Respiratory protection programs, including fit testing of respirators, have been inconsistently implemented; evidence of their long-term efficacy is lacking. We undertook a study to determine the short- and long-term efficacy of training for fit testing of N95 respirators in both untrained and trained healthcare workers (HCWs).

Prospective observational cohort study.

A group of at-risk, consenting HCWs not previously fit-tested for a respirator were provided with a standard fit-test protocol. Participants were evaluated after each of 3 phases, and 3 and 14 months afterward. A second group of previously fit-tested nurses was studied to assess the impact of regular respirator use on performance.

Of 43 untrained fit-tested HCWs followed for 14 months, 19 (44.2%) passed the initial fit test without having any specific instruction on respirator donning technique. After the initial test, subsequent instruction led to a pass for another 13 (30.2%) of the 43 HCWs, using their original respirators. The remainder required trying other types of respirators to acheive a proper fit. At 3 and 14 months' follow-up, failure rates of 53.5% (23 of 43 HCWs) and 34.9% (15 of 43 HCWs), respectively, were observed. Pass rates of 87.5%-100.0% were observed among regular users.

Without any instruction, nearly 50% of the HCWs achieved an adequate facial seal with the most commonly used N95 respirator. Formal fit testing does not predict future adequacy of fit, unless frequent, routine use is made of the respirator. The utility of fit testing among infrequent users of N95 respirators is questionable.

To evaluate the prevalence and transmission of methicillin-resistant Staphylococcus aureus (MRSA) nasal colonization, as well as risk factors associated with MRSA carriage, among residents of a long-term care facility (LTCF).

Prospective, longitudinal cohort study.

A 100-bed Veterans Administration LTCF

All current and newly admitted residents of the LTCF during an 8-week study period.

Nasal swab samples were obtained weekly and cultured on MRSA-selective media, and the cultures were graded for growth on a semiquantitative scale from 0 (no growth) to 6 (heavy growth). Epidemiologic data for the periods before and during the study were collected to assess risk factors for MRSA carriage.

Of 83 LTCF residents, 49 (59%) had 1 or more nasal swab cultures that were positive for MRSA; 34 (41%) were consistently culture-negative (designated “noncarriers”). Of the 49 culture-positive residents, 30 (36% of the total of 83 residents) had all cultures positive for MRSA (designated “persistent carriers”), and 19 (23% of the 83 residents) had at least 1 culture, but not all cultures, positive for MRSA (designated “intermittent carriers”). Multivariate analysis showed that participants with at least 1 nasal swab culture positive for MRSA were likely to have had previous hospitalization (odds ratio, 3.9) or wounds (odds ratio, 8.2). Persistent carriers and intermittent carriers did not differ in epidemiologic characteristics but did differ in mean MRSA growth score (3.7 vs 0.7; P < .001).

Epidemiologic characteristics differed between noncarriers and subjects with at least 1 nasal swab culture positive for MRSA. However, in this LTCF population, only the degree of bacterial colonization (as reflected by mean MRSA growth score) distinguished persistent carriers from intermittent carriers. Understanding the burden of colonization may be important when determining future surveillance and control strategies.

To describe the development of a guideline for the management of ventilator-associated pneumonia (VAP) based on local microbiologic findings and to evaluate the impact of the guideline on antimicrobial use practices.

Retrospective comparison of antimicrobial use practices before and after implementation of the guideline.

Intensive care units at Harborview Medical Center, Seattle, Washington, a university-affiliated urban teaching hospital.

A total of 819 patients who received mechanical ventilation and who underwent quantitative bronchoscopy between July 1, 2003, and June 30, 2005, for suspected VAP.

Implementation of an evidence-based VAP guideline that focused on the use of quantitative bronchoscopy for diagnosis, administration of empirical antimicrobial therapy based on local microbiologic findings and resistance patterns, tailoring definitive antimicrobial therapy on the basis of culture results, and appropriate duration of therapy.

