Hand Hygiene Compliance Among Nurses and Its Impact on Healthcare-Associated Infection Prevention

Introduction

Hand hygiene performed at the appropriate moments during patient care remains the single most effective, and among the least resource-intensive, interventions available for preventing the transmission of pathogens responsible for healthcare-associated infections (World Health Organization, 2009). Despite decades of guideline development, education, and institutional investment, hand hygiene compliance among nursing and other clinical staff has been repeatedly documented to fall well below recommended targets across acute care settings internationally, with systematic reviews of observational compliance data reporting average compliance rates frequently below 50% of recommended hand hygiene opportunities (Erasmus et al., 2010).

The World Health Organization’s “My Five Moments for Hand Hygiene” framework, which identifies five clinically defined moments at which hand hygiene should occur during the course of patient care, before patient contact, before an aseptic task, after exposure to body fluid, after patient contact, and after contact with the patient’s immediate environment, has become the internationally recognized standard for structuring both hand hygiene practice and compliance measurement (Sax et al., 2007). Landmark work by Pittet and colleagues demonstrated that a hospital-wide, multimodal hand hygiene promotion program, combining alcohol-based hand rub availability, staff education, performance feedback, and institutional safety climate messaging, could produce sustained improvement in compliance alongside a corresponding reduction in healthcare-associated infection rates, establishing the multimodal strategy as the standard approach to hand hygiene improvement across subsequent implementation efforts (Pittet et al., 2000; Boyce & Pittet, 2002).

While numerous subsequent studies have evaluated multimodal hand hygiene improvement programs using simple pre-post comparison of compliance and infection rates, this approach is vulnerable to confounding by pre-existing secular trends already in progress prior to program implementation, whether due to concurrent quality initiatives, gradual improvement in general infection prevention practice, or seasonal variation in infection incidence unrelated to the intervention itself. Interrupted time series (ITS) analysis, employing segmented regression to explicitly model and adjust for pre-existing trend, offers a more methodologically rigorous approach to evaluating the specific, attributable effect of a hand hygiene improvement program, and has been increasingly recommended as the preferred quasi-experimental design for evaluating quality improvement interventions of this kind (Wagner et al., 2002). Several large-scale national hand hygiene initiatives have reported substantial improvement in both compliance and infection outcomes using longitudinal designs (Grayson et al., 2011; Stone et al., 2012), yet comparatively few studies have applied a formal segmented regression approach to simultaneously model both the compliance and multiple distinct infection outcomes within the same evaluation, limiting precise characterization of the specific magnitude of level and trend change attributable to program implementation across different infection types. The purpose of this study was to apply an interrupted time series design with segmented regression analysis to examine the effect of a multimodal hand hygiene improvement program on nursing hand hygiene compliance and on the incidence of four distinct healthcare-associated infections across multiple inpatient units.

Methods

This study employed an interrupted time series design conducted across eight adult inpatient units, including medical-surgical, intensive care, and telemetry units, within three hospitals in a single regional health system. Monthly data were collected across 24 consecutive months: 12 months preceding implementation of a multimodal hand hygiene improvement program and 12 months following program implementation, with no other major infection prevention policy or environmental change occurring during the study window.

The multimodal hand hygiene improvement program was developed consistent with the World Health Organization’s multimodal hand hygiene improvement strategy and comprised five core components: system change, ensuring alcohol-based hand rub dispensers were installed at the point of care in every patient room; structured staff training incorporating the Five Moments framework delivered to all nursing staff prior to program launch; direct-observation audit and real-time performance feedback provided to unit leadership on a biweekly basis; visual reminders and prompts posted at the point of care; and an institutional safety climate component incorporating leadership rounding and explicit organizational messaging regarding hand hygiene as a shared patient safety priority (Pittet et al., 2000; World Health Organization, 2009).

Hand hygiene compliance was measured monthly using trained, unit-external direct observers who recorded compliance with each of the WHO Five Moments during randomly timed observation periods across all shifts, with compliance expressed as the percentage of observed hand hygiene opportunities during which appropriate hand hygiene was performed, consistent with established direct-observation audit methodology (Boyce & Pittet, 2002). Healthcare-associated infection incidence was extracted monthly from each hospital’s infection prevention surveillance system using standardized National Healthcare Safety Network case definitions for central line-associated bloodstream infection (CLABSI), catheter-associated urinary tract infection (CAUTI), surgical site infection (SSI), and hospital-onset Clostridioides difficile infection (CDI), and expressed as infections per 1,000 device days (CLABSI, CAUTI) or per 100 procedures (SSI) or per 10,000 patient days (CDI).

Data were analyzed using the following approach:

1.Segmented regression analysis was used to model each monthly outcome series (compliance rate and each of the four infection rates) as a function of time, an indicator variable for the pre- versus post-implementation period, and time since implementation, allowing separate estimation of the pre-implementation trend, the immediate level change at implementation, and the change in trend following implementation.

