| Friday, February 16 | |
| PS2 Poster Session 2 and Refreshments |
Fri, Feb 16, 5:15 PM - 6:30 PM
Salons F-I |
Data, Data Everywhere …, but Mind the Disclaimers: Benefits and Costs of Matching Large Cohorts to Individual US Mortality Case Data in the NDI, SSA Death Master File (DMF/SSDI), and More (303683)*Sigurd Wilson Hermansen, WestatKeywords: mortality ascertainment, SSA DMF, NDI, tracing, survival analysis, prospective cohort studies In prospective mortality studies tracing residents of the USA, big data present both opportunities and challenges. Mortality rates increase with study subject age, facilitating predictions of numbers of decedents and survivors. A very large cohort that includes diversity of subject risk factor behaviors for higher mortality offers natural contrasts among groups. The devil is in the details of coverage, accuracy, balance, availability, and affordability. A patchwork of State and Federal agencies record >2.5 million deaths each year at the individual level. Using mixes of mortality data and applying data science and econometric methods to very large cohorts, 100 million DMF records, NDI search results containing good and false matches, and/or web search results can benefit your mortality search strategy and your final results.
|
|
