|Year : 2013 | Volume
| Issue : 4 | Page : 221-228
Developing an integrated clinical risk management model for Hospitals
Fatemeh Rezaei1, Mohammad H Yarmohammadian2, Masoud Ferdosi2, Abbas Haghshenas3
1 Health Services Administration, Faculty of Management and Medical Information, Isfahan University of Medical Sciences, Isfahan, Iran
2 Health Management and Economics Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
3 Family Physician, Faculty of Health, University of Technology of Sydney, Sydney, Australia
|Date of Web Publication||16-Apr-2014|
Mohammad H Yarmohammadian
Health Management and Economics Research Center, Isfahan University of Medical Sciences, Isfahan
Source of Support: None, Conflict of Interest: None
Context: Improving patient safety in health systems is one of the main priorities in healthcare systems, for this reason clinical risk management in organizations has become increasingly significant. Although several tools have been developed for clinical risk management, each has its own limitations. Aims: This study aims to develop a comprehensive tool that can complete the limitations of each risk assessment and management tools with the advantage of other tools. Settings and Design: Procedure was determined in two main stages included development of an initial model during meetings with the professors and literature review, then implementation and verification of final model. Subjects and Methods: This study is a quantitative − qualitative research. In terms of qualitative dimension, method of focus groups with inductive approach is used. To evaluate the results of the qualitative study, quantitative assessment of the two parts of the fourth phase and seven phases of the research was conducted. Purposive and stratification sampling of various responsible teams for the selected process was conducted in the operating room. Final model verified in eight phases through application of activity breakdown structure, failure mode and effects analysis (FMEA), healthcare risk priority number (RPN), root cause analysis (RCA), FT, and Eindhoven Classification model (ECM) tools. This model has been conducted typically on patients admitted in a day-clinic ward of a public hospital for surgery in October 2012 to June. Statistical Analysis Used: Qualitative data analysis was done through content analysis and quantitative analysis done through checklist and edited RPN tables. Results: After verification the final model in eight-step, patient's admission process for surgery was developed by focus discussion group (FDG) members in five main phases. Then with adopted methodology of FMEA, 85 failure modes along with its causes, effects, and preventive capabilities was set in the tables. Developed tables to calculate RPN index contain three criteria for severity, two criteria for probability, and two criteria for preventability. Tree failure modes were above determined significant risk limitation (RPN > 250). After a 3-month period, patient's misidentification incidents were the most frequent reported events. Each RPN criterion of misidentification events compared and found that various RPN number for tree misidentification reported events could be determine against predicted score in previous phase. Identified root causes through fault tree categorized with ECM. Wrong side surgery event was selected by focus discussion group to purpose improvement action. The most important causes were lack of planning for number and priority of surgical procedures. After prioritization of the suggested interventions, computerized registration system in health information system (HIS) was adopted to prepare the action plan in the final phase. Conclusion: Complexity of health care industry requires risk managers to have a multifaceted vision. Therefore, applying only one of retrospective or prospective tools for risk management does not work and each organization must provide conditions for potential application of these methods in its organization. The results of this study showed that the integrated clinical risk management model can be used in hospitals as an efficient tool in order to improve clinical governance.
Keywords: Failure modes and effective analysis, risk management, root cause analysis
|How to cite this article:|
Rezaei F, Yarmohammadian MH, Ferdosi M, Haghshenas A. Developing an integrated clinical risk management model for Hospitals. Int J Health Syst Disaster Manage 2013;1:221-8
|How to cite this URL:|
Rezaei F, Yarmohammadian MH, Ferdosi M, Haghshenas A. Developing an integrated clinical risk management model for Hospitals. Int J Health Syst Disaster Manage [serial online] 2013 [cited 2018 Apr 24];1:221-8. Available from: http://www.ijhsdm.org/text.asp?2013/1/4/221/130740
| Introduction|| |
With regard to clinical risk management as an important part of hospital management and clinical governance, reducing the likelihood of risks occurrence has become an important issue in clinical quality improvement for hospitals.  Therefore, to achieve an acceptable level of risk, risk management and risk analysis should be based on reasonable tools for healthcare systems to evaluate and provide alternative strategies to reduce risks. Various tools have their own advantages and disadvantages.  So, a comprehensive risk assessment tool should be able to overcome to disadvantages of each tools with applying of advantages of other tools.
