ORLANDO, FLA. — Continuity of care is not a protective factor—and may be a significant predictive factor—for medical errors, despite being linked to many positive outcomes in health care delivery, Eric Wong, M.D., said at the annual meeting of the North American Primary Care Research Group.

In an effort to assess the connection between continuity of care and occurrences of medical errors in family practice, Dr. Wong and colleagues at the University of Western Ontario undertook a cross-sectional chart review using a stratified randomized sample of eligible electronic medical records from patients at an academic family medical center in southwestern Ontario.

The study enrolled 202 patients aged 18 years or older who had visited the family medical center two or more times between July 2002 and June 2003, excluding pregnant patients and residents of a nearby drug rehabilitation center.

The investigators identified 37 “preventable” medical errors that occurred in 1,376 visits, and classified the errors as being related to diagnosis, investigation, or treatment.

The diagnostic errors included misdiagnoses and delayed diagnoses. The investigation errors included incorrect or omitted tests.

And the treatment errors included incorrect drug and nondrug treatment based on the indication, incorrect drug dosage, delayed drug and nondrug treatment, and omitted drug and nondrug treatment. Seven of the errors (in five patients) resulted in nonfatal but physically or emotionally harmful consequences, according to Dr. Wong.

Three multivariate analyses—each using a different continuity of care index, with continuity of care defined as frequency of provider/patient interaction—showed a positive correlation between medical errors and continuity of care, and no correlation between disease complexity or severity and the occurrence of medical errors. All three analyses also showed alcohol abuse to be a significant independent risk factor for medical errors, Dr. Wong noted.

“The results did not support our hypothesis that good continuity of care on its own decreases the risk of medical errors,” he said.

“It's possible that our assumptions about continuity of care—particularly the idea that seeing patients more often will lead to a better relationship and better care—are wrong,” Dr. Wong added.

There are a number of possible explanations for the possible link between increased contact with the same provider and the increased risk of medical error, Dr. Wong explained.

“It might be that what we define as continuity of care is really a proxy measure for problem patient/doctor relationships, or that some of the patients are being seen so much because they have multiple, difficult complaints, which could increase the likelihood of error,” he said.

Additionally, physicians may be more likely to book frequent return visits for patients whose symptoms are vague, not as a way to optimize care but as a way to “keep the patient happy” without really having a solid care plan, Dr. Wong said.

Another possibility is that familiarity breeds complacency. When physicians see the same patient frequently, they are likely to engage in more social conversation than they would with a new patient or a patient who hasn't been seen recently, with less attention paid to the medical problems or routine care.

“Nearly 57% of the noted errors involved the omission of necessary investigations. This could be a function of the clinician just not paying close enough attention to what has and has not been done recently, or simply assuming that the necessary tests must have been taken care of at the previous visits,” he said.

Finally, the findings could be indicative of an overall system failure, in which the sometimes stilted nature of patient/physician communication precludes an easy assessment of a problem.

“The occurrence of medical errors and higher levels of continuity may represent parallel effects of the interaction of patient, physician, and system factors,” Dr. Wong said. As a result, “some patients may be seen a lot, without really being seen at all.”

With respect to the alcohol abuse finding, “the diagnosis of alcohol abuse could also be a proxy measure of a problem with the doctor/patient relationship, thus leading to less attentive care and an increase in medical errors,” Dr. Wong hypothesized. “Or it may be that these patients are more difficult to diagnose and treat, or that physicians' attitudes toward them [are] negative.”

Although the study findings are provocative, they are limited by a number of factors, Dr. Wong stressed, including the small number of errors overall, the chart review and audit process, and the classification system.

Additionally, “the provider continuity indices might not capture some aspects of continuous care that could decrease risk of error,” he said.

Despite the study limitations, Dr. Wong said, “the findings do suggest that clinicians should pay closer attention to patients they see frequently and to those with a diagnosis of alcohol abuse, to avoid preventable medical errors.”

Medical Errors More Likely In Frequently Seen Patients