- Email: [email protected]
Implementing a Hospitalwide Patient Safety Program for Cultural Change
Joint Commission
Journal on Quality and Safety
Patient Safety
Implementing a Hospitalwide Patient Safety Program for Cultural Change
Max M. Cohen, M.D., M.H.S.A. Nancy L. Kimmel, R.Ph. M. Kathleen Benage, R.N. Cuong C. Hoang Thomas E. Burroughs, Ph.D. Carolyn A. Roth, R.N., J.D.
he Institute of Medicine (IOM) report on medical errors1 markedly heightened awareness of patient safety among health care professionals, politicians, the media, and the general public. A number of initiatives followed, including the passage of the Healthcare Research and Quality Act (1999),2 the formation of the Leapfrog Group,3 the formation of the National Patient Safety Task Force4—which included the Food and Drug Administraton, the Agency for Healthcare Research and Quality, the Centers for Disease Control and Prevention, and the implementation of new Joint Commission on Accreditation of Healthcare Organizations patient safety standards in 2001.5 A review of the literature on patient safety initiatives suggests that although much work is under way to rectify current deficiencies (fear of reporting errors, assignment of blame rather than seeking the causes, inadequate investigation of mistakes), there is still less than universal agreement as to the goal or how it will best be achieved.6–13 There is some agreement that if health care providers operate within a nonpunitive environment, their willingness to point out system weaknesses and to report accidents and near misses is substantially improved.8,14,15 Missouri Baptist Medical Center (MBMC), a member of BJC HealthCare, St. Louis, embarked on a newly reenergized patient safety program in 1999. We first determined the kind of culture that existed. We initially assessed the state of error* reporting to identify and
T
* In this article the terms error and event are used interchangeably, reflecting the practice at Missouri Baptist Medical Center.
34
Article-at-a-Glance Background: After focus groups revealed that staff perceived a punitive culture, Missouri Baptist Medical Center (MBMC) embarked on a comprehensive patient safety program, which was initially directed at creating a just culture of patient safety. Interventions: A series of structures, processes, and initiatives were introduced to change the attitudes of management and staff toward human error, to communicate broadly with staff and the community, and to provide feedback on leadership’s responses to specific events. All events reported were tracked continuously and recorded each month on a spreadsheet. Results: Total medical events reported by staff increased significantly (p < .001) from 35 to 132 per 1,000 patient days. Reports to the hotline alone increased significantly (p < .001) from 3 to 23 per 1,000 patient days, and the proportion of callers who left their names increased significantly (p < .001) from 30% to 61%. Survey results from staff showed a small but significant increase in awareness of patient safety and in comfort with reporting. Conclusion: The implementation of a carefully planned and orchestrated series of interventions designed to improve a hospital’s culture of patient safety can, if led by senior hospital executives, lead to a substantial, profound, and lasting increase in error reporting and improvement in employee perceptions of the organization’s safety culture.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
address performance gaps. In late 1999, we conducted a series of six focus groups with physicians, nursing staff, unit clerks, and pharmacy staff. During these focus groups, in which 68 persons participated, participants were asked if they had ever reported an error made either by themselves or by another, and if not, what the reasons were. They were also asked what needed to be done to make reporting easier and what leadership needed to do to reassure staff that human error would not be punished. Three key themes emerged from the focus groups: ■ The existing error-reporting forms were too timeconsuming and intimidating. ■ Staff did not consider it appropriate to report errors that did not reach the patient. ■ There was a pervasive fear of negative consequences. On learning that we had a distinctly punitive culture, we set out to change the culture to a “just and fair” one. Our hypothesis was that this would encourage error reporting, which in turn would allow us to identify and improve our imperfect processes.* This article describes the development and implementation of a comprehensive patient safety program for establishing a culture of patient safety in a community hospital. The program’s main objectives were, as examined in a prospective, observational study, to determine this program’s impact on two specific putative measures of the safety culture: event-reporting rates and surveys of staff opinion.
Methods Setting Missouri Baptist Medical Center (MBMC) is a not-forprofit, 489-bed nonteaching suburban community hospital that is part of BJC HealthCare, a 13-hospital integrated health delivery system.
Interventions Setting the Goal. In response to the findings of the focus groups, we set a primary goal to establish a more effective culture of patient safety in the hospital, as defined by “the willingness of all staff members to report all safety events and near misses without fear * These efforts were recognized by MBMC’s receipt of the inaugural 2002 American Hospital Quest for Quality PrizeSM, which honored leadership and innovation in the creation of a culture of patient safety.
of retribution, but with an understanding of accountability.” We encountered an immediate barrier—the difficulty in persuading middle management that human error should not be punished. Many middle managers greeted the initial description of the new culture as “nonpunitive” skeptically, often asking “How can we have a nonpunitive culture but still demand accountability?” Because certain behaviors clearly demand disciplinary action, we concluded that the term nonpunitive was misleading and confusing. The culture to which we aspired was renamed “just,” reflecting an approach described by David Marx.16 Marx posits that a just culture is one in which discipline occurs only for reckless or criminal behavior and where repeated errors by the same individual call for a remedial action plan and possibly for a readjustment in that individual’s role and responsibilities in the organization. This approach drove the changes that we made in the way that supervisors responded to human error. The interventions, driven by the results of the focus groups and from a literature review, were designed to do the following: ■ Promote event reporting ■ Encourage individual accountability ■ Forgive human error and even rule-breaking ■ Discipline staff for reckless behavior Beginning in early 2000 and continuing through the winter of 2001–2002, we instituted a series of structures, processes, and initiatives (Table 1, page 36) designed to change the management’s and staff’s attitudes toward human error, communicate broadly with staff and the community, and provide feedback on leadership’s responses to specific events. Critical Success Factors. Although all the initiatives and interventions listed in Table 1 were important to the development of the patient safety program, the following critical success factors were evident: ■ Senior management not only voiced commitment but also demonstrated its commitment in the form of highly visible behavior. ■ Although everyone was held responsible for patient safety, a reality of organizations is captured in the statement “common property is nobody’s property.” To overcome this, safety champions were identified within the medical staff as well as among the front-line employees to lead most initiatives.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
35
Joint Commission
Journal on Quality and Safety
Table 1. Components of the Patient Safety Program* I Executive Leadership ■ Executive rounds ■ Executive participation on safety teams ■ Executives emphasize patient safety at all hospital meetings. ■ Safety is monthly agenda item at board and medical executive committee meetings.
