Feasibility Files Help Researchers Manage Scope
December 18, 2013
NCHS' record linkage program gives researchers a valuable resource, enabling them to take the "long view" and examine the factors that influence disability, chronic disease, health care utilization, morbidity, and mortality. Researchers also have another tool to determine the scope of available research, saving them time and even providing an opportunity to assess data before finalizing a research proposal: feasibility files.
To maximize the use of restricted-use linked files, NCHS has created publicly available Feasibility Study data files that can be downloaded directly from the NCHS website. These feasibility files provide a limited set of variables that can be used to determine the maximum available sample size for each linked file. Each feasibility file is survey-specific, and includes a NCHS survey public identification number, survey respondent eligibility status for linkage, final match status, and variables specifying on which data files the successfully linked survey respondent has a record. Although the feasibility files do not provide any information about benefits or payments received by the survey respondent, they can be merged with the public-use survey datasets to obtain demographic and health status information.
Dr. Jennifer Parker, Special Projects Branch Chief, says that feasibility files are invaluable to researchers contemplating submitting a proposal to NCHS' Research Data Center (RDC). While the files do not address every aspect of a study, they can give users the ability to determine how many respondents can be expected to be linked to Medicaid, Medicare, and Social Security files. This can save researchers time in the RDC, and can even help them avoid the inevitable disappointment of discovering that the linked files they have selected don't provide sufficient data to answer their research questions.
Dr. Parker says that using the feasibility files gives researchers an opportunity to think about their analysis beforehand and to fashion research based on the available data. Researchers can determine if they need to combine years of data to achieve the appropriate sample or rethink the scope of their study altogether. For example, if someone is looking at people in the 2004 National Health Interview Survey linked to 2004 Medicare files, he or she may want to consider also linking to 2005 or 2006 Medicare data to catch people too young for the 2004 data.
As research dollars become scarce, using feasibility files at the beginning of the RDC proposal process is an invaluable way to maximize those dollars. Free and easily accessible through NCHS' Public-Use Data Files program, feasibility files can contribute significantly to the quality of RDC users' research.