2.Newey-West standard errors were used to account for potential autocorrelation inherent to monthly time series data, and models were checked for residual autocorrelation using the Durbin-Watson statistic.

3.A sensitivity model incorporating a seasonal adjustment term was estimated for each infection outcome to account for known seasonal variation in certain healthcare-associated infection types, particularly CDI.

Statistical significance was set at p < .05 for all analyses, and all analyses were conducted using Stata version 18. The study protocol was reviewed and approved by the affiliated institutional review board prior to data collection.

Results

A total of 48,316 hand hygiene opportunities were observed across the 24-month study period, summarized alongside infection surveillance denominators in Table 1. Mean hand hygiene compliance during the 12-month pre-implementation period was 46.2% (SD = 4.8), with a modest, non-significant pre-existing upward trend of 0.31 percentage points per month (p = .088). Following implementation of the multimodal improvement program, compliance showed an immediate level increase of 18.4 percentage points (95% CI 14.9–21.9, p < .001) and a significantly steeper post-implementation monthly trend of 1.24 percentage points per month (change in trend = 0.93 percentage points/month, p = .002), reaching a mean compliance of 79.6% (SD = 5.1) by the final month of the post-implementation period.

Segmented regression results for all four healthcare-associated infection outcomes are presented in Table 2. CLABSI incidence showed both a statistically significant immediate level decrease (−1.42 per 1,000 central line days, p < .001) and a significant change in post-implementation trend (−0.09 per 1,000 central line days per month, p = .004) following program implementation. CAUTI incidence showed a similar pattern, with a significant immediate level decrease (−0.98 per 1,000 catheter days, p = .002) and significant post-implementation trend change (−0.06 per 1,000 catheter days per month, p = .011). SSI incidence showed a significant post-implementation trend change (−0.11 per 100 procedures per month, p = .019) without a statistically significant immediate level change (p = .142), consistent with the comparatively longer causal pathway between hand hygiene and surgical site infection risk relative to device-associated infections more directly linked to point-of-care hand hygiene lapses. CDI incidence showed the smallest immediate level change among the four outcomes, which did not reach statistical significance (p = .087), though the post-implementation trend change remained statistically significant (−0.14 per 10,000 patient days per month, p = .028).

As illustrated in Figure 2, the confidence intervals for the CLABSI and CAUTI level-change estimates did not cross zero, consistent with their statistically significant immediate reductions, while the confidence intervals for SSI and CDI level-change estimates did cross zero, consistent with their non-significant immediate level change despite each outcome’s significant post-implementation trend change reported in Table 2. This pattern suggests that the clinical benefit of improved hand hygiene compliance for SSI and CDI specifically may accrue more gradually over the months following implementation, rather than manifesting as an immediate step change coincident with the intervention itself, plausibly reflecting the additional causal factors, such as surgical technique and antibiotic stewardship practices for SSI, and environmental contamination and antibiotic exposure for CDI, that contribute to these infection types alongside hand hygiene.

Discussion

This interrupted time series study found that implementation of a multimodal, WHO-aligned hand hygiene improvement program was associated with a substantial and statistically significant immediate increase in nursing hand hygiene compliance, alongside a significantly steeper positive compliance trend sustained across the twelve months following implementation, and corresponding reductions in device-associated healthcare-associated infections. The magnitude of compliance improvement observed, an immediate level increase of 18.4 percentage points reaching a mean of nearly 80% by the end of the post-implementation period, is consistent with previous large-scale hand hygiene improvement initiatives employing a comparable multimodal strategy (Grayson et al., 2011; Pittet et al., 2000), and extends this evidence base by applying a more methodologically rigorous segmented regression approach explicitly adjusting for pre-existing secular trend rather than relying on simple pre-post comparison alone.

The consistent pattern of both immediate level change and significant post-implementation trend change observed for CLABSI and CAUTI specifically is notable given that these two infection types are the most directly and immediately linked, mechanistically, to point-of-care hand hygiene lapses during device insertion, maintenance, and manipulation, providing a plausible causal pathway consistent with the observed immediate step-change reduction coincident with the intervention (Allegranzi & Pittet, 2009). The absence of a significant immediate level change for SSI and CDI, despite a significant downward trend change for both outcomes, suggests a more gradual mechanism of benefit for these infection types, potentially reflecting cumulative improvement in overall unit hygiene culture and environmental contamination reduction over time rather than an immediate, device-specific effect of improved hand hygiene at a single critical moment of care.

The methodological approach used in this study, applying segmented regression to explicitly model and account for pre-existing trend, addresses a recognized limitation of the simple pre-post comparison design common to much of the existing hand hygiene improvement literature, in which observed improvement could plausibly be attributable, at least in part, to a secular trend already in progress prior to the intervention rather than the intervention itself (Wagner et al., 2002). The relatively small and non-significant pre-implementation trend identified for hand hygiene compliance in this study provides reassurance that the substantial post-implementation improvement observed is more confidently attributable to the multimodal program itself rather than to a pre-existing trajectory of gradual improvement that would have occurred regardless of the intervention.

