| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
TRACK to:
|
|
Electronic letters published:
![[Read Comment]](/icons/misc/sectionDOWN.gif){kind=icon}
Family Medicine Must Do Better
Response from the author
Just the tip of the iceberg - Comment on Elder et al.
Physician reports, and terminology, and the notion of harm
Seeking to further understand "real" medication error rates in primary care
Errors in primary care: What’s on the radar screen?
Comment on "The Identification of Medical Errors by Family Physicians During Outpatient Visits"
|
|
|||
|
Robert J. Riley, M.D., M.S.Ed, South Bend, IN Director, Family Medicine Residency Program, Memorial Hospital, South Bend, IN
Send response to journal:
|
The recent "Future of Family Medicine" report, accompanying this latest edition of the Annals, includes once again a call to the specialty to prioritize high quality research in family medicine. How ironic that the current article is an example of how far we have to go. The authors suggest in their introduction that their purpose is "to have physicians identify prospectively those errors and preventable adverse events that occur during office-based clinical encounters." But by having physicians serve as their own judges, the authors not only introduce an obvious source of bias, but ensure that they would almost never uncover medical errors that really matter. No one steps out of an examination room and says, "Boy, I just misdiagnosed that man's heart attack as reflux disease." Or, "I just prescribed ten times the appropriate dose of that medication--sure wish I hadn't done that." Or "I probably should have obtained a CT scan on that woman with the excruciating headache and the new focal finding." If physicians could recognize their own errors the moment they stepped from the examination room, our error rate would be near zero. While the study methodology doesn't allow us to uncover the types of errors that the specialty and the public care most about, it does allow us to discover a new kind of "pseudo-error." The doctor feels a little pressed for time? Count that as an error. The nurse knocks on the examination room door during the encounter? Another error. The doctor asks the nurse to call the lab to obtain a result which hasn't reached the chart yet? Error again. It's no wonder that the "pseudo-error" rate varied from 3% to 60% among physicians in the study. Does this reflect real differences in these physicians' practices? We can't tell from the study, but it seems more likely to be the result of poor study design. And how did the authors define "harm" in this article? Study participants could apparently define it however they wished. This allowed some physicians to count the time lost when the nurse called to check on that missing laboratory report as "harm," while others didn't, for example. This lack of standardization renders the data on harm uninterpretable. In summary, this study sheds no meaningful light on the error rate in family physicians' offices, nor the harm done to patients as a result. It compiles "pseudo-errors" instead and reports invalid data regarding their harm to patients. If we're to establish credibility for family medicine in the area of research, we need to do better, particularly in a journal which carries the name of our specialty. Competing interests: None declared |
|||
|
|
|||
|
Nancy C. Elder, Cincinnati, OH, USA Associate Professor, Department of Family Medicine, University of Cincinnati
Send response to journal:
|
The responses to our study, The identification of Medical Errors by Family Physicians during Outpatient Visits, raise many important and thought provoking questions. Drs. Kuzel and Fetters raise issues about definition and taxonomy of commonly used words like error. Dr. Main suggests that perhaps it is only a known bad outcome that changes an annoyance into an error for many physicians. The issues related to how we describe, count and report errors have far reaching implications. While the issues in defining error in primary care are complex, we must first acknowledge that there is still disagreement in the general medical community in what should be included in the discussion of medical error. Steve Woolf posits that much of what are called errors and lapses in patient safety are really lapses in quality and caring, and should not be confused with error.(1) This is in opposition to others who use error to describe a broad range of undesired practices. (2) Those currently trying to classify errors in family medicine, including Drs. Main, Kuzel and Galliher, rightly know the difficulties involved. Trying to make taxonomies developed from primary care error reports fit accepted classification systems such as the Eindhoven classification system(3) is often difficult. Trying to make a taxonomy from one primary care study understandable to another primary care researcher is even challenging. Bob Phillips, MD, of the American Academy of Family Physicians Graham Center is currently organizing an international meeting of primary care medical error researchers just to discuss taxonomy! In the case of our study, it based on our knowledge of how to define and classify errors as had been published prior to 2001. Our understanding of the complexity of error, multiple