During the baseline period, 168 (46.7%) of 360 patients had quantitative cultures that met the diagnostic criteria for VAP, compared with 216 (47.1%) of 459 patients in the period after the guideline was implemented. The pathogens responsible for VAP remained similar between the 2 periods, except that the prevalence of VAP due to carbapenem-resistant Acinetobacter species increased from 1.8% to 15.3% (P < .001), particularly in late-onset VAP. Compared with the baseline period, there was an improvement in antimicrobial use practices after implementation of the guideline: antimicrobial therapy was more frequently tailored on the basis of quantitative culture results (103 [61.3%] of 168 vs 150 [69.4%] of 216 patients; P = .034), there was an increase in the use of appropriate definitive therapy (135 [80.4%] of 168 vs 193 [89.4%] of 216 patients; P = .001), and there wasadecrease in the mean duration oftherapy (12.0vs 10.7days; P = .0014). The all-cause mortality rate was similar in the periods before and after the guideline was implemented (38 [22.6%] of 168 vs 46 [21.3%] of 216 patients; P = .756).

Implementation of a guideline for the management of VAP that incorporated the use of quantitative bronchoscopy, the use of empirical therapy based on local microbiologic findings, tailoring of therapy on the basis of culture results, and use of shortened durations of therapy led to significant improvements in antimicrobial use practices without adversely affecting the all-cause mortality rate.

To examine the relative proportions of central line-associated bloodstream infection (BSI) due to gram-negative bacteria and due to gram-positive bacteria among patients who had undergone surgery and patients who had not. The study also evaluated clinical predictive factors and unadjusted outcomes associated with central line-associated BSI caused by gram-negative bacteria in the postoperative period.

Observational, case-control study based on a retrospective review of medical records.

University of Chicago Medical Center, a 500-bed tertiary care center located on Chicago's south side.

Adult intensive care unit (ICU) patients who developed central line-associated BSI.

There were a total of 142 adult patients who met the Centers for Disease Control and Prevention National Nosocomial Infection Surveillance System definition for central line-associated BSI. Of those, 66 patients (46.5%) had infections due to gram-positive bacteria, 49 patients (34.5%) had infections due to gram-negative bacteria, 23 patients (16.2%) had infections due to yeast, and 4 patients (2.8%) had mixed infections. Patients who underwent surgery were more likely to develop central line-associated BSI due to gram-negative bacteria within 28 days of the surgery, compared with patients who had not had surgery recently (57.6% vs 27.3%; P = .002). On multivariable logistic regression analysis, diabetes mellitus (adjusted odds ratio [OR], 4.6 [95% CI, 1.2-18.1]; P = .03) and the presence of hypotension at the time of the first blood culture positive for a pathogen (adjusted OR, 9.8 [95% CI, 2.5-39.1]; P = .001 ) were found to be independently predictive of central line-associated BSI caused by gram-negative bacteria. Unadjusted outcomes were not different in the group with BSI due to gram-negative pathogens, compared to the group with BSI due to gram-positive pathogens.

Clinicians caring for critically ill patients after surgery should be especially concerned about the possibility of central line-associated BSI caused by gram-negative pathogens. The presence of diabetes and hypotension appear to be significant associated factors.

To compare the effectiveness for prevention of central venous and arterial catheter colonization of 3 skin antisepsis with 1 of 3 antiseptic solutions: 10% aqueous povidone iodine (aqueous PI), 2% aqueous chlorhexidine gluconate (aqueous CG), and 0.5% alcoholic chlorhexidine gluconate (alcoholic CG).

Prospective, randomized controlled trial.

Intensive care unit in a teaching hospital.

Patients were randomly assigned to 1 of the 3 skin antisepsis groups. The distal tips of catheters were semiquantitatively cultured when the catheters were no longer necessary or if there was a suspicion of catheter-related infection. Rates of catheter colonization, catheter-related sepsis, and catheter-related bacteremia were compared among the 3 groups.

A total of 631 catheters were included in the study (194 from the aqueous PI group, 211 from the aqueous CG group, and 226 from the alcoholic CG group). The incidence of catheter colonization was significantly lower in the alcoholic CG than in the aqueous PI group (14.2% vs 24.7%; relative risk, 0.5 [95% confidence interval, 0.3-0.8; P < .01]); it was also significantly lower in the aqueous CG group than in the aqueous PI group (16.1 % vs 24.7%; relative risk, 0.6 [95% confidence interval, 0.4-0.9; P = .03]). There were no significant differences between the aqueous CG and the alcoholic CG groups. Incidences of catheter-related bacteremia were similar for all 3 groups. The aqueous and alcoholic CG solutions were superior to the aqueous PI solution in preventing catheter colonization due to gram-positive bacteria.

