Could the Recent Change in Availability of Death Master File Data Affect Your Research?
The information on this page is archived and provided for reference purposes only.
Cancer epidemiologists and other health researchers have long relied on the Social Security Administration's (SSA) Death Master File (DMF) data for determining the vital status of participants. Researchers find the DMF helpful in their observational studies and prevention and clinical trials because it is a single, centralized source of information across the 50 states, low cost, quite current, frequently updated, and fairly complete.
The DMF covers all deaths reported to the SSA from a variety of sources, including state death records, death reports from family members, funeral homes, hospitals, states, Federal agencies, postal authorities, and financial institutions.
Many cancer epidemiologists use the DMF as part of their routine procedures for following up cohort study participants for vital status. For example, the NIH-AARP cohort of older Americans, which has more than half a million study participants across the United States, relies on linkages with the DMF and other sources to provide information about who in the cohort has died. Trialists involved in cancer screening in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial have used the DMF to help monitor adverse events that result in death and to gather information on deaths from malignancies.
Recently, SSA officials determined that Section 205r of the Social Security Act does not allow them to continue to disclose state death records that SSA receives through its contracts with the states, except in very limited circumstances. A fact sheet about this important change is available online.
What Does This Mean for Researchers?
Until November 1, 2011, information from state death records was included in the Public Death Master File. The DMF is still available through the National Technical Information System and other sources, but there is one big difference from before November 1, 2011. If the SSA knows of a death only from the states and not from any of its other sources of death information (which happens roughly more than one-third of the time), then those death data will not appear on the DMF.
While the DMF can be useful to researchers for deaths recorded in the file from some source other than through SSA's contracts with states, the absence of a record for a person on the DMF still could mean that the person has not died.
Fortunately, the National Death Index (NDI) still is available to researchers from the National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention (CDC). NCHS has the legislative authority and is mandated under the Public Health Service Act to collect vital statistics annually and is unaffected by the Social Security Act provisions.
The NDI is a centralized source of information for data on all deaths recorded in the 50 states. Researchers can link their data to the NDI, which is made available solely for medical and health research to ascertain vital status. Death records are added to the NDI file annually, approximately 12 months after the end of a particular calendar year.
Researchers in the extramural community, investigators at NCI and other NIH Institutes and Centers, and many other Federal health agencies have long used the DMF resource. NCI's Surveillance, Epidemiology and End Results (SEER) Program is one example. SEER is a key resource for cancer statistics in the United States. Individual cancer registries that are part of SEER use information from the DMF to search for missing social security numbers, check data (names, date of birth), and remove duplicate records from their cancer registry files, an important activity to ensure the completeness and quality of the SEER data.
Many individuals understand the impact of this SSA change and the importance and value of collecting death information from a single source that is timely and complete. NIH staff are working together with NCHS staff on a plan to accelerate the availability of the NDI and expedite the application process. NCI's Division of Cancer Control and Population Sciences (DCCPS) is interested in hearing from researchers. How have you used the DMF in your research?
For questions about the DMF issues, contact Debbie Winn in NCI's Division of Cancer Control and Population Sciences at email@example.com. We look forward to hearing from you.
Deborah Winn, Ph.D., is the Deputy Director of NCI's Division of Cancer Control and Population Sciences. Dr. Winn plays a central role in the planning, priority setting, development, and management of the Division's large integrated programs in cancer surveillance, epidemiology, behavioral science, health services, dissemination and diffusion, and cancer survivorship research. She is internationally recognized for her epidemiologic research on tobacco and head and neck cancer. Her other research interests include environmental influences on breast cancer and evaluating the impact of epidemiologic research.