IV Event Investigation ■ Root cause analysis ■ Performance improvement teams ■ Small tests of change ■ Event prioritization and risk assessment V Prevention of Error ■
II Structural Change ■ ■ ■ ■ ■ ■ ■
■
Medication safety team Patient safety council Safety manager (environmental) Patient advocate Patient safety specialist Safety briefings on nursing units Modification of written human resources policies
III Event Reporting ■ ■ ■ ■
Anonymous reporting Simplified check-box reporting forms Safety hotline Common database for safety and risk management
■ ■
Failure Modes and Effects Analyses Response to newsletters from Institute for Safe Medication Practices Response to Joint Commission Sentinel Event Alerts Monitor literature for best practices
VI Feedback and Communication ■ Celebration of reporting ■ Electronic newsletter describing changes made in response to reported events ■ Employee brochure on patient safety and culture change ■ Patient safety awards program for safety ideas ■ Patient safety grants program to fund safety initiatives ■ Mind Your MedsSM community education program ■ Community education program on quality and safety
* Joint Commission, Joint Commission on Accreditation of Healthcare Organizations.
Although the technical activities implemented to improve patient safety were similar to those reported in other patient safety initiatives in the literature, the executive team passionately delivered the message that patient safety was a strategic priority.17 A high-level patient safety council was formed that was composed of executives and key directors, including the director of risk management, the director of performance improvement, and the patient safety specialist; these three positions reported to the chief medical officer [M.M.C.], who also chaired the council. Patient Safety Specialist. The full-time patient safety specialist position, which was created in 2001, allowed the hospital to coordinate all of its patient safety improvements, which sent the message that safety was a strategic priority and provided the staff with a single contact for all patient safety issues. The patient safety specialist
36
reviewed all reports within 48 hours of receipt and prioritized them for action using a modification of the Safety Assessment Code described by Bagian et al.18 In addition, the patient safety specialist relayed all errors to the appropriate manager for possible immediate action. Hospital staff members, medical staff members, and patients or family members (using a hotline) were encouraged to report any kind of concern about any aspect of medical care or the safety of the hospital environment. Other Actions. The cultural changes were communicated not only to the employees and medical staff but also to patients, who were viewed as partners in the maintenance of their own health, and to the community in general. A “Mind Your Meds”SM brochure, which was given at discharge to every patient, was designed to educate the public about medication safety. It included a detachable wallet-sized card on which current prescribed
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
medications, over-the-counter medications, and even herbal remedies could be recorded.
Survey Instruments The ongoing monthly employee satisfaction survey mailed to a stratified random sample of all hospital staff was used to track cultural change on a longitudinal basis. According to the sampling strategy, samples were stratified by functional departments within the hospital, such that every hospital employee is selected for surveying once a year. The surveying process is conducted by an external research firm, and responses are completely anonymous. Information is gathered on staff’s clinical department and job type for tabulation purposes. Response-bias weighting is used to adjust for any systemic response differences in these two characteristics. In June 2001 two specific items pertaining to patient safety were added to the survey: 1. This hospital has appropriate measures in place to protect patients’ medical safety. 2. This hospital provides an environment where staff can report medical errors and concerns without fear of negative consequences. The instrument uses a six-point Likert scale for each statement (1, strongly disagree; 2, disagree; 3, slightly disagree; 4, slightly agree; 5, agree; 6, strongly agree). A mean score is calculated for the response to each statement on the survey. From June 2001 until April 2002, an average of 96 randomly selected employees were surveyed each month, and the response rate was 36%. From May 2002 through March 2003, an average of 238 employees were surveyed monthly, and the average response rate was 34%. With the increase in the number of employees surveyed, each member of the workforce was surveyed at least once in the course of a single year.
Patient Safety Event Tracking System All medical errors, as identified from all reporting methods, were tracked continuously and recorded each month on a spreadsheet. Staff members were offered a choice of reporting vehicles: ■ Staff could use the existing incident report forms but were no longer required to sign them. ■ A much simplified anonymous check-box form was introduced for reporting medication errors.
■ Pharmacy staff members were encouraged to leave
copies of prescribing/transcription errors that they identified and the interventions they implemented in a box housed in their department. ■ Staff could report errors and risky situations during the brief safety “huddles” conducted at the change of shift on nursing units. ■ Staff could report errors to executives during weekly rounds or directly to the patient safety specialist. In addition, a hospital hotline telephone system was introduced that accepted anonymous calls 24 hours a day. Callers were instructed to leave a voice message, which the patient safety specialist audited within 48 hours. Callers were encouraged to report errors made by themselves or others and even to report ideas for improvement of patient safety. Notices were placed in all patient care areas to encourage patients and family members to use the hotline. Events reported on the hotline were tracked separately. A record was kept as to whether the report was anonymous or identifiable. One person with several years’ experience [N.L.K.] in using the A-I classification of the National Coordinating Council for Medication Error and Reporting and Prevention19 (Table 2, page 38) prospectively coded the severity of all reported medication errors.
Analysis The entire program’s impact was measured on the basis of staff awareness and satisfaction, the rate of event reporting, and the harm caused to patients from medication errors. The study (January 2000–March 2003) was divided into three periods: ■ Baseline (January 2000–June 2001) ■ Transition (July 2001–March 2002) ■ Postintervention (April 2002–March 2003) The end of the baseline period was defined by the date on which the full-time patient safety specialist position started. The transition period was arbitrarily defined as lasting three quarters, by which time all the key interventions were put in place. The rate of medical events per 1,000 patient days, the rate of medication events per 10,000 doses dispensed by the hospital pharmacy, and the number of phone calls to the hotline per 1,000 patient days were compared across
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
37
Joint Commission
Journal on Quality and Safety
Table 2. National Coordinating Council (NCC) for Medication Error and Reporting and Prevention (MERP) Index for Categorizing Medication Errors* Category A Category B Category C Category D Category E Category F Category G Category H Category I
Circumstances or events that have the capacity to cause error An error occurred but the error did not reach the patient (an “error of omission” does reach the patient). An error occurred that reached the patient but did not cause patient harm. An error occurred that reached the patient and required monitoring to confirm that it resulted in no harm to the patient and/or required intervention to preclude harm. An error occurred that may have contributed to or resulted in temporary harm to the patient and required intervention. An error occurred that may have contributed to or resulted in temporary harm to the patient and required initial or prolonged hospitalization. An error occurred that may have contributed to or resulted in permanent patient harm. An error occurred that required intervention necessary to sustain life. An error occurred that may have contributed to or resulted in the patient’s death.
* Source: NCC MERP: NCC MERP Taxonomy of Medical Errors. http://www.nccmerp.org/pdf/taxo2001-07-31.pdf (last accessed May 10, 2004).
the three periods by use of the nonparametric KruskalWallis test. Differences in the proportion of callers who left their name or self-reported in the transition period compared with the postintervention period were compared by use of the chi-square test. Responses to the two questions in the patient safety questionnaire were compared using the nonparametric Mann-Whitney U test (the response data were not normally distributed).