errors happening sequentially or in parallel, contributing factors and harm have progressed dramatically in the last three years, and improved and more appropriate taxonomies for primary care certainly exist today. Dr. Scott-Cawiezall points out the drawbacks to self-report, and Dr. Main points out the weakness in not including non-physicians in our sample. Both of these are valid concerns and limitations to our study. I believe the range of physician responses within our study demonstrate just how difficult self-report can be in this area, and Dr. Fetters suggests using examples to assist with the ease of self report. Observation is an important technique, and, I believe when combined with self-report, chart reviews, patient input and other methods, will give us the most complete picture of error in family medicine. 1. Woolf SH. Patient safety is not enough: targeting quality improvements to optimize the health of the population. Ann Intern Med 2004;140(1):33-6. 2. Lee TH. A broader concept of medical errors. N Engl J Med 2002;347(24):1965-7. 3. Battles JB, Kaplan HS, Van der Schaaf TW, Shea CE. The attributes of medical event-reporting systems: experience with a prototype medical event-reporting system for transfusion medicine. Arch Pathol Lab Med 1998;122(3):231-8. Competing interests: None declared |
|||
|
|
|||
|
Deborah S Main, Denver, CO Department of Family Medicine
Send response to journal:
|
The study by Elder and colleagues(1) offers insight about errors and preventable adverse events identified by family physicians. Their findings highlight that errors are common in primary care, with a large percentage of visits resulting in “office administrative errors” (16.5%). These results are consistent with other primary care studies (2) including our Applied Strategies for Improving Patient Safety (ASIPS) study (3). We found that reports of lost or missing clinical and patient demographic or contact information were common, resulting in increased risk of harm for some patients. Even more remarkable is that these primary care studies may underestimate the problem. First, although Elder’s study used a well-focused reporting form that prompted study physicians to include charting and administrative errors and preventable adverse events, their study did not include non- physicians. Many non-physicians are the discoverers of administrative errors involving incorrect or missing forms or charts; lost or missing laboratory or x-ray reports; and incorrect phone numbers. That non- physicians are often excluded from primary care studies would result in underestimating the frequency of administrative errors. In addition, although our ASIPS study encourages anonymous and confidential reports from clinicians and office staff, the broad definition used in our study (similar to Dovey et al.[4] ): “any event you don’t wish to have happen again, that might represent a threat to patient safety” may have unwittingly deterred some people from reporting administrative errors or preventable adverse events because they are so common– yet are so different from usual incident reports required by local hospitals and medical errors publicized in national reports and studies. In fact, we learned from our narrative reports and confidential follow-up interviews that administrative errors “happen all the time” or “several times a week.” It would be just too overwhelming to report each and every one of these. Have office administrative errors become so common in primary care that they are just perceived (accepted?) as annoyances rather than as the potentially dangerous problems they are? Currently, it seems that much of what separates an error from an annoyance in primary care is just a KNOWN bad outcome. Elder et al.’s study reminds us that as we continue focusing our research and funding to solve prominent patient safety issues such as preventing wrong-site surgery and improving prescription drug dispensing, further investigation and prevention of these common office administrative errors and preventable adverse events in primary care should also receive our attention and resources. 1. Elder NC, Vonder Meulen M, Cassidy A. The identification of medical errors by family physicians during outpatient visits. Ann Fam Med 2004; 2: 125-129. 2. Elder NC, Dovey SM. Classification of medical errors and preventable adverse events in primary care: A synthesis of the literature. J Fam Pract 2002; 51: 927-932. 3. Fernald DH, Pace WD, Harris DM, West DR, Main DS, Westfall JM. Event reporting to a primary care patient safety reporting system: A report from the ASIPS Collaborative. Ann Fam Med (In Press) 4. Dovey SM, Meyers DS, Phillips RL, Green LA, Fryer GE, Galliher JM, Kappus J, Grob P. A preliminary taxonomy of medical errors in family practice. Qual Saf Health Care 2002; 11: 233-238 Competing interests: None declared |
|||
|
|
|||
|
Anton J. Kuzel, Richmond, Virginia, USA Chairman, Department of Family Practice, Virginia Commonwealth University
Send response to journal:
|
