Challenges using Safety Monitoring Systems A review of Integrating Incident Data from Five Reporting Systems to Assess Patient Safety: Making Sense of the Elephant David Peavy Carter Seward Jake Swinton
Reference Osnat Levtzion-Korach, M.D. M.H.A.; Allan Frankel, M.D.; Hanna Alcalai, B.P.T., M.H.A; Carol Keohane B.S.N., R.N.; John Orav, Ph.D.; Erin Graydon-Baker, M.S., R.R.T.; Janet Barnes, R.N, J.D., C.P.H.R.M.; Kathleen Gordon, M.S.N., M.S.; Anne Louise Puopulo, B.S.N, R.N.; Elena Ivanova Tomov, M.B.A.; Luke Sato, M.D.; David W. Bates, M.D., M.Sc. (September 2010). Integrating Incident Data from Five Reporting Systems to Assess Patient Safety: Making Sense of the Elephant. The Joint Commission Journal on Quality and Patient Safety, Vol. 36, No. 9. Retrieved from: PSNet-ID pdf. Accessed February 6, PSNet-ID pdf 2 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Introduction ●The article investigates the various reporting systems and how they correlate to one another o Use different methods of obtaining information o Place results using different categories o Different personnel fill our reports o Prioritize different events ●The article reviewed the quality reporting systems/processes used at Brigham and Women’s Hospital (BWH) located in Boston, MA. o 747-bed tertiary care academic medical center o 52,000 inpatient admissions/950,000 outpatient admissions per year o 12,000 employees; approximately 3000 are doctors 3 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Methods ●BWH uses a total of 5 Quality Reporting Systems: o Incident Reporting o Reports to Risk Management o Patient Complaints o Executive Walk Arounds o Malpractice Claims ●The authors wanted to: o Evaluate what type of information is received by each system o Develop a common framework for representing the identified safety issues o Assess the correlation between types of information collected o Evaluate the overall safety picture 4 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Data ●The data was collected from the source systems over a 22 month period between May 10th, 2004 and February 28th, 2006 ●The 22 month time period had a small number of malpractice claims, so to increase this number, the time period was extended to 10 years. 5 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Assessment Areas ●Incident Reporting System o Commercially available web- based incident reporting system o Usually filled out by nurses o Evaluated by Department Managers ●Hospital Risk Management o Physicians and Nurses call the team to report adverse events and poor patient outcomes o Even split of reporting between Nurses and Physicians o Evaluated by Risk Management staff 6 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Assessment Areas (Cont.) ●Patient Complaints o The Family and Patient Relations Department Responds to patient and family complaints, suggestions, and compliments ●Executive Walk Around o Executive leadership walk arounds occur semiweekly and engage with staff about safety concerns ●Malpractice Claims o The malpractice insurer has a system used for data collection called CMAPS. o This system takes information from potential claim reports risk managers, and formal malpractice claims 7 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Results 8 ETM Peavy, Seward & SwintonSafety Monitoring Systems (Page 404; Levtzion-Korach, M.D. et. al.)
Comparisons ●The five assessment areas did not have overlapping categorization schemes ●The experimenters developed a categorization scheme with 23 major categories to compare the 5 reporting systems. ●Examples: o Staff o Staff Training/Education o Technical Skills o Equipment/Supplies o Medication Errors o Identification Issues o Monitoring of Patient o Clinical Judgement o Medical Records o Test Reconciliation Issues 9 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Analysis ●Data collected by the five reporting systems was analyzed and classified into one of the new categories by a physician reviewer trained in patient safety. ●Investigators calculated the frequencies of each of the events ●Overlap between rankings of frequency of different categories were then assessed and correlations between the detection approaches were calculated 10 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Categorical Analysis ●The leading major categories across the reporting systems were: o Communication (11.6%) o Technical Skills (10.9% o Clinical Judgement (9%) ●However, each system had a different category that was most frequent 11 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Statistical Correlation ●Overall, across the five systems, Cronbach’s standardized alpha was 0.22, suggesting a low level of consistency 12 ETM Peavy, Seward & SwintonSafety Monitoring Systems (Page 408; Levtzion-Korach, M.D. et. al.)
Correlations ●Correlations presented: o Highest correlation observed was between risk management reports and malpractice claims o A high correlation was found between patient complaints and malpractice claims o The incident reporting system has little correlation with the other systems o Executive walk rounds had a negative correlation with all other systems 13 ETM Peavy, Seward & SwintonSafety Monitoring Systems
14 ETM Peavy, Seward & SwintonSafety Monitoring Systems (Page 403; Levtzion-Korach, M.D. et. al.)
15 ETM Peavy, Seward & SwintonSafety Monitoring Systems (Page 403; Levtzion-Korach, M.D. et. al.)
Discussion ●Some things to keep in mind about the study: o Only a single medical center was used. Results shown may not be generalizable to other facilities o Not all possible types of safety data were included. However, BWH allowed the authors to closely examine numerous independent data source 16 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Discussion (cont’d) ●The study allowed for the comparison and evaluation of each reporting system, both individually, and as a whole ●Each system provides a different set of partial conclusions, and there is little correlation between individual systems 17 ETM Peavy, Seward & SwintonSafety Monitoring Systems
Conclusion ●Studies or reports commenting on only one of the reporting systems are incomplete ●Much better, more complete conclusions may be drawn from a complete perspective that utilizes information from multiple reporting systems ●Hospitals should utilize a more broad approach to the classification of adverse events in order to make more informed decisions 18 ETM Peavy, Seward & SwintonSafety Monitoring Systems