O ver the course of my 45 or so years in the business, I have noticed that many patient charts contain errors. I’ve never stopped to count them, but some investigators at five leading academic centers recently did. Of 22,889 surveyed patients at three major US hospital systems who read their own records, 4,830 (25%) found mistakes. Almost 10% were classified as very serious, 42.3% as serious, and 32.4% as somewhat serious. My comments are in italics.

Patients were asked to give free text descriptions of mistakes. The most common type of error involved a current or previous diagnosis. Some diagnostic errors cited: a patient with BRCA-1 was labeled as non-BRCA-1; another without lung cancer was said to have the disease; a physical therapy referral was made for the wrong body part.

Other mistakes identified by patients included inaccurate medical histories, errors describing medications or allergies, records listing informed consent or counseling discussions that patients said did not occur, and “errors pertaining to the physical examination, including elements of the examination that, according to the patient, were documented but not done.”

The free text comments by the patients were most interesting. For example, “My cardiologist repeatedly says that I ‘deny’ symptoms (such as shortness of breath, etc.) that he never asked me about and that I never denied having.” This is likely caused by the templated review of systems in the electronic medical record.

“Doctor reported that I did not claim to have pain in my hand. I am a pianist and I went specifically because pain was in my hand.” This may be caused by copying and pasting from the previous visit’s note.

“I have been complaining of difficulty breathing [for over 3 months]. It has been a real and increasing problem for me, but is not mentioned in my notes. In fact, notes saying my breathing is normal are made for each visit.” Again, copying and pasting may be the culprit. Another common issue is the history and physical may say the patient complains of abdominal pain, but the review of systems template, hurriedly checked as negative, says “Denies abdominal pain.”

“A provider not only failed to mention risks and side effects of the injection he wanted to give, but sarcastically derided my concern about long-term risks. ‘Where’d you hear that, the Internet?’” I have seen informed consent discussion documentation that is half a page long and contains passages like “He/ she had no questions and agreed to the procedure.” One wonders if all 15 or 20 enumerated risks were actually explained to the patient.

Many patients reported the mistakes to their clinicians and some were dismayed to find that the errors were not corrected. In a few such cases, patients sought care elsewhere.

When you are a patient, take advantage of the opportunity to review your own records for accuracy and make sure errors are corrected. You may not understand everything you read, but do not let that discourage you.