AI · Tech Breaking
By mid-September, the Pennsylvania Department of Health had reported four confirmed measles-associated deaths, surpassing the highest annual number of measles-associated deaths in the United States since 1992. Pennsylvania public health professionals are sounding the alarm as the number of measles cases statewide exceeds 800 , with cases clustering among communities with a history of vaccine hesitancy. But if you look at the Centers for Disease Control and Prevention’s measles dashboard page , it shows just one measles-related death so far this year, with an asterisk.
Amid a tussle between Pennsylvania and the federal government, Health and Human Services Secretary Robert F. Kennedy Jr., recently stated that the CDC will only “report measles-caused deaths when the National Center for Health Statistics (NCHS) reports measles as the underlying cause based on death record information submitted by the states.” When we first read this, as former senior leaders at NCHS, we questioned the approach. This new method deviates from how CDC historically has relied on notifications from state medical officials, including information about symptoms, hospitalizations, and deaths, to immediately publish real-time “associated” death cases that include any deaths with a positive laboratory result for the outbreak condition.
By doing so during active outbreaks, CDC has played a key role in making real-time state-level public health surveillance information available for effective and timely responses. Shifting away from the rapid notification approach and relying strictly on mortality statistics generated by NCHS will not provide the surveillance information needed at the beginning of an active outbreak when counts are low. To see why, we need to understand the center’s mission and data processing and reporting procedures.
As the nation’s principal federal statistics agency, NCHS’s mission is to produce high-quality, trustworthy, statistical information for informing public health policy and actions. Collecting, compiling, and releasing information on death certificates is part of its mission. Before death counts are released by NCHS, death certificates are first completed by state and local medical officials including physicians, medical examiners, and coroners.
The medical officials must determine all causes of death — underlying and contributing — listed on the death certificate by adhering to established national guidelines. States’ vital registries then submit death certificate information to the NCHS National Vital Statistics System. The time between death occurrence and data submission to NCHS can vary from state to state.
When the data are received, NCHS verifies and codes the certificates using automated systems, with all causes of death coded with the International Classification of Diseases system. Once coded by NCHS, data are made available in CDC WONDER, a primary repository for disseminating NCHS mortality data within one to two weeks. A small portion of the certificates with inconsistent or complex information, or that identify rare causes of death, are manually adjudicated, sometimes by referring the certificate back to the state.
Manual adjudication for such cases can extend the time needed for release, ranging from weeks to even months in some cases. Data in CDC WONDER, including information on underlying and contributing causes, are available to anyone who needs the information, including HHS and infectious disease specialists at the CDC. Regarding the Pennsylvania measles cases, even when NCHS eventually verifies that those death certificates have measles as an underlying cause of death and make the data available in CDC WONDER, death counts on the CDC measles tracker dashboard will remain with an asterisk for some time.
This asterisk notation is a result of strict data suppression rules enforced by NCHS for protecting the confidentiality of the decedent. NCHS does not report death counts in any category from one to nine. The total number of measles-causing deaths in the U.S. without naming Pennsylvania or any other localities must surpass the threshold of 10 before CDC WONDER will show the actual death count for any use, including for the CDC measles dashboard.
The CDC now will shift to recognizing measles-causing deaths on its tracker dashboard only when measles is identified as the underlying cause, not contributing cause, of death on a death certificate. Although using only the underlying cause of death is consistent with NCHS’ standard practice for its official national mortality statistics for purposes of annual ranking and trending analysis, the new CDC approach effectively precludes the federal measles dashboard from being used to track “measles-associated deaths,” a traditional epidemiological outbreak definition that includes measles as either the underlying or a contributing cause essential for tracking the severity of the current outbreak. Nonetheless, CDC WONDER contains both underlying and contributing causes of death for researchers and decision-makers for use to make informed decisions.
Contrary to the new CDC plan, Pennsylvania needs to actively track measles-associated deaths because even if measles is not the underlying cause of death, measles severely weakens the immune system, substantially increasing a person’s risk of fatal secondary infections or complications. A discrepancy between how federal and state public health officials determine measles deaths will only increase public confusion about how dangerous the measles outbreaks have been and add to the public’s mistrust of conflicting information from governmental public health institutions. Furthermore, by not recognizing the broader death definition and relying only on NCHS reporting, HHS could refuse to act until the death counts with measles as the underlying cause rise to more than nine, compromising our ability to effectively contain the country’s deadliest measles outbreak in over two decades.
Denys T. Lau, Ph.D., is the AJPH editor-in-chief and was with NCHS for more than 11 years, first as director of the Division of Health Care Statistics and then as senior science adviser of the Division of Health and Nutrition Examination Surveys. Jennifer D.
Schoendorf, Ph.D., is the AJPH senior deputy editor and retired from NCHS in 2024 after nearly 30 years, most recently as the director of the Division of Research and Methodology.
Source: STAT