Results As shown in Table 3 (page 39), event reporting increased significantly from a median of 35 events/1,000 patient days in the baseline period to 125 events/1,000 patient days in the postintervention period (p < .001 for the three periods, Kruskall-Wallis test). Table 3 shows the results of reporting rates via the hospital hotline telephone system, which also indicate a significant increase in reporting—from a median 3 calls/1,000 patient days during the baseline period to 23 calls/1,000 patient days during the postintervention period (p < .001 for the three time periods of the study, Kruskal-Wallis test). Further evidence of the cultural change is also shown in Table 3, which documents a significant increase in the willingness of hotline callers to provide their names when reporting errors (30% in the transition period compared with 61% in the postintervention period, p < .001, chi-
38
square test). In addition, a significant increase in the proportion of hotline callers who self-reported errors was observed in the postintervention period (14%) compared with the transition period (7%, p < .001, chi-square test). The rate of medication events reported was compared for both total medication events (A-I) and serious medication events (E-I events). The total number of medication events reported per 10,000 doses dispensed by the pharmacy increased significantly from a median of 4 in the baseline period to 6/10,000 doses in the transition period and approximately 30/10,000 doses dispensed in the postintervention period (p < .001, Kruskal-Wallis test). The rate of serious medication events (E-I) also increased significantly following implementation of the patient safety initiative from a median of 0.06/10,000 doses dispensed in the baseline period to 0.18 in the transition period and 0.19/10,000 doses dispensed in the postintervention period (p = .001, KruskalWallis test). The employee survey results demonstrated an increase in staff’s awareness of the importance of patient safety and willingness to report safety events. Table 3 shows a small but significant (p < .001) increase in staff satisfaction with the safety measures in place and less fear of punishment or retribution for reporting medical error. This result indicates that there has been ongoing improvement
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
Table 3. Outcomes for Baseline, Transition, and Postintervention Periods* Baseline Transition Postintervention Monthly Employee Survey; Scale low(1) – high (6) [Mean score+S.D.] Question 1: “ . . . protect patients’ 4.65±1.09 4.99±0.95 medical safety” Question 2: “ . . . without fear of 4.68±1.10 4.93±1.03 negative consequences” Hotline phone calls / 1,000 patient 2.83 (1.86–4.28) 10.84 (8.56–11.92) 22.79(19.51–26.27) days [Median (interquartile range)] Proportion of callers who left their 30.0 61.0 name Proportion of callers who self6.8 14.3 reported Medical events / 1,000 patient days 34.74 (33.72–36.16) 47.49 (41.30–52.04) 125.22 (107.70–158.99) [Median (interquartile range)] Total medication events / 10,000 4.16 (2.53–5.36) 5.60 (4.74–7.74) 30.16 (24.65–48.41) doses [Median (interquartile range)] E to I medication events / 10,000 0.06 (0–0.13) 0.18 (0.12–0.23) 0.19 (0.09–0.42) doses [Median (interquartile range)]
p value
< .001 < .001 < .001 < .001 < .001 < .001 < .001 = .001
* S.D., standard deviation.
in the hospital’s safety culture even more than a year after the program was implemented.
Discussion The need for cultural change is frequently cited, and general characteristics (for example, open communication, nonpunitive environment, leadership involvement) theoretically needed for success are well described, but the specific details of how to develop and implement such a culture are seldom provided.16 Some nonhealth care organizations have been able to demonstrate effective cultural changes with respect to safety; perhaps the best example comes from the aviation industry, where the Aviation Safety Reporting System (ASRS) has grown across 20 years to the point that more than 300,000 reports of safety incidents have been made.20 Yet in health care, evidence of a cultural change has yet to be reported. Bagian et al. called for a “prevention not punishment process that focuses on learning that leads to improvement rather than the traditional name and blame techniques.”6(p. 529) They employed an expert panel to determine the elements of this systems-level understanding and offered the following elements as essential:
■ A nonpunitive system ■ Timely feedback about solutions implemented as a result of reporting ■ Analysis of near misses to promote prevention strategy ■ Training in the use of the systems adopted These authors claimed to have achieved a 30-fold increase in events reported and a 900-fold increase in close calls of “high-priority events” that were reported. (No actual data or analysis of data were provided to permit verification of this claim.) Wong et al.8 described the implementation of a safety program that entailed the creation of a safety board, demonstration of patient safety as a top leadership priority, and promotion of a nonpunitive culture. They assessed the impact of this program using a monthly survey of the opinions of the safety board’s members but did not report on its effect on event reporting or on the hospital staff’s perception of the hospital’s safety culture. Ketring and White14 provided education on patient safety to hospital leaders, introduced a blame-free approach to human error, and established a committee structure specifically for patient safety. They stated that staff believed that there was more support for reporting
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
39
Joint Commission
Journal on Quality and Safety
errors and less blame and that there was less fear of negative consequences for reporting errors but offered no data in support of those conclusions. The clearest evidence that our hospital’s culture was changed lies in the magnitude of the overall increase in error reporting. The major increase was in the rate of near-miss reporting, but there was also an increase in the reporting of serious events. In addition, there was a steady and progressive increase in reporting via the hotline. These findings contrast with the much smaller increases in reporting that have been reported to date at national meetings on patient safety. The most direct evidence of a cultural shift was found in the results of the employee survey, in which employees were more likely to perceive a patient safety culture that encourages open communication and nonpunitive reporting of actual errors and near misses such that leadership is committed to the highest level of patient safety. Although initiatives designed to improve the culture of safety have been reported, none has documented a cultural change in a health care organization.6,9,12,21 It is often difficult to overcome the tendency in health care to identify and assign blame rather than to develop safer systems. Morbidity and mortality rounds have taught generations of physicians to identify who was responsible for a particular death or complication.22 There is evidence that physicians do not consider process problems when assessing quality and complications of care.23 A number of methodological limitations of this study should be noted. The focus groups were not rigorously designed, and the participants were not randomly selected from our hospital staff population, with the result that the subsequent conclusions could be biased by self-selection. The program was not implemented as a controlled trial, making it impossible to determine which of the numerous interventions was primarily responsible for the improvements documented. All the interventions were prospectively studied insofar as the same data elements were collected before, during, and after implementation of each intervention. The urgency of the need to improve patient safety appeared to outweigh the desire to study the efficacy of each intervention serially. The response rate to the mailed surveys to hospital staff was modest
40
but was typical of the response rates usually obtained in single-phase employee satisfaction surveys.24 The anonymous design of this survey did not allow us to query nonresponders to determine possible bias among responders. Another limitation is that a small number of the events reported occurred in outpatients, even though the data were shown as reported events per 1,000 patient days. A further limitation is that the events were not coded for severity in a blinded manner, introducing the possibility of bias. It was not possible to blind the coder as to the time period because all events were coded concurrently. The same coder, however, reviewed all events and had many years of experience with this classification system. It will be important to assess the impact of the observed cultural change and increases in error reporting on the actual harm caused to patients. We have been unable to document a reduction in the harm, as documented by staff reports, probably because we have seen such a large increase in overall reporting of errors, particularly near misses. Methods are needed to more objectively assess the magnitude of patient harm from medical errors. Rozich et al.25 have described the adverse drug event trigger tool, but this method relies on a detailed manual chart review. Administrative data sets have been used but are only useful as screening tools and do not provide a meaningful measure of harm for performance improvement teams.26 We are currently developing a methodology for the accurate and objective measurement of harm which does not entail laborintensive manual chart review.
