Impact of patient safety climate on infection prevention practices and healthcare worker and patient outcomes

Abstract

Background:

Standard precautions may prevent patient health care associated infections and provider occupational exposures but are not often used by health care workers. A positive patient safety climate might contribute to improved adherence. The aim of this study was to determine the relationships among patient safety climate, standard precaution adherence, and health care worker exposures and HAIs.

Methods:

This multi-site, cross-sectional study included survey data from nurses on patient safety climate, observational data on adherence, and existing health care worker exposure and health care associated infections data. Data were aggregated to hospital unit level for correlational and multivariable regression analyses.

Results:

A total of 5,285 standard precaution observations and 452 surveys were collected across 43 hospital units. Observed adherence to all categories of standard precautions was 64.4%; there were significant differences by provider role. Multivariable models identified key predictors explaining sizeable variance in methicillin-resistant Staphylococcus aureus (41%), catheter associated urinary tract infections (23%), mucotaneous exposures (43%) and needlestick and sharps injuries (38%).

Discussion:

This study produced findings not previously published thus advancing the state of the science in patient and occupational health safety. These include identifying modifiable features of the safety climate and key organizational characteristics associated better outcomes.

Conclusions:

In this novel study we identified that a positive patient safety climate and adherence to standard precautions predict key HAI and occupational health outcomes.

Safe health care demands highly reliable delivery, including the most basic standards of care.^1–4 Hospitalized patients require advanced interdisciplinary clinical care in concert with current technological advancements under highly dynamic circumstances.^5,6 As such, health care delivery is an increasingly high-hazard occupation in a high-hazard setting, requiring that health care workers (HCW) have an adaptive skill set to operate under complex conditions.^7,8 Lapses are evident in complications of care, including HCW blood-borne pathogen exposures and patient health care associated infections (HAI).^9,10

The health care and social assistance industry sector covers over 19 million employed workers in the United States (U.S.), and 80% are in health care industries.¹¹ Of those, over 2 million are registered nurses (RNs) comprising the largest sector of the health care workforce, 70% of whom work in hospitals.^11,12 Therefore, keeping our health care workforce safe, particularly RNs in hospitals, is a national priority.

Unfortunately, in the U.S. the approximately 5.6 million HCW are at risk of exposure to blood-borne pathogens; of these, approximately 385,000 experience a sharps injury annually, or an average of 1,000 per day in hospitals alone, at a rate of 4.0/100 full time equivalents (FTE).^13,14 The majority of these, 56%−88%, are preventable.^13–15 Nevertheless, despite an immediate and dramatic decrease following passage of the Needlestick Safety and Prevention Act in 2001,¹⁶ rates remain unacceptably high, with HCW sharps injury incidence of 321,907 and blood-borne pathogen exposures of 441,344 annually at a rate of 2.48/100 FTEs.^16,17 Simultaneously, 5%−10% of hospitalized patients for whom at-risk HCW provide care acquire one or more HAI; that is approximately 2 million patients, of whom, an estimated 99,000 will die annually.^19–22 Estimates suggest between 10% and 70% of HAI are preventable.^19,21

Standard precautions are a fundamental set of actions HCW should take as a primary infection prevention strategy, designed to limit risk of blood-borne infections, other occupational infections, and patient HAI.²³ Components of standard precautions include indications and recommendations for hand hygiene, use of appropriate personal protective equipment (PPE), safe use and disposal of sharps, decontamination of environment and equipment, infectious patient placement, and linen and waste management. These precautions are federally regulated and establish a minimum standard of care for HCW and organizations.²⁴

Despite its importance, however, adherence to standard precautions remains grossly suboptimal and is adhered to less than 50% of the time.^25–28 Reasons are not completely clear, though there is evidence that features of the patient safety climate (defined as the group level perceptions of perceived leadership and management support, job hindrances, orderliness, and cleanliness), attitudes, and shared experiences of the organizational culture of safety may be related to safety practices, including adherence to standard precautions.^28–33 Nevertheless, the relationship between standard precaution adherence and important patient and HCW outcomes has not been fully elucidated. Consequently, the relationships among patient safety climate, standard precaution adherence and patient HAIs or HCW occupational exposures remains unknown.

