Restricting access to US death records could have serious consequences for long-term health studies. Government agencies should rethink their decision.
People who receive kidney transplants live twice as long as those who rely on dialysis. Trans fats are bad for your health. High alcohol consumption is linked with increased rates of breast cancer. High caffeine consumption is not. Daily aspirin use is associated with a reduced risk of death from cancer.
Large and long-term population studies have revealed these and other important findings. All required access to death records, which a policy from the US Social Security Administration (SSA) is making more difficult and more expensive.
The SSA produces the Death Master File, or Social Security Death Index, which includes more than 89 million records and lists names, death dates and social-security numbers. Data come from state governments, postal authorities, financial institutions, relatives of the deceased, health-care providers and funeral homes, and are updated weekly; the file has been publicly available since 1980. Last November, however, the SSA stopped making state death data generally available, saying that such information is protected by privacy laws.
Some 4 million entries have been removed from the public file because of privacy concerns. In future, an estimated 1 million of the 2.8 million records collected each year by the SSA will no longer be made public. That will make it much harder for researchers to track down whether a participant in a study has moved or died. It could also prevent researchers from collecting accurate details about circumstances of death while the memories of relatives and health providers are fresh.
Researchers involved in US longitudinal and cohort studies such as the Nurses’ Health Study and the Framingham Heart Study routinely consult the Death Master File, as do those at the Scientific Registry of Transplant Recipients. It was used to verify deaths in the Cancer Prevention Study II, and so has contributed to more than 200 published studies from the American Cancer Society. Health networks and funders must know who has died so that they can evaluate quality of care, the long-term benefits of a procedure or the risks of an intervention.
But fraudsters also have uses for death records: criminals can use the listed social-security numbers for identify theft. However, the benefits of restricting access to the database to curtail fraud must be weighed carefully against the real potential for harm to public health. Without access to death records, research in all areas of medicine will be hindered. Ironically, efforts to prevent other kinds of fraud will also be obstructed, because the records can be used to check whether the holder of a social-security number has died.
“ There is a lot of room between making data accessible indiscriminately and cutting off access. ”
There are other sources of death records. Investigators can submit the names of study participants to the Centers for Disease Control and Prevention (CDC), which will then search its National Death Index for matches in data supplied by the states. This index is more complete than the Death Master File, but it is updated less frequently — only every 18 months or so. It is also less convenient to use, and more costly. The CDC must pay states for their records, and these costs are passed on to researchers. The agency is working closely with state offices to adopt electronic filing systems that can keep lists more current, but the costs are unlikely to come down.
Even studies that already rely on the CDC still use the Death Master File for corroboration and extra information. Without access to this resource, researchers will need to spend further research dollars and effort on tracking down participants. They may have to send expensive signed-for letters to ensure that they reach someone, or chase through defunct contact information. The quality of studies will suffer. The more participants are lost to follow-up, the less certain will be the conclusions of long-standing, labour-intensive cohort studies.
The SSA should reconsider its decision to withhold information. Everyone involved should work to find a compromise that minimizes the risks of identify theft while maintaining researchers’ access to all the data in the Death Master File.
Before the latest policy took effect, nearly anyone could gain access to the records. Genealogy sites posted data for free or at minimal cost, and it is true that these records were easy pickings for identity thieves. However, there is a lot of room between making data accessible indiscriminately and cutting off access. The medical community is accustomed to working with sensitive, confidential information. To give accredited researchers access to death records, the SSA could adopt registration and permissions systems similar to those used for viewing and receiving medical information.
Given how many federal dollars will be wasted if access to the Death Master File is restricted — and how many public-health benefits will be lost or delayed — officials must work urgently to make these crucial records available for research once again.