulness and the experience of discipline or reprimand on the occasions of failing to carry out HH.
Conclusion
In conclusion, this study provided insights and understanding about nursing students’ self-perceptions of HH compliance, motivations to perform HH and the factors that impact HH practices. Students in this study had surprisingly high rates of self-reported HH compliance when considered with observations of postgraduate nurses’ HH compliance, and may have had difficulty with objective self-assessment, as had been suggested by Cole (2009). Considered with findings of greater HH compliance after patient contact when compared with compliance rates prior to patient contact, the findings suggest a need for specific education aimed at increasing awareness of the tendency to overestimate HH compliance in order to prevent the cross-transmission of pathogens. Increasing student awareness of the importance of proper HH compliance has been shown to impact students’ HH practices positively (Magaldi and Molloy, 2010; Salmon et al., 2013).
Focused education is necessary to address the specific barriers to HH compliance and the misconceptions surrounding the use of alcohol hand rub and skin damage. The decrease in HH compliance in relation to more clinical experience was an alarming finding, with the lowest HH compliance rate being in the final year of the nursing program. Declining compliance with more clinical experience may result from the formation of poor HH habits role modelled by other HCPs during clinical experiences, or the lack of reinforcement of HH concepts in successive years of the nursing program.
Clinical nursing instructors’ role modelling of proper HH practices was shown to have a positive influence on students’ HH practices. Clinical nursing instructors need to be aware of their influence and understand the importance of role modelling excellent HH compliance during student clinical experiences. Further research is necessary to explore study findings, which suggested that students in this study were motivated to perform HH by concerns about being reprimanded or disciplined if they did not follow HH guidelines.
The few studies that examined the impact of gender on HH have reported conflicting results (Korniewicz and El-Masri, 2010; Van de Mortel et al., 2001). As more male students enter nursing programs, the impact that gender might have on HH compliance merits greater consideration for future research studies. What also remains unknown is the optimal level of HH compliance necessary to effect change in the rate of HAI validation of an HH compliance questionnaire specifically for nursing students.
Key points for practice and/or research
Focused education for nursing students is necessary to increase awareness of the potential to overestimate HH compliance, the possibility of less HH compliance with greater clinical experience, and the need to perform HH prior to patient contact. Specific barriers to HH compliance need to be addressed: busyness, forgetfulness, and misconceptions about the use of alcohol hand rub and skin damage. Clinical nursing instructors need to be aware of their influence and the importance of role modelling excellent HH compliance during student clinical experiences. Opportunities for future research include further exploration of the experience of the perception of discipline or reprimand in the occasion of failing to perform HH; the impact of gender on HH compliance; and the development and validation of an HH compliance questionnaire specifically for nursing students.
Footnotes
Declaration of conflicting interest
None declared.
Funding
This study was supported by an award from the Registered Nurses’ Foundation of Ontario for the advancement of professional practice in infection control.