ORIGINAL ARTICLE
Does Employee Safety Matter for Patients Too? Employee Safety Climate and Patient Safety Culture in Health Care David C. Mohr, PhD,*† Jennifer Lipkowitz Eaton, MD, MPH,‡ Kathleen M. McPhaul, PhD, MPH, COHN-S,‡ and Michael J. Hodgson, MD, MPH§ Objective: We examined relationships between employee safety climate and patient safety culture. Because employee safety may be a precondition for the development of patient safety, we hypothesized that employee safety culture would be strongly and positively related to patient safety culture. Methods: An employee safety climate survey was administered in 2010 and assessed employees' views and experiences of safety for employees. The patient safety survey administered in 2011 assessed the safety culture for patients. We performed Pearson correlations and multiple regression analysis to examine the relationships between a composite measure of employee safety with subdimensions of patient safety culture. The regression models controlled for size, geographic characteristics, and teaching affiliation. Analyses were conducted at the group level using data from 132 medical centers. Results: Higher employee safety climate composite scores were positively associated with all 9 patient safety culture measures examined. Standardized multivariate regression coefficients ranged from 0.44 to 0.64. Conclusions: Medical facilities where staff have more positive perceptions of health care workplace safety climate tended to have more positive assessments of patient safety culture. This suggests that patient safety culture and employee safety climate could be mutually reinforcing, such that investments and improvements in one domain positively impacts the other. Further research is needed to better understand the nexus between health care employee and patient safety to generalize and act upon findings. Key Words: patient safety culture, employee safety, health care organizations, Veterans Health Administration (J Patient Saf 2015;00: 00–00)
P
atient safety culture is an increasingly important target of systems-based clinical improvement efforts.1,2 The publication of To Err Is Human brought the scope and cost of preventable medical errors to the forefront of national conversation.3 Research and innovation efforts focused on patient safety, particularly reduction of preventable medical errors to avoid patient morbidity and mortality, have become standard health care management practice since the report.4,5 The role of the larger organizational From the *Center for Healthcare Organization and Implementation Research (CHOIR), VA Boston Healthcare System; and †School of Public Health, Boston University, Boston, Massachusetts; ‡Office of Public Health, Occupational Health, Veterans Health Administration; and §Occupational Safety and Health Administration, Department of Labor, Washington, District of Columbia. Correspondence: David C. Mohr, PhD, VA Boston Healthcare System, 150 South Huntington Ave, Mailcode 152 M, Boston, MA 02130 (e‐mail: [email protected]). The authors disclose no conflict of interest. The work reported here was supported by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service (IIR 08–067). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the U.S. government. This document is not a standard or regulation, and it neither creates new legal obligations nor alters existing obligations created by OSHA standards or the Occupational Safety and Health (OSH) Act. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Patient Saf • Volume 00, Number 00, Month 2015
culture in the success and sustainability of clinical quality and patient safety gains is increasingly recognized.6–8 Furthermore, organizational climate has been shown to influence health care worker perceptions of safety climate and compliance with precautionary safety procedures.9,10 Previous studies have demonstrated relationships between positive hospital safety culture and patient safety outcomes, such as decreased mortality rates, decreased lengths of stay, decreased treatment and medication errors, and increased patient satisfaction.11–15 Similarly, the occupational safety sciences have examined and demonstrated the relationship between employee safety climate and/or culture with occupational accidents.16,17 Integrating these fields and theoretical models widens the lens to examine the effectiveness of patient safety systems in the context of the workplace safety systems. A collaboration between the Joint Commission and the National Institute for Occupational Safety and Health outlines a conceptual model and rationale for considering both the patient safety apparatus and the occupational safety systems, arguing that “principles, methods, and tools for creating safety are the same, regardless of the population where safety is the focus.”18 Furthermore, the same monograph points out that a safety culture that excludes some groups is neither “genuine nor legitimate.” In other words, it is not possible to sustain a strong culture of safety for patients when occupational safety is ignored or deemphasized. For a workplace to be safe for patients, it should also be safe for the employee workforce.19 A potential prerequisite for fostering patient safety culture is addressing occupational health and safety of employees.20 The greater use of management practices focused on promoting safety in health care has been associated with fewer employee injury rates.21 Research suggests a negative association between employee safety climate and safety behaviors, such as compliance and participation, and with safety performance, such as accidents and injuries.22,23 In some research, both employee and patient safety are both seen as end points of effective organizational work systems and structures or as having a reciprocal relation.24 An organizational climate that encourages and rewards safetyrelated behavior can lead to fewer accidents, whereas a climate emphasizing production over safety may lead to more accidents.25 Research, outside of health care settings, suggests a positive relationship between safe employee work practices and customer satisfaction.26 Thus, it may be important to ensure that employees perceive the environment as being safe and supportive before they are able to contribute effectively to patient safety by behaving in ways conducive to patient safety culture, such as reporting errors they or colleagues have made. In this article, we examine facilitylevel employee safety climate and its association with patient safety culture. We hypothesize that employee safety is a leading indicator and should be positively associated with patient safety domains.
METHODS Study Design We conducted analysis using 2 surveys administered to employees at medical centers within the Department of Veterans www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1
J Patient Saf • Volume 00, Number 00, Month 2015
Mohr et al
Affairs, Veterans Health Administration (VA). One survey focused on perceptions of employee safety and was administered a year before a second survey that assessed patient safety culture. Responses to both surveys were confidential and anonymous. The study was approved by the VA Boston Healthcare System Institutional Review Board.
supervisor participation, employee participation, safety support activities, safety support climate, and organizational climate. For our analysis, we excluded organizational commitment because the questions did not focus on safety, but rather general perceptions of job satisfaction or stress among employees. Items were rated on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” Cronbach α for the 6 dimensions ranged from 0.82 to 0.92. Intraclass correlation coefficient (ICC) values were computed to examine support for group-level aggregation. Statistics suggest the similarity among individual responses from a higher level unit. The ICC(1) values ranged from 0.02 to 0.04, and the ICC(2) values, influenced by number of respondents at each site, ranged from 0.65 to 0.82. These values are sufficient for aggregation at the facility level.28 Based on the high Pearson correlations among the 5 employee safety perception measures, which ranged from 0.84 to 0.95 at the facility level, we created a composite scale score by averaging across the 5 dimensions. We treated this new employee safety composite measure as the main predictor variable of interest.
