|Home | About | Journals | Submit | Contact Us | Français|
Standard precaution (SP) adherence is universally suboptimal, despite being a core component of healthcare-associated infection (HCAI) prevention and healthcare worker (HCW) safety. Emerging evidence suggests that patient safety climate (PSC) factors may improve HCW behaviours. Our aim was to examine the relationship between PSC and SP adherence by HCWs in acute care hospitals. A systematic review was conducted as guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis. Three electronic databases were comprehensively searched for literature published or available in English between 2000 and 2014. Seven of 888 articles identified were eligible for final inclusion in the review. Two reviewers independently assessed study quality using a validated quality tool. The seven articles were assigned quality scores ranging from 7 to 10 of 10 possible points. Five measured all aspects of SP and two solely measured needlestick and sharps handling. Three included a secondary outcome of HCW exposure; none included HCAIs. All reported a statistically significant relationship between better PSC and greater SP adherence and used data from self-report surveys including validated PSC measures or measures of management support and leadership. Although limited in number, studies were of high quality and confirmed that PSC and SP adherence were correlated, suggesting that efforts to improve PSC may enhance adherence to a core component of HCAI prevention and HCW safety. More clearly evident is the need for additional high-quality research.
Healthcare-associated infections (HCAIs) – largely preventable adverse events – are a global patient safety problem.1 Over the past decade literature continues to conclude that HCAIs are frequent, catastrophic, and costly.2–6 Despite estimates that 10–70% of HCAIs are preventable, the burden is staggering in developed and developing nations.1,7 In the USA 5–10% of acute care patients acquire one or more HCAIs; in lives directly affected this indicates that approximately two million US patients suffer an HCAI, resulting in an estimated 99,000 deaths annually.1,4,5,8 In European countries these statistics are similar with prevalence estimates of 6%, or 3.2 million patients per year with at least one HCAI.9,10 On any given day 80,000, or one in 18, patients in a European hospital have at least one HCAI, resulting in an estimated 37,000 attributable deaths annually.9,10 The estimates are more striking in developing countries where pooled prevalence estimates range between 10.1% and 15.5%.11,12 HCAI densities in intensive care units are up to three times greater than in developed countries at 47.9 per 1000 patient-days, with excess mortality attributed to HCAI at 18.5–29.3%.11,12 Moreover, the annual attributable direct costs of HCAI are $9.8 billion in the USA and €7 billion in Europe, and are estimated to be high also in developing countries.12–14 In sum, the prevention of HCAI is of significant and current importance, affecting all healthcare consumers with real direct and indirect consequences.2,7,9
Nearly 30 years ago the US Centers for Disease Control and Prevention (CDC) introduced universal precautions as a core component of HCAI prevention, deemed applicable to all healthcare workers (HCWs) in contact with all patients in all settings, regardless of the suspected or confirmed presence of an infectious agent.15 In 1996, CDC universal precautions guidelines were updated and termed ‘standard precautions’ (SPs). Specific components of SP include hand hygiene, use of appropriate personal protective equipment (PPE), safe use and disposal of sharps, decontamination of environment and equipment, patient placement and linen and waste management.16 These standards have been adopted internationally by European and other countries and are considered a common national-level guideline, in contradistinction to recommendations for prevention of specific types of HCAI, such as targeted device-related prevention bundles, surgical site infection procedures, or pharmacologic measures.13,16 The World Health Organization has declared it imperative that standard precautions be established prior to implementation of any specific measure or practice ‘bundle’ or targeted intervention.12 Thus, a longstanding and broad-reaching approach and primary strategy to prevent HCAI is adherence to SPs by HCWs.16
Over a decade of literature has demonstrated that HCW adherence to basic preventive practices such as SPs remains suboptimal, adhered to less than 50% of the time.17–19 A body of literature has also explored the relationships among individual-level factors such as intent, knowledge, attitudes, and experience and adherence to components of SP.20–24 Mixed findings from these studies demonstrate the complex and multidimensional nature of infection prevention behaviours, suggesting that important antecedents to SP adherence may also include organizational level characteristics in which the HCW performs.
Over a decade ago the Institute of Medicine’s To err is human landmark report recognized the importance of the safety culture of healthcare organizations in improved provider performance and adverse event reduction, and implored organizations to create a safety culture.25 Safety culture is considered broadly the managerial and HCW attitudes and values as they relate to the perception of risk and safety. Teamwork, leadership support, communication, non-punitive response to errors, perception of organizational commitment, work design, staffing and workload, resources, and emphasis on quality have been identified as important and common attributes of a positive safety culture in the literature.26–28 Patient safety climate (PSC), a related concept, has also been identified as an important antecedent of HCW behaviour.29–31 Although the terminology overlaps in the literature, one conceptual distinction is that safety culture is described as the overarching values, norms, and assumptions of the organization that drive the quality of care, and that safety climate is the collective reflection of the perception, attitudes, and shared experiences of the culture.26,32 Succinctly, safety climate comprises the group-level experiences of the overarching organization-level culture of safety.