Conclusion Cultural change is difficult to achieve because of the long time frame involved. The implementation of a carefully planned and orchestrated series of interventions designed to improve a hospital’s culture of patient safety, can, if led by senior hospital executives, lead to a substantial, profound, and lasting increase in medical event reporting and improvement in employee perceptions of the organization’s safety culture. J The authors are grateful to the staff of the Institute for Healthcare Improvement and the hospital participants in the Patient Safety Collaborative for freely sharing their ideas on patient safety; and Margie Olsen, Ph.D., M.P.H., for statistical advice and consultation.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
Max M. Cohen, M.D., M.H.S.A., is Vice President and Chief Medical Officer, Missouri Baptist Medical Center, BJC HealthCare, St. Louis; Nancy L. Kimmel, R.Ph., is Patient Safety Specialist; M. Kathleen Benage, R.N, is Director, Performance Improvement; and Cuong C. Hoang is Management Engineer, Missouri Baptist Medical Center. Thomas E. Burroughs, Ph.D., is Director, Center for Outcomes Research and Associate Professor, Department of Internal Medicine, St. Louis University, St. Louis. Carolyn A. Roth, R.N., J.D., is Director, Risk Management, Missouri Baptist Medical Center. Please address requests for reprints to Max M. Cohen, M.D., M.H.S.A., [email protected].
References 1. Institute of Medicine: To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press, 2000. 2. Public Law 106-129, 106th Congress: Healthcare Research and Quality Act of 1999. Sec 2. Amendment to the Public Health Service Act. Dec. 6, 1999. 3. The Leapfrog Group for Patient Safety: Rewarding Higher Standards. http://www.leapfroggroup.org (last accessed May 10, 2004). 4. Department of Health and Human Services (HHS) News: Secretary Thompson Announces HHS Patient Safety Task Force. Rockville, MD, Apr. 23, 2001. http://www.ahrq.gov/news/press/pr2001/pstfpr.htm (last accessed May 10, 2004). 5. Joint Commission on Accreditation of Healthcare Organizations: 2002 Comprehensive Accreditation Manual for Hospitals: The Official Handbook. Oakbrook Terrace, IL: Joint Commission Resources, 2001. 6. Bagian J., et al.: Developing and deploying a patient safety program in a large health care delivery system: You can’t fix what you don’t know about. Jt Comm J Qual Improv 27:522–532, Oct. 2001. 7. Larson E.B.: Measuring, monitoring, and reducing medical harm from a systems perspective: A medical director’s personal reflections. Acad Med 77:993–1000, Oct. 2002. 8. Wong P., Helsinger D., Perry J.: Providing the right infrastructure to lead the culture change for patient safety. Jt Comm J Qual Improv 28(7):363–372, Jul. 2002.
9. Mawji Z.,et al.: First do no harm: Integrating patient safety and quality improvement. Jt Comm J Qual Improv 28:373–386, Jul. 2002. 10. Leape L.L., Berwick D.M., Bates D.W.: What practices will most improve safety? Evidence-based medicine meets patient safety. JAMA 288:501–507, Jul. 21–27, 2002. 11. Shojania K.G., et al.: Safe but sound: Patient safety meets evidencebased medicine. JAMA 288:508–513, Jul. 14–31, 2002. 12. Layde P.M., et al.: Patient safety efforts should focus on medical injuries. 287:1993–1997, Apr. 17, 2002. Erratum in: JAMA 287(18):2363, May 8, 2002. 13. McNutt R.A., Abrams R., Aron D.C.: Patient safety efforts should focus on medical error. JAMA 287:1997–2001, Apr. 17, 2002. 14. Ketring S., White J.: Developing a systemwide approach to patient safety: The first year. Jt Comm J Qual Improv 28:287–295, Jul. 2002. 15. Mustard L.: The culture of patient safety. JONAS Healthc Law Ethics Regul 4:111–115, Dec. 2002. 16. Marx D.: Patient Safety and the “Just Culture”: A Primer for Healthcare Executives. http://www.mers-tm.net (last accessed May 10, 2004). 17. Cohen M.M., Eustis M.E., Gribbons R.: Changing the culture of patient safety: Leadership’s role in health care quality improvement. Jt Comm J Qual Saf 29:329–335, Jul. 2003. 18. Bagian J.P., Lee C.Z., Cole J.F.: A Method for Prioritizing Safety Related Actions. In Proceedings of Enhancing Patient Safety and Reduction of Errors in Health Care. Chicago: National Patient Safety Foundation, 1998 pp. 176–179. 19. National Coordinating Council for Medication Error and Reporting and Prevention: http://www.nccmerp.org (last accessed May 10, 2004). 20. Spencer F.C.: Human error in hospitals and industrial accidents: Current concepts. J Am Coll Surg 191:410–418, Oct. 2001 21. Frankel A., et al.: Patient Safety Leadership WalkRoundsTM. Jt Comm J Qual Saf 29:16–26, Jan. 2003. 22. Hamby L.S., et al.: Using prospective outcomes data to improve morbidity and mortality conferences. Curr Surg 57:384–388, Jul. 1, 2000. 23. Weingart S.N., et al.: Discrepancies between explicit and implicit review: Physician and nurse assessments of complications and quality. Health Serv Res 37:483–498, Apr. 2002. 24. Kaplowitz M.D., Hadlock T.D., Levine R.: A comparison of Web and mail survey response rates. Public Opin Q 68:94–101, 2004. 25. Rozich J.D., Haraden C.R., Resar R.K.: Adverse drug event trigger tool: a practical methodology for measuring medication related harm. Qual Saf Health Care 12:194–200, Dec. 2003 26. Romano P.S., et al.: A national profile of patient safety in U.S. hospitals. Health Aff (Millwood) 22:154–166, Mar.–Apr. 2000.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
41
Journal on Quality and Safety
Patient Safety
Implementing a Hospitalwide Patient Safety Program for Cultural Change
Max M. Cohen, M.D., M.H.S.A. Nancy L. Kimmel, R.Ph. M. Kathleen Benage, R.N. Cuong C. Hoang Thomas E. Burroughs, Ph.D. Carolyn A. Roth, R.N., J.D.