The aqueous and alcoholic CG solutions for cutaneous antisepsis were similarly effective in preventing colonization of central venous catheters and arterial catheters. Both had significantly lower incidences of colonization than did the aqueous PI solution; this effect seems to be related to the CG solutions' more efficacious prevention of colonization with gram-positive bacteria.

To estimate the level of hand or glove contamination with vancomycin-resistant enterococci (VRE) among healthcare workers (HCWs) who touch a patient colonized with VRE and/or the colonized patient's environment during routine care.

Structured observational study.

Medical intensive care unit of a 700-bed, tertiary-care teaching hospital.

VRE-colonized patients and their caregivers.

We obtained samples from sites on the intact skin of 22 patients colonized with VRE and samples from sites in the patients' rooms, before and after routine care, during 27 monitoring episodes. A total of 98 unique HCWs were observed during 131 HCW observations. Observers recorded the sites touched by HCWs. Culture samples were obtained from HCWs' hands and gloves before and after care.

VRE were isolated from a mean (±SD) of 55% ± 24% of patient sites (n = 256) and 17% ± 12% of environmental sites (n = 1,572). Most HCWs (131 [56%]) touched both the patient and the patient's environment; no HCW touched only the patient. Of 103 HCWs whose hand samples were negative for VRE when they entered the room, 52% contaminated their hands or gloves after touching the environment, and 70% contaminated their hands or gloves after touching the patient and the environment (P = .101). In a univariate logistic regression model, the risk of hand or glove contamination was associated with the number of contacts made (odds ratio, 1.1 [95% confidence interval, 1.01-1.19). In a multivariate model, the effect of the number of contacts could not be distinguished from the effect of type of contact (ie, touching the environment alone or touching both the patient and the environment). Overall, 37% of HCWs who did not wear gloves contaminated their hands, and 5% of HCWs who wore gloves did so (an 86% difference).

HCWs were nearly as likely to have contaminated their hands or gloves after touching the environment in a room occupied by a patient colonized by VRE as after touching the colonized patient and the patient's environment. Gloves were highly protective with respect to hand contamination.

To investigate the cause(s) of an increased incidence of clinical cultures growing Mycobacterium abscessus at a hospital in Florida.

Outbreak investigation.

University-affiliated, tertiary-care hospital.

A site visit was done during the first week of September 2006. We reviewed the medical records of patients from whom M. abscessus was recovered during the period from January 1, 2003, through June 30, 2006. We collected environmental samples from various sites and evaluated specimen processing procedures in the microbiology laboratory. Isolates of M. abscessus recovered from the environment and from 12 randomly selected patients who sought medical care in 2006 were compared by pulsed-field gel electrophoresis (PFGE). Follow-up case surveillance was continued through March 31, 2007.

Specimens from 143 patients obtained from various anatomical sites grew M. abscessus on culture in 2005-2006, compared with specimens from 21 patients in 2003-2004. The 12 isolates from patients that were selected for molecular typing had indistinguishable PFGE patterns. Observations revealed no major breaches in the processing of mycobacterial specimens in the laboratory. Isolates grew only after prolonged incubation (mean ± SD, 45 ± 15 days) in test tubes containing diagonally oriented Middlebrook and Cohn 7H10 agar or Lowenstein-Jensen medium. Environmental samples obtained from the inside of the specimen incubator grew M. abscessus on culture. A test tube containing diagonally oriented, uninoculated Middlebrook and Cohn 7H10 agar that was incubated in the same incubator as clinical specimens grew M. abscessus with a PFGE pattern that matched the pattern of the patient isolates. Cases of M. abscessus infection decreased to baseline after the hospital changed suppliers of mycobacterial media and cleaned the incubator.

Although the source was never confirmed, our investigation suggests that this was a pseudo-outbreak of M. abscessus infection that resulted from contamination of mycobacterial cultures during incubation. Our findings emphasize the need for guidance on the disinfection of specimen incubators.

To implement a system for monitoring of rare events based on statistical process control charts.

Statistical process control plotting by g chart of clinical microbiology laboratory data.

Primary and secondary care Norwegian hospital with a 9-bed intensive care unit.

During the winter of 2001–2002 in Norway, there was a national monoclonal nosocomial outbreak of Pseudomonas aeruginosa infection mainly affecting patients in intensive care units. In the present work, we demonstrate how the use of SPC at one of the affected hospitals would have detected this outbreak several weeks before the alert from the Norwegian National Public Health Institute (NIPH). By plotting the monthly incidence rate of P. aeruginosa infection (with a c chart), we found that the hospital would have been alerted in February, by plotting the number of days between events (with a g chart), we found that the hospital would have detected a process already out of control in early January 2002. Not until 9 weeks later (ie, mid-March) did the NIPH declare the P. aeruginosa outbreak to be national, and a commercially produced mouth swab contaminated during the manufacturing process was found to be the source.

