search text   Advanced Search

This Article Abstract Full Text (PDF) P3Rs: Submit a response Alert me when this article is cited Alert me when P3Rs are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via ISI Web of Science (25) Citing Articles via Google Scholar Google Scholar Articles by Taylor, J. A. Articles by Shafii, J. Search for Related Content PubMed PubMed Citation Articles by Taylor, J. A. Articles by Shafii, J.

Use of Incident Reports by Physicians and Nurses to Document Medical Errors in Pediatric Patients

James A. Taylor, MD^*, Dena Brownstein, MD, Dimitri A. Christakis, MD, MPH, Susan Blackburn, RN, PhD, Thomas P. Strandjord, MD, Eileen J. Klein, MD, MPH and Jaleh Shafii, RN, MS^||

^* Developmental Center for Evaluation and Research in Pediatric Patient Safety
Department of Pediatrics, University of Washington and Children's Hospital and Regional Medical Center, Seattle, Washington
University of Washington School of Nursing, Seattle, Washington
^|| Children's Hospital and Regional Medical Center, Seattle, Washington

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Objectives. To describe the proportion and types of medical errors that are stated to be reported via incident report systems by physicians and nurses who care for pediatric patients and to determine attitudes about potential interventions for increasing error reports.

Methods. A survey on use of incident reports to document medical errors was sent to a random sample of 200 physicians and nurses at a large children's hospital. Items on the survey included proportion of medical errors that were reported, reasons for underreporting medical errors, and attitudes about potential interventions for increasing error reports. In addition, the survey contained scenarios about hypothetical medical errors; the physicians and nurses were asked how likely they were to report each of the events described. Differences in use of incident reports for documenting medical errors between nurses and physicians were assessed with ² tests. Logistic regression was used to determine the association between health care profession type and likelihood of reporting medical errors.

Results. A total of 140 surveys were returned, including 74 from physicians and 66 by nurses. Overall, 34.8% of respondents indicated that they had reported <20% of their perceived medical errors in the previous 12 months, and 32.6% had reported <40% of perceived errors committed by colleagues. After controlling for potentially confounding variables, nurses were significantly more likely to report 80% of their own medical errors than physicians (odds ratio: 2.8; 95% confidence interval: 1.3–6.0). Commonly listed reasons for underreporting included lack of certainty about what is considered an error (indicated by 40.7% of respondents) and concerns about implicating others (37%). Potential interventions that would lead to increased reporting included education about which errors should be reported (listed by 65.4% of respondents), feedback on a regular basis about the errors reported (63.8%) and about individual events (51.2%), evidence of system changes because of reports of errors (55.4%), and an electronic format for reports (44.9%). Although virtually all respondents would likely report a 10-fold overdose of morphine leading to respiratory depression in a child, only 31.7% would report an event in which a supply of breast milk is inadvertently connected to a venous catheter but is discovered before any breast milk goes into the catheter.

Conclusions. Medical errors in pediatric patients are significantly underreported in incident report systems, particularly by physicians. Some types of errors are less likely to be reported than others. Information in incident reports is not a representative sample of errors committed in a children's hospital. Specific changes in the incident report system could lead to more reporting by physicians and nurses who care for pediatric patients.

Key Words: medical errors • incident reports

Abbreviations: CHRMC, Children's Hospital and Regional Medical Center • OR, odds ratio • CI, confidence interval

Medical errors have been defined as a major public health problem in the United States. In a comprehensive review conducted by the Institute of Medicine, it was estimated that 44 000 to 98 000 deaths occur each year as a result of medical errors.¹ If these estimates are accurate, then medical errors rank as the eighth leading cause of death in the United States.¹

For understanding and, ultimately, preventing medical errors, data on the occurrence, frequency, types, and results of such events are crucial. Undoubtedly, the most commonly used method for detecting adverse patient events is through the use of incident reports. Incident reports are used for a myriad of purposes, including personnel credentialing and review.² Because of this and other features of the system, there is significant underreporting of medical errors through this mechanism.^3,4 The impact of error type and severity on the rate of underreporting is unknown, further complicating interpretation of these already incomplete data sets.