The aims of this research were to: (1) describe the relationship between patient safety climate and self-reported or observed standard precaution adherence; (2) identify the relationship between standard precaution adherence and HAIs or HCW exposures; and (3) determine the relationships among patient safety climate, observed standard precaution adherence, and HAIs or HCW exposures.

METHODS

Research design

This multi-site, cross-sectional study design included surveys of nurses in U.S. hospital units on patient safety climate and standard precaution adherence, collection of standard precaution observational adherence data on those same units, and unit level HAI and HCW sharp and splash exposure data.

Sample and setting

Study sites were identified, and lead personnel were recruited through professional infection control (Association for Professionals in Infection Control and Epidemiology) and occupational health (Association of Occupational Health Professionals) organizations. Hospitals were screened to assure eligibility including use and availability of data following National Health care Safety Network (NHSN)³⁴ and OSHA 300⁸ surveillance methodology and definitions. Hospitals meeting criteria were selected based on timing of response with the goal to maximize generalizability by geographic region and bed size. Ultimately recruitment and selection of study sites entailed multiple communications with potential study sites over several months to assure their administrative permissions to participate as well as consideration of competing demands on IP staff, this limited exclusion by geography or bed size though remained a consideration in analyses. To limit the potential effect of health care delivery differences across varying patient care settings medical-surgical units were included and critical care (eg, emergency department, intensive care units) were excluded. Hospital units that did not submit all 3 types of data (observations, surveys and outcomes) were excluded from analyses.

Survey inclusion criteria were RNs who currently worked in a direct care capacity for at least 16 hours per week and work history for a minimum of 6 months on the selected unit. Exclusion criteria were nurses who are not RNs or did not work in a direct care capacity either currently or for the minimum time to be considered a part of the unit climate (eg, float nurses, nurses in orientation). Rationale for selecting RNs to survey was that they comprise the largest sector of the unit level workforce representing patient safety climate and are routinely available, supporting study feasibility.

Measures

Survey on patient safety and standard precautions

This survey measured HCW perceptions of patient safety climate in the hospital unit on which they worked, reported adherence to standard precaution practices, and factors that influenced that adherence. The development and testing of the tool has been reported elsewhere.³⁵ Forty-four items measured 12 dimensions of patient safety climate: supervisor expectations and actions promoting safety (4 items), organizational learning (3 items), teamwork within (4 items) and across units (4 items), communication openness (3 items), feedback and communication about errors (3 items), non-punitive responses to incidents (3 items), staffing (4 items), hospital management support for patient safety (3 items), handoffs and transitions (4 items), frequency of events reported (3 items) and overall perceptions of safety (4 items), and 2 items measured patient safety grade and number of events reported. Items are measured using a 5-point Likert scales so that a 1 represents a low score and a 5 a high score, and a composite score per dimension is obtained. An additional 22 items measured 2 dimensions, the perception of work environment barriers and facilitators to perform standard precautions (10 items on a 5-point Likert scale from “strongly disagree” to “strongly agree”), and self-reported standard precaution behaviors performed (12 items on a 5-point Likert scale from “never” to “always).” Unit level patient safety climate was operationalized as a composite frequency scores of positive responses (rated 4 or 5). In sum there were 14 dimensions measured; 13 (excluding self-reported practices) were aggregated to create a unit level average as a composite unit safety score.

Standard precautions observation tool

The Standard Precaution Observation Tool (SPOT) is a hard copy observation tool that was designed to unobtrusively observe HCW encounters with patients to measure observed HCW standard precaution behaviors in hospital settings. The tool development and testing has been reported elsewhere.³³ Though in brief, no identifiers are collected, and each form can be used to record up to a total of 9 HCW-patient encounters and up 10 standard precaution indications per encounter. The 10 indications represented categories of hand hygiene, PPE, needles or other sharps, or soiled linen and are measured dichotomously (action taken or missed).⁵ Site liaisons were trained by the research team and after 100% inter-rater reliability was established with each site data on HCW patient interactions was collected. Data were aggregated to create a unit level average (an index for unit level standard precaution adherence).

HCW outcomes

Blood-borne pathogen exposure via sharps injuries and/or mucotaneous exposures were reported by hospitals using OSHA 300⁸ surveillance methodology and definitions on a modified form to collect unit level data and 12 months of data were collected to create an average rate per outcome per unit. These outcomes were expressed as: (1) nurse and all staff sharps injury incidence rate/100 RN FTE, and (2) nurse and all staff mucotaneous exposure incidence rate/100 RN FTE.

