PEDIATRICS Vol. 111 No. 5 May 2003, pp. 1108-1109
COMMENTARY |
Don’t Believe Everything You Read in the Patient’s Chart
Abbreviations: NICU, neonatal intensive care unit.
A medical error is defined as the ‘failure of a planned action to be completed as intended or use of a wrong plan to achieve an aim.’1 Some types of errors are familiar to most clinicians, while others are not. Errors that result in catastrophic harm to the patient and those that are egregious deviations from the standard of care, such as surgery on the wrong side of the body, are easy to recognize as errors and receive much attention from the media as well as hospital accreditation organizations. However it is important to realize that errors can occur in a wide variety of domains in patient care, and many such errors are unrecognized and underreported. Errors of commission are often easier to recognize than errors of omission. In this month’s issue, Carroll et al2 report how frequently residents working in the neonatal intensive care unit (NICU) committed errors of documenting patient weights, intravascular catheters, and medications. They found at least 1 documentation error in 62% of randomly selected progress notes in their unit over a 4-month period. What are the lessons and implications of this study?
The methodology used in this study is a good example of a rigorous approach to operationally defining and measuring errors. To assess the incidence or prevalence of medical errors it is essential to obtain both numerators and denominators. The numerator is the type of error chosen for measurement. The denominator is the number of opportunities that arise for that particular error to occur, during the course of study. Obtaining numerators is often much easier to obtain than obtaining the denominators and therefore methods of data collection should be carefully designed to yield precise and accurate data. Subjectivity is required in the labeling of some clinical events as errors and there can be variability among chart reviewers in their determination of medical errors.3 Carroll et al chose to measure deviations in the process of care that had easily defined and quantifiable reference standards (daily weights, vascular catheters, and medications). They found a significantly high number of documentation errors (the authors call them ‘discrepancies’). Of note, for vascular catheters, errors of omission were much fewer than errors of commission. In general, errors of omission in any clinical process are likely to remain undetected, unless specifically looked for. Errors in this unit occurred despite the residents’ use of a computer-generated, structured form. It is plausible that the prevalence of errors would be higher with handwritten or dictated notes because such notes do not have the prompts or checklists contained in structured forms.
Why did these errors of documentation occur? In understanding the causation of errors, experts differentiate between the person approach and the system approach.4 The person approach, focusing on the ‘sharp end’ and blaming an individual for being careless or insincere is considered less desirable because ‘To Err Is Human.’1 Most errors are committed by well-intentioned individuals working in systems of care that are rife with latent conditions that predispose to the occurrence of errors. A system is defined as ‘a set of interdependent elements interacting to achieve a common aim. These elements may be both human and nonhuman.’1 In a NICU, health care providers are just one of the many components in a system that is comprised of several human and nonhuman components such as physicians, nurses, respiratory therapists, patients, patient equipment, ancillary staff, the physical environment, work schedules, and computers. These components are continuously interacting in constantly shifting patterns. To prevent errors, a more fruitful approach than blaming individuals is thought to be the system approach, so that the system is redesigned to eliminate the conditions that permit or encourage errors to happen.4
Carroll et al do not describe any adverse events resulting from the documentation errors. Other studies of errors have also found that many errors in the process of care do not result in harm to the patient.8 This does not diminish the need to eliminate such errors but does highlight the importance of prioritizing specific types of errors to target for prevention. In their attempt to identify factors that contribute to the occurrence of documentation errors, the authors studied relatively few factors. Several additional factors can be hypothesized to play a role. These include the census of the NICU, the degree of sickness of the patients, the number of patients assigned to each resident, the other clinical demands on the residents (such as attending deliveries), the residents’ previous duration of clinical training, stress, long duty hours, and sleep deprivation. The last 2 factors especially have received much attention from the Accreditation Council for Graduate Medical Education.5 To truly understand the various factors contributing to an error, a detailed analysis of the systems of care, such as a root cause analysis6, is necessary. Such detailed analyses are time-consuming and require considerable effort. Therefore, they are currently reserved for errors that actually or potentially result in serious harm to the patient (sentinel events).
Finally, documentation errors are 1 type of communication error. Communication is a pervasive and ubiquitous factor in all interactions between health care providers, patients, parents, and all other components of the ‘NICU-as-a-system’ mentioned above. In a study of the first 1000 medical errors reported anonymously by NICU personnel to the Vermont Oxford Network using Web-based reporting,8 we found that a communication problem was present in 21% of reported errors. A specific error of charting or documentation was reported in 13% of reported errors. Improving written, spoken, and visual communication among the team members caring for NICU patients is likely to go a long way in the prevention of medical errors.
Department of Pediatrics, Given Building
University of Vermont College of Medicine
Burlington, VT 05405
-->
FOOTNOTES
Received for publication Mar 14, 2003; Accepted Mar 14, 2003.
Address correspondence to Gautham Suresh, MD, DM, MS, Department of Pediatrics, Given Building, University of Vermont College of Medicine, Burlington, VT 05405. E-mail: gautham.suresh{at}vtmednet.org
 |
REFERENCES |
|---|
 TOP REFERENCES |
|---|
- Kohn LT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington DC: National Academy Press; 2000
- Carroll AE, Tarczy-Hornoch P, O’Reilly E, Christakis DA. Resident documentation discrepancies in a neonatal intensive care unit.
Pediatrics.2003; 111
:976
–980
[Abstract/Free Full Text] - Thomas EJ, Studdert DM, Brennan TA. The reliability of medical record review for estimating adverse event rates.
Ann Intern Med.2002; 136
:812
–816
[Abstract/Free Full Text] - Reason J. Human error: models and management.
BMJ.2000; 320
:768
–770
[Free Full Text] - Philibert I, Friedman P, Williams WT. New requirements for resident duty hours.
JAMA.2002; 288
:1112
–1114
[Free Full Text] - Joint Commission on Accreditation of Healthcare Organizations. Root Cause Analysis in Health Care: Tools and Techniques. Oakbrook Terrace, IL: Joint Commission on Accreditation of Healthcare Organizations; 2000:109–128
- Kaushal R, Bates DW, Landrigan C, et al. Medication errors and adverse drug events in pediatric inpatients.
JAMA.2001; 285
:2114
–2120
[Abstract/Free Full Text] - Suresh G, Horbar J, Plsek P, et al. Voluntary anonymous reporting of medical errors for neonatal intensive care [abstract]. Pediatr Res.2003 . In press
PEDIATRICS (ISSN 1098-4275). ©2003 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