To characterize better the issues that lead to underreporting of medical error in hospitalized pediatric patients, we conducted a survey of physicians and nurses who work in a children's hospital about their attitudes and use of the incident report system. Areas specifically surveyed included the frequency and completeness of reporting medical errors, reasons for underreporting, possible methods for increasing reporting, and types of errors that were more or less likely to be reported.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Setting
The study was conducted at Children's Hospital and Regional Medical Center (CHRMC), Seattle, Washington. CHRMC is both a community hospital for children and the primary pediatric referral center for the Northwest and provides both medical and surgical care. CHRMC has 220 beds and is the primary training site for pediatric residents at the University of Washington. At CHRMC, staff members are responsible for completing incident reports when there is a significant patient injury and are encouraged to report all medical errors, including close-call events. Typically, nurses complete the actual form, either self-initiated or at the request of another member of the health care team. Data for this study were collected in 2002; in 2001, 2506 incident reports were completed at CHRMC.

Participants
For the study, a random sample of 100 nurses and 100 physicians were mailed a survey regarding their attitudes and experiences using the incident report system for conveying information on medical errors in hospitalized children. Only registered nurses who provide direct patient care were eligible for selection to receive the survey; nurses in managerial or administrative positions, nurse practitioners, and licensed practical nurses were excluded. Pediatric residents and regular faculty physicians were eligible to receive the survey; private practice physicians were excluded unless they were also members of the University of Washington faculty.

Survey
The survey was developed by the University of Washington Developmental Center for Evaluation and Research in Pediatric Patient Safety. Organizationally, the developmental center consists of a core group of researchers that includes 4 academic pediatricians and an academic nurse. This core research group is advised by a multidisciplinary advisory team that consists of individuals with specific expertise. Membership on the team includes pediatric nurses, pediatricians, medical information technology experts, epidemiologists, attorneys, risk management specialists, and an aviation safety expert.

In addition to items that characterized the respondent, such as profession (physician or nurse), year of graduation from nursing/medical school, and years of employment at CHRMC, the survey was divided into 5 sections. Section 1 included items regarding the number of times in the past year that the responding physician or nurse had completed an incident report (or asked for one to be completed) regarding a medical error, the percentage of errors committed by the respondent in the past year that he or she had reported, and the percentage of errors committed by others that he or she had reported. Several possible responses were provided for each item; physicians and nurses were asked to indicate which response most accurately reflected their experience. The second section of the survey dealt with reasons for underreporting of medical errors. Several possible explanations were listed, and the respondent was asked to check all that were applicable. A response indicating that the nurse or physician reported all of his or her errors was also listed. Finally, the respondent was asked to record any reason for underreporting that was important to him or her but not listed. A similar format was used for the third section of the survey, dealing with possible interventions for increasing the reporting of medical errors via the incident report system. For section 4, the responding nurse or physician was asked his or her attitudes about a confidential or anonymous system for reporting medical errors.

The final section of the survey included hypothetical clinical scenarios describing medical errors in pediatric patients. After each scenario, the respondent was asked how likely she or he would be to complete an incident report on the described event, using a 6-point Likert scale, with responses ranging from "very likely" to "very unlikely." A 3-step process was conducted to develop the scenarios used in the survey. First, with the use of a 3 x 3 matrix for classifying medical errors by severity (serious, moderately serious, and trivial) and occurrence to the patient (happened, close-call, detected early), each of the 5 members of the core research group for the developmental center composed 1 or more scenarios for each of the 9 possible categories of medical errors. A medical error was considered to have "happened" when it reached the patient (eg, wrong medication administered), a "close-call" when it was discovered just before reaching the patient (eg, an incorrect medication is almost hung at the bedside), or "detected early" when it was detected before beginning implementation (eg, a nurse or a pharmacist notes that an incorrect medication dose has been ordered). Each of the scenarios was classified independently by the members of the multidisciplinary advisory team and rated for realism. The final survey included the highest rated scenario from each of the 9 categories of errors. The scenarios are listed, along with their classification, in Fig 1.