A retrospective methodology for risk assessment in healthcare approved by JCAHO in 1997 is root cause analysis (RCA). Now, potential limitations of RCA have been identified by researchers and application of this mode faced with some constraint. ,, Then, in 2001, JCAHO required healthcare organization to implement failure mode and effects analysis (FMEA) according to Standard LD 5.2 Accreditation Manual. This methodology prioritizes identified failure modes to prevent adverse events of a selected process. , But, noncompliance of FMEA model for healthcare systems reduced its effectiveness.
RCA has a reactive and event-based analysis approach. Applying this model as the only approach to risk analysis is an important challenge because only information about an event is considered.  But FMEA has a preventive process-based analysis approach. Dr Tim Sandle compared risk assessment tools with five criteria included risk identification, risk analysis, risk evaluation, risk reduction, and risk acceptance. This study declared that FMEA has capability in doing all criteria except risk identification. On the contrary, FTA as a tool for RCA model has the most capability for risk identification.  So, we can conclude that applying just one model can not be sufficient for risk management and highlights requirement for considering an integrated model. Senders  also confirm that FMEA and RCA can not be separated from each other.
One of limitation of using RCA in healthcare organizations is focusing only on the most basic causes to purpose intervention, while all causes contributing in the event are crucial. On the contrary, a systematic RCA need to be done instead of considering each event separately. , So, we concluded that process-biased RCA would be more effective in reducing risk and purposed interventions could improve all contributing factors in event. In our study, we integrated process-biased FMEA with RCA in seeking to resolve this problem.
If FMEA implemented completely, it could identify all failure modes, their causes, and AEs. But any systems deal with human, confront with unpredictable behavior, and occurrence probability of an event could not be ensured. , With integrating RCA and FMEA, realities are checked and reliability of FMEA model enhances. In addition, reviewing all valuable scientific documentation in the field of selected process was purposed as a result of limitations of experiences and awareness of responsible individuals involved in each steps of the process. In this study, we use SURPASS checklist along with focus discussion group (FDG) comments.
One of optional steps in RCA model is prioritization of reported events as lots of events should be reviewed to determine their root causes. Other studies emphasized on prioritization of event through risk Matrix , Therefore, this study identified prioritized failure modes through FMEA and then events in areas of high prioritized failure modes were being explored through RCA model.
Despite many researches has been done in the field of RCA and FMEA models, applying these two models together has not been considered before. In our study, the accuracy of predicted failure modes in selected process put to the test with RCA model. So a fact-oriented logic is used and over time corrective actions modify with regard to reality through clarified weak points of our plan. Our study seeking to determine surgical AEs as 51%−79% of all AEs related to surgical wards and 43% of these AEs are preventable. ,
| Subjects and Methods|| |
Procedure was determined in two main stages included development of an initial model during meetings with the professional professors and literature review, then implementation and verification of final model in eight following phases:
Determine the most vulnerable and important process in the scope of study that has potentiality for AEs occurrences.
FDGs, a multidisciplinary team with involvement of responsible groups in each steps of the process, with considering inclusion criteria for this group included experiencing * seniority.
Mapping selected process in phase according to one of project management tools named activity breakdown structure (ABS) with considering FDG opinion, interviewing with responsible groups in each steps of the process, and scientific evidence in scope of study. 
Applying FMEA methodology for each steps of selected process:
- Prioritization of process phases according to its role in promoting patient safety and reducing the risk of nosocomial events. It should be done in FDG meeting with consensus of responsible groups and researcher made checklist
- Determine failure modes, its causes, effects, and preventable actions of the prioritized phase of the process according to FDG opinion, brain storming, interviewing, and considering scientific documentation
- Determine criteria for calculating severity, occurrence, and preventability through FDG meetings, interview, and literature review
- Calculating risk priority number (RPN) for failures of prioritized phase of the process
- Applying RCA methodology for reported events.
- Determine AEs resulted from failures of prioritized phase of the selected process, after 3 months of RPN calculation
- Revising RPN index for events result from failure modes which have gotten high score RPN number
- Determine root causes of AEs in previous step through fault tree (FT).
Classification of root causes through Eindhoven Classification model (ECM).
Proposing required interventions for the most important identified roots cause.
