Making patient‐level estimates from medical encounter records using a multiplicity estimator

Abstract

The National Ambulatory Medical Care Survey (NAMCS) is a nationally representative survey of medical encounters in physician offices in the United States. Data from this survey and its counterpart in hospitals, the National Hospital Ambulatory Medical Care Survey (NHAMCS), have been used to investigate physician treatment and prescribing patterns. A limitation of these data, however, is that they represent visits rather than patients. Starting in 2001, the survey questionnaires began collecting information on the number of past visits the patient had to the sample provider during the one‐year period prior to the sampled visit. This information was used to estimate number of patients from the NAMCS and NHAMCS visit data using a multiplicity estimator. The resulting distribution of patients by the number of annual visits is similar to the distribution of persons in the U.S. making ambulatory care visits from a population‐based survey. This estimation technique may be useful in estimating patients with clinical characteristics that are difficult to collect from a population‐based survey. Published in 2007 by John Wiley & Sons, Ltd.