Charting a Different Course for Death Data
Surveillance and Data — Blogs and Stories
Updated June 30, 2020
By Caron Peper, MPH, CPH, Vital Records Data Analytics State Program Director and Rachel Ragland, MPH, Vital Records Epidemiologist, Indiana State Department of Health
When plans don’t go as expected, the best resource you’ll ever have is the right people. Here at the vital records office in Indiana, we encountered some unexpected twists and turns in trying to speed up our death reporting, but we always kept moving in the right direction through the sheer ingenuity of our team.
A Non-Natural Death search reveals the number of coroner cases we see in a month, based on records where the manner of death was Accident, Homicide, Suicide, or Undetermined. Source: Indiana Death Data Quality Assessment, 7/2019
The best laid plans
Faced with the growing epidemic of deaths from drug overdoses, we knew we needed to get death data moving faster to get a clearer picture of the crisis. We joined a project led by the National Center for Health Statistics (NCHS) to help states improve their death reporting systems.
When we joined the project, we did what most states do first, which is to look for better technology. We knew that getting a more modern electronic death reporting system (EDRS) would solve a lot of the problems we had with quickly and accurately reporting overdose deaths. For example, a newer system would be more interoperable with systems at the national level. It would also increase the speed of death records by allowing funeral directors, medical examiners, and physicians to work on the same death certificate at the same time.
NCHS’ goal is for 80% of all death records to be submitted within 10 days. Source: Indiana Death Data Quality Assessment, 7/2019
But then our plans to get a new system unexpectedly fell through.
Going “old school”
Since we couldn’t get the new technology, we had to go “old school” and figure out how to move the data faster using the resources we already had. Fortunately for us, one of those resources was a small team of data analysts stationed within the vital records office. The team had been put in place by our new director, who had determined that there needed to be more knowledge of data and data quality within the vital records group.
NCHS’ goal is for 90% of all drug overdose death records to be submitted within 90 days. Source: Indiana Death Data Quality Assessment, 7/2019
Tapping into the team’s expertise, we decided to analyze ourselves. We began putting together data about our own processes. A new data dashboard pulled together all the information that could help us see where our operations could be improved. It helped us find our “pinch points,” or bottlenecks in the system that were slowing down death reporting.
Listening to the data
The dashboard revealed a lot – some things we expected, and some we didn’t.
From January-June 2019, physicians certified almost 85% of records and coroners certified about 15%. A large percentage of coroners and a quarter of physicians were taking longer than 5 days to certify death records. Source: Indiana Death Data Quality Assessment, 7/2019
First, we saw that only about half of our death records were reaching NCHS within a 10-day timeframe, which is an important part of public health surveillance. Because our electronic system isn’t fully interoperable with NCHS, we were manually submitting our records in batches once a week. We changed the process to twice a week, and now more than two-thirds of our records reach NCHS within 10 days.
The dashboard also revealed that we had too many “pending” death records in the queue. When a medical examiner or coroner needs to wait for test results or other investigation, they submit the death certificate to the state vital records office with the cause of death marked “pending” and complete the record when the missing information is received. However, coroners get very busy, and death certificates often go without being finalized for a long time. We assigned a person in our office to call the coroners with a friendly reminder to check on specific records that are more than 45 days old, and it’s appreciated by our coroners.
The other big thing the dashboard told us – contrary to what we had assumed – was that medical examiners and coroners were not the main source of delay for most of our death records. Physicians turned out to be slower at filling out death certificates. We’re working with them now to figure out how we can make the process easier for them.
A course for the future
Today, we use the data dashboard regularly to track progress and guide the way – to tell us what’s working well and what still needs improvement. In the meantime, we’ve kept working toward getting that new EDRS we needed. We expect to have it by the end of 2020.
The contributions of our data analysts and our state’s participation in the NCHS project have improved our death data and prepared us to better handle any crisis that comes our way. The COVID-19 pandemic has reinforced the importance of this work: faster and more accurate data can help Indiana monitor, understand, and prevent deaths in our state, no matter the cause.