The plotting of rare events, such as an outbreak of nosocomial infection, with a g chart may be used for early detection of a process out of control.

To determine the rate of influenza vaccination and the factors associated with the vaccination's acceptance among nurses in Hong Kong.

Cross-sectional survey.

Nurses practicing between 2003 and 2007.

A questionnaire was sent to all nurses registered with any of the 3 nursing associations that participated in this study.

A total of 941 completed questionnaires were available for analysis, though not all nurses responded to every question (response rate, 33.5%-36.3%). Vaccination rates in 2006 and 2007 were 57.2% and 46.2%, respectively. Nurses who were vaccinated in 2006 were more likely to get vaccinated in 2007 (P<.01); 56% of the nurses perceived influenza vaccine as being effective against influenza. The perceived effectiveness of influenza vaccine was a consistent predictor of rates of vaccination in 2006 (odds ratio [OR], 8.47 [95% confidence interval {CI}, 6.13-11.70]; P<.01) and 2007 (OR, 6.05 [95% CI, 3.79-9.67]; P<.01). Concern about contracting avian influenza was a predictor of the vaccination rate in 2006 but not in 2007 (OR, 1.47 [95% CI, 1.03-2.09]; P<.05), as was the perceived lack of control over avian influenza infection (OR, 1.52 [95% CI, 1.06-2.18]; P<.05).

The overall influenza vaccination rate for nurses in Hong Kong was about 50%. It was affected by the perceived threat of an impending outbreak. The attitudes of nurses toward the effectiveness of and rationale for vaccination were a major barrier to increasing the rate of vaccination.

To evaluate educational level as a contributing factor in handwashing compliance.

Observation of hand washing opportunities was performed for approximately 12 weeks before an announced Joint Commission on Accreditation of Healthcare Organizations (JCAHO) visit and for approximately 10 weeks after the visit. Trained observers recorded the date, time, and location of the observation; the type of healthcare worker or hospital employee observed; and the type of hand hygiene opportunity observed.

University of Toledo Medical Center, a 319-bed teaching hospital.

A total of 2,373 observations were performed. The rate of hand washing compliance among nurses was 91.3% overall. Medical attending physicians had the lowest observed rate of compliance (72.4%; P < .001). Nurses showed statistically significant improvement in their rate of hand hygiene compliance after the JCAHO visit (P = .001), but no improvement was seen for attending physicians (P = .117). The compliance rate in the surgical intensive care unit was more than 90%, greater than that in other hospital units (P = .001). Statistically, the compliance rate was better during the first part of the week (Monday, Tuesday, and Wednesday) than during the latter part of the week (Thursday and Friday) (P = .002), and the compliance rate was better during the 3 PM-1 1 PM shift, compared with the 7 AM-3 PM shift (P < .001). When evaluated by logistic regression analysis, non-physician healthcare worker status and observation after the JCAHO accreditation visit were associated with an increased rate of hand hygiene compliance.

An inverse correlation existed between the level of professional educational and the rate of compliance. Future research initiatives may need to address the different motivating factors for hand hygiene among nurses and physicians to increase compliance.

To measure the incidence of nosocomial infection (NI) among patients with septic shock according to the place of septic shock acquisition and to evaluate the increase in the risk of pulmonary infection associated with septic shock.

Prospective cohort study.

TWO intensive care units (ICUs) of a French university hospital.

The study included a total of 209 septic shock patients during the period December 1, 2001 through April 30, 2005. The place of septic shock acquisition for 108 patients was the community; for 87, the hospital; and for 14, the ICU. To evaluate the impact of septic shock on the development of pulmonary infection, a competitive and adjusted hazard ratio (aHR) model was applied to nontrauma ICU patients.

Among the 209 study patients, 48 (23%) experienced 66 NIs after septic shock. There was no significant difference in the NI attack rates according to place of acquisition: for the community acquisition group, 24 cases per 100 patients (95% confidence interval [CI], 16-32); for the hospital acquisition group, 20 cases per 100 patients (95% CI, 11-28); and for the ICU acquisition group, 36 cases per 100 patients (95% CI, 11-61) (P = .3). For nontrauma ICU patients, the presence of community-acquired septic shock was found to be independently associated with a higher incidence of pulmonary infection, compared with the absence of septic shock (aHR, 2.12 [95% CI, 1.08-4.16]; P = .03).