he Institute of Medicine (IOM) report on medical errors1 markedly heightened awareness of patient safety among health care professionals, politicians, the media, and the general public. A number of initiatives followed, including the passage of the Healthcare Research and Quality Act (1999),2 the formation of the Leapfrog Group,3 the formation of the National Patient Safety Task Force4—which included the Food and Drug Administraton, the Agency for Healthcare Research and Quality, the Centers for Disease Control and Prevention, and the implementation of new Joint Commission on Accreditation of Healthcare Organizations patient safety standards in 2001.5 A review of the literature on patient safety initiatives suggests that although much work is under way to rectify current deficiencies (fear of reporting errors, assignment of blame rather than seeking the causes, inadequate investigation of mistakes), there is still less than universal agreement as to the goal or how it will best be achieved.6–13 There is some agreement that if health care providers operate within a nonpunitive environment, their willingness to point out system weaknesses and to report accidents and near misses is substantially improved.8,14,15 Missouri Baptist Medical Center (MBMC), a member of BJC HealthCare, St. Louis, embarked on a newly reenergized patient safety program in 1999. We first determined the kind of culture that existed. We initially assessed the state of error* reporting to identify and
T
* In this article the terms error and event are used interchangeably, reflecting the practice at Missouri Baptist Medical Center.
34
Article-at-a-Glance Background: After focus groups revealed that staff perceived a punitive culture, Missouri Baptist Medical Center (MBMC) embarked on a comprehensive patient safety program, which was initially directed at creating a just culture of patient safety. Interventions: A series of structures, processes, and initiatives were introduced to change the attitudes of management and staff toward human error, to communicate broadly with staff and the community, and to provide feedback on leadership’s responses to specific events. All events reported were tracked continuously and recorded each month on a spreadsheet. Results: Total medical events reported by staff increased significantly (p < .001) from 35 to 132 per 1,000 patient days. Reports to the hotline alone increased significantly (p < .001) from 3 to 23 per 1,000 patient days, and the proportion of callers who left their names increased significantly (p < .001) from 30% to 61%. Survey results from staff showed a small but significant increase in awareness of patient safety and in comfort with reporting. Conclusion: The implementation of a carefully planned and orchestrated series of interventions designed to improve a hospital’s culture of patient safety can, if led by senior hospital executives, lead to a substantial, profound, and lasting increase in error reporting and improvement in employee perceptions of the organization’s safety culture.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
address performance gaps. In late 1999, we conducted a series of six focus groups with physicians, nursing staff, unit clerks, and pharmacy staff. During these focus groups, in which 68 persons participated, participants were asked if they had ever reported an error made either by themselves or by another, and if not, what the reasons were. They were also asked what needed to be done to make reporting easier and what leadership needed to do to reassure staff that human error would not be punished. Three key themes emerged from the focus groups: ■ The existing error-reporting forms were too timeconsuming and intimidating. ■ Staff did not consider it appropriate to report errors that did not reach the patient. ■ There was a pervasive fear of negative consequences. On learning that we had a distinctly punitive culture, we set out to change the culture to a “just and fair” one. Our hypothesis was that this would encourage error reporting, which in turn would allow us to identify and improve our imperfect processes.* This article describes the development and implementation of a comprehensive patient safety program for establishing a culture of patient safety in a community hospital. The program’s main objectives were, as examined in a prospective, observational study, to determine this program’s impact on two specific putative measures of the safety culture: event-reporting rates and surveys of staff opinion.
Methods Setting Missouri Baptist Medical Center (MBMC) is a not-forprofit, 489-bed nonteaching suburban community hospital that is part of BJC HealthCare, a 13-hospital integrated health delivery system.
Interventions Setting the Goal. In response to the findings of the focus groups, we set a primary goal to establish a more effective culture of patient safety in the hospital, as defined by “the willingness of all staff members to report all safety events and near misses without fear * These efforts were recognized by MBMC’s receipt of the inaugural 2002 American Hospital Quest for Quality PrizeSM, which honored leadership and innovation in the creation of a culture of patient safety.
of retribution, but with an understanding of accountability.” We encountered an immediate barrier—the difficulty in persuading middle management that human error should not be punished. Many middle managers greeted the initial description of the new culture as “nonpunitive” skeptically, often asking “How can we have a nonpunitive culture but still demand accountability?” Because certain behaviors clearly demand disciplinary action, we concluded that the term nonpunitive was misleading and confusing. The culture to which we aspired was renamed “just,” reflecting an approach described by David Marx.16 Marx posits that a just culture is one in which discipline occurs only for reckless or criminal behavior and where repeated errors by the same individual call for a remedial action plan and possibly for a readjustment in that individual’s role and responsibilities in the organization. This approach drove the changes that we made in the way that supervisors responded to human error. The interventions, driven by the results of the focus groups and from a literature review, were designed to do the following: ■ Promote event reporting ■ Encourage individual accountability ■ Forgive human error and even rule-breaking ■ Discipline staff for reckless behavior Beginning in early 2000 and continuing through the winter of 2001–2002, we instituted a series of structures, processes, and initiatives (Table 1, page 36) designed to change the management’s and staff’s attitudes toward human error, communicate broadly with staff and the community, and provide feedback on leadership’s responses to specific events. Critical Success Factors. Although all the initiatives and interventions listed in Table 1 were important to the development of the patient safety program, the following critical success factors were evident: ■ Senior management not only voiced commitment but also demonstrated its commitment in the form of highly visible behavior. ■ Although everyone was held responsible for patient safety, a reality of organizations is captured in the statement “common property is nobody’s property.” To overcome this, safety champions were identified within the medical staff as well as among the front-line employees to lead most initiatives.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
35
Joint Commission
Journal on Quality and Safety
Table 1. Components of the Patient Safety Program* I Executive Leadership ■ Executive rounds ■ Executive participation on safety teams ■ Executives emphasize patient safety at all hospital meetings. ■ Safety is monthly agenda item at board and medical executive committee meetings.
