A 61-year-old asymptomatic man sees his internist for a preventive medicine evaluation. The internist has cared for the patient for 10 years. The examination and laboratory studies are normal, except for the prostate specific antigen (PSA), which is 11.8 ng/mL. Surprised by this finding, the internist reviews the patient's previous blood work. A year ago, the PSA was 8.2 ng/mL and three years ago it was normal. There is no evidence in the medical record that the patient was informed of, or that the internist acted upon, the prior abnormal PSA result. However, the internist is certain she never saw it. She wonders what to do next.
According to the American Medical Association (AMA) Code of Ethics, “an error is an unintended act or omission, or a flawed system or plan that harms or has the potential to harm a patient.” Mishandling of an abnormal test result is a common health care-related error that results in diagnosis and treatment delays. Evidence suggests that many clinically relevant abnormal test results, including abnormal blood and imaging test results, lack evidence of clinician awareness in patients' medical records.
In fact, 83% of internists who responded to a survey administered by Poon et al recalled at least one recent test result that they wished they had known about sooner. In another study, Wahls and Cram found that 30% of primary care providers reported encountering one or more patients with delays in diagnosis or treatment because of mishandled test results within the 2 previous weeks. The commonest delayed diagnoses were cancers and endocrine and cardiac disorders.
Mishandling of abnormal PSA results is also common. In a Veterans Affairs study, Nepple et al identified 327 men who had an abnormal PSA result before they were diagnosed with prostate cancer. The time between the first abnormal PSA result and a documented clinician response ranged from 0 to 1,342 days. Overall, 253 men (77%) had a timely clinician response to the abnormal PSA (≤30 days). However, 23 men (7%) had a response in 31 to 180 days, 24 (7%) had a response in 181 to 360 days, and 27 (8%) had a response after 360 days. For men who experienced a timely clinician response, the median number of days between the abnormal PSA result and a request for urological consultation was 1 day; the median time to prostate biopsy was 69 days. However, for men who experienced a delayed clinician response, the median number of days between the abnormal PSA result and the request for urological consultation or prostate biopsy was 342 days and 526 days, respectively.
Not surprisingly, clinicians experience negative emotions when they realize they have committed an error. Nevertheless, they are ethically obligated to disclose errors to patients. According to the College's Ethics Manual, clinicians “should disclose to patients information about procedural or judgment errors made in the course of care if such information is material to the patient's well-being. Errors do not necessarily constitute improper, negligent or unethical behavior, but failure to disclose them may.” Indeed, the ethical rationale for disclosing errors to patients is strong.
First, clinicians should act in the best interests of patients. In this case, withholding the prior abnormal PSA result from the patient does not serve the patient. Instead, the internist should explain the nature of the error and its implications and continue to provide professional and compassionate care. Notably, nondisclosure of errors damages trust as many patients eventually learn of errors. If the internist in the case refers the patient for prostate biopsy without disclosing the error, it is possible the patient will learn of the error from the consulting urologist.
Secondly, respect for patient autonomy requires that clinicians disclose errors to patients to allow for informed decision making. The patient in the case has the right to know about the previous abnormal PSA result so that he can act upon it according to his health care-related goals and values.
Finally, justice requires that patients be given what is due to them, such as information about their medical condition and, if injured, appropriate compensation. Clinicians' duties regarding health care-related errors, however, go beyond disclosure. Clinicians should encourage and participate in efforts to prevent errors. Indeed, many errors, including mishandling of test results, are caused by systems problems. Clinicians should play a central role in identifying, reducing and preventing errors.
What do patients want? Virtually all want to know about errors, even minor ones. Patients desire disclosure about what happened, why it happened, the implications for their health and future care, and strategies for preventing future errors. Notably, patients are more likely to consider litigation if errors are not disclosed.
There are benefits to disclosing errors to patients. For patients, disclosing errors informs them, allows for proper diagnosis and treatment, resolves uncertainty and promotes trust. For clinicians, disclosing errors may relieve stress, foster patient forgiveness, promote trust, improve practice (such as by adopting systems that prevent errors) and reduce litigation. In fact, health care organizations that have implemented programs to fully disclose errors and provide appropriate compensation to patients have experienced neutral or beneficial effects related to litigation.
The internist resolved to disclose the error to the patient. She met with the patient and his wife at her office. She learned that the patient did not recall hearing about the prior PSA result. She then disclosed the prior and current PSA results and acknowledged that the prior result was mishandled. The patient appeared upset by the news and expressed feeling angry and helpless. The internist acknowledged these emotions. However, the patient later stated that he appreciated the internist's honesty. Together, the internist and patient formulated a plan for further assessment, treatment and follow-up. The internist also notified the sentinel event group at her institution to engage in an analysis of the error.