IT is well known that hand washing is one of the most important infection control procedures in health care. However, the rate of hand hygiene in hospital settings by health care staff and non-health care workers remains unsatisfactory. Previous research has identified low rates of hand washing in hospital settings,1 but studies have failed to find methods of successfully improving these rates. Most efforts have focused on staff education to improve compliance with hand hygiene. Monitoring and feedback can also be used in the attempt to increase hand hygiene, but there is scarce research pertaining to this method. A single hospital, multiprofessional test of this method was performed.
LITERATURE REVIEW
There is a large volume of literature that acknowledges the importance of hand washing by hospital staff.2-4 Hand hygiene has been linked to the reduction in the transmission of pathogens5 and a decrease in the rates of infection.6 However, compliance to hand hygiene protocols by hospital staff has been traditionally low.1 Research examining the topic of hand hygiene practices has yielded little on how to effectively increase and sustain hand hygiene compliance.7-11
A vast number of interventions designed to increase hand hygiene compliance have focused on education. These interventions have met with some moderate short-term success, but positive long-term effects are seldom seen.3 The utilization of multimodal approaches that include more than only education is viewed as a more effective means to increase hand hygiene compliance.5
Approaches that use more than education to increase hand hygiene compliance have demonstrated some promising outcomes. The use of monitoring and feedback has been noted as a means to increase compliance in hand hygiene.7 Benton2 evaluated a project designed to improve physician compliance with hand washing. In the project, nurses monitored physicians and provided compliance reports to each physician. Results demonstrated a significant increase in the physician compliance.
Naikoba and Hayward6 reviewed 9 studies evaluating the effects of using performance feedback on the hand washing compliance. The studies suggested that feedback can improve the frequency of hand washing. The use of feedback is seen as a key to the success of any hand hygiene improvement program.1 However, most studies used nonpersonal feedback, with statistics and educational material as the primary sources of feedback to staff.12
The goal of this research was to increase compliance through both nonpersonal and personal feedback to staff. For the purposes of this study, personal feedback was defined as any feedback stated directly and verbally from one person to another about the individual's behavior. Feedback that did not use the previous method was considered nonpersonal.
METHODS
The intervention tested in this article was based on an audit tool developed at the University of North Carolina Hospitals Burn Center to address the presence of a multidrug resistant organism in the hospital. The audit tool is designed to allow all staff to observe the hand hygiene practices of other health care professionals and when hand hygiene is not performed appropriately, to provide feedback to that staff member. Initial results revealed the success of this audit tool in the Burn Center. However, a more systematic research methodology was designed to determine the ability of this audit procedure to increase the hand hygiene compliance in other settings.
In an attempt to evaluate the effectiveness of this intervention, a simple pre- and postintervention methodology was used in the surgery intensive care, neurosurgery intensive care, and surgical intermediate care units at University of North Carolina Hospitals in Chapel Hill, North Carolina. This methodology included a 3-step process. The first step (preintervention) involved the use of objective observers to collect baseline data on the frequency of hand hygiene on these hospital units. Preintervention observations were completed 1-week before the audit tool was introduced to staff to provide baseline data. The second step was the actual intervention. During this step, audit tools were given to staff to implement as described later. The third step (postintervention) used the same process as the first step. Postintervention observations were completed the week immediately after the end of the 2-month intervention period to assess for possible changes from preintervention observations.
For both the pre- and postintervention observations, 4 observers from the Surgical Service Research Committee, who were all registered nurses (RNs), were trained in effective monitoring techniques for hand hygiene compliance. Research indicates that the use of direct observation to measure the compliance of hand hygiene is the best method to evaluate hand hygiene compliance.3
Hand hygiene compliance was defined as any attempt to use water and soap or the use of alcohol based gels by health care professionals after the contact (with a patient or nonpatient object) in a patient's room and before the contact with another patient or object outside the patient's room.13 This definition was selected because of its reliability and ease of monitoring.13 Initial education on the observation technique demonstrated a 100% interrater reliability between observers.
Observers monitored the 3 hospital units unobtrusively. Each observer monitored randomly selected rooms on these units in increments of 30 minutes over a 2-hour period. There were 2 periods of observations: 8:30 to 10:30 and 8:30 to 10:30 . All units were observed on the same day for the morning observation and night observation.
Observations were sorted by profession and type of contact by the health care provider. The providers observed were RNs, nursing assistants (NAs), physicians, ancillary staff such as physical and occupational therapists, and support staff such as housekeeping and nutritional services. All observations were recorded. Multiple observations of 1 staff member were not distinguished during observations.
Intervention
After the preintervention baseline data were collected, staff on each unit were educated on the implementation of the hand hygiene audits and proper methods to give personal feedback. The hand hygiene audit tool asks the auditor to provide information in the following 4 areas for each observation made: (1) title of the person audited (eg, RN, MD); (2) if hand hygiene was performed after contact with a patient (yes/no/na); (3) if hand hygiene was performed after contact with anything in the patient's room other than the patient (yes/no/na); and (4) what feedback was given if the person being audited did not properly perform hand hygiene.
