|Grant Number:||5R01CA160738-03 Interpret this number|
|Primary Investigator:||Davila, Jessica|
|Organization:||Baylor College Of Medicine|
|Project Title:||Guideline-Based Surveillance and Treatment of Hepatocellular Carcinoma|
DESCRIPTION (provided by applicant): Hepatocellular carcinoma (HCC) is the fastest rising cause of cancer-related deaths. Survival following diagnosis with HCC is poor unless it is detected early and stage-appropriate treatment is applied. Clinical practice guidelines outline recommendations for HCC surveillance and treatment of HCC. No study has examined the implementation or outcomes of guideline-based HCC surveillance in high risk patients or stage- appropriate treatment in patients who develop HCC. OBJECTIVES: Using data obtained from VA administrative databases combined with information abstracted from national VA electronic medical records using the VA Compensation and Pension Record Interchange application, we propose to address the following Aims: 1) to examine the implementation of guideline-based HCC surveillance in a national sample of patients with cirrhosis, and to examine patient, physician and facility factors associated with appropriate or inappropriate implementation of surveillance; 2) to evaluate the implementation of guideline-based stage-appropriate treatment among patients with HCC, and to examine patient, physician, and facility factors associated with receipt of stage-appropriate treatment; and 3) to examine the effect of guideline-based HCC surveillance and stage-appropriate treatment on survival. METHODS: We propose a retrospective cohort study using two stratified random samples of approximately 1,500 patients with cirrhosis and 1,500 patients with HCC diagnosed during 2006-2009 in VA facilities nationwide. For Aim 1, the outcome will be appropriate (guideline-based surveillance, receipt of an imaging test unrelated to HCC surveillance, no surveillance) versus inappropriate (non-guideline based surveillance, overutilization of surveillance) implementation of HCC surveillance among patients with one of the four primary etiologies for cirrhosis (hepatitis C, non-alcoholic fatty liver disease, hepatitis B, alcoholic liver disease). Patient, physician, and facility factors associated with appropriate implementation of HCC surveillance will be examined using logistic regression. For Aim 2, the outcome will be receipt of stage-appropriate HCC treatment versus stage-inappropriate treatment or no treatment among patients with HCC. Each patient will be assigned to a stage-appropriate treatment category according to current guidelines, which will be compared against the treatment received. Logistic regression will be used to examine predictors of receiving stage-appropriate treatment. For Aim 3, the outcome will be survival among patients with cirrhosis who developed HCC. Receipt of appropriate HCC surveillance and stage-appropriate treatment will be the primary predictors of interest. Marginal structural modeling will be used to examine these associations. IMPACT: This study will examine an important and timely issue in the management of HCC that has not been examined. Findings from this study will provide a strong foundation for developing and implementing a future intervention to improve the delivery of guideline-based HCC surveillance and treatment.
Validation of Case Finding Algorithms for Hepatocellular Cancer From Administrative Data and Electronic Health Records Using Natural Language Processing.
Authors: Sada Y, Hou J, Richardson P, El-Serag H, Davila J
Source: Med Care, 2013 Aug 6;null, p. null.
EPub date: 2013 Aug 6.