Kudos to Nancy Elder and colleagues for demonstrating that it is possible to engage family physicians in routine reporting of errors and perceived harms in office practice. They used a different method than other investigators – quickly completed check list survey, followed by interviews to characterize harms. Judging by the apparent response rate, this method should be considered for future studies that aim to characterize the relative incidence of kinds of errors in primary care. I know that Nancy Elder, John Hickner, Wilson Pace, Susan Dovey, and others are engaged in ongoing studies of physician, staff, and patient reports. I will invite further discussion on a couple of points. Nancy Elder has previously implied her working definitions of medical error and preventable adverse event (Elder NC, Dovey SM. Classification of medical errors and preventable adverse events in primary care: A synthesis of the literature. Journal of Family Practice • November 2002 • Vol. 51, No. 11). In that paper, it seems that “preventable adverse event” means “what went wrong,” that is, what goal failed to be achieved (diagnosis, treatment, prevention). “Medical error” and “process error” seem to be used interchangeably in the earlier paper, and are felt to be “why” something happened (e.g., inadequate history, lab result or chart missing). I am inferring from this understanding that the Figure 1 in the current paper includes a top section that are “preventable adverse events” since they are failures of diagnosis or treatment, while the lower section is understood as a “medical error” or “process error,” since they get at purported causes (organized according to Susan Dovey’s taxonomy). I think clarifying this distinction is important, and, if I have it right, I think some would choose to label preventable failures to diagnose, treat, or prevent illness as medical errors, rather than “preventable adverse events.” I acknowledge that I may have Nancy’s definitions exactly backwards, but if so, it underscores the importance to continuing to provide definitions in our papers – I think the nomenclature is still in a state of flux. The other point is around physician perceptions of harms (actual or potential). The current study suggests that physicians are able to perceive both physical and emotional harms, which is in contrast to the results of the US arm of an international study of physician reports of errors (Woolf SH, Kuzel AJ, Dovey SM, Phillips RL. A string of mistakes: the importance of cascade analysis in describing, counting, and preventing medical errors. Accepted for publication, Annals of Family Medicine). In Elder, et al’s study, physicians were interviewed about this issue for all encounters in which they reported a preventable adverse event or medical error, whereas in the study coming soon in the Annals, physicians responded to a yes/no question on a survey as to whether a patient was harmed. It may be that some definitions of harms (as Elder and colleagues suggest), and specific inclusion of emotional or psychological harms, can be incorporated in future work using purely survey methods. It seems relevant to include the psychological harms both for their own sake, as well as for what they may be doing to physician- patient relationships and patient participation in their ongoing health care. Again, my compliments to the authors. I look forward to their reply, and to the input of other readers. Competing interests: None declared |
|||
|
|
|||
|
Jill R Scott-Cawiezell, Columbia, MO USA PhD University of MO-Columbia
Send response to journal:
|
In the United States, between 44,000 and 98,000 patients die each year from preventable errors, suggesting that preventable error across the health care system is the eighth most common case of death (Sexton, Thomas, & Helmreich, 2000). Furthermore, research suggests that medication errors average 10% or more of administered doses, excluding wrong time errors (Barker and Allan, 1995). There is no doubt that the problem of medication error is still not yet completely understood, particularly outside the hospital setting. The study used a method of voluntary reporting, that has been challenged as an underestimate of the actual extent of medication errors (Barker & Allan, 1995; Cullen et al., 1995; Gandhi, Seger, & Bates, 2000). While certainly a place to begin the daunting task of quantifying error in the primary care setting, the authors need to further clarify the limitations of the study beyond “Errors were captured only during the patient encounter…”. This leads the reader to believe that a self-reporting method can be a complete assessment of medication error and adverse drug events. The gold standard for detecting error is the observational method, detecting over 1000-fold more errors than voluntary report methods, even if health care providers are more careful when observed (Barker & Allan, 1995). Although cumbersome and expensive, a systematic approach to capturing baseline of medication rates in the primary care setting would require this type of assessment. The authors provide a great start to the discussion about seeking to clarify and understand medication errors in primary care. However, methodological limitations and the lack of direction about the next logical steps to explicating “real medication error rates” in primary care must be noted. I look forward to learning more from their next steps in quantifying medication error in the primary care setting. Competing interests: None declared |
|||
|
|
|||
|
