Opportunities for emergency medical services intervention to prevent opioid overdose mortality

Abstract

Objectives:

The opioid crisis is a growing cause of mortality in the United States and may be mitigated by innovative approaches to identifying individuals at-risk of fatal opioid overdose. We examined Emergency Medical Services (EMS) utilization among a cohort of individuals who died from opioid overdose in order to identify potential opportunities for intervention.

Methods:

Individuals who died of unintentional opioid overdose in a large North Carolina county between 01/01/2014 and 12/31/2016 were studied in a retrospective cohort. Death records obtained from North Carolina Vital Records were linked to EMS patient care records obtained from the county EMS System in order to describe the EMS encounters of each decedent in the year preceding their death. Patient demographics and EMS encounters were assessed to identify encounter characteristics that may be targeted for intervention. Chi-square tests and odds ratios were used to evaluate and characterize the statistical significance of differences in EMS utilization.

Results:

Of the 218 individuals who died from unintentional opioid overdose in the study interval, 30% (n=66) utilized EMS in the year before their death and 17% (n=38) had at least one EMS encounter with documented drug or alcohol use (i.e. “drug-related encounter”). The mean age at death was 38 (range 19–74) years, 30% were female, 89% were White, and 8% were Black/African American. Factors associated with higher incidence of EMS utilization included age (P<.001), gender (P=.006), and race (P<.001). Decedents aged 56–65 had the highest EMS utilization (47%) and patients aged <25 and 25–35 had more drug-related EMS encounters (29% and 20%, respectively). The most common reasons for EMS utilization were “other medical” (27%), “non-traumatic pain” (20%), “traumatic injury” (16%), and “poisoning/drug ingestion” (14%). Drug or alcohol use was documented by EMS in 33% of all encounters and an opioid prescription was reported in 22% of encounters.

Conclusions:

Nearly one-third of individuals who died from accidental opioid overdose utilized EMS in the year before their death and nearly one-fifth had a drug-related encounter. EMS encounters may present an opportunity to identify individuals at-risk of opioid overdose and, ultimately, reduce overdose mortality.

Introduction

The United States (U.S.) is currently in the midst of an alarming opioid abuse crisis. Between 2000 and 2014, opioid-related sdeaths increased by 200%; in 2016 alone, over 40,000 deaths were attributed directly to opioid abuse/misuse (1,2). In October of 2017, Health and Human Services (HHS) Deputy Secretary Eric D. Hargan declared the opioid crisis a public health emergency. In 2018, U.S. Surgeon General Jerome M. Adams, M.D, M.P.H., released an advisory emphasizing the importance of public access to the overdose-reversing drug naloxone (3,4). While a national issue, specific regions of the U.S. experience a higher rate of opioid misuse and death for a myriad of socioeconomic and cultural reasons. North Carolina (NC) is home to four of the 25 U.S. cities with the highest opioid abuse rates, with a 464% increase in opioid-related overdoses from 2000–2016 (2,5).

Prior research has sought to delineate and categorize the numerous risk factors associated with opioid overdose and related mortality. The most consistent predictors of overdose include concurrent substance use disorders, middle-age, male sex, history of depression or other mental illness, high dosage opioid prescriptions, initiation of extended-release/long-acting prescription regimens, and individuals suffering from chronic pain (6). The extensive heterogenicity in persons at risk for opioid abuse adds an additional layer of complexity to an already enigmatic problem (7).

Individuals who experience a non-fatal opioid overdose are at an increased risk of death within one year (8–10). Individuals at high-risk for opioid overdose and related mortality are often in contact with health services in the time preceding their death. Gjersing et al. (11) found that patients who died from an overdose utilized ambulance services and acute care clinics in the year before their death. However, this study included drug overdoses of all types and did not differentiate between EMS and acute care clinics, characterize the ambulance encounters (e.g., reason for encounter, alcohol or drug use, opioid prescription), or describe the disposition of the ambulance encounters.