Data Collection and Participants Both surveys were administered to full-time employees in all sections and departments. The VA Employee Safety Perception survey, also referred to as the “Safety Barometer,” developed by the National Safety Council, a nonprofit, congressional charter organization, was administered to employees through the internet. An e-mail from local medical center leaders was sent to employees with a link to complete the survey online. Facility survey coordinators sent weekly reminders along with response rates. The survey was administered during October through November 2010. A total of 37,971 employees of the health care administration replied, for an estimated 14% response rate. The Patient Safety Culture Questionnaire was administered to employees in a similar fashion. The VA National Center for Patient Safety sponsored survey administration. Items were similar to surveys developed by the Agency for Healthcare Research and Quality.27 The survey was fielded in the spring of 2011. A total of 48,225 employees responded, for an estimated 21% response rate. Table 1 represents scale properties and example items for both the employee safety and patient safety survey.
Patient Safety Culture The survey consisted of 68 items about safety perceptions in addition to demographics and other nonsafety items. The survey assessed the following dimensions using safety-oriented language: overall perceptions of patient safety, nonpunitive response to an error, education/training/resources, communication openness, organizational learning, feedback and communication about error, patient safety at the facility, senior management awareness, and a patient safety grade. Items were rated on a 5-point scale, ranging from “strongly disagree” to “strongly agree.” The patient safety grade provided 5 options ranging from “A” to “E.” Cronbach α ranged from 0.70 to 0.96. The ICC(1) ranged from 0.01 to 0.03; ICC(2) ranged from 0.56 to 0.85.
Instruments Employee Safety Climate The survey consisted of 50 questions that represented 6 fundamental program categories as follows: management participation,
TABLE 1. Descriptive Statistics of Employee Safety Perceptions and Patient Safety Culture (n = 132 Facilities) k Mean SD Employee safety perceptions Management participation Supervisor participation Employee participation Safety support activities Safety support climate Safety composite Patient safety culture Overall safety perceptions
3.41 3.63 3.66 3.47 3.41 3.52
Sample Items
0.02 0.02 0.02 0.02 0.02 0.02
0.74 0.65 0.71 0.78 0.76 0.75
Management including safety in job promotion reviews Supervisors behaving in accordance with safe job procedures Employees using standardized precautions for hazardous materials Occurrence of emergency response procedures testing Belief that hazards not fixed right away will still be addressed
6 4.04 0.10 0.89
0.02
0.69
6 3.56 0.10 0.76 5 3.63 0.11 0.70
0.02 0.02
0.70 0.71
Communication openness
11 3.57 0.08 0.81
0.01
0.56
Organizational learning Feedback and communication Patient safety at facility Senior management awareness Patient safety grade
3 3 11 7 1
0.02 0.02 0.03 0.02 0.03
0.67 0.72 0.84 0.79 0.85
Our systems and procedures are good at preventing mistakes from happening. There are incentives in the work environment to hide mistakes. Safety education and training provided to me are enough to accomplish my job safely. Staff feel free to question the decisions or actions of those with more authority. Mistakes have led to positive changes here. We are informed about errors that happen in this unit. Patients often complain about unsafe practices. Senior managers successfully communicate patient safety goals. Please give your facility an overall grade on patient safety
3.83 3.51 3.80 3.63 3.87
0.12 0.09 0.11 0.12 0.13 0.11
ICC(1) ICC(2)
0.85 0.92 0.84 0.87 0.90 0.96
Nonpunitive response to an error Education/training/resources
7 9 9 10 10 45
α
0.10 0.14 0.13 0.16 0.17
0.88 0.86 0.91 0.96 ..
All items are based on a 1- to 5-point Likert scale, where higher scores indicate a more positive or favorable response. Mean and SD values reported are for the 132 valid medical centers in the sample. k indicates the number of items in the scale. ICC represents the intraclass correlation coefficient used as an indicator of group effects.
2
www.journalpatientsafety.com
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
Employee Safety Climate and Patient Safety Culture
Control Variables We modeled context variables that could influence findings based on examples from similar thematic studies.29 We obtained values from VA internal databases and other sources. Teaching affiliation was modeled dichotomously based on membership in the Council of Teaching Hospitals. We also included 2 variables for geographic characteristics, urban or rural area based on medical center zip code. We modeled for geographical region, which consisted of 4 regions. We modeled size using the total number of authorized beds.
Data Analysis We first merged survey data sets and reconciled medical center matching. We then examined descriptive statistics among variables and conducted correlation analysis among survey measures. We examined the correlation between the employee safety composite measure with each of the patient safety culture scores. We regressed each patient safety culture measures on the composite employee safety measures using ordinary least squares models. Standardized coefficient values, P values, and R2 model results are provided in tables.