IV Event Investigation ■ Root cause analysis ■ Performance improvement teams ■ Small tests of change ■ Event prioritization and risk assessment V Prevention of Error ■
II Structural Change ■ ■ ■ ■ ■ ■ ■
■
Medication safety team Patient safety council Safety manager (environmental) Patient advocate Patient safety specialist Safety briefings on nursing units Modification of written human resources policies
III Event Reporting ■ ■ ■ ■
Anonymous reporting Simplified check-box reporting forms Safety hotline Common database for safety and risk management
■ ■
Failure Modes and Effects Analyses Response to newsletters from Institute for Safe Medication Practices Response to Joint Commission Sentinel Event Alerts Monitor literature for best practices
VI Feedback and Communication ■ Celebration of reporting ■ Electronic newsletter describing changes made in response to reported events ■ Employee brochure on patient safety and culture change ■ Patient safety awards program for safety ideas ■ Patient safety grants program to fund safety initiatives ■ Mind Your MedsSM community education program ■ Community education program on quality and safety
* Joint Commission, Joint Commission on Accreditation of Healthcare Organizations.
Although the technical activities implemented to improve patient safety were similar to those reported in other patient safety initiatives in the literature, the executive team passionately delivered the message that patient safety was a strategic priority.17 A high-level patient safety council was formed that was composed of executives and key directors, including the director of risk management, the director of performance improvement, and the patient safety specialist; these three positions reported to the chief medical officer [M.M.C.], who also chaired the council. Patient Safety Specialist. The full-time patient safety specialist position, which was created in 2001, allowed the hospital to coordinate all of its patient safety improvements, which sent the message that safety was a strategic priority and provided the staff with a single contact for all patient safety issues. The patient safety specialist
36
reviewed all reports within 48 hours of receipt and prioritized them for action using a modification of the Safety Assessment Code described by Bagian et al.18 In addition, the patient safety specialist relayed all errors to the appropriate manager for possible immediate action. Hospital staff members, medical staff members, and patients or family members (using a hotline) were encouraged to report any kind of concern about any aspect of medical care or the safety of the hospital environment. Other Actions. The cultural changes were communicated not only to the employees and medical staff but also to patients, who were viewed as partners in the maintenance of their own health, and to the community in general. A “Mind Your Meds”SM brochure, which was given at discharge to every patient, was designed to educate the public about medication safety. It included a detachable wallet-sized card on which current prescribed
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
medications, over-the-counter medications, and even herbal remedies could be recorded.
Survey Instruments The ongoing monthly employee satisfaction survey mailed to a stratified random sample of all hospital staff was used to track cultural change on a longitudinal basis. According to the sampling strategy, samples were stratified by functional departments within the hospital, such that every hospital employee is selected for surveying once a year. The surveying process is conducted by an external research firm, and responses are completely anonymous. Information is gathered on staff’s clinical department and job type for tabulation purposes. Response-bias weighting is used to adjust for any systemic response differences in these two characteristics. In June 2001 two specific items pertaining to patient safety were added to the survey: 1. This hospital has appropriate measures in place to protect patients’ medical safety. 2. This hospital provides an environment where staff can report medical errors and concerns without fear of negative consequences. The instrument uses a six-point Likert scale for each statement (1, strongly disagree; 2, disagree; 3, slightly disagree; 4, slightly agree; 5, agree; 6, strongly agree). A mean score is calculated for the response to each statement on the survey. From June 2001 until April 2002, an average of 96 randomly selected employees were surveyed each month, and the response rate was 36%. From May 2002 through March 2003, an average of 238 employees were surveyed monthly, and the average response rate was 34%. With the increase in the number of employees surveyed, each member of the workforce was surveyed at least once in the course of a single year.
Patient Safety Event Tracking System All medical errors, as identified from all reporting methods, were tracked continuously and recorded each month on a spreadsheet. Staff members were offered a choice of reporting vehicles: ■ Staff could use the existing incident report forms but were no longer required to sign them. ■ A much simplified anonymous check-box form was introduced for reporting medication errors.
■ Pharmacy staff members were encouraged to leave
copies of prescribing/transcription errors that they identified and the interventions they implemented in a box housed in their department. ■ Staff could report errors and risky situations during the brief safety “huddles” conducted at the change of shift on nursing units. ■ Staff could report errors to executives during weekly rounds or directly to the patient safety specialist. In addition, a hospital hotline telephone system was introduced that accepted anonymous calls 24 hours a day. Callers were instructed to leave a voice message, which the patient safety specialist audited within 48 hours. Callers were encouraged to report errors made by themselves or others and even to report ideas for improvement of patient safety. Notices were placed in all patient care areas to encourage patients and family members to use the hotline. Events reported on the hotline were tracked separately. A record was kept as to whether the report was anonymous or identifiable. One person with several years’ experience [N.L.K.] in using the A-I classification of the National Coordinating Council for Medication Error and Reporting and Prevention19 (Table 2, page 38) prospectively coded the severity of all reported medication errors.
Analysis The entire program’s impact was measured on the basis of staff awareness and satisfaction, the rate of event reporting, and the harm caused to patients from medication errors. The study (January 2000–March 2003) was divided into three periods: ■ Baseline (January 2000–June 2001) ■ Transition (July 2001–March 2002) ■ Postintervention (April 2002–March 2003) The end of the baseline period was defined by the date on which the full-time patient safety specialist position started. The transition period was arbitrarily defined as lasting three quarters, by which time all the key interventions were put in place. The rate of medical events per 1,000 patient days, the rate of medication events per 10,000 doses dispensed by the hospital pharmacy, and the number of phone calls to the hotline per 1,000 patient days were compared across
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
37
Joint Commission
Journal on Quality and Safety
Table 2. National Coordinating Council (NCC) for Medication Error and Reporting and Prevention (MERP) Index for Categorizing Medication Errors* Category A Category B Category C Category D Category E Category F Category G Category H Category I
Circumstances or events that have the capacity to cause error An error occurred but the error did not reach the patient (an “error of omission” does reach the patient). An error occurred that reached the patient but did not cause patient harm. An error occurred that reached the patient and required monitoring to confirm that it resulted in no harm to the patient and/or required intervention to preclude harm. An error occurred that may have contributed to or resulted in temporary harm to the patient and required intervention. An error occurred that may have contributed to or resulted in temporary harm to the patient and required initial or prolonged hospitalization. An error occurred that may have contributed to or resulted in permanent patient harm. An error occurred that required intervention necessary to sustain life. An error occurred that may have contributed to or resulted in the patient’s death.
* Source: NCC MERP: NCC MERP Taxonomy of Medical Errors. http://www.nccmerp.org/pdf/taxo2001-07-31.pdf (last accessed May 10, 2004).
the three periods by use of the nonparametric KruskalWallis test. Differences in the proportion of callers who left their name or self-reported in the transition period compared with the postintervention period were compared by use of the chi-square test. Responses to the two questions in the patient safety questionnaire were compared using the nonparametric Mann-Whitney U test (the response data were not normally distributed).