Prioritizing interventions according to two criteria included their ability to mitigate the contributing factor and team's belief that the intervention will be implemented and executed. ,
Developing an action plan for introduced prioritized intervention determined in seventh phase.
| Result|| |
The first phase
The most important process in the field of surgery determined all steps of patient's journey from getting hospitalization admission to discharge of ward surgery. To distinguish this process, we reviewed World Health Organization (WHO) guidelines, literature, and interview with FDG members. Result indicated that safety improvement of surgical processes should not be limited to the operating room. In addition, several studies have shown that the majority of surgical errors (53-70%) occur outside the operating room, before or after surgery. So, we considered entire surgical pathway. 
In the present study, stakeholders including matron (director of clinical governance unit) and two senior specialist of clinical governance unit, director of accreditation unit, head of operating room, head of anesthesiology, recovery head nurse, head of D. clinic, two general surgeons, and two nurses from the operating room and recovery all were enrolled as members.
According to the SURPASS checklist (a scientific documentation about preparing process of a patient for surgery), we provided a checklist for FDG members with making changes needed with regard to general condition of Iran operating rooms. They were asked to develop necessary steps for performing surgery with ABS tool. Finally, 5 main phases, 24 activities, and 108 tasks developed [Figure 1] and [Table 1].
|Figure 1: Process mapping of the 1-3 phase of surgical patient's journey according to activity breakdown structure tool|
Click here to view
|Table 1: Patient's journey process from getting hospitalization admission to discharge of ward surgery|
Click here to view
Applying FMEA methodology
1. The most important phases according to its role in promoting patient safety and reducing the risk of nosocomial events determined 3-1 phase. The justification for the selection was its closeness to main phase of the operation and more probability of preventing occurrence of events or unintentional hazards due to this phase contained the last defensive barriers
2. Twelve failure modes determined for 3-1 phase with regard to its activities and tasks. These include:
(1) Displacement operation report, name, and documents of patients; (2) Failure to identify patient by surgical team; (3) Incorrect counting of swaps and sponge, the needles and other retained foreign body; (4) Losing pathological specimens; (5) Inadequate equipment and supplies during surgery; (6) Mismatch of brought surgical instrument by companies with surgeon preference; (7) Not preparation of required surgical instrument; (8) Incomplete documentation of this step of the process; (9) Failure to complete delivery of the radiographic documents to anesthesia technician; (10) Failure to separate connectors to patients decontamination from blood and patient privacy; and (11) Changing patient positions without coordinating with the anesthesia team. These failures were analyzed in the form [Table 2]
|Table 2: Failure modes relevant to actions for patients before starting procedure in the scope of identification|
Click here to view
3. Criteria for calculating severity, occurrence, and preventability.
- For grading the severity tree domains of clinical injury, legal consequences, and wasted tme and costs were considered for each failure modes. The score of 10 indicates the most negative impacts and number 1 indicates the lowest one.
- For grading, the risk of occurrence two domains of occurrence per unit of time and exposure to risk were considered. The score of 10 indicates the most occurrences and number 1 indicates the lowest one.
- For grading preventability, two domains of likelihood of future occurrence and defensive barriers were considered. The score of 10 indicates the most preventability and number 1 indicates the lowest one.
4. Calculating RPN
Applying RCA methodology for reported events.
- Determine AEs − Among the 13 cases reported, 11 cases were related to 1-3 phase which indicates our forecast, 1-3 phase most is important phase of the patient journey, is considerable. Also, the most events occurred related to misidentification of patient. Moreover, as can be seen in [Table 3], failure to identify patient by surgical team (failure (2), incorrect counting of swaps and sponge, the needles and other retained foreign body (failure (3), and changing patient positions without coordinating with the anesthesia team (failure 11) were gotten the highest scores up to the defined obvious risk score (250). However, due to time constraints, we developed chronology of only events related to misidentification of patient. These events included blood transfusion mistake, surgery on the wrong patient, and wrong side surgery.
- Revising RPN index − during verification of RPN scores it was found that some effects, causes, and of preventable barriers can not be predicted through FMEA model. This is also true in relation to risk of patient identification. Items in [Table 2] were predictable but others appeared during incident occurrence. In terms of clinical injury and wasted time and costs, scores of only one event predicted the value. In terms of legal consequences, predicted values were consistent with what actually occurred in all three events. In terms of risk occurrence in two domains of occurrence per unit of time and exposure to risk, because there was no official reporting system and even near miss would not reported orally to head nurse, comparing RPN score what actually occurred in all three events was impossible. In terms of likelihood of future occurrence and defensive barriers predicted score was less than the actual amount. This could be due to the nature of misidentification failure because possibility of its occurrence exits in all phases of patient journey and as a result many defensive barriers should be gone through.