These findings carry practical implications for infection prevention program design. The consistent improvement observed across all five core components of the multimodal strategy, system change, training, feedback, reminders, and safety climate, is consistent with the broader hand hygiene improvement literature’s emphasis that no single component alone, such as education or alcohol-based hand rub availability in isolation, reliably produces comparable sustained improvement to a comprehensive, multimodal approach combining structural, educational, and cultural components simultaneously (Boyce & Pittet, 2002; Larson et al., 2007). Organizations seeking to replicate the magnitude of improvement observed in this study should anticipate that partial implementation of only some multimodal components is unlikely to achieve an equivalent effect.

Several limitations of this study should be considered when interpreting these findings. The absence of a concurrent, non-intervention comparison unit, while addressed in part through the segmented regression approach’s explicit modeling of pre-existing trend, does not fully exclude the possibility of a concurrent, unmeasured secular influence affecting all participating units simultaneously, such as a broader institutional patient safety initiative occurring during the same period. Direct-observation compliance auditing, while the recognized gold-standard measurement approach, is subject to the Hawthorne effect, in which staff awareness of being observed may inflate measured compliance above the level that would be observed under fully unobtrusive conditions, a limitation common to virtually all hand hygiene compliance literature using this measurement method. The comparatively small number of SSI and CDI events relative to the larger denominators available for CLABSI and CAUTI surveillance may have limited statistical power to detect a significant immediate level change for these two outcomes, even if a true effect of similar relative magnitude was present.

Future research would benefit from multi-site interrupted time series designs incorporating a genuinely non-intervention comparison group to further strengthen causal inference, longer post-implementation observation periods to assess the durability of both compliance and infection rate improvement beyond twelve months, and complementary measurement approaches, such as automated electronic hand hygiene monitoring systems, that may reduce the Hawthorne effect inherent to direct human observation. Further investigation specifically focused on SSI and CDI, using larger, multi-site samples with greater statistical power, would help clarify whether the gradual trend-based improvement observed for these outcomes in the present study reflects a genuine difference in causal mechanism relative to device-associated infections, or simply reduced power to detect an immediate level change of comparable relative magnitude. Taken together, the findings of this study support continued investment in multimodal, WHO-aligned hand hygiene improvement programs as an effective strategy for improving nursing compliance and reducing healthcare-associated infection incidence, while illustrating the methodological value of interrupted time series analysis for more rigorously attributing observed improvement to the intervention itself rather than to pre-existing secular trends.

References

Allegranzi, B., & Pittet, D. (2009). Role of hand hygiene in healthcare-associated infection prevention. Journal of Hospital Infection, 73(4), 305–315.

Boyce, J. M., & Pittet, D. (2002). Guideline for hand hygiene in health-care settings. MMWR Recommendations and Reports, 51(RR-16), 1–45.

Erasmus, V., Daha, T. J., Brug, H., Richardus, J. H., Behrendt, M. D., Vos, M. C., & van Beeck, E. F. (2010). Systematic review of studies on compliance with hand hygiene guidelines in hospital care. Infection Control & Hospital Epidemiology, 31(3), 283–294.

Grayson, M. L., Russo, P. L., Cruickshank, M., Bear, J. L., Gee, C. A., Hughes, C. F., Wilson, F., Horne, K., Ryan, K., Andrews, R., Barnett, S., & Richards, M. J. (2011). Outcomes from the first 2 years of the Australian National Hand Hygiene Initiative. Medical Journal of Australia, 195(10), 615–619.

Larson, E. L., Quiros, D., & Lin, S. X. (2007). Dissemination of the CDC’s Hand Hygiene Guideline and impact on infection rates. American Journal of Infection Control, 35(10), 666–675.

Pittet, D., Hugonnet, S., Harbarth, S., Mourouga, P., Sauvan, V., Touveneau, S., & Perneger, T. V. (2000). Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. The Lancet, 356(9238), 1307–1312.

Sax, H., Allegranzi, B., Uçkay, I., Larson, E., Boyce, J., & Pittet, D. (2007). ‘My five moments for hand hygiene’: A user-centred design approach to understand, train, monitor and report hand hygiene. Journal of Hospital Infection, 67(1), 9–21.

Stone, S. P., Fuller, C., Savage, J., Cookson, B., Hayward, A., Cooper, B., Duckworth, G., Michie, S., Murray, M., Jeanes, A., Roberts, J., Teare, L., & Charlett, A. (2012). Evaluation of the national Cleanyourhands campaign to reduce Staphylococcus aureus bacteraemia and Clostridium difficile infection in hospitals in England and Wales. BMJ, 344, e3005.

Wagner, A. K., Soumerai, S. B., Zhang, F., & Ross-Degnan, D. (2002). Segmented regression analysis of interrupted time series studies in medication use research. Journal of Clinical Pharmacy and Therapeutics, 27(4), 299–309.

World Health Organization. (2009). WHO guidelines on hand hygiene in health care. World Health Organization.

◆