RESULTS We had a total of 132 medical centers of a population of 140 with matching data for the analysis. Eight medical centers were not included because of a small sample of less than 30 respondents on both surveys. Descriptive statistics for medical centers are reported in Table 2 for the study. Half of the medical centers had a teaching affiliation, three-fourths were located in urban areas, and the average bed size was 365. Pearson correlation coefficients and regression analysis results are presented in Table 3. The employee safety composite measure was significantly positively related to all measures of patient safety culture in both the correlation and multiple regression model estimates. The standardized coefficients ranged from 0.44 to 0.64 in multivariate regression models, which accounted for control variables. The patient safety culture measures most strongly explained by employee safety were patient safety grade (β = 0.64, P < 0.001), patient safety at the facility (β = 0.64, P < 0.001), and overall safety perceptions (β = 0.61, P < 0.001). These measures seem to suggest that the broader patient safety culture measures are significantly explained by employee safety. In contrast, specific measures of safety practices, such as communication openness, displayed smaller, although still significant, standardized coefficients (β = 0.44, P < 0.001). TABLE 2. Characteristics of Medical Centers n (%) Operating beds, mean (SD) Teaching affiliation Yes No Geographic region Northeast Midwest South Western Setting Rural Urban
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
364.7 (261.3) 66 (50) 66 (50) 28 (21.1) 29 (22.0) 48 (36.4) 27 (20.5) 28 (21.2) 104 (78.8)
TABLE 3. Employee Safety Perceptions Association With Patient Safety Culture Dimensions: Bivariate Correlation (r) and Standardized Multiple Regression Results Patient Safety Culture Overall safety perceptions Nonpunitive response to an error Education/training/resources Communication openness Organizational learning Feedback and communication Patient safety at facility Senior management awareness Patient safety grade
r
β
Model R2
0.60* 0.51* 0.50* 0.44* 0.52* 0.46* 0.64* 0.50* 0.64*
0.61* 0.50* 0.49* 0.44* 0.52* 0.45* 0.64* 0.51* 0.64*
0.40 0.36 0.31 0.27 0.35 0.30 0.43 0.36 0.44
Models adjust for covariates including operating beds, geographic region, and urban or rural setting. * P < 0.01.
Differences between the correlation estimates and standardized coefficients were minimal, suggesting limited influence of control variables on the patient safety culture measures. Specifically regarding control variables, size and urban/rural region were nonsignificant in all the regression models. Teaching affiliation was significant in the model for communication openness (β = 0.25, P < 0.001) and senior management awareness (β = 0.20, P = 0.01).
DISCUSSION The findings highlight a very strong association between employee safety climate perceptions and patient safety culture as assessed through 2 independent facility-level surveys. Our findings highlight patient safety and employee safety as being similar but distinct concepts and suggest exploring the unique variance that employee safety climate may contribute in explaining adverse patient safety events beyond patient safety culture itself. Our analysis provides basis for developing and testing more specific models using employee and patient outcomes data. For example, it is not clear whether patient and employee safety is part of a larger system for total health care safety, which may also contain distinct, but strongly associated, constructs of data or digital safety and environmental safety. Quality improvement and systems redesign efforts are currently being used in health care, resulting in a rapidly changing health care environment. Practices, such as the accountable care organization, electronic health record, and the patient-centered medical home, are driving many delivery system reforms.30 We suggest that total health care safety is also an area that can benefit from systems redesign efforts and activities. Each aspect of the total health care safety may be synergistically connected. Interventions on one area may impact the other safety areas. For example, patient and employee safety are influenced by many of the same human factor– related issues, thus making them amenable to similar interventions. Total health care safety will depend on multidimensional analysis of various data streams for each domain, which could allow for a more targeted policy and systems intervention.
Future Directions and Limitations Future directions from the research findings should include a focus on longitudinal data to see whether the relationship between employee and patient safety is maintained over time. Although we used a lagged cross-sectional analysis, finding more time points www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
3
J Patient Saf • Volume 00, Number 00, Month 2015
Mohr et al
would be useful for consideration to better understand the complex relations among variables. Second, the study was conducted within the VA, an integrated care system that has a common set of policies, practices, and infrastructure. Testing the pattern in other health care facilities would be helpful to better understand the relations among constructs. Research can also explore whether employee safety can explain additional variance above what patient safety culture alone explains. Qualitative research involving interviews with senior management, clinical service leaders, and employees, on the role of employee safety and its association with patient safety may yield valuable insights as found in studies of patient safety improvement efforts.31,32 There are noteworthy limitations of the study. First, the low response rate to the survey may suggest nonresponse bias. Although we are unable to report the difference and similarities between respondents and nonrespondents because of limited demographic questions asked in both surveys, previous results from unpublished internal organizational survey analysis suggest small, but statistically significant, demographic differences between VA survey respondents and nonrespondents. Research suggests that response rate may not necessarily relate to the quality of survey responses.33,34 A better understanding of how demographic and safety perceptions may differ between groups would be an area for further hypothesis-driven survey research.35 Intraclass correlation coefficients were shown to be adequate for the study purposes however, potentially alleviating concerns. Our study examined each medical center as a unit of observation rather than focusing on relationships within specific clinical areas. For example, there may be a much stronger relation between employee and patient safety culture in the emergency departments, surgical suites, and intensive care, where there is a greater risk of preventable adverse outcomes related to unsafe processes. Further research on associations across settings would be useful in understanding the relation between the constructs.
CONCLUSIONS Our study investigated the association between employee safety climate as measured in 2010 with patient safety culture measured in 2011 using medical centers within the VA. Our results showed that employee safety demonstrated a large and positive association with multiple aspects of patient safety culture, suggesting that concepts are highly related but distinct from one another. We see findings as supportive of an overall total health care safety dimension. REFERENCES 1. Weaver SJ, Lubomski LH, Wilson RF, et al. Promoting a culture of safety as a patient safety strategy: a systematic review. Ann Intern Med. 2013; 1585:369–374. 2. Morello RT, Lowthian JA, Barker AL, et al. Strategies for improving patient safety culture in hosptials: a systemic review. BMJ Qual Saf. 2013;22:11–18. 3. Kohn KT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999. 4. Olden PC, McCaughrin WC. Designing healthcare organizations to reduce medical errors and enhance patient safety. Hosp Top. 2007;85:4–9. 5. Kaissi A. An organizational approach to understanding patient safety and medical errors. Health Care Manag (Frederick). 2006;25:292–305. 6. Burnett S, Benn J, Pinto A, et al. Organisational readiness: exploring the preconditions for success in organisation-wide patient safety improvement programmes. Qual Saf Health Care. 2010;19:313–317. 7. Nieva VF, Sorra J. Safety culture assessment: a tool for improving patient safety in healthcare organizations. Qual Saf Health Care. 2003;12(suppl 2): ii17–ii23.
