There is very limited information on the burden of oral and pharyngeal cancers for specials needs populations, including the frail elderly, the disabled, and those with AIDS/HIV. In addition, significant racial disparities exist, as black males have a much lower 5-year survival rate compared with white males. Medicaid is an important source of insurance coverage for low-income racial and ethnic minorities, and also for the special needs populations. Medicaid administrative data linked with cancer registry data can provide a powerful data source to understand health disparities among these subgroups. This study will analyze 1999 through 2004 Medicaid administrative data and statewide all-payer discharge data (to identify comparison groups, including those insured by private payers and Medicare) linked with cancer registry databases to address three specific objectives: (1) assess the ability to study the special needs population and comparison cohorts with oral and pharyngeal cancers in the linked databases;(2) generate descriptive statistics on the incidence of cancer, stage at diagnosis, and cancer mortality among priority populations (based on findings from the data quality assessment performed under objective 1) and compare with other populations;and (3) assess the utility of the linked database to study treatment patterns among priority populations. Data from the states of Georgia, Illinois, South Carolina, and Texas were selected based on specific selection criteria, including large Medicaid fee-for-serve populations and a high burden of oral and pharyngeal cancers. To address objective 1, we will assess the quality of linkage between Medicaid claims and the statewide discharge data with the cancer registry data, the ability to identify specific subgroups, and availability of adequate sample sizes to perform assessments. For objective 2, we will produce unadjusted and age-adjusted incidence and mortality rates. We will also generate the annual percentage change and perform joinpoint regression analysis to identify points where a significant change in the linear slope of the trend occurs. For objective 3, we will assess the number of months of continuous enrollment available in Medicaid, evaluate the percentage of claims with missing values, and assess the ability to identify relevant procedures and treatments using ICD-9 and CPT-4 codes. If this proposed study shows that claims data are of adequate quality to systematically study oral cancer treatments and address health disparities, a follow-on application will be submitted to perform comparative assessments of treatment patterns using cancer registry data linked with Medicaid, Medicare, and private payer claims data.
Medicaid is an important source of insurance coverage for low-income racial and ethnic minorities, as well as special needs populations, including the frail elderly, the disabled, and those with HIV/AIDS. Medicaid administrative data linked with cancer registry data can provide a powerful data source to understand health disparities among these subgroups. In this study, the utility of this linked database will be evaluated to assess disparities related to oral and pharyngeal cancers, and incidence, stage of diagnosis, and mortality rates will be derived for the special needs population enrolled in Medicaid and compared with non-Medicaid cohorts.