The risk of NI did not differ by the place of septic shock acquisition. The risk of pulmonary infection was higher for ICU patients with community-acquired septic shock who were admitted for underlying nontrauma disease. Studies are needed to investigate the pathogenic mechanisms that facilitate pulmonary infection in this population, taking into account exposure to invasive devices and immunosuppression after the initial phase of septic shock.

To evaluate the antimicrobial efficacy of and risk of organisms developing resistance to 2 novel hand soaps: (1) a soap containing triclosan, polyhexamethylene biguanide, and benzethonium chloride added to a soap base (TPB soap); and (2) a soap containing farnesol, polyhexamethylene biguanide, and benzethonium chloride added to a soap base (FPB soap). Tests also included soaps containing only triclosan.

The risk of emergence of resistant bacterial mutants was investigated by determining the susceptibility changes after repeated exposure of bacteria to the drugs and soaps in vitro. The effectiveness of the soaps was evaluated using an in vitro tube dilution method, a volunteer method (the ASTM standard), and 2 pig skin methods.

The minimum inhibitory concentration and minimum bactericidal concentration of triclosan against Staphylococcus, aureus increased 8- to 62.5-fold, whereas those of TPB and FPB (both alone and in soap) were unchanged. In vitro, TPB and FPB soaps produced higher log10 reductions in colony-forming units of all tested organisms (4.95-8.58) than did soaps containing triclosan alone (0.29-4.86). In the test using the pig skin and volunteer methods, TPB soap produced a higher log10 reduction in colony-forming units (3.1-3.3) than did the soap containing triclosan alone (2.6-2.8).

The results indicate that TPB and FPB soaps may provide superior rapid and broad-spectrum efficacy with a lower risk of organisms developing resistance than do soaps containing triclosan alone. Pig skin methods may be used to predict the efficacy of antibacterial soaps in the rapid disinfection of contaminated hands. Hand washing with TPB and FPB soaps by healthcare workers and the general population may reduce the transmission of pathogens, with a lower risk of promoting the emergence of resistant organisms.

With childhood varicella vaccination in the United States have come concerns that the incidence of herpes zoster may increase, because of diminishing natural exposure to varicella and consequent reactivation of latent varicella zoster virus. We wanted to estimate the rate of herpes zoster-related hospitalizations and the associated hospital charges before and during the promotion of varicella vaccination in the United States.

A retrospective study of patients from the Nationwide Inpatient Sample for the years 1993–2004 who were hospitalized due to herpes zoster infection.

We searched for diagnoses of herpes zoster (using the International Classification of Diseases, Ninth Revison, Clinical Modification codes starting with 053) in all 15 diagnostic-code fields included for hospital discharges in the Nationwide Inpatient Sample during 1993–2004. We designed our analysis to examine the rates of severe illness due to herpes zoster that resulted in hospitalization, as measured by the rates of herpes zoster-related hospital discharges (HZHDs). The annual population-adjusted rate of HZHDs (per 10,000 US population) and the annual inflation-adjusted total charges for HZHDs were the primary outcomes. Secondary outcomes included mean charges for HZHDs and the distribution of total charges for HZHDs by expected primary payer. Varicella-related hospital discharges (VRHDs) were identified by use of similar diagnosis-based methods, which were described in our previous study.

Population-adjusted rates of HZHDs did not change significantiy from the prevaccination years (1993–1995) through the initial 5 years of the varicella vaccination period. Beginning in 2001, however, the rate of HZHDs overall began to increase, and by 2004 the overall rate was 2.5 HZHDs (95% confidence interval, 2.38–2.62) per 10,000 US population, significantly higher than any of the rates calculated during the years prior to 2002. Hospital charges for HZHDs overall increased by more than $700 million annually by 2004; in particular, we found that the herpes zoster vaccine–eligible population (ie, persons aged 60 years or older) accounted for 74% of the total annual hospital charges in 2004. The annual rate of VRHDs and the associated hospital charges decreased significantly from 1993 through 2004, but the decrease in hospitalizations and charges for VRHDs was less than the increase in hospitalizations and charges for HZHDs.