All RNs, NAs, and unit coordinators on each unit were required to complete 10 audits every week over a 2-month period. These staff members were free to audit any other staff (including other RNs, NAs, unit coordinators, physicians, ancillary staff, and support staff) at any time during their shift. During each audit, staff were encouraged to give personal feedback to other staff who were audited as not performing hand hygiene after patient or nonpatient contact. Staff were also instructed that they could enlist the assistance of the nurse manager to give feedback if there were concerns in giving feedback to other staff members. A total of 428 audits were completed over the 2-month intervention period.
Managers of the units on which the study took place and the researcher kept track of all audit tools on a weekly basis. Managers were provided information on which staff were not completing weekly audits. Managers provided feedback to staff to encourage completion of required weekly hand hygiene audits as they saw fit. Compliance rates were generally high with more than 90% of all staff completing audits on a consistent basis.
RESULTS
Preintervention
A total of 263 preintervention observations were made by observers. Of these observations, 110 observations of the hand hygiene compliance were made after direct patient contact and 152 were made after nonpatient contact. Table 1 shows the percentage of hand hygiene compliance by provider after both patient and nonpatient contact.
Pre- and postintervention comparisons
The postintervention observations yielded a total of 253 observations. Of these observations, 125 observations of hand hygiene compliance were after direct patient contact and 128 observations were after nonpatient contact. Student t test analyses were performed to determine whether there were differences between pre- and postintervention hand hygiene compliance. Because of an inadequate number of observations of support staff, they were not included in the analysis. As shown in Table 1, there were no significant differences between pre- and postintervention hand hygiene compliance after patient contact overall or for any particular type of provider.
Analysis of pre- and postintervention hand hygiene compliance after nonpatient contact revealed a significant increase in compliance (P = .006). As shown in Table 1, there was also a significant increase (16.9%) in hand hygiene compliance for RNs after nonpatient contact (P = .003).
DISCUSSIONS
The intervention showed a significant increase in an overall hand hygiene compliance after nonpatient contact in the RN subgroup. This is consistent with other research that found monitoring2,7 and feedback1,6 to be effective tools in increasing hand hygiene compliance.
The intervention showed marked improvement in hand hygiene compliance for the NAs after nonpatient contact but not significantly so. This may be due to the smaller sample size for this population.
Physicians and support staff hand hygiene compliance was comparatively low pre- and postintervention. However, the limited number of observations for these groups makes it difficult to make any conclusions. Further study is needed for these populations.
It was interesting that the intervention's effect was limited to increasing hand hygiene practices after the nonpatient contact but not after direct patient contact. It is possible that the already high level of compliance to hand hygiene practices after patient contact preintervention (83% for RNs and 91% for NAs) created a ceiling effect to mask the possible significant changes in hand hygiene. The use of nominal data points limited the types of statistical procedures able to be used. A much greater level of observations would have been necessary to show the significance of more subtle changes from the pre- to postintervention.
CONCLUSIONS
The use of peer monitoring and feedback appears to be effective in increasing hand hygiene compliance after nonpatient contact. This was particularly true for the RN subgroup. The intervention increased RN hand hygiene compliance after nonpatient contact (82.8%) to that of hand hygiene compliance after direct patient contact (86.5%). The significant increase in hand hygiene after the nonpatient contact was viewed as an important finding especially because research has consistently shown that the transmission of infections in the hospital setting is often through a staff's contact with objects in a patient's room.14-17
The audit tool was straightforward and easy for the staff to complete. Managers appreciated the ease and speed with which it could be implemented. All the 3 managers noted that the audit tool would be an effective means to increase hand hygiene compliance.
Anecdotal feedback from staff also suggested an appreciation of the simplicity of the audit tool and how it made them more aware of the hand hygiene practices. As 1 staff member noted, "I could complete it [the audit tool] easily over 1 shift, and I am sure it made me more aware of my own hand washing practices." Staff also noted that the implementation of the tool gave them a sense of empowerment in giving feedback to the other staff. Some comments from staff included "I would never have told an MD to wash his or her hands without this tool" and "It [the audit tool] gave me justification to tell others to follow [hand hygiene] policy."
There were several weaknesses of the study. First, during pre- and postobservation periods, there were multiple observations of the same people. All of these observations were given the same weight. Some attempts were made to control this through the use of randomization techniques and the same methodology during both pre- and postintervention observations. However, future studies should attempt to give each staff member observed equal weight in the data collection and analysis.
The long-term effects of this intervention were not studied. Follow-up research is planned to determine any long-term effects of the intervention. Future research should concentrate on obtaining larger samples to capture more subtle changes in the data. Possibly the use of more observers and several times for observation would assist in this endeavor. Research should also include ancillary, support, and MD staff in the actual intervention to determine whether this methodology has a significant effect on each of those groups' hand hygiene practices.
REFERENCES