4
www.journalpatientsafety.com
8. Blouin AS, McDonagh KJ. Framework for patient safety, part 1: culture as an imperative. J Nurs Adm. 2011;41:397–400. 9. DeJoy D, Gershon R, Schaffer B. Safety climate: assessing management and organizational influences on safety. Profess Safety. 2004;49:50–57. 10. DeJoy DM, Schaffer BS, Wilson MG, et al. Creating safer workplaces: assessing the determinants and role of safety climate. J Safety Res. 2004;35:8190. 11. Huang DT, Clermont G, Kong L, et al. Intensive care unit safety culture and outcomes: a US multicenter study. Int J Qual Health Care. 2010;22:151–161. 12. Katz-Navon T, Naveh E, Stern Z. Safety climate in health care organizations: a multidimensional approach. Acad Manage J. 2005;48: 1075–1089. 13. Hofmann DA, Marks B. An investigation of the relationship between safety climate and medication errors as well as other nurse and patient outcomes. Personnel Psychol. 2006;59:847–869. 14. Sorra J, Khanna K, Dyer N, et al. Exploring relationships between patient safety culture and patients' assessments of hospital care. J Patient Saf. 2012;8:131–139. 15. Pronovost P, Weast B, Rosenstein B, et al. Implementing and validating a comprehensive unit-based safety program. J Patient Saf. 2005;1:33–40. 16. Fernández-Muñiz B, Montes-Peón JM, Vázquez-Ordás CJ. Safety climate in OHSAS 18001-certified organisations: antecedents and consequences of safety behaviour. Accid Anal Prev. 2012;45:745–758. 17. Olsen E. Exploring the possibility of a common structural model measuring associations between safety climate factors and safety behaviour in health care and the petroleum sectors. Accid Anal Prev. 2010;42:1507–1516. 18. The Joint Commission. Improving Patient and Worker Safety: Opportunities for Synergy, Collaboration, and Innovation. Oakbrook Terrace, IL: The Joint Commission; 2012. 19. Lucian Leape Institute. Through the Eyes of the Workforce: Creating joy, Meaning and Safer Health Care. Boston, MA: National Patient Safety Foundation; 2013. 20. Clarke S. An integrative model of safety climate: linking psychological climate and work attitudes to individual safety outcomes using meta-analysis. J Occup Organ Psychol. 2010;83:553–578. 21. Vredenburgh AG. Organizational safety: which management practices are most effective in reducing employee injury rates. J Safety Res. 2002;33:259–276. 22. Jiang L, Yu G, Li Y. Perceived colleagues' safety knowledge/behavior and safety performance: safety climate as a moderator in a multilevel study. Accid Anal Prev. 2010;42:1468–1476. 23. Clarke S. The relationship between safety climate and safety performance: a meta analytic review. J Occup Health Psychol. 2006;11:315–327. 24. Carayon P, Schoofs Hundt A, Karsh BT, et al. Work system design for patient safety: the SEIPS model. Qual Saf Health Care. 2006;15(suppl 1): i50–i58. 25. Oliver A, Cheyne A, Tomas JM, et al. The effects of organizational and individual factors on occupational accidents. J Occup Organ Psychol. 2002;75:473–488. 26. Willis PG, Brown KA, Prussia GE. Does employee safety influence customer satisfaction? Evidence from the electric utility industry. J Safety Res. 2012;43:389–396. 27. Agency for Healthcare Research and Quality. Surveys on Patient Safety Culture. Available at: http://www.ahrq.gov/professionals/quality-patientsafety/patientsafetyculture/index.html. Accessed September 17, 2014. 28. Bliese PD. Within-group agreement, non-independence, and reliability. In: Klein K, Kozlowski SW, eds. Multilevel Theory, Research, and Methods in Organizations. San Francisco, CA: Jossey-Bass; 2000:349–381. 29. Longo DR, Hewett JE, Ge B, et al. Hospital patient safety: characteristics of best-performing hospitals. J Healthc Manag. 2007;52:188–204.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
Employee Safety Climate and Patient Safety Culture
30. Rittenhouse DR, Shortell SM, Fisher ES. Primary care and accountable care—two essential elements of delivery-system reform. N Engl J Med. 2009;361:2301–2303.
33. Templeton L, Deehan A, Taylor C, et al. Surveying general practitioners: does a low response rate matter? Br J Gen Pract. 1997;47:91–94.
31. Krein SL, Damschroder LJ, Kowalski CP, et al. The influence of organizational context on quality improvement and patient safety efforts in infection prevention: a multi-center qualitative study. Soc Sci Med. 2010;71:1692–1701.
34. Hikmet N, Chen SK. An investigation into low mail survey response rates of information technology users in health care organizations. Int J Med Info. 2003;72:29–34.
32. Benning A, Ghaleb M, Suokas A, et al. Large scale organisational intervention to improve patient safety in four UK hospitals: mixed method evaluation. BMJ. 2011;342:d195.