EMS personnel are uniquely positioned to identify patients in need of intervention and addiction services. The conclusions of several contemporary retrospective investigations indicate that EMS may be able to provide an avenue through which opioid abuse/misuse interventions and strategies can be employed (12–14).

Despite the growing amount of research investigating the role of emergency medical services in combatting the opioid epidemic, little research has explicitly investigated and categorized prior EMS encounters among a cohort of individuals who later died specifically from opioid overdose. As such, the purpose of our study was to identify patients in Wake County, NC who died of unintentional opioid overdose between 2014 and 2016 and catalogue the frequency with which EMS resources encountered these individuals in the year prior to death in order to identify potential opportunities for prehospital intervention. These findings have the potential to pave the way for policy development and improve identification of individuals at risk of fatal opioid overdose encountered by EMS.

Methods

Study Setting

Wake County is an urban county located in central North Carolina, with an area of 854 square miles and a population of 1,000,000+ residents. The Wake County EMS System is comprised of the Wake County Department of EMS and two contracted provider agencies, Cary EMS and Eastern Wake EMS. The System operates at an advanced life support (ALS) level, with at least one paramedic on each ambulance. The system received ~107,000 calls for service in 2018.

Study Design and Data

This is a retrospective cohort study of subjects who died from opioid overdose in Wake County, NC between 1/1/2014 and 12/31/2016. We accessed death record data provided by the North Carolina State Center for Health Statistics and patient care records provided by Wake County Emergency Medical Services. Patient care records were accessed through the electronic health record (EHR) system, ESO Solutions, which houses all Wake County patient care reports. The University of North Carolina Institutional Review Board (IRB) Committee reviewed this study’s procedures and determined that this study did not constitute human subjects research and did not require IRB oversight.

Inclusion Criteria

Subjects were persons of any age whose death certificate reported unintentional opioid overdose as the cause of death in Wake County, NC between 1/1/2014 and 12/31/2016. Opioid overdose deaths were defined as having an International Classification of Diseases, Tenth Revision (ICD-10) underlying cause of death code of unintentional poisoning (X40-X44) or poisoning of undetermined intent (Y12, Y14) and at least one opioid specific contributing cause of death ICD-10 code of T40.0-T40.4 or T40.6 assigned on the recorded death certificate filed with the North Carolina State Center for Health Statistics.

Data Matching

Identified death records were linked to Wake County EMS patient care records to define the cohort of subjects with EMS encounters in the year before death. EMS patient care records were linked using two variables: subject name and date of birth. The subject name was matched to EMS data by searching for an exact name match, a phonetic name match, and a common alias name match. For records matching on name, date of birth was manually verified to ensure at least two of three elements matched (i.e., month, day, and/or year). If date of birth was not available in EMS data, this step was excluded and the EMS data was only used if the name match resulted in a single match (i.e., exact, phonetic, or common alias) and no other patients existed with a similar name. The date of each EMS encounter was then compared to the decedent’s date of death. The encounter was included if it occurred within 365 days prior to the date of their death. If these three criteria were not met, the EMS data were excluded. All encounters classified as “cardiac arrest” were excluded to ensure that EMS encounters ultimately related to the subject’s death were not included in this study as potential opportunities for intervention.

Data Collection

Data extracted from death records included county of death, name, gender, race, date of birth (DOB), date of death (DOD), and cause of death classified using the ICD-10 codes. Data from EMS patient care records included date of encounter, reason for EMS encounter, naloxone administration, opioid administration, existing opioid prescriptions, disposition of encounter, and transport destination. Raw data were indexed into a pre-designed spreadsheet using Microsoft Excel (version 16.14.1, Redmond, WA).