Dr. Winn received an M.S.P.H. and her Ph.D. in epidemiology from the School of Public Health at the University of North Carolina at Chapel Hill. She also received a B.A. in government from Radcliffe College. She is a Fellow of the American College of Epidemiology and a member of the Delta Omega Public Health Honor Society. She has received numerous awards in recognition of her research and professional service, including from NIH and the U.S. Department of Health and Human Services (HHS), and the H.A. Tyroler Distinguished Epidemiologist Alumni Award from the University of North Carolina at Chapel Hill.
Gary Chase - May 30, 2012 at 3:33 PM (UTC -4)
I oversee respondent retention for the Nurses’ Health Study at Harvard. Until access to the Death Master File was curtailed, we used it daily to search for deceased participants. With over 260,000 individuals that we are following in our cohorts, we are always looking for cost-efficient ways to locate missing participants and to determine which of them are no longer living.
Being able to search the DMF has two primary benefits over other options for learning about deaths. First it is typically much more up-to-date than the National Death Index, which is comprehensive, but typically a year or more behind. Learning about the deaths of study members in a timely manner helps to reduce unnecessary mailings. It also expedites our ability to follow-up with next-of-kin to retrieve critical information about the cause of death. The longer the lag between a participant's death and the start of our follow-up, the less likely that we will be able to reach a family member. (Spouses often move or die themselves.)
The other major benefit of using the DMF is the ease and low cost of checking. If we have an older respondent with a good record whose mail suddenly becomes undeliverable, our first step would typically be to look for them in the DMF. Rather than waste a lot of resources trying to regain contact, if we can easily confirm a participant's death, we save a lot of time and money. Being able to punch a SSN into a website and know right away makes us much more efficient.
Having been at this job since before we had Internet-based search resources, I can say with great confidence that our access to the DMF was one of the biggest changes in our follow-up procedures. Access to the death file supports our scientific work and enhances what we are able to achieve. I look forward to renewed access to the DMF.
Meir Stampfer - June 2, 2012 at 4:46 PM (UTC -4)
As PI of the Nurses’ Health Study, and investigator in several other large cancer cohort studies, I can confirm that lack of access to DMF has significantly impaired the efficiency of our work. Complete followup of participatns is key to the integrity of findings from cohort studies, and timely identification of deaths enhances our abioity to obtain the relevant medical records and tumor tissue that are critical for cancer research. Without access to DMF, we must resort to slower and much more costly means to confirm deaths. I understand the abuses the arose from public access to DMF, but it should be easy to arrange for a system of restricted access, limited to bona fide researchers who are required to maintain confidentiality, similar to the highly successful system used by the National Death Index (NDI). These two resources are complementary: NDI provides causes of death, but is delayed, whereas DMF is up-to date, but with limited information. At a time when research dollars are so limited, it is a pity to withdraw access to DMF which was a highly cost-effective means to enhance our research capability.
Frances Ross - October 9, 2012 at 9:41 AM (UTC -4)
The Kentucky Cancer Registry strives to provide accurate and up-to-date cancer incidence information to the state legislature as they direct limited resources to efforts aimed at cancer control and prevention in Kentucky. These data are also shared with the National Cancer Institute and the Centers for Disease Control and Prevention in order to support cancer surveillance efforts nationwide. The NDI and the SSDI are essential tools for us to correctly identify patients seen in multiple facilities, in order to produce correct counts of cancer cases. They are also essential to correctly identify patients for follow up information, including date of death and cause of death. These data items are essential to a variety of cancer research efforts, and they are core to estimating the burden of cancer in Kentucky as well as the US.
Debbie Winn - October 23, 2012 at 4:50 PM (UTC -4)
Thank you Gary, Meir, and Frances for describing how you have used Death Master File (DMF) data and National Death Index (NDI) data, and how the Social Security Administration's (SSA) changes to the death data in the DMF will impact your research.
Additionally, I would also like to thank Gary for his contributions to and great quote in an October 8, 2012 New York Times article discussing this very issue titled "Researchers Wring Hands as U.S. Clamps Down on Death Record Access."
The information on this page is archived and provided for reference purposes only.