Several studies have demonstrated that safety climate factors are a significant predictor of safe work behaviours. Findings by DeJoy et al. indicate that a positive PSC may facilitate the creation of a work environment that will enable, support, and reinforce HCWs to comply with safe practices.33 These findings are supported in a review that examined the relationship between PSC and nurses’ health and safety behaviours and outcomes.34 Similarly, Gershon et al. demonstrated that SP compliance was strongly correlated with organizational commitment to safety.21 Advancing this knowledge, DeJoy et al. found that a negative safety climate was the strongest predictor of job hindrances, which in turn were the strongest predicators of lower SP adherence.35 Most recently, support for the PSC antecedent of SP adherence was demonstrated by Nichol et al., and also by Brevidelli and Cianciarullo who identified that factors of management support for ‘safe work practices’ and ‘safety performance feedback’ were correlated with SP adherence.36,37
Despite the significant burden of HCAI, persistent evidence of suboptimal SP adherence, and the growing body of evidence of the importance of PSC factors to HCW behaviours such as SP adherence, there has been no systematic review specifically examining the relationship between PSC and SP adherence. A systematic review is required to summarize this research evidence, thereby accelerating the translation of evidence into practice and guiding future research as appropriate. There is an urgent need to systematically identify and appraise this body of research and information, synthesize the results, and provide an assessment of the evidence that may support decision-making and guide allocation of scarce resources. To address this gap a systematic review was conducted to identify, critically review and synthesize literature regarding the evidence of a relationship between PSC and SP adherence in acute care hospitals.
This systematic review addressed the question: ‘What is the relationship between PSC and SP adherence in healthcare professionals working in acute care hospital settings?’ The analytic framework is presented in Figure 1. The primary outcome of interest is adherence to SP protocols; the secondary outcome of interest is the occurrence of HCAI and HCW exposures. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement and 27-item checklist guided this review.38
All eligible studies had to meet the following inclusion criteria: a quantitative study that examines the relationship between PSC dimensions and adherence to components of SP by HCWs in acute care hospital settings, published between January 2000 and September 2014, and available in English language. This 14-year time frame was selected as it encompasses the recent literature following the 1999 Institute of Medicine landmark report and the subsequent focus of safety culture and climate in healthcare settings.25 HCWs must include direct care providers who work in the setting to represent those whose behaviours have an impact on patient outcomes and who are adequately assimilated in the environment to rate the PSC. Studies solely including students or trainees or laboratory workers are excluded. Studies that included outcomes of HCAI and HCW exposure injury were of particular interest as a secondary outcome. The study had to explicitly use a PSC scale or measure leadership, peer support, or teamwork to more fully capture the breadth of the PSC construct in contradistinction to articles that focused solely on need for resources, patient equipment availability, education, or training. Quality improvement and studies that included an intervention such as education were excluded to isolate the specific relationship of interest between PSC and SP adherence. Whereas these studies indicated an organizational interest, the focus of this review was the relationship between measured features of the PSC and SP, not the intervention effect. Therefore these studies were excluded to allow for a more rigorous and focused review addressing the study question. Qualitative studies, reviews, letters and articles that did not report primary data were excluded, as were studies that were not conducted in acute care hospitals.
Three databases explicitly selected that include primary biomedical research were searched: PubMed, CINAHL, and Embase. The search strategy was developed following consultation with an information specialist in an iterative process. Initially, titles and abstracts were searched for key terminology. Search terms included Medical Subject Headings (MeSH), related text word, entry terms or major headings as appropriate for each database structure (Appendix A). Search terminology was broad and specifically selected to capture the evolving terminology in fields of infection prevention and patient safety. These included ‘universal precautions’, and ‘universal precaution’ (this was selected as there is no MeSH term for standard precautions – the terminology recommend by the CDC in 1996), and ‘safety climate’, ‘patient safety culture’, ‘patient safety’ (introduced in MeSH in 2012). Search terminology from both safety culture and climate literature were included to ensure an exhaustive review, though the inclusion criteria for abstract selection and final review was that articles include the PSC factors of HCW perceptions, attitudes, or shared experiences of the organizational features of the work environment. That is, the key discerning factor for inclusion was the HCW view of elements of the safety culture. Additionally, several terms to capture the population of all HCWs were selected including ‘healthcare worker’, ‘healthcare professional,’ ‘nurse,’ and ‘physician’. The terminology search yielded a total of 2,147,731 titles and abstracts that contained any one of the population, exposure, or outcome terms of interest (Appendix A). The search strategy was then applied to yield only articles that included all three (population, exposure and outcomes) components. This search yield was then de-duplicated and abstracts and titles were screened for inclusion and exclusion criteria. Disagreements were resolved by consensus (A.H. and E.L.L.).