Data Analysis

Descriptive statistics were used to describe subject demographics (e.g., age, sex, race) and EMS encounter characteristics (e.g., reason for encounter, naloxone administration). Chi- squared analysis was used to evaluate the statistical significance of differences in group proportions. Odds ratios were used to characterize the associations between subject demographics and EMS utilization. Number of EMS encounters per decedent was compared directly against demographics and was delineated into no encounters, one or more encounters, or one or more drug-related EMS encounters.

EMS encounter characteristics were categorized based on information provided by prehospital personnel under the “primary impression” and “report narrative” variables located in EMS patient care records. Encounter characteristics of interest included number of EMS encounters per patient within 365 days of death, reason for each EMS encounter, elements of drug or alcohol involvement, and days between last EMS encounter and death. “Reason for EMS encounter” was further subdivided into “altered level of consciousness (LOC)/altered mental status (AMS),” “behavior/psychiatric,” “other medical,” “pain (non-traumatic),” “poisoning/drug ingestion,” “respiratory,” or “traumatic injury.” “Elements of drug involvement” was also subdivided to highlight patients whose encounters included elements of drug abuse and/or opioid use including “suspected drug/alcohol use,” “naloxone administered,” “known opioid prescription,” and “opioid administration by EMS.” “Days between last EMS encounter and death” utilized data obtained from patient death records cross-referenced with last documented EMS encounter.

Prehospital opioid administration by EMS was compared between those patients whose EMS encounter records indicated either suspected drug/alcohol abuse or pre-existing opioid prescriptions. Additionally, the relationship between hospital transport and elements of drug involvement were examined. All analyses were completed using Microsoft Excel (version 16.14.1, Redmond, WA).

Results

Two hundred eighteen subjects met inclusion criteria as fatal opioid overdoses in Wake County, NC from 1/1/2014 through 12/31/2016. The mean age at death was 38 (range 19–74) years and individuals aged “25–35,” “36–45,” and “46–55” accounted for the highest percent of decedents (32.1%, 27.1%, 17.9%, respectively). Of the 218 decedents, 30.3% (n=66) were female, 89.0% (n=194) were White, 8.3% (n=18) were Black/African American, and 2.8% (n=6) were “other” (Table 1).

Table 1.

Number of EMS encounters by decedent demographics

	No EMS Encounters (n=152)		≥ 1 EMS Encounter (n=66)		≥1 EMS Drug-related Encounter (n=38)		Total (n=218)
Age	%	no.	%	no.	%	no.	%	no.
<25	61.29	19	38.71	12	29.03	9	14.22	31
25–35	71.43	50	28.57	20	20.00	14	32.11	70
36–45	72.88	43	27.12	16	11.86	7	27.06	59
46–55	74.36	29	25.64	10	12.82	5	17.89	39
56–65	52.94	9	47.06	8	17.65	3	7.80	17
>65	100.00	2	0.00	0	0.00	0	0.92	2
Sex
Male	75.00	114	25.00	38	14.47	22	69.72	152
Female	57.58	38	42.42	28	24.24	16	30.28	66
Race
White	73.71	143	26.29	51	14.95	29	88.99	194
Black/African American	33.33	6	66.67	12	38.89	7	8.26	18
Other	50.00	3	50.00	3	33.33	2	2.75	6

Note. Results of chi-squared analysis comparing decedent demographics by EMS utilization (i.e. no EMS encounters vs. ≥ 1 EMS Encounter) were significant for age (P<.001) and sex (P=.006) and race (P<.001)

Of the 218 decedents, 30.3% (n=66) utilized EMS in the year before their death. Of the decedents who utilized EMS at least once, 42.4% (n=28) were female, 77.3% (n=51) were White, 18.2% (n=12) were Black/African American, and the average age was 38 (range 21–65) years. Of the decedents who did not utilize EMS, 25.0% (n=38) were female, 94.1% (n=143) were White, 4.0% (n=6) were Black/African American, and the mean age was 38 (range 21–74) years.