- Determine root causes − FT applied as a tool for RCA in our study. We found 25 causes for blood transfusion mistake, 19 causes for wrong side surgery, and 25 causes for wrong patient surgery. Five levels causes, with different numbers of causes for each level was depicted as shown in [Figure 2] and [Figure 3].
|Figure 3: Percentage of root causes in each of the classes of Eindhoven Classification model model|
Click here to view
|Table 3: Risk priority number of 13 failure modes of 3‑1 phase in order of priority|
Click here to view
Classification of root causes through ECM
As it was shown in the above chart, the most noticeable root causes of patient misidentification associated with active human error (about 53%). Than among active human error, rule-based behavior encompasses more than half of human errors. So, for prevention of bias due to RCA model use to analyze individual events separately, we classified root causes of each event into main categories and subcategories separately according to ECM model. Result indicated that HRV subcategory (human rule-Biased error for verification) was the most causes for blood transfusion mistake and OM subcategory (organizational management priority) was the most causes for wrong patient surgery.
Proposing required interventions for the most important identified roots cause.
Time constraint and FDG member consensus lead us to propose interventions just for wrong patient surgery related event. Among the well-known root causes, what was agreed between all members of FDG and had gotten the highest frequency according to subcategories of ECM models have been considered to propose interventions. These causes included lack of prioritization of patients registered in waiting list and lack of scheduling for a determined number of procedures in previous days, failure to inform surgery department and other clinical departments about priority of patients that should be prepared for surgery by a doctor, large number of surgeries, and excessive number of procedures in 1 day and not care about necessity to determine the number of procedures. Proposed interventions to solve these causes are visible in [Table 4]. ,
Developing an action plan
AP developed for the first strategy contained goal (improving the quality of health services), objective (risk management and improving patient safety), strategy (implementation of a computerized record system in HIS software in which surgeons are main user with receiving specified user code), and five phases of analysis, design, implementation, maintenance, and validation. Then, we defined activities of each phases with their responsible groups, necessary resource (time and cost), and success criteria.
| Discussion|| |
Among the many processes that occurring in the surgical department, FDG members consensus on patient's journey process from getting hospitalization admission to discharge of ward surgery. SURPASS checklist has designed in five phases started from preoperative measurement in surgical department or ward until measurement before discharge patient from hospital. But in our study, the last phase has been eliminated because most of possible failure modes could be encompassed until transfer phase [fifth phase - [Table 1]]. WHO surgical safety checklist has considered surgery process just in operating room and surgical and anesthetic team members are as the target group to complete checklist. So, our study is different with WHO vision as considered wider scope for surgery process. ,
We developed the process mapping of patient journey in 5 main phases, 24 activities, and 108 tasks [Figure 1] and [Table 1] by ABS tool. Cagliano et al.,  also applied ABS tool in their study to develop drug management process in 3 main phases and 22 activities. In addition, they used risk breakdown structure (RBS) approach to identify risk sources and then lowest RBS level are linked to the activities in the lowest ABS level by means of the risk breakdown matrix. Failure modes and waste analysis were done through designed tables with considering previous information. But, first of all we selected the most important phase and then applied risk analysis tools like FMEA.
In our study, preventability criteria for calculating RPN distinguished between preventability and detectability. This differentiation is considered also in Nobari et al.,  study as these researches is done for healthcare subject and preventability concept is more comprehensible for involving personnel. On the contrary, HFMEA model has eliminated detectability criteria and just considered severity and occurrence possibility, because it was believed that occurrence possibility criterion encompass concept of detectability by itself and there is low ability to detect failures in healthcare sector. 
The most important failure obtained from RPN index was failure to identify patient by surgical team [Table 2]. The misidentification failure could occur during all phases of surgical patient journey and if responsible team of each phases trust in identification of previous teams, any mistakes would continue during all phases of process, as far as lead to an AE. In addition, occurrence of this failure is possible for many proceedings like injection of medication and blood, application of implants and other surgical instruments, positions, and others.  Our study indicated that over period of 3 months, the highest occurred failure was misidentification errors. This implies its important role in improving the quality of care provided to patients.
