Understanding Death Can Improve Health - Making Mortality Reporting More Timely and Useful
Paula Braun joined the IDEA Lab in November 2014, as part of the Entrepreneur-in-Residence (EIR) program at the U.S. Department of Health and Human Services (HHS). The EIR Program brings together talent from within and outside government to tackle high priority issues in health, health care, and the delivery of human services. This is Paula's first blogpost in this EIR series documenting her project and experience.
Before I left public service in 2011, I lived and worked in active war zones in Baghdad, Iraq and Kabul, Afghanistan. I was compelled to return last fall when I heard that the Centers for Disease Control and Prevention (CDC) was looking to hire an Entrepreneur-in-Residence. Through the application process, I learned that the role would entail working with the National Center for Health Statistics (NCHS) on a "next-generation" electronic death registration system that will allow mortality data to flow more timely and accurately from the field to CDC. My friends and family ask, "What's so exciting about counting dead people?" My answer surprises them. Mortality statistics help us understand health and how it can be improved. Or, as stated more eloquently in a World Health Organization interview with Professor Prabhat Jha, we can "save lives by counting the dead". Knowing what kills us helps us measure progress toward important health goals and answer questions, such as: Which diseases are on the rise? What are their underlying causes? How can they be prevented? Who is at risk? Which interventions worked, and what disparities exist? The benefits of improved mortality reporting also extend far beyond public health. Birth and death data, also known as vital statistics, form the bedrock of our nation's statistical infrastructure. Nearly every major economic projection depends, at least in part, on vital statistics. The data help answer important resource allocation questions, such as, "What will the future demand for Social Security, Medicare, and Medicaid be?" and "What changes in tax policy are needed to meet this future demand?" The list goes on. Most importantly, though, the data can help save lives. That's why, in this effort, time is of the essence. In the United States, death reporting is a complex and decentralized system. The process begins when a heartbeat stops and a person is pronounced dead. Basic demographic information, such as the decedent's legal name, age, social security number, address, marital status, and place of death are verified, typically by a funeral home. A medical certifier pieces together the decedent's medical history and circumstances around the death to determine an immediate cause, underlying factors, and other significant conditions that contributed. If there are reasons to suspect unnatural causes, a medical examiner or coroner meticulously combs through the forensic evidence. The death is then registered. A certificate is issued, and the data flow continues from the field, through the States, territories, and other reporting jurisdictions to NCHS. NCHS compiles data from all 57 reporting jurisdictions (i.e., the 50 states, 5 territories, the District of Columbia and New York City) through the Vital Statistics Cooperative Program and makes this data available through the National Death Index, public use data sets, and other data for official use. Since death registration is not a Federal function, NCHS does not finalize and release national mortality data until all 57 jurisdictions have reported. Timeliness of the overall effort, therefore, depends on when the last jurisdiction submits data to NCHS. As timeliness, accuracy, and location granularity improve, the value of the data increases exponentially. A near real-time death registration system would allow CDC to identify emerging virulent diseases, predict where they are likely to spread next, and take steps to prevent further losses. With these benefits in mind, CDC developed an agency-wide strategy to improve CDC's public health surveillance activities. As part of CDC's Surveillance Strategy, NCHS established a goal for the 57 jurisdictions that participate in the Vital Statistics Cooperative Program to report 80% of deaths within 10 days of the event. Electronic death registration systems play an essential role in achieving this goal, however electronic systems alone will not address all of the challenges in the mortality data information flow. The National Association for Public Health Statistics and Information System recently issued a report on strategies for improving the timeliness of vital statistics. An integrated approach is needed to work across data silos and within existing constraints while keeping pace with advancements in technology. My role as CDC's Entrepreneur-in-Residence is to catalyze and promote a transition path to help move mortality reporting from its current state to one that provides more actionable insight, on a near real-time basis, about the progression of specific diseases, in specific locations, across time. Initial pilots indicate that vital statistics are a viable option for real-time disease surveillance of some infectious diseases. Flu surveillance provides a good example. During the 2013 - 2014 Influenza season, CDC ran parallel surveillance to compare data in real-time from CDC's traditional 122 Cities Mortality System and CDC's National Vital Statistics Systems. The pilot demonstrated that the vital statistics data offered more complete coverage than the 122 Cities Mortality System as well as other benefits, such as better adherence to case definitions for cause of death reporting, and more detailed data about the decedent. Additional benefits will be realized when the difference between the date of death and the date CDC receives the cause of death is minimized. To address these challenges, we're looking to the front line for ideas about specific functional needs and how best to transform the system. Much of the data required for the death certificate is already contained in electronic health records or other systems. Lack of integration between these systems and vital records puts additional burden on primary data providers, especially physicians, who must either enter identical information in multiple systems, or resort back to paper forms. This gap between "work as imagined" and "work as performed" is not unique to mortality data reporting; it results in workarounds that impede the flow and hinder the quality of public health data. While, under the current system, the mortality registration process starts when a heartbeat stops, there are clear advantages to capturing data entered into other electronic systems before the heartbeat stopped. My current ideas about "next generation" electronic death registration systems are nascent and evolving. It's clear that we need to get the user experience right for capturing and transmitting data. Also, instead of re-inventing the wheel, we'll need to leverage both existing standards as well as new approaches for integrating data. Ideally, the systems will allow data to be captured and maintained in a patient-centered way. Other important technical benefits include security, data integrity, auditability, usability across multiple platforms, extensibility, cost, and feasibility. As such, the "next-generation" system will likely be a hybridization of current systems that bridges existing standards and technologies, while making room for future innovations. Some examples of these innovations could include:
- A redesigned user interface for cause of death reporting that reduces burden on medical certifiers. The user interface would integrate with hospitals' electronic health record systems and help physicians better understand and adhere to World Health Organization standards for certifying the cause of death.
- Predictive models that can track disease trajectories to determine which early indicators are most likely to lead to premature death and what can be done to save lives.
- A medical device capable of providing a non- or minimally-invasive autopsy to aid medical certifiers in cause of death determinations.
Have ideas about how to make electronic death reporting more timely or useful? Send them to me - pabraun(AT)cdc.gov. EIRs are brought on for a fixed period of time to leverage their skill-set to develop solutions while also bringing new skills, experience and capabilities to HHS. Learn more about the EIR program.