HAI outcomes

Existing HAI data utilizing CDC NHSN definitions for: (1) central line associated bacteremia (CLABSI), (2) catheter associated urinary tract infections (CAUTI), and (3) hospital-onset methicillin-resistant Staphylococcus aureus (MRSA) bacteremia.^34,36 were collated and submitted by each site. These outcome data were expressed as rates following numerator definitions and corresponding location specific denominator data consisting of device days (CAUTI and CLABSI) or patient days (MRSA). Twelve months of data (pre-and post survey) were collected to create an average rate per outcome per unit. Most states in the U.S. require hospitals to report HAI rates through NHSN for public reporting purposes, and the data are subject to validation procedures to check the accuracy and quality of the data.³⁴ Therefore, we anticipated high data quality and minimal burden as the sites selected collect this information as part of their routine infection prevention and control department activities.

Potential confounders

The analytic models included several important hospital and provider characteristics identified a priori as potential confounders: (1) hospitals without post graduate medical residents or fellows (non-teaching) distinguished from teaching hospitals; (2) bed size stratified as <100 beds, 101–250 beds, and >251 beds; (3) hospitals categorized into 1 of 5 geographic categories based on U.S. Rural-Urban Continuum Codes of the county where the hospital is located; (4) hospital Magnet status (Yes/No, defined as Magnet designation through American Nurse Credentialing Center); (5) nurse hours per patient day (the number of productive hours worked by all nursing staff with direct patient care responsibilities divided by in-patient days); and (6) nurse staffing (defined as occupied RN full time equivalent).^18,29,37,38

Data collection procedures

Following Institutional Review Board approvals all data were collected between January 2017 and October 2018. Study site liaisons included infection preventionists, occupational health nurses, and clinical nurses who were incentivized to participate in the study (entered into a professional conference registration raffle) and were trained by the research team to collect and collate outcomes and observational data. Unit participation was designed to be initiated in cohorts of up to 10 units every 2 weeks to allow for early identification of any issues and related adjustments in accordance with National Occupational Research Agenda Research to Practice guidance.¹¹ Administration of the survey was initiated in an approximate 2–4 month time period either preceding or following the observation data collection so that it was contemporaneous, not simultaneous with the observational data to minimize any potential bias. Following administrative permission from each unit management the electronic survey or pen and paper survey (depending on site preference) was administered over 6–8 weeks with reminders sent every 2 weeks.

Analytic approach

All hospital unit data were aggregated, and standard descriptive statistics and techniques were employed to gain familiarity with the distributions and frequencies. Bivariate analyses were conducted to describe relationships among the key variables and include correlations as appropriate. Pearson correlation coefficients were computed to assess the relationships between unit observed adherence and reported adherence, patient safety climate and standard precaution adherence, and standard precaution adherence and HCW or HAI outcomes (Aims 1 and 2). Multivariable regression models were then conducted, following model assumption verifications, at the unit level of analysis to assess the relationship among the group of standard precaution adherence, patient safety climate, and potential confounders as determined by a priori selection and prior procedures (such as Magnet status, teaching status, licensed bed size, nurse staffing) and each HAI or HCW outcome (Aim 3). To account for hierarchal data structure all data were aggregated at the unit level to yield consistent effect estimates and standard errors. Models were also run with robust standard error procedures to address any potential issues of heteroscedasticity, which would make it likely for the model to identify statistical significance when it does not exist. We estimated we would require a sample of 87 units to detect these relationships if they existed. Finally, we also ran post-hoc power analyses on the models as sample size was a concern. Data was analyzed using STATA/MP13.1 (StataCorp.).

RESULTS

In total, 2,139 health care worker patient encounters that included 6,518 standard precaution indications were observed, and 500 surveys were collected from nurses on 54 units in 15 hospitals from 6 states. Excluding sites that did not submit all 3 types of data yielded a total of 5,285 standard precaution observations and 452 surveys collected across 43 units in 13 hospitals from 6 states used for analyses.

Descriptive results

Demographic distributions of hospitals and nurses are shown in Tables 1 and 2. The majority of standard precaution observations included nurses (43.1%) (Fig 1A), and the most frequent indication observed was hand hygiene (72.6%). Overall observed standard precaution adherence at the individual level was 64.4%. Overall adherence for nurses was highest (69.1%), followed by the other provider category (62.1%), and lastly physicians (58.4%). As shown in Figure 1B, in descending order, adherence rates were: PPE (81.8%), sharps handling (80.9%), linen handing (68.3%) and hand hygiene (58.3%).