The highest proportion of EMS utilization were among those aged 56–65 and <25 years (47.1% and 38.7%, respectively). Drug or alcohol use was documented most often among patients aged <25 and 25–35 (29.0% and 20.0%, respectively). Females were more likely to utilize EMS than males (OR: 2.2; 95% CI: 1.2–4.1). A larger proportion of Black/African American decedents utilized EMS than White decedents (66.7% and 26.3%, respectively), but the sample size of Black/African American decedents was comparatively small with only 18 decedents in total (Table 1).

The average number of encounters per individual who utilized EMS was 2.3 (SD 3.8; range 1–28). The majority of these individuals had one or two encounters (54.5% and 27.3%, respectively). Only 4.5% (n=3) had more than six encounters in the year preceding death with the highest number of encounters being 28. The majority of EMS encounters occurred within 180 days of death (77.3%; n=51), 31.8% (n=21) occurred within 30 days, 13.6 % (n=9) occurred within 5 days of death, and one encounter occurred the same day as death (Table 2).

Table 2.

EMS encounter characteristics

	Mean	Range
Number of EMS Encounters per patient	2.32	1–28
	Patients (n=66)
	%	n
1	54.55	36
2	27.27	18
3	6.06	4
4	4.55	3
5	3.03	2
≥6	4.55	3
Reason for EMS Encounters	Encounters (n=158)
	%	n
Other Medical	26.58	42
Pain (Non-traumatic)	20.25	32
Traumatic Injury	15.82	25
Poisoning/Drug Ingestion	13.92	22
Respiratory	8.23	13
Altered LOC/AMS	7.59	12
Behavioral/Psychiatric	7.59	12
Elements of Drug Involvement
Suspected Drug/Alcohol Use	32.91	52
Naloxone administered	10.76	17
Known opioid prescription	22.15	35
Opioid Administration by EMS	8.86	14
	Mean	Range
Days between last EMS Encounter and Death	106.08	0–365
	%	n
<5	13.64	9
5–30	18.18	12
31–180	45.45	30
180–365	22.73	15

EMS encounter types included “other medical” (26.6%; n=42), “non-traumatic pain” (20.3%; n=32), “traumatic injury” (15.8%; n=25), “poisoning/drug ingestion” (13.9%; n=22), “respiratory” (8.2%; n=13), “altered LOC/AMS” (7.6%; n=12), and “behavioral/psychiatric” (7.6%; n=12). Drug or alcohol use was documented by EMS in 32.9% (n=52) of encounters and naloxone was administered in 10.8% (n=17) (Table 2). These 52 drug-related encounters belonged to 38 (57.6%) of the individuals who utilized EMS (Table 1).

Of the remaining individuals (n=28) who utilized EMS but had no drug-related encounters, 42.9% (n=12) had at least one encounter in which an existing opioid prescription was documented by EMS. EMS administered opioids in 8.9% (n=14) of all encounters (Table 2). Of the encounters in which opioids were administered, EMS documented a prescription for opioids in 42.9% (n=6) of cases and suspicion of drug or alcohol use in 21.4% (n=3) of cases.

Of all EMS encounters, 81.7% (n=129) resulted in transport to the hospital. This included 82.7% (n=43) of encounters with documented drug or alcohol use, 81.1% (n=86) of encounters with no documented drug or alcohol use, and 83.3% (n=15) of encounters in which EMS administered naloxone (Table 3). Eleven individuals (16.7%) utilized EMS at least once and were never transported for any of their encounters.

Table 3.

Transport status by elements of drug involvement

	Transported% (n)	Not Transported % (n)	Totaln
Drug-related encounters	82.69 (43)	17.31 (9)	52
Non-drug-related encounters	81.13 (86)	12.66 (20)	106
Naloxone Administered	83.33 (15)	16.67 (3)	18

Discussion

This study combines local population demographics and opioid-specific mortality statistics from a high-risk region of the U.S. with EMS encounter data to form a better understanding of the role that prehospital resources may play in reducing the impact of opioid abuse and related mortality. Our analysis revealed that approximately one-third of individuals who died from opioid overdose employed EMS resources in the year prior to death. Several common encounter characteristics may provide opportunities for prehospital intervention of individuals at risk of a fatal opioid overdose.

Higher incidence of EMS utilization was associated with age, gender, and race. EMS utilization was highest among individuals aged 56–65 and <25 years. Encounter types were predominantly drug-related for younger decedents (individuals aged <25 and 26–35), while older decedents demonstrated lower proportions of drug-related encounters. None of the decedents in the 65 years and over age group utilized EMS. These findings suggest that while many younger at-risk individuals present to EMS with a readily identifiable substance abuse problem, EMS providers may need to employ alternate methods to identify older individuals at risk for future opioid overdose. Female decedents demonstrated markedly higher proportions of EMS utilization (OR: 2.2; 95% CI: 1.2–4.1). However, similar proportions of females and males had at least one drug-related encounter (OR: 1.9; 95% CI: .9–3.9). A higher proportion of Black/African American decedents utilized EMS but only 8.3% (n=18) of all decedents were Black/African American. A larger sample size is needed to draw accurate conclusions regarding utilization. Overall, the demographics of individuals who died of opioid overdose corroborate the results observed in similar studies, indicating middle-age white males as the population at highest risk for opioid-related mortality (1,15,16).

The most common reasons for EMS utilization among overdose decedents were “other medical,” “non-traumatic pain,” “traumatic injury,” and “poisoning/drug ingestion.” Decedents were more likely to have an encounter where EMS recognized and documented drug or alcohol use than they were to be seen primarily for “poisoning/drug ingestion.” Only 14% of encounters were documented as “poisoning/drug ingestion” while EMS recognized drug or alcohol use in 32.9% of all encounters. These results suggest that offering intervention exclusively during “poisoning/drug ingestion” encounters is insufficient for reaching the majority of individuals at- risk for opioid abuse and overdose. EMS identification of drug or alcohol use outside of “drug ingestion” as a primary complaint is critical and may enable EMS to reach over half (57.6%) of the at-risk individuals whom they encounter.

Further development and standardization of EMS documentation of drug and alcohol use would enable more accurate and efficient analysis of missed opportunities for interventions. The patient care reports reviewed in this study included a standardized “alcohol/drugs” drop-down data entry point where EMS providers select “none,” “unknown,” “patient admits to alcohol use,” “patient admits to drug use,” or “alcohol and/or drug paraphernalia at scene.” Due to the inconsistency of documentation among EMS providers and the unspecific nature of the categories, we also reviewed the “narrative” portion of the patient care record for documentation of suspected and/or patient-reported drug or alcohol use.

The reason for a high proportion of EMS encounters for non-traumatic pain (20%) is likely multifactorial and highlights the complex relationship between pain and opioid use. Dunn et al. (17) explored the connection between chronic pain and increased opioid overdose risk and found multiple contributing factors to include long-term exposure to opioids, combining long- acting opioids with short-acting for breakthrough pain, and comorbid medical conditions. Without further investigation, it is unknown if these individuals presented to EMS for acute medical pain, poorly controlled chronic pain, or drug-seeking behavior related to opioid use disorder. The potential for EMS to intervene during these encounters is variable but given the frequency in which at-risk individuals present for non-traumatic pain, EMS must make an effort to further explore the circumstances surrounding their complaints to recognize underlying opioid misuse.

It is of little surprise that in this study, a significant proportion of primary complaints were for traumatic injury given that alcohol and drug use have been well established as risk factors for injury (18,19). It is common for trauma centers to screen trauma patients for substance abuse, conduct a brief intervention, and refer them to appropriate resources (18). EMS is the only health care provider to see first-hand the circumstances of these injuries and has the opportunity to recognize signs of drug use while on scene. With training to recognize drug use outside of overdose, a larger proportion of at-risk individuals could be identified and reached. A well-established model called Screening, Brief Intervention, Referral to Treatment (SBIRT) already exists but has not yet gained traction in EMS (20). Maragh-Bass et al. (21) conducted a qualitative study to gauge the interest of EMS providers in participating in a pilot program to implement this technique. They concluded that EMS providers recognized their unique position to deliver brief motivational messages and were inclined to participate.

Despite the potential for EMS to identify a substantial portion of at-risk individuals through recognition of drug and/or alcohol use, a large proportion (42.4%) of the individuals seen by EMS had no encounters with documented drug or alcohol use. This suggests that additional measures are necessary to elucidate individuals at-risk for future opioid-related mortality. Nearly half (42.9%; n=12) of the individuals with a history of only non-drug related EMS encounters had opioids documented on their medication list by EMS personnel at some point prior to overdose expiration. Hirsch et al. (22) illuminated the contribution of prescription medications in drug overdoses and found that nearly half of overdose decedents in North Carolina filled a prescription for at least one of the drugs that contributed to their death within 60 days of death.

Although we cannot elucidate the circumstances of these individuals’ respective deaths without further information, these data suggest the existence of a subset of individuals who may benefit from safe opioid prescription use counseling or screening for substance abuse or misuse by EMS during encounters. Upon recognition of a current opioid prescription, EMS personnel could inquire about the individual’s use of opioids and potentially identify vulnerable individuals who may not initially be perceived as at-risk for overdose.

The vast majority of subjects utilized EMS within six months of their overdose death. These findings suggest that at-risk individuals may have a higher propensity to utilize EMS in the months leading up to their death, when they are most vulnerable. Maeng et al. (23) reached similar conclusions and found an escalation of utilization of acute care and primary emergency departments as early as two years prior to opioid overdoses. For the individuals who do not receive transport to the hospital, EMS personnel may be the last health care providers they encounter prior to lethal overdose, adding significant weight to the critical role these providers can play in averting disastrous outcomes.

Additionally, EMS personnel will typically only have a few opportunities to intervene in the time prior to fatal overdose. Upwards of eighty percent of individuals who utilized EMS resources prior to death had between one and two encounters prior to overdose expiration. These findings indicate that a small window of opportunity exists in which EMS resources can have the most significant impact on the reduction of opioid-related overdose and associated mortality. In order to take advantage of this brief but critical opportunity, EMS personnel must have adequate training to recognize drug use and agencies must be willing to devote the necessary time and resources to the cause.

The prehospital nature of EMS care presents a novel opportunity to distribute overdose reversal agents without the need for patients to be brought in for standardized inpatient observation. Current protocols in Wake County, NC allow for distribution of naloxone kits to persons as risk of experiencing an opioid overdose or in a position to assist a person at risk of overdose (24). In 2016 alone, EMS personnel administered naloxone over 13,000 times for opioid-related encounters in NC (16). Often, patients who have been successfully revived in the field with naloxone refuse medical transport and/or further aid, leaving against medical advice.

As overcrowded emergency departments have become inundated with opioid overdose patients, new strategies have arisen which seek to capitalize on data suggesting that patients who leave against medical advice following reversal of opioid/heroin overdose with naloxone are in no immediate life-threatening danger secondary to the event and, as such, do not require additional hospital resources (26–28). Such methods, colloquially dubbed “treat and release” protocols, have gained significant popularity under the mounting pressure to reduce the economic burden of caring for opioid abuse patients at the inpatient level (29,30). A comprehensive meta-analysis by Kolinsky et al. (25) critically reviewed five studies in which patients left against medical advice following opioid overdose reversal in an effort to establish the potential efficacy of a treat and release system. Out of almost four-thousand patients identified across all five studies, only three deaths were attributed directly to rebound toxicity directly related to naloxone administration, suggesting that patients are not at an increased risk for life-threatening events following naloxone reversal and may not benefit from additional hospital-led observation. Additionally, data gathered from EMS systems regarding in-field administration of naloxone following opioid overdose could potentially provide area-specific geographic and demographic statistics that could significantly enhance targeted administration of prophylactic drug and social resources to populations at highest risk for fatal abuse.