35. Listyowardojo TA, Nap RE, Johnson A. Demographic differences between health care workers who did or did not respond to a safety and organizational culture survey. BMC Res Notes. 2011;4:328.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
5
Does Employee Safety Matter for Patients Too? Employee Safety Climate and Patient Safety Culture in Health Care David C. Mohr, PhD,*† Jennifer Lipkowitz Eaton, MD, MPH,‡ Kathleen M. McPhaul, PhD, MPH, COHN-S,‡ and Michael J. Hodgson, MD, MPH§ Objective: We examined relationships between employee safety climate and patient safety culture. Because employee safety may be a precondition for the development of patient safety, we hypothesized that employee safety culture would be strongly and positively related to patient safety culture. Methods: An employee safety climate survey was administered in 2010 and assessed employees' views and experiences of safety for employees. The patient safety survey administered in 2011 assessed the safety culture for patients. We performed Pearson correlations and multiple regression analysis to examine the relationships between a composite measure of employee safety with subdimensions of patient safety culture. The regression models controlled for size, geographic characteristics, and teaching affiliation. Analyses were conducted at the group level using data from 132 medical centers. Results: Higher employee safety climate composite scores were positively associated with all 9 patient safety culture measures examined. Standardized multivariate regression coefficients ranged from 0.44 to 0.64. Conclusions: Medical facilities where staff have more positive perceptions of health care workplace safety climate tended to have more positive assessments of patient safety culture. This suggests that patient safety culture and employee safety climate could be mutually reinforcing, such that investments and improvements in one domain positively impacts the other. Further research is needed to better understand the nexus between health care employee and patient safety to generalize and act upon findings. Key Words: patient safety culture, employee safety, health care organizations, Veterans Health Administration (J Patient Saf 2015;00: 00–00)
P
atient safety culture is an increasingly important target of systems-based clinical improvement efforts.1,2 The publication of To Err Is Human brought the scope and cost of preventable medical errors to the forefront of national conversation.3 Research and innovation efforts focused on patient safety, particularly reduction of preventable medical errors to avoid patient morbidity and mortality, have become standard health care management practice since the report.4,5 The role of the larger organizational From the *Center for Healthcare Organization and Implementation Research (CHOIR), VA Boston Healthcare System; and †School of Public Health, Boston University, Boston, Massachusetts; ‡Office of Public Health, Occupational Health, Veterans Health Administration; and §Occupational Safety and Health Administration, Department of Labor, Washington, District of Columbia. Correspondence: David C. Mohr, PhD, VA Boston Healthcare System, 150 South Huntington Ave, Mailcode 152 M, Boston, MA 02130 (e‐mail: [email protected]). The authors disclose no conflict of interest. The work reported here was supported by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service (IIR 08–067). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the U.S. government. This document is not a standard or regulation, and it neither creates new legal obligations nor alters existing obligations created by OSHA standards or the Occupational Safety and Health (OSH) Act. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Patient Saf • Volume 00, Number 00, Month 2015
culture in the success and sustainability of clinical quality and patient safety gains is increasingly recognized.6–8 Furthermore, organizational climate has been shown to influence health care worker perceptions of safety climate and compliance with precautionary safety procedures.9,10 Previous studies have demonstrated relationships between positive hospital safety culture and patient safety outcomes, such as decreased mortality rates, decreased lengths of stay, decreased treatment and medication errors, and increased patient satisfaction.11–15 Similarly, the occupational safety sciences have examined and demonstrated the relationship between employee safety climate and/or culture with occupational accidents.16,17 Integrating these fields and theoretical models widens the lens to examine the effectiveness of patient safety systems in the context of the workplace safety systems. A collaboration between the Joint Commission and the National Institute for Occupational Safety and Health outlines a conceptual model and rationale for considering both the patient safety apparatus and the occupational safety systems, arguing that “principles, methods, and tools for creating safety are the same, regardless of the population where safety is the focus.”18 Furthermore, the same monograph points out that a safety culture that excludes some groups is neither “genuine nor legitimate.” In other words, it is not possible to sustain a strong culture of safety for patients when occupational safety is ignored or deemphasized. For a workplace to be safe for patients, it should also be safe for the employee workforce.19 A potential prerequisite for fostering patient safety culture is addressing occupational health and safety of employees.20 The greater use of management practices focused on promoting safety in health care has been associated with fewer employee injury rates.21 Research suggests a negative association between employee safety climate and safety behaviors, such as compliance and participation, and with safety performance, such as accidents and injuries.22,23 In some research, both employee and patient safety are both seen as end points of effective organizational work systems and structures or as having a reciprocal relation.24 An organizational climate that encourages and rewards safetyrelated behavior can lead to fewer accidents, whereas a climate emphasizing production over safety may lead to more accidents.25 Research, outside of health care settings, suggests a positive relationship between safe employee work practices and customer satisfaction.26 Thus, it may be important to ensure that employees perceive the environment as being safe and supportive before they are able to contribute effectively to patient safety by behaving in ways conducive to patient safety culture, such as reporting errors they or colleagues have made. In this article, we examine facilitylevel employee safety climate and its association with patient safety culture. We hypothesize that employee safety is a leading indicator and should be positively associated with patient safety domains.
METHODS Study Design We conducted analysis using 2 surveys administered to employees at medical centers within the Department of Veterans www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1
J Patient Saf • Volume 00, Number 00, Month 2015
Mohr et al
Affairs, Veterans Health Administration (VA). One survey focused on perceptions of employee safety and was administered a year before a second survey that assessed patient safety culture. Responses to both surveys were confidential and anonymous. The study was approved by the VA Boston Healthcare System Institutional Review Board.
supervisor participation, employee participation, safety support activities, safety support climate, and organizational climate. For our analysis, we excluded organizational commitment because the questions did not focus on safety, but rather general perceptions of job satisfaction or stress among employees. Items were rated on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” Cronbach α for the 6 dimensions ranged from 0.82 to 0.92. Intraclass correlation coefficient (ICC) values were computed to examine support for group-level aggregation. Statistics suggest the similarity among individual responses from a higher level unit. The ICC(1) values ranged from 0.02 to 0.04, and the ICC(2) values, influenced by number of respondents at each site, ranged from 0.65 to 0.82. These values are sufficient for aggregation at the facility level.28 Based on the high Pearson correlations among the 5 employee safety perception measures, which ranged from 0.84 to 0.95 at the facility level, we created a composite scale score by averaging across the 5 dimensions. We treated this new employee safety composite measure as the main predictor variable of interest.
