This is an application for a K08 award for Dr. L. Elizabeth Goldman, a hospitalist at the University of California, San Francisco. Dr. Goldman is establishing herself as a young investigator in health services research who aims to improve the quality and safety of hospital care. This K08 award will provide Dr. Goldman with the support necessary to accomplish the following career development goals: (1) to become an expert in performance measurement;(2) to improve the assessment of hospital quality and patient safety using administrative data;(3) to learn advanced statistical techniques in health services research;and (4) to acquire skills and policy context to translate research findings into policy. To achieve these goals, Dr. Goldman has assembled a mentoring team comprised of a primary mentor. Dr. Andrew Bindman, Director of the General Medicine Fellowship at UCSF and an expert in the use of administrative data for performance measurement, and two co-mentors: Dr. R. Adams Dudley, who studies public reporting of hospital quality and financial incentives as strategies to improve care, and Dr. Peter Bacchetti, a biostatistician who is an expert in study design and complex modeling. Having accurate assessments of hospital quality is a critical step to improving the quality of hospital care. Dr. Goldman's research will evaluate the accuracy of present-on-admission indicators, a strategy that two states are using to improve hospital assessments based on administrative data and Medicare and other payers are using to withhold payments for """"""""preventable adverse events"""""""".
In Aim 1, Dr. Goldman will use the California Patient Discharge Data Audit in which 1658 medical records from 2005 were externally reviewed as a gold standard against which she will compare the accuracy of present-on-admission coding by hospitals.
In Aim 2, she will use the California Patient Discharge Data and its audit to determine if certain hospital types such as teaching hospitals or small rural hospitals are associated with inaccuracies in coding of present-on- admission and simulate the effect of inaccuracies on hospital assessments comparing hospital death rates for heart attacks.
In Aim 3, she will explore the use of clinical laboratory data to help create datachecks that states and other auditors can use to assess accuracy of present-on-admission coding without extensive chart review. This research will form the basis for an R01 application, to be proposed at the end of this K award, to assess whether this strategy to use clinical data to assess the accuracy of present-on-admission coding will be more clinically acceptable to providers and stakeholders than current tools.
Present on admission coding serves as one of the cornerstones of new policies by Medicare and other payers to withhold payments for """"""""preventable adverse events"""""""" and to improve hospital assessments and patient safety. The degree to which present-on-admission coding is accurate and unbiased will have a major influence on the acceptability and utility of these policy initiatives and performance assessments using present-on-admission.
Showing the most recent 10 out of 14 publications
