Mortality Data: Death Certificates

and Vital Statistics

Last Updated: March 2022

Local staff at funded health departments in states and Washington, DC abstract data from ME/C reports, death certificates, and postmortem toxicology reports into CDC’s SUDORS.

In some states, ME/Cs release their own drug overdose reports which are often of higher quality and more detailed than data in vital statistics.

In a growing number of states, state vital statistics systems release preliminary data on drug overdose deaths. Example reports provided below:

State Unintentional Drug Overdose Reporting System

SUDORS abstracts data from ME/C reports and integrates with death certificate data and toxicology reports. Data include demographics, toxicology results, and overdose-specific context and circumstances. Preliminary data can be reviewed by the state and new cases can be entered into the system at an accelerated rate during an outbreak response. Example reports are below:

For more information about SUDORS see the CDC OD2A website.

Glossary

Toolkit Resources

Get more insights by using our toolkit resources.
Go to Resources

Glossary

Learn the definition of the key words being used.
Go to Glossary

Thanks

Thank you, to all of our contributors.
View our contributors

Using the power of epidemiology to improve
the public’s health.

Contact Us

Have any questions or recommendations, you can contact us at overdose@cste.org

Swift Toxicology for Opioid Related Mortalities (STORM)

The STORM system is a rapid death toxicology testing program that operates out of the Western Michigan University Medical School that leverages post-mortem biological samples submitted by participating medical examiner offices. New data are available on a weekly basis. A 4-week rolling average and standard deviation for fentanyl, heroin, xylazine, cocaine, amphetamine, and methamphetamine are calculated by county. If a given county is three standard deviations above the mean for that rolling average, an alert will be generated. Especially lethal substances such as carfentanil generate an alert for one or more associated fatalities in the state.

Michigan staff run SAS code every week based on updated data to generate alerts.

Partner Systems/Systems External to the Health Department

The poison control center will be especially useful in working with these data as you are likely to have some drug-specific questions that only a toxicologist can adequately answer. Law enforcement and drug seizure data, if possible, should be used to add context. Changes in drug seizure data, in some states, have been shown to correlate with fatal overdoses (New Drug Threats: Latest on the Stimulant Crisis PPT).

How to modify parameter settings if results are not useful

In addition to the modification described in the first section, the adjustments below can be made to STORM alerts:

Multiple ICD codes may be assigned to a single visit, using patient visits for use rather than counts of ICD codes will provide a more robust threshold.
Consider drug lethality when setting an alert to ensure that you do not receive more alerts than desired for less lethal substances.

Swift Toxicology for Opioid Related Mortalities (STORM)

Michigan staff run SAS code every week based on updated data to generate alerts.

Partner Systems/Systems External to the Health Department

How to modify parameter settings if results are not useful

In addition to the modification described in the first section, the adjustments below can be made to STORM alerts:

Multiple ICD codes may be assigned to a single visit, using patient visits for use rather than counts of ICD codes will provide a more robust threshold.
Consider drug lethality when setting an alert to ensure that you do not receive more alerts than desired for less lethal substances.