Data Collection and Participants Both surveys were administered to full-time employees in all sections and departments. The VA Employee Safety Perception survey, also referred to as the “Safety Barometer,” developed by the National Safety Council, a nonprofit, congressional charter organization, was administered to employees through the internet. An e-mail from local medical center leaders was sent to employees with a link to complete the survey online. Facility survey coordinators sent weekly reminders along with response rates. The survey was administered during October through November 2010. A total of 37,971 employees of the health care administration replied, for an estimated 14% response rate. The Patient Safety Culture Questionnaire was administered to employees in a similar fashion. The VA National Center for Patient Safety sponsored survey administration. Items were similar to surveys developed by the Agency for Healthcare Research and Quality.27 The survey was fielded in the spring of 2011. A total of 48,225 employees responded, for an estimated 21% response rate. Table 1 represents scale properties and example items for both the employee safety and patient safety survey.
Patient Safety Culture The survey consisted of 68 items about safety perceptions in addition to demographics and other nonsafety items. The survey assessed the following dimensions using safety-oriented language: overall perceptions of patient safety, nonpunitive response to an error, education/training/resources, communication openness, organizational learning, feedback and communication about error, patient safety at the facility, senior management awareness, and a patient safety grade. Items were rated on a 5-point scale, ranging from “strongly disagree” to “strongly agree.” The patient safety grade provided 5 options ranging from “A” to “E.” Cronbach α ranged from 0.70 to 0.96. The ICC(1) ranged from 0.01 to 0.03; ICC(2) ranged from 0.56 to 0.85.
Instruments Employee Safety Climate The survey consisted of 50 questions that represented 6 fundamental program categories as follows: management participation,
TABLE 1. Descriptive Statistics of Employee Safety Perceptions and Patient Safety Culture (n = 132 Facilities) k Mean SD Employee safety perceptions Management participation Supervisor participation Employee participation Safety support activities Safety support climate Safety composite Patient safety culture Overall safety perceptions
3.41 3.63 3.66 3.47 3.41 3.52
Sample Items
0.02 0.02 0.02 0.02 0.02 0.02
0.74 0.65 0.71 0.78 0.76 0.75
Management including safety in job promotion reviews Supervisors behaving in accordance with safe job procedures Employees using standardized precautions for hazardous materials Occurrence of emergency response procedures testing Belief that hazards not fixed right away will still be addressed
6 4.04 0.10 0.89
0.02
0.69
6 3.56 0.10 0.76 5 3.63 0.11 0.70
0.02 0.02
0.70 0.71
Communication openness
11 3.57 0.08 0.81
0.01
0.56
Organizational learning Feedback and communication Patient safety at facility Senior management awareness Patient safety grade
3 3 11 7 1
0.02 0.02 0.03 0.02 0.03
0.67 0.72 0.84 0.79 0.85
Our systems and procedures are good at preventing mistakes from happening. There are incentives in the work environment to hide mistakes. Safety education and training provided to me are enough to accomplish my job safely. Staff feel free to question the decisions or actions of those with more authority. Mistakes have led to positive changes here. We are informed about errors that happen in this unit. Patients often complain about unsafe practices. Senior managers successfully communicate patient safety goals. Please give your facility an overall grade on patient safety
3.83 3.51 3.80 3.63 3.87
0.12 0.09 0.11 0.12 0.13 0.11
ICC(1) ICC(2)
0.85 0.92 0.84 0.87 0.90 0.96
Nonpunitive response to an error Education/training/resources
7 9 9 10 10 45
α
0.10 0.14 0.13 0.16 0.17
0.88 0.86 0.91 0.96 ..
All items are based on a 1- to 5-point Likert scale, where higher scores indicate a more positive or favorable response. Mean and SD values reported are for the 132 valid medical centers in the sample. k indicates the number of items in the scale. ICC represents the intraclass correlation coefficient used as an indicator of group effects.
2
www.journalpatientsafety.com
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
Employee Safety Climate and Patient Safety Culture
Control Variables We modeled context variables that could influence findings based on examples from similar thematic studies.29 We obtained values from VA internal databases and other sources. Teaching affiliation was modeled dichotomously based on membership in the Council of Teaching Hospitals. We also included 2 variables for geographic characteristics, urban or rural area based on medical center zip code. We modeled for geographical region, which consisted of 4 regions. We modeled size using the total number of authorized beds.
Data Analysis We first merged survey data sets and reconciled medical center matching. We then examined descriptive statistics among variables and conducted correlation analysis among survey measures. We examined the correlation between the employee safety composite measure with each of the patient safety culture scores. We regressed each patient safety culture measures on the composite employee safety measures using ordinary least squares models. Standardized coefficient values, P values, and R2 model results are provided in tables.