Results As shown in Table 3 (page 39), event reporting increased significantly from a median of 35 events/1,000 patient days in the baseline period to 125 events/1,000 patient days in the postintervention period (p < .001 for the three periods, Kruskall-Wallis test). Table 3 shows the results of reporting rates via the hospital hotline telephone system, which also indicate a significant increase in reporting—from a median 3 calls/1,000 patient days during the baseline period to 23 calls/1,000 patient days during the postintervention period (p < .001 for the three time periods of the study, Kruskal-Wallis test). Further evidence of the cultural change is also shown in Table 3, which documents a significant increase in the willingness of hotline callers to provide their names when reporting errors (30% in the transition period compared with 61% in the postintervention period, p < .001, chi-
38
square test). In addition, a significant increase in the proportion of hotline callers who self-reported errors was observed in the postintervention period (14%) compared with the transition period (7%, p < .001, chi-square test). The rate of medication events reported was compared for both total medication events (A-I) and serious medication events (E-I events). The total number of medication events reported per 10,000 doses dispensed by the pharmacy increased significantly from a median of 4 in the baseline period to 6/10,000 doses in the transition period and approximately 30/10,000 doses dispensed in the postintervention period (p < .001, Kruskal-Wallis test). The rate of serious medication events (E-I) also increased significantly following implementation of the patient safety initiative from a median of 0.06/10,000 doses dispensed in the baseline period to 0.18 in the transition period and 0.19/10,000 doses dispensed in the postintervention period (p = .001, KruskalWallis test). The employee survey results demonstrated an increase in staff’s awareness of the importance of patient safety and willingness to report safety events. Table 3 shows a small but significant (p < .001) increase in staff satisfaction with the safety measures in place and less fear of punishment or retribution for reporting medical error. This result indicates that there has been ongoing improvement
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
Table 3. Outcomes for Baseline, Transition, and Postintervention Periods* Baseline Transition Postintervention Monthly Employee Survey; Scale low(1) – high (6) [Mean score+S.D.] Question 1: “ . . . protect patients’ 4.65±1.09 4.99±0.95 medical safety” Question 2: “ . . . without fear of 4.68±1.10 4.93±1.03 negative consequences” Hotline phone calls / 1,000 patient 2.83 (1.86–4.28) 10.84 (8.56–11.92) 22.79(19.51–26.27) days [Median (interquartile range)] Proportion of callers who left their 30.0 61.0 name Proportion of callers who self6.8 14.3 reported Medical events / 1,000 patient days 34.74 (33.72–36.16) 47.49 (41.30–52.04) 125.22 (107.70–158.99) [Median (interquartile range)] Total medication events / 10,000 4.16 (2.53–5.36) 5.60 (4.74–7.74) 30.16 (24.65–48.41) doses [Median (interquartile range)] E to I medication events / 10,000 0.06 (0–0.13) 0.18 (0.12–0.23) 0.19 (0.09–0.42) doses [Median (interquartile range)]
p value
< .001 < .001 < .001 < .001 < .001 < .001 < .001 = .001
* S.D., standard deviation.
in the hospital’s safety culture even more than a year after the program was implemented.
Discussion The need for cultural change is frequently cited, and general characteristics (for example, open communication, nonpunitive environment, leadership involvement) theoretically needed for success are well described, but the specific details of how to develop and implement such a culture are seldom provided.16 Some nonhealth care organizations have been able to demonstrate effective cultural changes with respect to safety; perhaps the best example comes from the aviation industry, where the Aviation Safety Reporting System (ASRS) has grown across 20 years to the point that more than 300,000 reports of safety incidents have been made.20 Yet in health care, evidence of a cultural change has yet to be reported. Bagian et al. called for a “prevention not punishment process that focuses on learning that leads to improvement rather than the traditional name and blame techniques.”6(p. 529) They employed an expert panel to determine the elements of this systems-level understanding and offered the following elements as essential:
■ A nonpunitive system ■ Timely feedback about solutions implemented as a result of reporting ■ Analysis of near misses to promote prevention strategy ■ Training in the use of the systems adopted These authors claimed to have achieved a 30-fold increase in events reported and a 900-fold increase in close calls of “high-priority events” that were reported. (No actual data or analysis of data were provided to permit verification of this claim.) Wong et al.8 described the implementation of a safety program that entailed the creation of a safety board, demonstration of patient safety as a top leadership priority, and promotion of a nonpunitive culture. They assessed the impact of this program using a monthly survey of the opinions of the safety board’s members but did not report on its effect on event reporting or on the hospital staff’s perception of the hospital’s safety culture. Ketring and White14 provided education on patient safety to hospital leaders, introduced a blame-free approach to human error, and established a committee structure specifically for patient safety. They stated that staff believed that there was more support for reporting
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
39
Joint Commission
Journal on Quality and Safety
errors and less blame and that there was less fear of negative consequences for reporting errors but offered no data in support of those conclusions. The clearest evidence that our hospital’s culture was changed lies in the magnitude of the overall increase in error reporting. The major increase was in the rate of near-miss reporting, but there was also an increase in the reporting of serious events. In addition, there was a steady and progressive increase in reporting via the hotline. These findings contrast with the much smaller increases in reporting that have been reported to date at national meetings on patient safety. The most direct evidence of a cultural shift was found in the results of the employee survey, in which employees were more likely to perceive a patient safety culture that encourages open communication and nonpunitive reporting of actual errors and near misses such that leadership is committed to the highest level of patient safety. Although initiatives designed to improve the culture of safety have been reported, none has documented a cultural change in a health care organization.6,9,12,21 It is often difficult to overcome the tendency in health care to identify and assign blame rather than to develop safer systems. Morbidity and mortality rounds have taught generations of physicians to identify who was responsible for a particular death or complication.22 There is evidence that physicians do not consider process problems when assessing quality and complications of care.23 A number of methodological limitations of this study should be noted. The focus groups were not rigorously designed, and the participants were not randomly selected from our hospital staff population, with the result that the subsequent conclusions could be biased by self-selection. The program was not implemented as a controlled trial, making it impossible to determine which of the numerous interventions was primarily responsible for the improvements documented. All the interventions were prospectively studied insofar as the same data elements were collected before, during, and after implementation of each intervention. The urgency of the need to improve patient safety appeared to outweigh the desire to study the efficacy of each intervention serially. The response rate to the mailed surveys to hospital staff was modest
40
but was typical of the response rates usually obtained in single-phase employee satisfaction surveys.24 The anonymous design of this survey did not allow us to query nonresponders to determine possible bias among responders. Another limitation is that a small number of the events reported occurred in outpatients, even though the data were shown as reported events per 1,000 patient days. A further limitation is that the events were not coded for severity in a blinded manner, introducing the possibility of bias. It was not possible to blind the coder as to the time period because all events were coded concurrently. The same coder, however, reviewed all events and had many years of experience with this classification system. It will be important to assess the impact of the observed cultural change and increases in error reporting on the actual harm caused to patients. We have been unable to document a reduction in the harm, as documented by staff reports, probably because we have seen such a large increase in overall reporting of errors, particularly near misses. Methods are needed to more objectively assess the magnitude of patient harm from medical errors. Rozich et al.25 have described the adverse drug event trigger tool, but this method relies on a detailed manual chart review. Administrative data sets have been used but are only useful as screening tools and do not provide a meaningful measure of harm for performance improvement teams.26 We are currently developing a methodology for the accurate and objective measurement of harm which does not entail laborintensive manual chart review.
