Last week I had the opportunity to attend the PfS https://www.cdc.gov/drugoverdose/states/state_prevention.html and DDPI https://www.cdc.gov/drugoverdose/foa/ddpi.html Awardee Meeting in Atlanta, Georgia. These meetings were held concurrently with another awardee meeting, the Enhanced State Surveillance of Opioid-Involved Morbidity and Mortality - ESOOS https://www.cdc.gov/drugoverdose/foa/state-opioid-mm.html. These three programs support various initiatives across 44 states to address the opioid overdose epidemic. Presenters shared various approaches to surveillance for opioid overdoses and I have compiled a brief list of my takeaways, questions and next steps.
Several states participating in ESOOS are providing emergency department (ED) data on opioid-related ED visits to the CDC Division of Unintentional Injury Prevention via the National Syndromic Surveillance Program ESSENCE platform. This is an exciting approach to sharing ED data with CDC programs not directly affiliated with NSSP and hopefully one that can be expanded in the future as time and resources permit.
The CDC Division of Unintentional Injury Prevention is working with NSSP ESSENCE to develop opioid-related indicators to use in ESSENCE. The first definition added is for heroin overdose (https://www.cdc.gov/nssp/news.html#heroin) and additional indicators are under development. In addition, this collaboration plans to collect and incorporate locally-developed definitions into NSSP ESSENCE for all ESSENCE participants to use as they see fit.
There are tremendous opportunities for similar collaborations among syndromic surveillance and injury epidemiologists at state and local levels. Please share any best practices or barriers for collaborations across program areas with the ISDS Community of Practice.
The transition to ICD-10-CM on October 1, 2015 has created additional challenges in conducting surveillance for opioid overdoses in emergency department (ED) data in most states. While the initial expectation was that hospitals would use the “T codes” for opioid overdoses (Poisoning by drugs, medicaments and biological substances), many ED visits with a chief complaint or triage note mentioning overdose do not have any poisoning diagnosis codes; instead they receive an “F code” (Mental and behavioral disorders due to psychoactive substance use). Additional analyses are needed to identify why F codes are used instead of T codes; one hypothesis is that overdose patients who received naloxone prior to arrival in the ED and are stable upon arrival may receive an F code while those receiving naloxone in the ED receive a T code.
Additional work is needed with natural language processing methods to take full advantage of the data in ED triage notes and EMS narratives. While these data elements can provide invaluable contextual information related to overdoses and can help to identify overdose encounters that may not be identified through other submitted data elements, techniques that improve specificity and account for negation and historical information must be applied.
Several states are creating publicly-available dashboards that present a variety of opioid overdose indicators. Links to a few examples are below.
Rhode Island: http://preventoverdoseri.org/see-the-data/
If you are interested in overdose surveillance issues, please consider joining the ISDS Overdose Surveillance Committee. More information is available at http://www.healthsurveillance.org/members/member_engagement/groups.aspx?code=Overdose and a kick-off call is scheduled for June 2, 2017.
NC DETECT Program Director
ISDS Board Member
Amy Ising is Immediate Past-President of ISDS