RESULTS We had a total of 132 medical centers of a population of 140 with matching data for the analysis. Eight medical centers were not included because of a small sample of less than 30 respondents on both surveys. Descriptive statistics for medical centers are reported in Table 2 for the study. Half of the medical centers had a teaching affiliation, three-fourths were located in urban areas, and the average bed size was 365. Pearson correlation coefficients and regression analysis results are presented in Table 3. The employee safety composite measure was significantly positively related to all measures of patient safety culture in both the correlation and multiple regression model estimates. The standardized coefficients ranged from 0.44 to 0.64 in multivariate regression models, which accounted for control variables. The patient safety culture measures most strongly explained by employee safety were patient safety grade (β = 0.64, P < 0.001), patient safety at the facility (β = 0.64, P < 0.001), and overall safety perceptions (β = 0.61, P < 0.001). These measures seem to suggest that the broader patient safety culture measures are significantly explained by employee safety. In contrast, specific measures of safety practices, such as communication openness, displayed smaller, although still significant, standardized coefficients (β = 0.44, P < 0.001). TABLE 2. Characteristics of Medical Centers n (%) Operating beds, mean (SD) Teaching affiliation Yes No Geographic region Northeast Midwest South Western Setting Rural Urban
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
364.7 (261.3) 66 (50) 66 (50) 28 (21.1) 29 (22.0) 48 (36.4) 27 (20.5) 28 (21.2) 104 (78.8)
TABLE 3. Employee Safety Perceptions Association With Patient Safety Culture Dimensions: Bivariate Correlation (r) and Standardized Multiple Regression Results Patient Safety Culture Overall safety perceptions Nonpunitive response to an error Education/training/resources Communication openness Organizational learning Feedback and communication Patient safety at facility Senior management awareness Patient safety grade
r
β
Model R2
0.60* 0.51* 0.50* 0.44* 0.52* 0.46* 0.64* 0.50* 0.64*
0.61* 0.50* 0.49* 0.44* 0.52* 0.45* 0.64* 0.51* 0.64*
0.40 0.36 0.31 0.27 0.35 0.30 0.43 0.36 0.44
Models adjust for covariates including operating beds, geographic region, and urban or rural setting. * P < 0.01.
Differences between the correlation estimates and standardized coefficients were minimal, suggesting limited influence of control variables on the patient safety culture measures. Specifically regarding control variables, size and urban/rural region were nonsignificant in all the regression models. Teaching affiliation was significant in the model for communication openness (β = 0.25, P < 0.001) and senior management awareness (β = 0.20, P = 0.01).
DISCUSSION The findings highlight a very strong association between employee safety climate perceptions and patient safety culture as assessed through 2 independent facility-level surveys. Our findings highlight patient safety and employee safety as being similar but distinct concepts and suggest exploring the unique variance that employee safety climate may contribute in explaining adverse patient safety events beyond patient safety culture itself. Our analysis provides basis for developing and testing more specific models using employee and patient outcomes data. For example, it is not clear whether patient and employee safety is part of a larger system for total health care safety, which may also contain distinct, but strongly associated, constructs of data or digital safety and environmental safety. Quality improvement and systems redesign efforts are currently being used in health care, resulting in a rapidly changing health care environment. Practices, such as the accountable care organization, electronic health record, and the patient-centered medical home, are driving many delivery system reforms.30 We suggest that total health care safety is also an area that can benefit from systems redesign efforts and activities. Each aspect of the total health care safety may be synergistically connected. Interventions on one area may impact the other safety areas. For example, patient and employee safety are influenced by many of the same human factor– related issues, thus making them amenable to similar interventions. Total health care safety will depend on multidimensional analysis of various data streams for each domain, which could allow for a more targeted policy and systems intervention.
Future Directions and Limitations Future directions from the research findings should include a focus on longitudinal data to see whether the relationship between employee and patient safety is maintained over time. Although we used a lagged cross-sectional analysis, finding more time points www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
3
J Patient Saf • Volume 00, Number 00, Month 2015
Mohr et al
would be useful for consideration to better understand the complex relations among variables. Second, the study was conducted within the VA, an integrated care system that has a common set of policies, practices, and infrastructure. Testing the pattern in other health care facilities would be helpful to better understand the relations among constructs. Research can also explore whether employee safety can explain additional variance above what patient safety culture alone explains. Qualitative research involving interviews with senior management, clinical service leaders, and employees, on the role of employee safety and its association with patient safety may yield valuable insights as found in studies of patient safety improvement efforts.31,32 There are noteworthy limitations of the study. First, the low response rate to the survey may suggest nonresponse bias. Although we are unable to report the difference and similarities between respondents and nonrespondents because of limited demographic questions asked in both surveys, previous results from unpublished internal organizational survey analysis suggest small, but statistically significant, demographic differences between VA survey respondents and nonrespondents. Research suggests that response rate may not necessarily relate to the quality of survey responses.33,34 A better understanding of how demographic and safety perceptions may differ between groups would be an area for further hypothesis-driven survey research.35 Intraclass correlation coefficients were shown to be adequate for the study purposes however, potentially alleviating concerns. Our study examined each medical center as a unit of observation rather than focusing on relationships within specific clinical areas. For example, there may be a much stronger relation between employee and patient safety culture in the emergency departments, surgical suites, and intensive care, where there is a greater risk of preventable adverse outcomes related to unsafe processes. Further research on associations across settings would be useful in understanding the relation between the constructs.
CONCLUSIONS Our study investigated the association between employee safety climate as measured in 2010 with patient safety culture measured in 2011 using medical centers within the VA. Our results showed that employee safety demonstrated a large and positive association with multiple aspects of patient safety culture, suggesting that concepts are highly related but distinct from one another. We see findings as supportive of an overall total health care safety dimension. REFERENCES 1. Weaver SJ, Lubomski LH, Wilson RF, et al. Promoting a culture of safety as a patient safety strategy: a systematic review. Ann Intern Med. 2013; 1585:369–374. 2. Morello RT, Lowthian JA, Barker AL, et al. Strategies for improving patient safety culture in hosptials: a systemic review. BMJ Qual Saf. 2013;22:11–18. 3. Kohn KT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999. 4. Olden PC, McCaughrin WC. Designing healthcare organizations to reduce medical errors and enhance patient safety. Hosp Top. 2007;85:4–9. 5. Kaissi A. An organizational approach to understanding patient safety and medical errors. Health Care Manag (Frederick). 2006;25:292–305. 6. Burnett S, Benn J, Pinto A, et al. Organisational readiness: exploring the preconditions for success in organisation-wide patient safety improvement programmes. Qual Saf Health Care. 2010;19:313–317. 7. Nieva VF, Sorra J. Safety culture assessment: a tool for improving patient safety in healthcare organizations. Qual Saf Health Care. 2003;12(suppl 2): ii17–ii23.
4
www.journalpatientsafety.com
8. Blouin AS, McDonagh KJ. Framework for patient safety, part 1: culture as an imperative. J Nurs Adm. 2011;41:397–400. 9. DeJoy D, Gershon R, Schaffer B. Safety climate: assessing management and organizational influences on safety. Profess Safety. 2004;49:50–57. 10. DeJoy DM, Schaffer BS, Wilson MG, et al. Creating safer workplaces: assessing the determinants and role of safety climate. J Safety Res. 2004;35:8190. 11. Huang DT, Clermont G, Kong L, et al. Intensive care unit safety culture and outcomes: a US multicenter study. Int J Qual Health Care. 2010;22:151–161. 12. Katz-Navon T, Naveh E, Stern Z. Safety climate in health care organizations: a multidimensional approach. Acad Manage J. 2005;48: 1075–1089. 13. Hofmann DA, Marks B. An investigation of the relationship between safety climate and medication errors as well as other nurse and patient outcomes. Personnel Psychol. 2006;59:847–869. 14. Sorra J, Khanna K, Dyer N, et al. Exploring relationships between patient safety culture and patients' assessments of hospital care. J Patient Saf. 2012;8:131–139. 15. Pronovost P, Weast B, Rosenstein B, et al. Implementing and validating a comprehensive unit-based safety program. J Patient Saf. 2005;1:33–40. 16. Fernández-Muñiz B, Montes-Peón JM, Vázquez-Ordás CJ. Safety climate in OHSAS 18001-certified organisations: antecedents and consequences of safety behaviour. Accid Anal Prev. 2012;45:745–758. 17. Olsen E. Exploring the possibility of a common structural model measuring associations between safety climate factors and safety behaviour in health care and the petroleum sectors. Accid Anal Prev. 2010;42:1507–1516. 18. The Joint Commission. Improving Patient and Worker Safety: Opportunities for Synergy, Collaboration, and Innovation. Oakbrook Terrace, IL: The Joint Commission; 2012. 19. Lucian Leape Institute. Through the Eyes of the Workforce: Creating joy, Meaning and Safer Health Care. Boston, MA: National Patient Safety Foundation; 2013. 20. Clarke S. An integrative model of safety climate: linking psychological climate and work attitudes to individual safety outcomes using meta-analysis. J Occup Organ Psychol. 2010;83:553–578. 21. Vredenburgh AG. Organizational safety: which management practices are most effective in reducing employee injury rates. J Safety Res. 2002;33:259–276. 22. Jiang L, Yu G, Li Y. Perceived colleagues' safety knowledge/behavior and safety performance: safety climate as a moderator in a multilevel study. Accid Anal Prev. 2010;42:1468–1476. 23. Clarke S. The relationship between safety climate and safety performance: a meta analytic review. J Occup Health Psychol. 2006;11:315–327. 24. Carayon P, Schoofs Hundt A, Karsh BT, et al. Work system design for patient safety: the SEIPS model. Qual Saf Health Care. 2006;15(suppl 1): i50–i58. 25. Oliver A, Cheyne A, Tomas JM, et al. The effects of organizational and individual factors on occupational accidents. J Occup Organ Psychol. 2002;75:473–488. 26. Willis PG, Brown KA, Prussia GE. Does employee safety influence customer satisfaction? Evidence from the electric utility industry. J Safety Res. 2012;43:389–396. 27. Agency for Healthcare Research and Quality. Surveys on Patient Safety Culture. Available at: http://www.ahrq.gov/professionals/quality-patientsafety/patientsafetyculture/index.html. Accessed September 17, 2014. 28. Bliese PD. Within-group agreement, non-independence, and reliability. In: Klein K, Kozlowski SW, eds. Multilevel Theory, Research, and Methods in Organizations. San Francisco, CA: Jossey-Bass; 2000:349–381. 29. Longo DR, Hewett JE, Ge B, et al. Hospital patient safety: characteristics of best-performing hospitals. J Healthc Manag. 2007;52:188–204.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Patient Saf • Volume 00, Number 00, Month 2015
Employee Safety Climate and Patient Safety Culture
30. Rittenhouse DR, Shortell SM, Fisher ES. Primary care and accountable care—two essential elements of delivery-system reform. N Engl J Med. 2009;361:2301–2303.
33. Templeton L, Deehan A, Taylor C, et al. Surveying general practitioners: does a low response rate matter? Br J Gen Pract. 1997;47:91–94.
31. Krein SL, Damschroder LJ, Kowalski CP, et al. The influence of organizational context on quality improvement and patient safety efforts in infection prevention: a multi-center qualitative study. Soc Sci Med. 2010;71:1692–1701.
34. Hikmet N, Chen SK. An investigation into low mail survey response rates of information technology users in health care organizations. Int J Med Info. 2003;72:29–34.
32. Benning A, Ghaleb M, Suokas A, et al. Large scale organisational intervention to improve patient safety in four UK hospitals: mixed method evaluation. BMJ. 2011;342:d195.
35. Listyowardojo TA, Nap RE, Johnson A. Demographic differences between health care workers who did or did not respond to a safety and organizational culture survey. BMC Res Notes. 2011;4:328.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.journalpatientsafety.com
Copyright © 2015 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
5