A summary of select study characteristics was created and completed (Table I). Two reviewers (A.H. and E.L.L.) independently assessed study quality using a modified Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) tool.45 This 22-item validated measure guides the reviewer to assess the quality of the study title and abstract, introductions, methods, results, discussion and other information. This was modified to retain 10 items in four categories: introduction (one item); methods (five items); results (two items); and discussion (two items) (Table II). Items relevant to this review included: background and rationale, setting, variables, data sources, measurement, statistical methods, main results and results summary, discussion strengths and limitations, and interpretation. Each item was scored one point if sufficiently reported; thus each study was assigned a score ranging from 0 to 10 by each reviewer. Inter-rater reliability was then established by discussion of the scoring assigned. Quality scores were considered in agreement if they were within two points of each other based on prior use of the tool. In this review, the two raters reached consensus as all studies were scored within one point of each other.
The search yielded 888 articles; of these, 22 articles were duplicates and excluded, yielding 866 that were screened for eligibility. Following title screening and abstract review 855 were excluded, yielding 11 articles retrieved for full review and data extraction. The main reasons for excluding studies was precaution behaviours not related to patient care or outcomes of interest (609), study design (150), study setting (43), or non-direct care provider subjects or providers in training and not considered a part of the climate (42) (e.g. laboratory workers, nursing students, dentists, rural settings, tobacco exposure, handling of chemotoxic agents). During full review four articles were excluded from data extraction as they did not meet the inclusion criteria. One was a quality improvement intervention study two did not include SP behaviour, and one described SP adherence and features of the organizational climate but did not measure the relationship between the variables.46–49 Seven articles were included in the final review (Figure 2).
All seven studies were cross-sectional surveys conducted in acute care hospitals. Sample sizes ranged from 266 to 2287. All studies included nurses; two included nurses only, and five included physicians, technicians, laboratory workers and other HCWs with direct patient or specimen contact. Five studies were conducted in geographically distinct regions of the USA, one was conducted in an urban location in Brazil and one in rural north India. Two were single site studies, the remaining five were multi-site. Multi-site studies were conducted at three, seven, 22, or 84 hospitals. Hospital settings were heterogeneous, and included large (~1000 beds) and small (<100 beds), teaching and non-teaching hospitals, research hospitals, those with and without specialty certification such as The Joint Commission Accreditation or Magnet Designation, and private and public hospitals.
All of the studies reported statistically significant results in support of a relationship between features of the PSC and SP adherence (Table I). Overall, better PSC was related to greater SP adherence in heterogeneous hospital settings. Five of the studies measured items in all categories of SP adherence; these studies, however, analytically examined SP adherence as a global construct and did not discern between types of behaviours. The remaining two studies measured a single component of SP adherence: one measured the use of gloves when performing needlestick procedures and the other measured recapping needles when using sharps equipment. There was no standardized measure of PSC used in all studies, although three studies did employ an adaptation of the Gershon tool.21 All studies minimally measured features of management support and leadership as specified in the inclusion criteria. Other factors measured included job demands and feedback, equipment availability, commitment to safety and training, physical environment, transmission knowledge, risk-taking personality and non-punitive environment.29,35,37,41,42,44,48 Three of the seven studies examined the relationship between secondary outcomes of HCW (occupational) outcomes. Of note, one study examined HCWs’ prior exposure to blood and body fluids as a reinforcing factor or independent variable as opposed to an outcome. No study included patients’ HCAI outcomes.