View larger version (56K): [in this window] [in a new window]   Fig 1. Clinical scenarios selected for survey with classification based on consensus.

Analysis
For the analysis, the rate of reporting perceived medical errors via incident reports, proportion of perceived errors reported, and proportion of perceived errors by colleagues that were reported by nurses or physicians were compared with the use of ² tests. The independent association between professional status (physician vs nurse) on likelihood of reporting a high proportion of perceived errors was determined by dichotomizing responses indicating reporting of 80% of perceived errors made by the respondent or a colleague and those reporting <80%. Logistic regression was performed on both of these outcomes. The odds ratio (OR) and 95% confidence interval (95% CI) for nurses who reported 80% of perceived errors, as compared with physicians, was calculated after adjusting for year of graduation from nursing/medical school and number of years at CHRMC.

The frequency of each potential reason for underreporting medical errors and for suggestions to increase reporting indicated by the respondents was determined. The denominator for these calculations was the number of valid responses for each item, defined as a nurse or a physician indicating that 1 or more reasons were applicable to him or her. The proportion of respondents who indicated that a confidential or an anonymous reporting system would make them more or less likely to report medical errors was also calculated.

The proportion of responding nurses and physicians who indicated that they would be "likely" or "very likely" to report each of the medical errors described in the clinical scenarios was computed. The Likert scale responses were also transformed to an ordinal scale, ranging from 1 for a response of "very unlikely" to 6 for a response of "very likely." The likelihood of reporting by nurses and physicians of each medical error described was compared with the use of Mann-Whitney tests. For all statistical tests, differences were considered to be significant at P < .05 or when the 95% CI did not include 1.0. The study was approved by the CHRMC Institutional Review Board.

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 
Of the initial 200 physicians and nurses who received the survey, 16 had left CHRMC, were no longer involved in patient care, or were on extended leave. An additional 34 potential respondents declined participation in the study. Replacements to receive the survey were selected for each of these individuals. A total of 140 completed surveys were returned, 74 by physicians and 66 by nurses; 12 of the responding physicians were residents. Thus, 70% of eligible participants responded to the survey.

Data on survey responses regarding reporting of medical errors are summarized in Tables 1 and 2. Nurses reported more frequent use of the incident report system than physicians (P < .001). As is shown in Table 1, 45.9% of physicians completed 0 incident reports during the previous 12 months (or asked to have one completed) as compared with 10.5% of the responding nurses. Nurses also reported higher proportions of their own perceived medical errors and errors made by others than physicians (P < .001 and P = .007, respectively). As is seen in Table 2, 52.2% of physicians indicated that they reported <20% of their perceived medical errors via the incident report system, whereas 56.9% of the nurses responded that they reported 80% of their errors. After adjusting for year of graduation and years of service at CHRMC, nurses were significantly more likely to report 80% of their own perceived medical errors than physicians (OR: 2.8; 95% CI: 1.3–6.0). However, after controlling for these characteristics, there was no significant difference between nurses and physicians in reporting 80% of perceived medical errors committed by colleagues (OR: 1.2; 95% CI: 0.6–5.7)

The percentage of respondents who indicated that they would be "likely" or "very likely" to report the medical errors described in hypothetical scenarios is shown in Table 5. Overall, the physicians and nurses who completed surveys indicated that they would be more likely to report serious events than trivial and were more likely to report events that happened to the patient than close calls or errors that were detected early. Thus, 99.3% of the doctors and nurses who completed surveys indicated that they would be "likely" or "very likely" to report a morphine overdose leading to respiratory depression in a young child (serious/happened medical error), whereas only 6.4% would report a event in which a pharmacist catches an overdose of amoxicillin before sending it to the floor (trivial/detected early). However, only 31.7% of respondents indicated that they would report an event in which a nurse determines that she or he has connected a supply of breast milk to a central venous catheter instead of the nasogastric tube in a premature infant (serious/close call). Nurses were significantly more likely than physicians to report 7 of the 9 medical errors described in the scenarios; there was no difference between the 2 groups in the likelihood of reporting an amoxicillin dosing error that occurs and is detected in the pharmacy (P = .85) or in a nurse's detecting that he or she has connected the breast milk supply to a central venous catheter (P = .07).

