application) Poor children with chronic health conditions are a vulnerable population at risk for poor quality of care. Asthma has been identified as particularly suited to assessing quality of care because: (1) it is the most prevalent pediatric chronic condition; (2) quality of care guidelines are available for asthma thanks to the National Asthma Education and Prevention Program (NAEPP); and (3) it is an ambulatory-care-sensitive condition. This project will test the feasibility and validity of a new measure of quality of asthma care for poor inner-city children with asthma, the Asthma Visit Questionnaire (AVQ), to assess quality of asthma care in the primary care setting. The following specific aims will be addressed: (1) How accurate is the AVQ relative to an audio taped record of the primary care visit? (2) What factors are associated with accuracy of the AVQ? Three categories of factors will be examined: patient/family characteristics, practitioner characteristics, and visit characteristics. (3) What proportion of parents will not be able to self-administer the AVQ? What are the sociodemographic characteristics of parents who are not able to self-administer the AVQ? A trained research assistant will audiotape 102 clinic visits with 2-12 year old inner-city children with asthma. Immediately following the audio taped visit, the parent will complete the AVQ, which is a list of 29 questions that ask the parent to report what was discussed and what happened during the visit. The items reflect quality of care elements abstracted from the NAEPP consensus guidelines by an expert panel. These elements will be combined to compute a total quality of care score for each audio taped visit, as a summary measure of care delivered during the visit. The audiotapes will be coded by trained physicians using the same elements of care. The level of accuracy of the AVQ will be measured by the level of agreement between the transcript of the audiotape and the AVQ for each individual quality of care element and for a summary quality of care score. The degree of concordance between the AVQ and the audiotape will be established through cross-tabulations of responses for each data element. Sensitivity and specificity of the AVQ for each element will also be calculated. Chi-square and Fisher?s exact tests will assess the effect of independent variables on the accuracy of report, followed by regression analysis with indicated variables.