Conclusion Cultural change is difficult to achieve because of the long time frame involved. The implementation of a carefully planned and orchestrated series of interventions designed to improve a hospital’s culture of patient safety, can, if led by senior hospital executives, lead to a substantial, profound, and lasting increase in medical event reporting and improvement in employee perceptions of the organization’s safety culture. J The authors are grateful to the staff of the Institute for Healthcare Improvement and the hospital participants in the Patient Safety Collaborative for freely sharing their ideas on patient safety; and Margie Olsen, Ph.D., M.P.H., for statistical advice and consultation.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
Joint Commission
Journal on Quality and Safety
Max M. Cohen, M.D., M.H.S.A., is Vice President and Chief Medical Officer, Missouri Baptist Medical Center, BJC HealthCare, St. Louis; Nancy L. Kimmel, R.Ph., is Patient Safety Specialist; M. Kathleen Benage, R.N, is Director, Performance Improvement; and Cuong C. Hoang is Management Engineer, Missouri Baptist Medical Center. Thomas E. Burroughs, Ph.D., is Director, Center for Outcomes Research and Associate Professor, Department of Internal Medicine, St. Louis University, St. Louis. Carolyn A. Roth, R.N., J.D., is Director, Risk Management, Missouri Baptist Medical Center. Please address requests for reprints to Max M. Cohen, M.D., M.H.S.A., [email protected].
References 1. Institute of Medicine: To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press, 2000. 2. Public Law 106-129, 106th Congress: Healthcare Research and Quality Act of 1999. Sec 2. Amendment to the Public Health Service Act. Dec. 6, 1999. 3. The Leapfrog Group for Patient Safety: Rewarding Higher Standards. http://www.leapfroggroup.org (last accessed May 10, 2004). 4. Department of Health and Human Services (HHS) News: Secretary Thompson Announces HHS Patient Safety Task Force. Rockville, MD, Apr. 23, 2001. http://www.ahrq.gov/news/press/pr2001/pstfpr.htm (last accessed May 10, 2004). 5. Joint Commission on Accreditation of Healthcare Organizations: 2002 Comprehensive Accreditation Manual for Hospitals: The Official Handbook. Oakbrook Terrace, IL: Joint Commission Resources, 2001. 6. Bagian J., et al.: Developing and deploying a patient safety program in a large health care delivery system: You can’t fix what you don’t know about. Jt Comm J Qual Improv 27:522–532, Oct. 2001. 7. Larson E.B.: Measuring, monitoring, and reducing medical harm from a systems perspective: A medical director’s personal reflections. Acad Med 77:993–1000, Oct. 2002. 8. Wong P., Helsinger D., Perry J.: Providing the right infrastructure to lead the culture change for patient safety. Jt Comm J Qual Improv 28(7):363–372, Jul. 2002.
9. Mawji Z.,et al.: First do no harm: Integrating patient safety and quality improvement. Jt Comm J Qual Improv 28:373–386, Jul. 2002. 10. Leape L.L., Berwick D.M., Bates D.W.: What practices will most improve safety? Evidence-based medicine meets patient safety. JAMA 288:501–507, Jul. 21–27, 2002. 11. Shojania K.G., et al.: Safe but sound: Patient safety meets evidencebased medicine. JAMA 288:508–513, Jul. 14–31, 2002. 12. Layde P.M., et al.: Patient safety efforts should focus on medical injuries. 287:1993–1997, Apr. 17, 2002. Erratum in: JAMA 287(18):2363, May 8, 2002. 13. McNutt R.A., Abrams R., Aron D.C.: Patient safety efforts should focus on medical error. JAMA 287:1997–2001, Apr. 17, 2002. 14. Ketring S., White J.: Developing a systemwide approach to patient safety: The first year. Jt Comm J Qual Improv 28:287–295, Jul. 2002. 15. Mustard L.: The culture of patient safety. JONAS Healthc Law Ethics Regul 4:111–115, Dec. 2002. 16. Marx D.: Patient Safety and the “Just Culture”: A Primer for Healthcare Executives. http://www.mers-tm.net (last accessed May 10, 2004). 17. Cohen M.M., Eustis M.E., Gribbons R.: Changing the culture of patient safety: Leadership’s role in health care quality improvement. Jt Comm J Qual Saf 29:329–335, Jul. 2003. 18. Bagian J.P., Lee C.Z., Cole J.F.: A Method for Prioritizing Safety Related Actions. In Proceedings of Enhancing Patient Safety and Reduction of Errors in Health Care. Chicago: National Patient Safety Foundation, 1998 pp. 176–179. 19. National Coordinating Council for Medication Error and Reporting and Prevention: http://www.nccmerp.org (last accessed May 10, 2004). 20. Spencer F.C.: Human error in hospitals and industrial accidents: Current concepts. J Am Coll Surg 191:410–418, Oct. 2001 21. Frankel A., et al.: Patient Safety Leadership WalkRoundsTM. Jt Comm J Qual Saf 29:16–26, Jan. 2003. 22. Hamby L.S., et al.: Using prospective outcomes data to improve morbidity and mortality conferences. Curr Surg 57:384–388, Jul. 1, 2000. 23. Weingart S.N., et al.: Discrepancies between explicit and implicit review: Physician and nurse assessments of complications and quality. Health Serv Res 37:483–498, Apr. 2002. 24. Kaplowitz M.D., Hadlock T.D., Levine R.: A comparison of Web and mail survey response rates. Public Opin Q 68:94–101, 2004. 25. Rozich J.D., Haraden C.R., Resar R.K.: Adverse drug event trigger tool: a practical methodology for measuring medication related harm. Qual Saf Health Care 12:194–200, Dec. 2003 26. Romano P.S., et al.: A national profile of patient safety in U.S. hospitals. Health Aff (Millwood) 22:154–166, Mar.–Apr. 2000.
2004 Global Supplement Copyright 2004 Joint Commission on Accreditation of Healthcare Organizations
41