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

Perhaps more troubling, we found that the medical errors that are likely to be reported via incident reports are not a representative sample of all errors committed. Physicians and nurses were more likely to report some types of errors than others. In general, serious errors were more likely to be reported than those that were less serious, and events that occurred to the patient were more likely to be reported than close calls. Reporting close-call events has been cited as a key element in improving patient safety.⁵ Perhaps the most striking finding in our survey was that only 31.7% of respondents would be likely or very likely to report a close-call event in which a supply of breast milk is inadvertently connected to a central venous catheter. This medical error, if completed, is potentially fatal, anecdotally is more common than realized, and is totally preventable by making different types of connectors for the breast milk supply and intravenous fluids. However, no correction can be made unless close-call events are reported.

We found that physicians reported only a limited proportion of their errors and were less likely to report than nurses. This suggests that a disproportionate number of errors committed by nurses are included in the incident reports. Conversely, there may be virtually no information on physician-to-physician errors or physician-only errors, when a nurse is not involved.

The results of our study should be interpreted cautiously. Although we randomly selected physicians and nurses for the survey, it is possible that those who responded had different attitudes about incident reports than nonrespondents, thus biasing our results. Our survey was conducted in 1 children's hospital with a specific incident reporting system; some of findings may not be generalizable to other institutions. Finally, we measured the stated behaviors of physicians and nurses in regard to use of incident reports; actual use of this system may be different.

Much of the previous research on the attitudes of health professionals about reporting medical errors has been done in other specialties or focused on specific types of events, such as medication errors.^6,7 Vincent et al⁸ surveyed midwives and obstetricians regarding their attitudes about reporting adverse events. Like us, these investigators found that different types of events were likely to be reported at different rates, with serious events the most likely to be documented. The responding obstetricians indicated that they were less likely to report adverse events than midwives who completed the survey. This is consistent with our finding that pediatricians were less likely to report a medical error than nurses. Many of the reasons listed by the responding obstetricians and midwives for not reporting adverse events and recommendations for increasing reporting were similar to what we found in our survey. However, in the study by Vincent et al, 23% of respondents indicated that they underreported adverse events because of worries about litigation. In our survey of both nurses and pediatricians, we did not specifically ask about malpractice fears as a barrier to reporting medical errors; none of 47 respondents who wrote in reasons for underreporting mentioned concerns about possible litigation.

The results of our survey highlight some of the inherent problems with using incidents reports to monitor medical errors. Several methods for measuring errors have been used; the various techniques have different strengths and weaknesses.⁹ In the early studies on patient safety, detailed review of a sample of medical records was used to detect errors.^10,11 This method has advantages in that a rate of medical errors can be determined. In addition, retrospective review provides an opportunity to assess the patient outcomes that result from a medical error. However, the review of medical records is significantly limited by the completeness of the written record. It is a virtual certainty that, because of a lack of documentation, many errors are missed with this technique, particularly close-call events. Self-report of specific events has been used as a method for collecting information about medical errors, particularly in fields such as anesthesiology.¹² The applicability of this method to other areas in medicine is unclear. Direct observation of clinical activities, such as medication administration, is a particularly powerful tool for detecting errors.¹³ Unfortunately, this technique is extremely labor intensive and may make some health care professionals uncomfortable.⁹ Finally, regardless of the method used, merely documenting errors will not result in improved pediatric patient safety unless the data collected contain accurate and complete information and changes in the systems of care provided are implemented on the basis of the reports.