As mentioned above, revising RPN was done by implementing RCA model 3 months later. The results indicated lack of generalizability of RPN scores in all tree misidentification AEs derived from one specific failure. Events had gotten their own specific RPN scores in triple criteria and in some cases they were in consistent with predictions [Table 2]. So, we concluded that anticipating just failures with their causes and effect in FMEA process may not be effective for introducing risk management strategies, because the relationship between a determined caused and an events is not clear. In addition, a retrospective study helps us to realize the inefficacy of our prevention programs despite existing defensive barriers like WHO surgical safety checklist and other identification instructions in our study. Therefore, in this study, the necessity of using both retrospective and prospective risk management approach is recommended.
However, the study conducted by Senders  noted that "If FMEA could do exactly what it is claimed to be, there would be no need for RCA. A complete proactive analysis would have identified all the causal sets and the outcomes that would have occurred. Unfortunately, things do not work out that way." This limitation is inevitable aspect of any analysis method dealing with human failures because human factors are unpredictable.
In the present study in order to determine the root causes, FT was drawn in five levels by FDG members and then edition done by individuals involved in the incident. Wagtendonk et al., study also conducted to analyze the root causes of unintended events (UEs) though causal tree. Among 881 reported UEs, two-thirds of them had a single root cause and only ·6% had four root causes. But in our study, the average number of root causes was found 23 for 3 AEs. According to their claims, the reason of small casual tree is event types they reviewed (UNs) as RCA for AEs are more complicated. The second reason for the relatively small causal trees is that only members of staff in the participating units were interviewed about reported events and not staff from other collaborating departments, but our study interviewed with all member involving in patient journey process. In addition, they excluded causes present in other units classiﬁed as external. It is possible that an external factor had more underlying root causes, but these were not examined. Finally, assumptions of reporters were not also considered but our study did not put on any restrictions on inclusion criteria for designing FTs. Due to the above study, the event types and the people who are interviewed to develop the root causes are significant factors in determining the root causes.
Wagtendonk et al., applied PRISMA − Medical tool in order to RCA and ECM model to classify causes. In consistent with our study, they prioritized causes to introduce interventions by referring to their frequency. In terms of main categories, the most frequent was related to human factors, organizational factors, patient-related factors, others, and technological factors and these arrangement is in consistent with our study. But subcategories showed different result. In our study HRV, OM, and technology − design (TD) subcategories have gotten the most frequency in arrangement. In Wagtendonk et al., study HRI, HRC, HRV, and H_ex (Human − rule-based intervention, coordination, veriﬁcation, and external) have gotten the most frequency in human-related categories. Organizational-culture and TD have gotten the most frequency in their own categories. 
In this study, the decision was made about how to schedule and plan for surgery department. Hereof, decision delineation should clarify items such as assignment of a date, a time indication, and an operating room or the allocation of capacity in tree levels contained medical disciplines, surgeon, and the patient level. At discipline level, two researchers report on an integer programming model in which operating room types and days of the week determines for each specialty (i.e. a decision concerning date and room). In our study, decision delineation was biased on surgeon level. So to implement selected intervention, like Cardoen et al.,  we introduced a software tool in which decisions made by surgeons, instead of disciplines. For each surgeon, as main user, the software imposes some restrictions to decide on what day and in which room surgeries have to be performed
One of the most important problems in implementing the RCA is a team that has been put in charge to do this task. The team is composed mostly of healthcare providers and is not able to master some nonclinical issues. If both expert team and clinical team collaborate together to explore for how to conduct analysis, design, implementation, maintenance, and validation phases, applicability of purposed interventions will be enhanced. So, we recommend that action plan must be developed when RCA method is used with considering views of other stakeholder groups and experts in the implementation of the intervention.
| Conclusion|| |
The increasing complexity of health care requires manager to have a multifaceted perspective. As mentioned in literature, so far, a lot of studies are conducted in healthcare related to risk management and they try to make innovation through completion of previous incompatible research for health system. So, their inefficiency becomes more and more clear with applying them in health system and motivates researchers to providing meaningful and understandable terminology by modifying logic of these models.