Michael D. Fetters, Ann Arbor, USA Assistant Professor, University of Michigan
Send response to journal:
|
In the current issue of the Annals, Elder and colleagues present their study results on error identification by family physicians during outpatient visits.(1) This study helps advance the field of inquiry on errors and quality in primary care specifically due to their prospective use of a mixed methods, triangulation design,(2) and a specific error taxonomy to prompt physicians to identify errors.(3) The most common cause of error reported from 351 out patient visits, office administration errors, are probably the most amenable to a systems approach for improvement.(4) The variation in reporting of errors (3.2% to 60%) by the participating physicians merits discussion. Elder et al standardized the measurement tool, and trained participants on completion of the form. In spite of these steps, it appears physicians have varying thresholds for detecting errors and adverse events on their radar screens. While not statistically significant, the authors report different rates in reported ranges of errors by male and female participants, 21.2% vs. 26.3%, respectively. Are there gender differences among physicians in occurrence of errors, or ability to recognize errors? Further agreement of what constitutes an error, and what needs to be detected on the radar screen, is needed. A codebook with specific examples for each type of error might improve physician recognition of error.(5) The classification of errors in primary care is becoming more sophisticated, but the debate about the classification of error in primary care should continue. My own bias is to include patient- related factors—these were not included in the current study. Improvement of the system as a whole requires the involvement of all parties including patients. Patient perspectives could be particularly valuable for judging events, e.g., potential error or harm, that physicians can’t agree upon among themselves. A limitation of the Elder et al approach for assessing the incidence of errors in primary care is the difficulty of detecting errors of omission, arguably the most important in primary care. Errors of omission occur because they are not on the physician’s radar screen. If a physician is unaware of the need for a medication in a particular patient, eg, Beta blocker following a myocardial infarction, then it is unlikely the same physician would detect its omission as an error. Further refinement and acceptance of a primary care error taxonomy with specific examples for interpretation, involvement of patients, and mechanisms for moving errors of omission onto radar screens is needed. 1. Elder NC, Vonder Meulen, M, Cassedy A. “The identification of medical errors by family physicians during outpatient visits. Ann of Fam Med, 2(2):125-129, 2004. 2. Creswell JW, Fetters MD, Ivan KN. Designing a mixed methods study in family medicine. Ann of Fam Med, 2(1):7-12, 2004. 3. Elder NC. Dovey SM. Classification of medical errors and preventable adverse events in primary care: a synthesis of the literature. J Fam Pract. 51(11):927-32, 2002. 4. Bogner MS.Human Error in Medicine. Lawrence Erlbaum Associates, pp. 1-411, 1994. 5. Fetters MD. Medical error in primary care. In What do We Know About Medical Mistakes?, edited by Marilyn Rosenthal and Kathleen Sutcliffe. Jossey Bass, pp. 58-83, 2002. Competing interests: None declared |
|||
|
|
|||
|
James M. Galliher, Leawood, KS Director of Research, AAFP National Research Network
Send response to journal:
|
This is a wonderful and welcome contribution to the literature on medical errors in primary care settings. Among other things, this research used physicians’ self reports during the patient-physician encounter to record recognized errors that occurred during the encounter. The research also is the first (to my knowledge) to investigate physicians’ evaluations of “harm” (rather than perceived consequences) caused by the errors reported. The authors are to be commended for recognizing that similar phenomena are the subject of investigation in other disciplines (e.g., criminology) and that primary care researchers might learn from that body of research. Among the questions raised for further research are: 1) how do physicians and patients conceptualize errors and adverse events; 2) what factors account for variability in the recognition (and reporting) of errors and adverse events; 3) what do physicians and their patients mean by the harm associated with medical errors; 4) how are the observed different types of errors and/or the occurrence of specific errors evaluated by physicians, other clinicians, and their patients in terms of harm or seriousness; and 5) what is the distinction between the perceived harm and seriousness of errors and how are these evaluations related to the consequences of these errors? While the reported research has recognized limitations, the authors address and raise important questions for investigators interested in medical errors in primary care settings. Answers to these and related questions should prove useful in developing strategies for enhancing patient safety and quality outcomes. Competing interests: None declared |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
TRACK: