Search

Objective To describe the use of multiple data sources to monitor overdoses in near real-time in order to evaluate response to the provincial overdose emergency Introduction On April 14, 2016, British Columbia (BC)’s Provincial Health Officer declared a public health emergency due to a significant increase in drug-related overdoses and deaths in the Province. Despite the declaration, 161 suspected drug overdose deaths were reported across the Province in December 2016, a 137% increase over the number of deaths occurring in the same month of 2015 [1]. In response to the surge overdoses, Vancouver Coastal Health Authority (VCH), one of 5 health regions within BC, rapidly implemented a number of novel harm reduction initiatives. Overdose Prevention Sites (OPS) were opened on December 8, 2016. At these sites, people using illicit drugs are supervised by peers who can provide rapid intervention if an overdose occurs. The Mobile Medical Unit (MMU), a temporary state-of-art medical facility, was deployed on December 13, 2016 to reduce the congestion for the BC Ambulance Service (BCAS) and a major urban emergency department (ED) [2]. Following deployment of the MMU, services were transitioned to a permanent program at the Downtown Eastside Connections Clinic (DTES Connections) in the spring of 2017. DTES Connections was created to provide rapid access to addiction treatment [3]. In order to keep pace with the rapidly increasing number of novel harm reduction initiatives, enhanced surveillance programs were implemented at VCH to monitor and evaluate these innovative harm reduction activities, including development of new surveillance programs for the MMU, OPS and DTES Connections, along with existing routine surveillance system from EDs and a Supervised Injection Site (Insite). Methods Since 2011, after a spike of heroin-related deaths was reported in the Vancouver region, VCH started weekly monitoring of overdoses at nine EDs and Insite. Daily data extracts from EDs are automatically transferred to a secure driver by secure file transfer protocols. Groups of ICD 9/10 codes and keywords were refined to identify overdoses from EDs. A formal epidemiological evaluation was conducted to measure the algorithm’s accuracy in 2013. A live connection with Insite database was set up in 2011. Overdose events at Insite are clinically determined by clinical staff. Substance injected, characteristics of overdose event and emergency interventions are entered in the database. With the implementation of MMU, OPS and DTES Connections, a series of protocols were developed to monitor visitors' information and overdose events from each site. Demographic information, visit information, clinical presentations and substance used are collected from MMU and DTES Connections. A subset of data fields, including client handle, visit information, substance involved, overdose occurrence, naloxone intervention and ED transfer, are collected from OPS to minimize impact on peers and community partners who run the sites. Results Between November 2016 and January 2017, a sharp increase in overdoses was identified from EDs and Insite. Opioids, especially fentanyl and analogs, most likely contributed to the sudden increase. Weeks with government income assistance payment showed an even greater increase in overdoses. Since December 2016, six OPS opened in Vancouver. Four of them are still operating and one received federal approval to become a supervised consumption site. By September 2017, there were 184,760 visits to the OPS. 1,017 overdoses were reversed. A total of 2,798 patients visited the MMU during the whole operation period. 589 (21%) presented from treatment of overdose. The highest number of overdose visits occurred on December 21, 2016 after that month’s income assistance payment. Since then, the number of visits fluctuated with most visits driven by non-overdose related reasons. 89% of overdose visits arrived by BCAS and 79% of overdoses needed emergent and urgent care. 108 patients were transferred to DTES Connections by BCAS for treatment of overdoses by September 2017. All patients presented with opioid addiction issues. As of the end of September 2017, no deaths were reported from OPS, MMU and DTES Connections since operations. Conclusions As VCH continues responding to the drug overdose emergency in face of increasing drug overdoses, enhanced surveillance data have been widely used by the VCH Emergency Overdose Response Committee for decision making on harm reduction activities, such as expanding operation hours at OPS and Insite on income assistance payment days; examining the impact on EDs of opening the MMU; encouraging users to avoid using alone; opening new supervised injection service and women’s only OPS; and referring ED patients with non-fatal overdose to rapid access opioid agonist treatment and outreach follow-up.

Objective To describe the potential impact of using toxicology data to support drug overdose mortality surveillance. Introduction Although Marin County ranks as the healthiest county in California, it ranks poorly in substance abuse indicators, including drug overdose mortality. 1 Death certificates do not always include specific detail on the substances involved in a drug overdose. 2 This lack of specificity makes it difficult to identify public health issues related to specific prescription drugs in our community. We analyzed 2013 drug overdose death toxicology reports to determine if they could improve the description of drug overdose deaths in our community and to describe associated data characteristics. Methods Toxicology reports were requested from the Office of the Sheriff- Coroner for 37 drug overdose deaths among Marin County residents, comprising 95% of the 39 total drug overdose deaths in 2013. The remaining two deaths were excluded as they were associated with inhalation of therapeutic gases. Select information from toxicology reports was entered into a database for aggregate analyses. Drug overdose deaths were considered “fully detailed” if they included the specific types of drugs involved in the death and did not use any broad language to describe the death (i.e. narcotic, multiple drugs). Student’s T-tests ( α =0.05) were used to identify significant differences between groups of interest. Results Of the 37 drug poisoning deaths analyzed, 34 (92%) had available toxicology information. The remaining three (8%) deaths occurred outside of Marin County and were thus investigated by another jurisdiction. A basic toxicology panel was ordered on 17 (50%) of the 34 drug overdose deaths, while an expanded toxicology panel was ordered on the remaining 17 (50%). Alcohol was identified in the toxicology screen of 15 (44%); Amphetamines were identified in 8 (24%); and opiates were identified in 25 (74%) drug overdose deaths. Among the 25 deaths with at least one opiate identified on the toxicology screen, the majority (52%, n=13) also had alcohol present. The majority of drug overdose deaths, 18 (53%), did not have full information about the type of drug involved. The average number of drugs identified on the toxicology screen of all 34 drug overdose deaths was 6 (SD: 3). The average number of drugs identified in the toxicology screen significantly differed (p=0.0001) between causes of death that were fully detailed (Mean: 4; 95% CI: 3-5) and those that were not fully detailed (Mean: 8; 95% CI: 7-10). Conclusions Data from the Sheriff-Coroner’s office provided detail on the types of drugs involved in overdose deaths; however, it is difficult for local public health practitioners to make decisions about causality or contributions of these drugs to the death. These data may be useful in understanding the difference between fully detailed and non-detailed drug overdose deaths, and a broader context of drug combinations associated with these deaths. Less drugs were identified in the toxicology screen of deaths that were fully detailed, suggesting that overdose deaths that are not fully detailed may be exceedingly complex, making it difficult for medical examiners and coroners to assess causality. Approximately three-quarters of 2013 drug overdose deaths contained opiates on the toxicology screen, indicating that opiates may be a significant contributor to overdose deaths in our community. Our results are descriptive in nature; therefore, even though alcohol or opiates were identified on the toxicology screen, they may not be responsible for the overdose death. Given that over half of our 2013 overdose deaths were not fully detailed with drug type, local jurisdictions should work closely with their corner and/or medical examiner to fully detail death certificates with drugs involved in overdose deaths.

Substance abuse results in considerable morbidity each year in the United States. The relationship between drug overdose emergency department visits and deaths that opiate use contributed to or caused in Virginia was assessed for 2012-2013 to determine if syndromic surveillance data can be used to provide a real-time and accurate picture of the population at risk for drug overdose. Drug overdose emergency department visits showed divergent demographics from opiate deaths with respect to sex and age distributions. This may indicate a shift in the population previously identified as at risk for drug use morbidity.

We determined if the Rhode Island Real-time Outbreak and Disease Surveillance (RODS) system identified an increase in emergency department (ED) overdose visits during a known cluster of illicit-drug overdose deaths in RI and characterized ED overdose visits. When stratified by ED there was a significant increase in overdose chief complaints from one ED during March - May 2013 compared to the previous year. This coincides with the cluster of drug overdose deaths in March 2013. Despite most chief complaints for overdose not specifying cause, syndromic surveillance systems provide an existing platform that could be used to better assess ED overdose visits.