Table 1.

Characteristic of units (n = 43)

Characteristic	n	%
Magnet Designated
Yes	28	65.1
No	15	34.9
Teaching Status
Teaching	28	65.1
Non-teaching	15	34.9
Hospital Ownership
Private	36	83.7
Public	7	16.3
Hospital Bed Size
Large (>400)	29	67.4
Medium (216–400)	7	16.3
Small (1–215)	7	16.3

Bold face indicates the characteristic category.

Table 2.

Characteristics of nurses (n = 452)

Characteristic	n*	%
Years in current profession
0–5	224	49.6
6–10	81	17.9
≥11	138	30.5
Years worked in current hospital
0–5	242	53.5
6–10	68	15.0
≥11	131	29.0
Primary work unit
Combined Medical/Surgical	279	61.7
Medicine	48	10.6
Surgery	15	3.3
Pediatrics	17	3.8
Other	72	15.9
Many different units/No specific unit	13	2.9
Years worked on current unit
0–5	313	69.2
6–10	43	9.5
≥11	86	19.0
Hours worked per week
≥40 (Full-time)	417	92.3
16–39 (Part-time)	25	5.5

Bold face indicates the characteristic category.

^*Numbers may not total 452 due to missing data

Fig 1.

Standard Precautions Adherence Summary.

When aggregated for unit level analyses, overall observed adherence was 62.6%; adherence for nurses was highest (69.1%), followed by the other provider category (56.7%), and lastly physicians (46.1%). In descending order, adherence rates were: PPE (81.1%), sharps handling (63.2%), linen handing (46.3%) and hand hygiene (56.4%). A one-way ANOVA was conducted to compare the effect of provider role on observed standard precaution adherence, and significant differences were identified (P < .001). Differences were identified between nurses and physicians (P < .001) and nurses and others (P = .01), but not physicians and others (P = .08).

The average unit response rate for survey completion was 38.7%. The distribution of perceptions of patient safety climate dimensions is shown in Table 3. The majority of nurses surveyed (95.8%) reported they often or always perform the 14 precaution behaviors included in the survey and (77.3%) rated their unit environment positively, or conducive to following standard precautions.

Table 3.

Associations among patient safety climate dimensions and reported standard precaution adherence (N = 43)

Dimension	Mean (SD)	r2	P value
Reported Standard Precaution Practice	.96 (.04)
Teamwork Within Units	.85 (.12)	.113	.47
Organizational Learning − Continuous Improvement	.79 (.15)	.522	<.001*
Supervisor/Manager Expectations & Actions Promoting Patient Safety	.78 (.17)	.386	.01*
Standard Precaution Environment	.77 (.12)	.435	<.001*
Feedback & Communication About Error	.74 (.18)	.504	<.001*
Frequency of Events Reported	.73 (.17)	.513	<.001*
Communication Openness	.67 (.18)	.321	.04*
Management Support for Patient Safety	.63 (.18)	.402	.01*
Teamwork Across Units	.57 (.17)	.364	.02*
Overall Perceptions of Patient Safety	.56 (.18)	.333	.03*
Nonpunitive Response to Errors	.48 (.22)	.070	.66
Handoffs & Transitions	.46 (.16)	.334	.03*
Staffing	.43 (.18)	.261	.09*
Composite Safety Score	.64 (.14)	.442	<.001*

Notes: Patient safety climate measured as composite frequency scores of positive responses (rated 4 or 5). Composite safety score excludes reported adherence.

^*= statistically significant at P <.05

Regarding outcomes, unit HAI mean rates in descending order were: 0.76 CAUTIs per 1000 device days (SD = 0.76); 0.69 CLABSIs per 1000 device days (SD = 1.22), and 0.04 MRSA infections per 1000 patient days (M = 0.04, SD = 0.08). Unit needlestick injury rate for all staff was 12.54 per 100 RN FTE per year (SD = 24.95), and specific to nurses was 5.35 per RN FTE per year (SD = 5.34); mucotaneous blood or other potentially infectious material rate for all staff was 2.30 per 100 RN FTE per year (SD = 5.18) and nurses only was 0.77 per 100 RN FTE per year (SD = 1.60).

Aim 1 results: Association between patient safety culture and standard practice adherence

Analysis of the relationship between patient safety climate in aggregate and reported standard precaution adherence revealed a positive correlation (P < .01). Each dimension, except Teamwork Within Units and Non-Punitive Response to Errors, was independently correlated with reported standard precaution adherence (P < .05). There was also a correlation between positive (agree or strongly agree) perceptions of a work environment that is conducive to the conduct of standard precautions and reported adherence (P < .001). Details are shown in Table 3 The correlation between unit observed adherence and reported adherence was not significant (P = .879). The relationship between positive perceptions of patient safety climate (ratings of 4 and 5) and observed standard precaution adherence was not significant in aggregate. In summary, patient safety climate was correlated with reported standard precaution adherence.

Aim 2 results: Relationship between standard precaution adherence and HAIs or HCW exposures

Analysis of the relationship between reported standard precaution adherence and HCW exposures or HAI events revealed that there were no significant correlations. Similarly, there was no significant association between the aggregate of overall observed standard precaution adherence and HCW exposures or HAI events.

Aim 3 results: Relationships among patient safety climate, observed standard precaution adherence, and HAIs or HCW exposures

Regarding HAIs multivariable models demonstrated the group of variables (standard precaution adherence, aggregate patient safety climate and potential confounders) reliably predicted CAUTI (P = .02) and MRSA (P = .03). Regarding HCW outcomes the group of variables reliably predicted nurse mucotaneous exposures (P = .004), all staff mucotaneous exposures (P = .007), and all staff sharps and needlestick injuries (P = .001). In these models, the covariates that independently added significantly to the prediction of the outcome of interest were Magnet status, nurse staffing, hospital ownership and teaching status. Patient safety climate and observed standard precaution adherence were not significant independent predictors in these models. Details are shown in Table 4. Finally, as we may have been underpowered to identify a relationship between the primary predictors and outcomes, if indeed they did exist, we ran post-hoc power analyses given the sample size of 43 units, number of predictors in each model, observed R² and P <.05. We found the observed statistical power for CAUTI was 0.74, CLABSI was 0.88, MRSA was 0.99, nurse needlestick/sharps injury 1.00, nurse mucotaneous exposures was 0.80, all needlestick/sharps injury 0.88 and all mucotaneous exposures was 0.99. In summary, important relationships among patient safety climate, standard precaution adherence, key covariates and HAI or occupational outcomes were identified.

Table 4.

Multivariable regression models of predictors of unit HAIs and occupational exposures (N = 43)

	HAI outcomes
CAUTI	Omnibus P = .023*, R2 = .233
Predictors	β Coefficient	SE	P value
Observed SP Adherence	−.120	.014	.641
Patient Safety Climate	.009	1.19	.952
Magnet Designated Hospital	.082	.336	.607
Teaching Status	.282	.314	.067
Nurse Staffing	.356	.008	.003*
CLABSI	Omnibus P = .357, R2 = .278
Observed SP Adherence	.097	.007	.406
Patient Safety Climate	.194	1.48	.235
Magnet Designated Hospital	−.277	.442	.121
Hospital Ownership	−.419	.812	.101
MRSA	Omnibus P = .034 * , R2 = .412
Observed SP Adherence	.042	.000	.727
Patient Safety Climate	.077	.070	.498
Teaching Status	.201	.017	.058
Nurse Staffing	.555	.001	.030*
	Occupational exposures
Nurse needlestick/sharps injury	Omnibus P = .345, R2 = .082
Predictors	β Coefficient	SE	P value
Observed SP Adherence	−.103	.048	.541
Patient Safety Climate	.211	5.56	.133
Teaching Status	.240	1.83	.154
Nurse Staffing	−.000	.069	.999
Nurse mucotaneous exposures	Omnibus P = .004 * , R2 = .362
Observed SP Adherence	−.167	.017	.401
Patient Safety Climate	−.084	1.22	.406
Magnet Designated Status	−.441	.692	.041*
Licensed Hospital Bed Size	.371	.001	.055
Teaching Status	.258	.333	.014*
All needlestick/sharps injuries	Omnibus P = .001 * , R2 = .378
Observed SP Adherence	.266	.198	.074
Patient Safety Climate	.262	28.6	.091
Hospital Ownership	−.577	15.25	.016*
Average Daily Census	−.041	.593	.813
All mucotaneous exposures	Omnibus P = .007 * , R2 = .431
Observed SP Adherence	.098	.032	.394
Patient Safety Climate	.217	6.31	.184
Average Daily Census	−.050	.105	.733
Hospital Ownership	−.440	2.82	.037*
Magnet Designated Hospital	−.414	1.43	.004*

SP, standard precautions; CAUTI, Catheter-associated urinary tract infection; CLABSI, Central line-associated bloodstream infection; MRSA, Methicillin-resistant Staphylococcus aureus. Nurse staffing defined as occupied RN full time equivalent.

^*= statistically significant p < .05., Robust regression approach with robust standard errors (SE). standardized Beta coefficients reported.

DISCUSSION

The overarching aim of this project was to determine if hospital units that had a stronger patient safety climate also demonstrated higher levels of adherence to standard precaution practices, and in turn lower HAI and occupational infectious exposure outcomes. This project produced the following key findings.

First, to our knowledge this study is the first to document that features of a stronger patient safety climate were correlated with higher levels of self-reported standard precaution behaviors. Second, observed adherence to standard precautions, as a set of practices, was low across all professions and disciplines. This was discordant with high levels of reported adherence on the same unit but consistent with the respondents’ reports of an unfavorable practice environment to perform standard precautions. Hand hygiene adherence was particularly poor across all roles. However, within the remaining standard precaution categories interesting differences in performance by role were noted. Physicians had lower levels of adherence using PPE when indicated as compared to both nurses and other providers. This is concerning as physicians routinely provide hands-on evaluations of patients, which might warrant use of PPE, if by example examining wounds, lesions, or respiratory mucosa. The “other” category of providers, largely comprised of nursing assistants, did not handle contaminated linens safely compared to nurses, though adherence to sharps safety was higher for the “other” category than nurses. By nature of their professional work and scope of practice, nurses handle sharps more often than nursing aides, and aides handle soiled linens more often than nurses. It might be that with some action’s providers have higher levels of comfort and confidence and perhaps move more quickly and are less careful and meticulous than with actions they are less experienced performing.

These findings provide valuable insight into role specific risks and opportunities to improve standard precaution adherence using focused prevention interventions. This may also suggest an entrenched and role-based normalization of deviance; HCW are socialized into the “way we do things here”.³⁹ Individual HCW should be held accountable for their professional actions, but not in isolation from the system and structural factors that facilitate or impede standard precaution adherence, such as infrastructure and resources, work-flow and work-force, and effective training and enforcement.^40,41 Organizations might leverage these role differences and encourage the high-reliability principles to “defer to the expertise of others,” those who are performing well consistently. For example, how might the nursing aide prompt the physician to use PPE when indicated, or the nurse guide the aide to handle contaminated linens without further environmental contamination? Team training and enhanced communication skills will be essential to coalesce our care for patients and fellow HCWs, and developers of these training modalities would be well served to take into consideration role-based hierarchies and needs of educationally disadvantaged HCWs.

Third, the multivariable models identified for the first time that in combination a stronger patient safety climate, better standard precaution adherence (as measured by observation), and key hospital characteristics (such as nurse staffing, daily census, teaching status) predict key HAI and occupational health outcomes. These models explain 41% of the variance in MRSA, 23% of the variance in CAUTI, 43% of the variance in all staff mucotaneous exposures, and 38% of the variance in all staff needlestick and sharps injuries. Finally, potentially modifiable variables of nurse staffing and hospital Magnet designation explained substantial variance in the multivariable models for outcomes of MRSA, CAUTI, nurse, and all staff mucotaneous exposures

Emerging evidence has identified a relationship between nurse staffing and HAIs, including bloodstream infections, pneumonia, and urinary tract infections (with and without a catheter).^42,43 Our study confirms these findings and extends our knowledge by identifying that unit level nurse staffing predicts unit level CAUTI and MRSA rates, independent of patient safety climate and other organizational factors. Literature has also documented that Magnet facilities have better patient outcomes, including lower incidence of HAIs (CLABSI, CAUTI, and MRSA), length of stay, and mortality and reported benefits of increased nurse satisfaction and retention and decreased staff turnover.^38,44–46

Magnet status characterizes and includes nurse participation in hospital affairs, nursing foundations for quality care; nurse manager ability; leadership; support of nurses, staffing and resource adequacy; and collegial nurse-physician relations and is measured in part through the nurse practice environment.⁴⁶ Thus, the nurse practice environment captures distinct, but similar, constructs to the dimensions of the safety climate.⁴⁶ In this study, hospital Magnet status may be considered a proxy measure for nursing practice environment.