Given the difficulties with some of the other techniques, it is likely that passive reporting systems, such as incident reports, will remain the most common source of information about medical errors in pediatrics. Thus, it is important that efforts be made to maximize reporting. Our survey provides insights on several approaches to accomplish this. First, a substantial educational effort, aimed at nurses and, in particular, physicians, about which types of events should be reported and how to report errors is needed. The mechanisms used to report errors should be easily accessible and require only a small amount of time; electronic formats are preferable. It is crucial that health care professionals receive regular feedback both on the types and frequencies of errors that are reported and on information on specific examples of system changes being made in response to the identified sources of medical errors. Finally, our survey results suggest that continuing efforts to change the culture around medical errors are needed. Ultimately, reporting medical errors should be seen not as a punitive exercise but as an essential ingredient in providing optimal patient care.¹³

	ACKNOWLEDGMENTS

 
This study was funded by a grant from the Agency for Healthcare Research and Quality.

	FOOTNOTES

 
Accepted Apr 26, 2004.

Reprint requests to (J.A.T.) Child Health Institute, University of Washington, Box 354920, Seattle, WA 98915-4920. E-mail: uncjat{at}u.washington.edu

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION REFERENCES

 

1. Kohn LT, Donaldson MS, eds. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 2000
2. McDonough WJ. Systems for risk identification. In: Carroll R, ed. Risk Management Handbook for Health Care Organizations. 3rd ed. San Francisco, CA: Josey-Bass Inc; 2001:171–189
3. Cullen DJ, Bates DW, Small SD, Cooper JB, Nemeskal AR, Leape LL. The incident reporting system does not detect adverse drug event: a problem for quality improvement. Jt Comm J Qual Improv. 1995;21 :541 –548[Medline]
4. Flynn EA, Barker KN, Pepper GA, Bates DW, Mikeal RL. Comparison of methods for detecting medication errors in 36 hospitals and skilled-nursing facilities. Am J Health Syst Pharm. 2002;59 :436 –446[Abstract/Free Full Text]
5. Stalhandske E, Bagian JP, Gosbee J. Department of Veteran Affairs patient safety program. Am J Infect Control. 2002;30 :296 –302[CrossRef][ISI][Medline]
6. Wakefield DS, Wakefield BJ, Uden-Holman T, Blegen MA. Perceived barriers in reporting medication administration errors. Best Pract Benchmarking Healthc. 1996;1 :191 –197[Medline]
7. Wakefield DS, Wakefield BJ, Uden-Holman T, Borders T, Blegen M, Vaughn T. Understanding why medication administration errors may not be reported. Am J Med Qual. 1999;14 :81 –88[Abstract]
8. Vincent C, Stanhope N, Crowley-Murphy M. Reasons for not reporting adverse incidents: an empirical study. J Eval Clin Pract. 1995;5 :13 –21
9. Thomas EJ, Petersen LA. Measuring errors and adverse events in health care. J Gen Intern Med. 2003;18 :61 –67[CrossRef][ISI][Medline]
10. Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med. 1991;324 :370 –376[Abstract]
11. Thomas EJ, Studdert DM, Burstin HR. Incidence and types of adverse events and negligent care in Utah and Colorado. Med Care. 2000;38 :261 –271[CrossRef][ISI][Medline]
12. Webster CS, Merry AF, Larsson L, McGrath KA, Weller J. The frequency and nature of drug administration error during anaesthesia. Anaesth Intensive Care. 2001;29 :494 –500[ISI][Medline]
13. Barker KN, Flynn EA, Pepper GA, Bates DW, Mikeal RL. Medication errors observed in 36 health care facilities. Arch Intern Med. 2002;162 :1897 –1903[Abstract/Free Full Text]

This article has been cited by other articles:

				J. Garbutt, A. D. Waterman, J. M. Kapp, W. C. Dunagan, W. Levinson, V. Fraser, and T. H. Gallagher Lost Opportunities: How Physicians Communicate About Medical Errors Health Aff., January 1, 2008; 27(1): 246 - 255. [Abstract] [Full Text] [PDF]

				C. S. Hankin, J. Schein, J. A. Clark, and S. Panchal Adverse events involving intravenous patient-controlled analgesia Am. J. Health Syst. Pharm., July 15, 2007; 64(14): 1492 - 1499. [Abstract] [Full Text] [PDF]

				J. L. Costello, D. L. Torowicz, and T. S. Yeh Effects of a pharmacist-led pediatrics medication safety team on medication-error reporting Am. J. Health Syst. Pharm., July 1, 2007; 64(13): 1422 - 1426. [Abstract] [Full Text] [PDF]

				R. S. Gerstle, C. U. Lehmann, and the Council on Clinical Information Technology Electronic Prescribing Systems in Pediatrics: The Rationale and Functionality Requirements Pediatrics, June 1, 2007; 119(6): e1413 - e1422. [Abstract] [Full Text] [PDF]

				M. Hirose, S. E Regenbogen, S. Lipsitz, Y. Imanaka, T. Ishizaki, M. Sekimoto, E.-H. Oh, and A. A Gawande Lag time in an incident reporting system at a university hospital in Japan Qual. Saf. Health Care, April 1, 2007; 16(2): 101 - 104. [Abstract] [Full Text] [PDF]

				N. C. Elder, D. Graham, E. Brandt, and J. Hickner Barriers and Motivators for Making Error Reports from Family Medicine Offices: A Report from the American Academy of Family Physicians National Research Network (AAFP NRN) J Am Board Fam Med, March 1, 2007; 20(2): 115 - 123. [Abstract] [Full Text] [PDF]

				G. Haller, P. S. Myles, J. Stoelwinder, M. Langley, H. Anderson, and J. McNeil Integrating Incident Reporting into an Electronic Patient Record System J. Am. Med. Inform. Assoc., March 1, 2007; 14(2): 175 - 181. [Abstract] [Full Text] [PDF]

				J. Garbutt, D. R. Brownstein, E. J. Klein, A. Waterman, M. J. Krauss, E. K. Marcuse, E. Hazel, Wm. C. Dunagan, V. Fraser, and T. H. Gallagher Reporting and Disclosing Medical Errors: Pediatricians' Attitudes and Behaviors Arch Pediatr Adolesc Med, February 1, 2007; 161(2): 179 - 185. [Abstract] [Full Text] [PDF]

				J. Braithwaite, M. T Westbrook, N. A Mallock, J. F Travaglia, and R. A Iedema Experiences of health professionals who conducted root cause analyses after undergoing a safety improvement programme Qual. Saf. Health Care, December 1, 2006; 15(6): 393 - 399. [Abstract] [Full Text] [PDF]

				M. A. Ghaleb, N. Barber, B. D Franklin, V. W. Yeung, Z. F Khaki, and I. C. Wong Systematic Review of Medication Errors in Pediatric Patients Ann. Pharmacother., October 1, 2006; 40(10): 1766 - 1776. [Abstract] [Full Text] [PDF]

				D. C. Stockwell and A. D. Slonim Quality and Safety in the Intensive Care Unit. J Intensive Care Med, July 1, 2006; 21(4): 199 - 210. [Abstract] [PDF]

				K E Walsh, R Kaushal, and J B Chessare How to avoid paediatric medication errors: a user's guide to the literature Arch. Dis. Child., July 1, 2005; 90(7): 698 - 702. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) P3Rs: Submit a response Alert me when this article is cited Alert me when P3Rs are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via ISI Web of Science (25) Citing Articles via Google Scholar Google Scholar Articles by Taylor, J. A. Articles by Shafii, J. Search for Related Content PubMed PubMed Citation Articles by Taylor, J. A. Articles by Shafii, J.

	Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved.