Five-year study shows improvement in all aspects of notes
Physicians have many complaints about electronic health records’ (EHRs) impact on the way they practice medicine. But a new study suggests that EHRs are having a beneficial effect in at least one area-the quality of clinical notes.
The study compares the quality of handwritten and EHR-generated outpatient clinical notes for 100 patients with Type 2 diabetes over a five-year period. The authors examined notes made six months before the physician began using an EHR, then again six months and five years after the start of EHR use, for a total of 300 notes.
To assess note quality, the authors developed and validated a quantitative instrument known as QNOTE. Notes were measured in terms of 12 clinical components, ranging from chief complaint and history of present illness to assessment, plan of care, and follow-up information. The components were, in turn, graded on seven criteria, including whether the note was clear, complete, concise, current, organized, prioritized, and contained sufficient information.
The notes were evaluated by eight internists and eight family physicians in the Washington, D.C. area. Evaluators rated each criterion as either fully acceptable, partially acceptable, or unacceptable. Those ratings were assigned scores of 100, 50, and 0, respectively, with the average of the criteria scores used as the overall score for a given component of the note. Scores were reported as means, ranging from 0 to 100.
The study found that the overall, or grand mean score for the 12 clinical components rose from 52.0 for the pre-EHR assessment to 61.2 for the first (six-month) post-EHR assessment to 80.4 for the five-year assessment. Among the individual components, the biggest rise was in past medical history, where the mean score went from 29.4 pre-EHR to 84.7 five-year-post-EHR.
The authors note that most physicians focus on six “core” components, or elements, during a patient visit: chief complaint, history of present illness, physician findings, assessment, plan of care, and follow-up. The average score among these components rose from 64.4 pre-EHR to 83.7 five years-post EHR. “Because the core elements are not a result of auto population or other core automated systems it is reasonable to conclude that the EHR’s effect on physician performance is responsible for the observed 30% improvement in core note quality,” they write.
The study’s findings have implications in two areas of importance to physicians, the authors say. First, because the American Medical Association is considering a proposal requiring doctors to provide more documentation of their medical decision-making when selecting an evaluation and management billing code. “If this is approved, more attention will have to be paid to the quality of information the note,” they say.
Second, the International Classification of Diseases-10th revision (ICD-10) code set-use of which is scheduled to become mandatory next year-will require more extensive and detailed documentation than ICD-9.
The study, “Electronic health records improve clinical note quality,” was published online first October 23 on the website of the Journal of the American Medical Informatics Association.