An IPRO Medicaid study evaluated charts throughout New York State with a principal diagnosis of community acquired pneumonia (CAP)-period: 9/1/01-3/31/02. A risk-adjusted mortality rate of 14.8% was found at Elmhurst Hospital Center. Attendings felt that x-rays reviewed at morning report and resident evaluations were at odds. Our aim was to review the resident’s effectiveness in correlating the x-ray and the clinical course.
Retrospective chart review of 55 IPRO reviewed Medicaid charts between 9/1/01-3/31/02. Data abstracted included demographics, medical history, clinical presentation and review of every chest x-ray and CT scan of chest. We looked at the official x-ray report and had access to prior hospital/clinic x-rays and follow up admission/clinic x-rays.We compared initial impressions with progress note impressions and final diagnoses as documented on discharge summaries.
All 55 Medicaid charts were available for review. Fourteen charts met the criteria for CAP, including radiographic evidence for consolidation. Forty one charts did not meet the criteria. Misdiagnoses included congestive heart failure, COPD with acute exacerbation, interstitial lung disease, pleural effusion, alelectasis, bronchogenic carcinoma.CONCLUSIONS: Coding by medical records is dependent on accurate discharge diagnoses. Proper billing depends upon accurate principal diagnoses.Residents had been encouraged to complete charts in a timely fashion and were noted to be writing their summaries as early as night one. Residents did not correct treatments or their notes for CAP, even after a better diagnosis had been established.
A quality control panel has been established to review concurrent charts for accuracy in establishing the principal diagnosis of CAP prior to coding is done on a chart.Attending physicians are being encouraged to review discharge summaries before they sign off on a chart with the principal diagnosis of CAP.Accurate National Data for CAP depends on accurate principal diagnoses. This study questions national data and statistics, including morbidity, mortality and prevalence of disease.
I.M. Berlin, None.