Unfortunately, there is a dearth of literature that examines the organizational impact, operationalized by Magnet designation, on needle stick injury and mucotaneous exposures in HCW. Moreover, sentinel studies, while informative, are limited largely by decades old publication dates, occur prior to the national strengthening of voluntary adverse event reporting systems, or include self-reported incidence of needle stick injuries^36,47 In fact, a recent systematic review that examined the impact of hospital Magnet status on patient, nurse, and organizational outcomes identified only 1 article that included needle stick exposures, and this was a business case assessment for small hospitals.⁴⁵

This is the first study to our knowledge to document the impact of Magnet designation status on unit level nurse and staff mucotaneous exposure rates. While our study did not identify Magnet status as an independent predictor of HAIs, we generated new evidence of the relationship of Magnet designation and important occupational health outcomes. When these results are considered in context of extant literature, it appears both patients and HCWs benefit in terms of outcomes when seeking care or working in a Magnet designated organization.

2020 marked the 20th anniversary of the Needlestick Safety and Prevention Act. Unfortunately, our findings reveal there has been little progress in improvement, and dishearteningly, this issue has garnered little attention in occupational and health services research. Moreover, the focus of published work is largely percutaneous, not mucotaneous exposures, which is concerning as estimates suggest only 12% of mucotaneous exposures are reported.⁴⁸ Findings from this study amplify the recently published Moving the Sharps Safety in Health care Agenda Forward in the United States: 2020 Consensus Statement and Call to Action, which declares the risk of occupational exposure is greater today than at the time of the initial report and calls to redouble our efforts.⁴⁸

Limitations

This is a cross-sectional study and as such, though it was possible to show significant relationships among several key variables, causality cannot be established. While our models identified important predictors and explained substantial variance in outcomes, we were limited by sample size on the number of predictors we could include, and by design did not include all possible important factors for each outcome. Despite the post-hoc power analyses findings for the multivariable models, it is possible this study was underpowered with 43 units (rather than the aim of 87) to detect additional meaningful relationships if they existed.

Survey data were only collected from nurses; therefore, we do not know if self-reported and observed adherence data is better aligned for other provider roles. Despite careful planning and observer selection and training, the possibility of a Hawthorne effect exists; therefore, actual adherence may be even lower than we report. It is possible that our sample includes hospitals with higher safety awareness and participation in improvement initiatives to decrease HCW and HAI adverse outcomes. Finally, reliance on secondary data from hospital departments may also be a limitation, though using established metrics and definitions may have limited any inconsistencies in data.

CONCLUSIONS

This study produced 4 major findings that advance the state of the science in patient and occupational health safety. First, patient safety climate is correlated with higher levels of both self-reported and observed standard precaution behaviors. Second, observed adherence to standard precautions is low across all professions and disciplines, and key role distinctions are noted. Third, we identify, and document specific standard precaution actions associated with HAI and occupational pathogen exposures. Finally, by employing innovative methods, strategies, and approaches to collect and analyze multiple complex sources of data, we identified direct and indirect relationships among patient safety climate, observed and reported standard precaution adherence, and HCW and HAI outcomes. In combination, these results indicate that a stronger patient safety climate, better standard precaution adherence, and key organizational characteristics, predict key HAI and occupational health outcomes.

The translation of these findings at the point of care can be realized through targeted interventions and cross-cutting surveillance methodology that captures risks at the intersection of patient and occupational health and safety. Leaders can identify and implement prevention strategies based on local surveillance data and other organizational information. These breakthroughs also provide future orientation for policy makers to support research that further clarifies and disentangles the factors that meaningfully contribute to high-reliability organizations and positive patient and HCW outcomes. In so doing, this study contributes to stopping the transmission of infectious diseases in health care and social assistance industry sector settings among workers, patients, and visitors; promoting safe and healthy workplaces and optimizing safety culture in health care organizations; and reducing sharps and mucotaneous injuries and their impacts among all health care personnel.