The scored quality of the studies ranged from 7 to 10 of 10 possible points; four were at or above the median of nine (Table I). Because the STROBE scoring system assigns the same weight to each criterion, the specific reason points were not assigned for each of the seven studies, detailed by category and item as follows. All studies sufficiently reported the introduction section as rated by the background/rationale; the results section, as rated by the adequacy of reporting main results for each specified objective/aim in the introduction; and the discussion section, as rated by sufficiently reporting the interpretation of the results. Studies by Clarke et al.41 and DeJoy et al. were not assigned a setting point for sufficiently describing the setting locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data collection.35,41 Studies by Brevidelli and Cianciarullo and by DeJoy et al. were not assigned a data sources point for sufficiently reporting sources of data and details for each variable of interest.35,37 Clarke et al.’s study was not assigned a point for measurement, indicating that methods of assessment (measurement) and description of measure validation were insufficiently reported.41 Clarke et al.’s study was not assigned a point for statistical methods, including a description of methods to control for confounding, confounders included, and rationale.41 Brevidelli and Cianciarullo’s study was not assigned a results point for clearly summarizing and presenting results with appropriate graphics.37 Finally, Anderson et al.39 and Dejoy et al.’s studies were not assigned a discussion point for strengths and limitations including discussion of any potential bias, and direction and magnitude of any bias, and external validity.35,37,39
CDC guidelines exist for HCWs to adhere to SPs in all settings, for all patients, all the time, as a primary strategy to prevent HCAI and HCW injuries and exposures. However, these guidelines are not embedded in practice, and reports of suboptimal adherence persist in the literature. An emerging body of evidence suggests that PSC factors are important to improve HCW behaviours. This focused systematic review was conducted to elucidate the relationship between PSC and SP adherence in healthcare professionals working in acute care hospital settings. Fourteen years of research was synthesized and important gaps in our knowledge were identified. These gaps are perhaps unsurprising given the dearth of research that has been conducted on this topic. Of the nearly 1000 articles identified through an exhaustive search, fewer than ten studies have been published that aim to address this important topic in whole or in part. Notably, no studies contained HCAI outcomes which would justify efforts to link PSC with SPs and patient outcomes.
The interest in PSC is evident; PSC has been identified as an important antecedent of HCW behaviour in general, and appears extensively in the infection prevention literature, typically in context of secondary strategies such as ‘bundled’ or targeted interventions to prevent device-related or surgical site infections.26–28,50,51 This was exemplified in our systematic review; the search strategy yielded 175,185 articles on PSC, more than three times the number (53,369) published that included SPs. Hence, an important gap identified in this review is the examination of the relationship between shared core values (PSC) and core practices (SPs).
Although limited in number, the studies identified were of high quality and therefore some critical insight into these relationships can be gained from this review, specifically the consistent correlation reported between better PSC and greater SP adherence in heterogeneous hospital settings. However, these studies largely focused on nurses, used a variety of self-report measures, and a limited number included HCW outcomes. Notably, none included patient outcomes; therefore the potential preventability of HCAIs by adherence to SPs, or the value of SPs, is not evident. Additionally, the majority of studies measured SPs as a global construct even though there may be different antecedent barriers and facilitators for different behavioural actions of SPs. For example, equipment availability may not be a barrier to performing hand hygiene but is a barrier to appropriate gown and glove use.
Identifying the modifiable features of the PSC that improve practices such as SPs will likely support the provision of more efficient and effective patient care. Notwithstanding the strengths of the identified studies, the gaps revealed by the lack of studies identified in this systematic review in and of itself points to the need for standardized, psychometrically sound measures and for more research that specifically examines the relationship between PSC and SP in acute care settings.
This study used three databases: PubMed, CINAHL, and EMBASE, and focused on terms used to capture complex concepts and practices such as PSC and SP. Searching additional databases or using additional search terms may have identified more publications. The search strategy, however, was comprehensive and identified more than two million articles that included a population, exposure, or outcome of interest, and 866 that included all concepts of interest. The criteria that the article be written or available in the English language may also have led to omissions of studies published in other languages, particularly since studies that were screened and included were international. Additionally, PSC and SPs were operationalized differently across studies, making it impossible to conduct a meta-analysis. The final sample of seven studies limits the external validity of the study results. Finally, publication bias is a possible limitation; studies are less likely to be published if the findings are negative. For these reasons, and despite the high quality ratings of all studies, there are limits to our confidence in the correlation between PSC and SPs.
Despite decades of discordance between SP guidelines and practice, little attention has been paid, and subsequently little progress made, to close that gap. The importance of the organizational culture and climate has been suggested as an important variable in patient safety and HCW behaviour for nearly 15 years, but to our knowledge this is the first review to examine the specific relationship between PSC and SP adherence. Although limited in number, the seven studies identified were of high quality and confirmed that PSC and adherence to SP were correlated. Implications of this systematic review for administrators and clinical practice suggest that efforts to improve PSC may enhance adherence to a core component of HCAI prevention and HCW safety. Implications for researchers are indisputably evident; more high-quality studies are needed.
The authors would like to acknowledge L. Falzon, Information Specialist at Columbia University Libraries, for consultation in developing the search strategy and the Columbia University School of Nursing Writing Workshop Initiative members in refining this manuscript.
The corresponding author was supported as a postdoctoral trainee by the National Institute of Nursing Research, National Institutes of Health (Training in Interdisciplinary Research to Prevent Infections, T32 NR013454). The funder of the study had no role in the study design, data collection, analysis, interpretation of data, or writing of the report.
Total articles identified in PubMed, Embase, and CINAHL, N = 888
Articles excluded based on removal of duplicates, N = 22
Abstracts screened for eligibility, N = 866
Articles excluded following title screening and abstract review, N = 855
Full-text articles retrieved for full review, N = 11
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of interest statement