AS the rates of VRHDs and the associated charges have decreased, there has been a significant increase in HZHDs and associated charges, disproportionately among older adults. Herpes zoster vaccine may mitigate these trends for HZHDs.

Although urinary tract infection (UTI) is the most common hospital-acquired infection, there is little information about why hospitals use or do not use a range of available preventive practices. We thus conducted a multicenter study to understand better how US hospitals approach the prevention of hospital-acquired UTI.

This research is part of a larger study employing both quantitative and qualitative methods. The qualitative phase consisted of 38 semistructured phone interviews with key personnel at 14 purposefully sampled US hospitals and 39 in-person interviews at 5 of those 14 hospitals, to identify recurrent and unifying themes that characterize how hospitals have addressed hospital-acquired UTI.

Four recurrent themes emerged from our study data. First, although preventing hospital-acquired UTI was a low priority for most hospitals, there was substantial recognition of the value of early removal of a urinary catheter for patients. Second, those hospitals that made UTI prevention a high priority also focused on noninfectious complications and had committed advocates, or “champions,” who facilitated prevention activities. Third, hospital-specific pilot studies were important in deciding whether or not to use devices such as antimicrobial-impregnated catheters. Finally, external forces, such as public reporting, influenced UTI surveillance and infection prevention activities.

Clinicians and policy makers can use our findings to develop initiatives that, for example, use a champion to promote the removal of unnecessary urinary catheters or exploit external forces, such public reporting, to enhance patient safety.

In September 2006, the Centers for Disease Control and Prevention was notified of cases of gram-negative bloodstream infection (BSI) occurring among outpatients who received an intravenous formulation of the prostanoid treprostinil. An investigation was conducted to determine rates of prostanoid-associated BSI in this patient population and possible risk factors for infection.

We performed a retrospective cohort study of patients who had received intravenous formulations of at least 1 of the 2 approved prostanoids (epoprostenol and treprostinil) from January 1, 2004, through late 2006. Chart reviews were conducted at 2 large centers for pulmonary arterial hypertension, and a survey of infection control practices was conducted at 1 center.

A total of 224 patients were given intravenous prostanoid treatment, corresponding to 146,093 treatment-days during the study period. Overall, there were 0.55 cases of BSI and 0.18 cases of BSI due to gram-negative organisms per 1,000 treatment-days. BSI rates were higher for patients who received intravenous treprostinil than for patients who received intravenous epoprostenol (1.13 vs. 0.42 BSIs per 1,000 treatment-days; P < .001), as were rates of BSI due to gram-negative organisms (0.81 vs. 0.04 BSIs per 1,000 treatment-days; P < .001). Adjusted hazard ratios for all BSIs and for BSIs due to gram-negative organisms were higher among patients given treatment with intravenous treprostinil. The survey identified no significant differences in medication-related infection control practices.

At 2 centers, BSI due to gram-negative pathogens was more common than previously reported and was more frequent among patients given treatment with intravenous treprostinil than among patients given treatment with intravenous epoprostenol. Whether similar results would be found at other centers for pulmonary arterial hypertension warrants further investigation. This investigation underscores the importance of surveillance and evaluation of healthcare-related adverse events in patients given treatment primarily as outpatients.

To assess published hand hygiene behavioral interventions that employed a social marketing framework and to recommend improvements to future interventions.

We performed a systematic literature review by searching the PubMed database and the Cumulative Index to Nursing and Allied Health Literature for published articles about hand hygiene behavioral interventions in healthcare facilities, schools, and community settings. Our analysis included articles that describe multifaceted interventions and evaluated them with predefined social marketing benchmark criteria.

Of 53 interventions analyzed in this review, 16 (30.2%) employed primary formative audience research, 5 (9.4%) incorporated social or behavioral theories, 27 (50.9%) employed segmentation and targeting of the audience, 44 (83.0%) used components of the “marketing mix,” 3 (5.7%) considered the influence of competing behaviors, 7 (13.2%) cultivated relationships with the target audience, and 15 (28.3%) provided simple behavioral messages. Thirty-five (66.0%) of the interventions demonstrated a significant improvement in performance, but only 21 (39.6%) were considered to have a strong evaluative design. The median duration of the interventions was 8.0 months.

From a social marketing perspective, the promotion of hand hygiene could be improved in several ways. The effectiveness of social marketing in hand hygiene promotion should be tested in future interventions.