| References|| |
|1.||Chiozza ML, Ponzetti C. FMEA: A model for reducing medical errors. Clin Chim Acta 2009;404:75-8. |
|2.||Approaching risk assessment: Tools and methods [Internet]. Global Biopharmaceutical Resource, Inc; [Last accessed on 2013 June 11]. |
|3.||Wu AW, Lipshutz AK, Pronovost PJ. Effectiveness and efficiency of root cause analysis in medicine. JAMA 2008;299:685-7. |
|4.||Nicolini D, Waring J, Mengis J. Policy and practice in the use of root cause analysis to investigate clinical adverse events: Mind the gap. Soc Sci Med 2011;73:217-25. |
|5.||Palmieri PA, Peterson LT, Ford EW. Technological iatrogenesis: New risks force heightened management awareness. J Healthc Risk Manag 2007;27:19-24. |
|6.||Chan DT, Ng SS, Chong YH, Wong J, Tam YH, Lam YH, et al. Using ′failure mode and effects analysis′ to design a surgical safety checklist for safer surgery. Surg Pract 2010;14:53-60. |
|7.||Ahluwalia J, Marriott L. Critical incident reporting systems. Semin Fetal Neonatal Med 2005;10:31-7. |
|8.||Senders JW. FMEA and RCA: The mantras of modern risk management. Qual Saf Health Care 2004;13:249-50. |
|9.||Ursprung R, Gray J. Random safety auditing, root cause analysis, failure mode and effects analysis. Clin Perinatol 2010;37:141-65. |
|10.||Barkai J. Automatic generation of a diagnostic export system from Failure Modes and Effects Analysis (FMEA) Information, SAE Paper 99MF-24 (1998). |
|11.||Mengis J, Nicolini D. Root cause analysis in clinical adverse events. Nurs Manag (Harrow) 2010;16:16-20. |
|12.||van Wagtendonk I, Smits M, Merten H, Heetveld MJ, Wagner C. Nature, causes and consequences of unintended events in surgical units. Br J Surg 2010;97:1730-40. |
|13.||Walker IA, Reshamwalla S, Wilson IH. Surgical safety checklists: Do they improve outcomes? Br J Anaesth 2012;109:47-54. |
|14.||Cagliano AC, Grimaldi S, Rafele C. A systemic methodology for risk management in healthcare sector. Safety Science 2011;49:695-708. |
|15.||Zegers M, de Bruijne MC, de Keizer B, Merten H, Groenewegen PP, vander WG, et al. The incidence, root-causes, and outcomes of adverse events in surgical units: Implication for potential prevention strategies. Patient Safe Surg 2011;5:13. |
|16.||Pham JC, Kim GR, Natterman JP, Cover RM, Goeschel CA, Wu AW, et al. ReCASTing the RCA: An improved model for performing root cause analyses. Am J Med Qual 2010;25:186-91. |
|17.||de Vries EN, Prins HA, Crolla RM, den Outer AJ, van Andel G, van Helden SH, et al. SURPASS Collaborative Group. Effect of a comprehensive surgical safety system on patient outcomes. N Engl J Med 2010;363:1928-37. |
|18.||Park KW, Smaltz D, McFadden D, Souba W. The operating room dashboard. J Surg Res 2010;164:294-300. |
|19.||Cardoen B, Demeulemeester E, Beliën J. Operating room planning and scheduling: A literature review. Eur J Oper Res 2010;201:921-32. |
|20.||WHO. Surgical safety checklist; 2008. [last accessed on 2012 Nov 5]. |
|21.||de Vries EN, Hollmann MW, Smorenburg SM, Gouma DJ, Boermeester MA. Development and validation of the SURgical PAtient Safety System (SURPASS) checklist. Qual Saf Health Care 2009;18:121-6. |
|22.||Nobari FA, Tofighi S, Hafezimoghadam P, Maleki MR, Goharinezhad S. Risk assessment of processes of rasoule akram emergency department by the Failure Mode and Effects Analysis (FMEA) Methodology. Hakim Res J 2010;13:165-76. |
|23.||DeRosier J, Stalhandske E, Bagian JP, Nudell T. Using health care failure mode and effect analysis: The VA national center for patient safety′s prospective risk analysis system. Jt Comm J Qual Improv 2002;28:248-67, 209. |
|24.||24. Youngberg BJ. Principles of risk manage ment and patient safety. Sudbury: 2 nd ed. Jones and Bartlett Publishers; 2011. |
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4]