IMPROVING QUALITY OF CARE IN ADULTS WITH DIABETES

Johns Hopkins University, Baltimore MD

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Diabetes is the seventh leading cause of death in the U.S.A. and the leading cause of blindness, non-traumatic amputations and end-stage renal disease. Most adults with diabetes are not adequately controlled, despite evidence-based Quality of Care (QOC) guidelines which have been developed and validated as improving process of care, morbidity, and mortality. However, these tools are not being systematically applied in practice, resulting in poor control of risk factors for complications of diabetes, especially in the African American community. The Health Plan Employer Data and Information Set (HEDIS) is the industry standard for monitoring quality of care used in 90% of managed care organizations. HEDIS? Comprehensive Diabetes Care is the proportion of adults with annual HbAlc, lipid and urine testing, eye exam, plus poor HbAlc control and lipid control. However, HEDIS measures do not segregate the components of the process of care-visit intervals, laboratory investigations/clinical data, and intensification of therapy where indicated. These investigators believe that a detailed investigation of the circumstances and predictors of these processes will reveal specific problems underlying the observed deficiencies. This study aims to predict which patients are at highest risk for missed opportunities for timely intensification of therapy for high HbAlc and lipids. They hypothesize that patient characteristics associated with the failure to intensify therapy are physician distraction, patient obesity, African-American race and older age. To test these hypotheses, these investigators propose primary data collection from patients utilizing the 2000 HEDIS methodology. The study population will be a HEDIS sample who receive their primary care from Johns Hopkins Outpatient Center, Bayview, and the Johns Hopkins Community Physicians Network of 19 Primary Health Centers in Maryland. Linking medical chart abstraction data and administrative and pharmacy databases, an innovative review tool will be used to determine the circumstances and predictors of poor HbAlc, lipid and hypertension control and failure to intensify therapy in a timely manner for abnormal HbAlc, lipid and blood pressure. The main outcomes (dichotomous) are poor HbAlc control and failure to intensify therapy in a timely manner versus appropriate intensification. Simple and multiple logistic regression will be utilized to determine the crude and independent patient, physician and visit characteristics which predict poor HbAlc control and failure to intensify therapy. If these hypotheses are correct, the circumstances of failed intensification of therapy will be determined, and the results can be utilized to propose specific remedies, which may be potentially modifiable with modest effort.