Despite the currently utilized “treat and release” protocol for opioid overdose patients in Wake County, the transport rate for encounters in which EMS administered naloxone was still higher than the reported transport rate of all EMS encounters within Wake County during the timeframe of this study (83.3% and 70%, respectively). The stringency of EMS protocol requirements that must be met for a patient to be safely “released” likely contributes to this high transport rate. EMS protocols vary but many are similar to Wake County EMS by requiring: 1. the patient disclose isolated intravenous opioid use and not have consumed oral opioids; 2. the dosage of naloxone required to restore their respiratory and mental status must not exceed 2 mg; and 3. the patient must not have been in cardiac arrest. In addition, the patient must consent to the administration of an additional 2 mg naloxone intramuscularly to lower the likelihood of re-overdose and not have any other medical conditions requiring treatment (24). When indicated, “treat and release” protocols reduce strain on hospital resources and are safe for patients. It is especially important that EMS offer resources to patients who decline transport and are precluded from resources that may have been offered in the emergency department.

Wake County EMS has implemented a new program that encapsulates this concept well. The day after a patient overdoses and receives naloxone, Advance Practice Paramedics (APPs) visit the patient with a certified peer support counselor to provide resources. Peer support counselors are individuals living in recovery from substance use who are trained to assess readiness for recovery and provide person-centered peer support services (31). This resource was not available during the timeframe of this study.

Interestingly, none of the encounters included in this study resulted in transport to an alternate destination (e.g., community mental health and substance abuse treatment facility). In Wake County, NC, medical protocol allows EMS to offer transport to an alternate destination if the patient is experiencing an acute mental health crisis, to include substance abuse crises (32). It is unknown what proportion of our encounters met criteria for transport to an alternate destination or were provided with this option. A recent study characterized patients diverted to WakeBrook, one of the alternate destinations utilized by Wake County EMS. This study found that “substance abuse” was the second leading reason for patient transport to this facility (32). It is notable that none of the individuals who were transported to alternate destinations died from opioid overdose during the timeframe of this study. This suggests that non-hospital community resources specializing in substance use could provide a valid, though currently underutilized, resource for this patient population.

Based on this initial research, it can be reasonably concluded that emergency medical services interact with a large proportion of individuals at-risk of fatal opioid overdose and are well-positioned for targeted prehospital intervention. This study highlights the ongoing need for additional investigation of the feasibility and efficacy of EMS intervention in suspected opioid related encounters. Our research indicates the possibility of a role for EMS resources in opioid abuse screening and safe-use.

Limitations

The limitations encountered during this study were primarily those inherent to a retrospective design, including unintended reporting bias and missing or incomplete data. Additionally, the number of EMS encounters that opioid overdose decedents had in the year before death could have conceivably been underestimated, as some individuals may have encountered EMS outside of Wake County, NC. The analysis of encounter characteristics (e.g., reason for encounter, drug or alcohol use, opioid prescription) was dependent on documentation by EMS providers; variability in reporting between providers could have led to misclassification. Furthermore, it is possible that some encounters classified as “drug-related” could have involved alcohol and not drugs since encounters were included if “alcohol and/or drug paraphernalia at scene” was selected as a drop-down option on the patient care report. Finally, it is possible that a small number of EMS encounters could have been for the same incident that ultimately led to the patient’s death, as hospital records were not linked to EMS and death records.

Conclusion

This research suggests that EMS is well-positioned to contribute to the nationwide effort to reduce overdose mortality, beyond immediate life-saving overdose reversal. We concluded that EMS came into direct contact with nearly one-third of all individuals who died of opioid overdose in the year before their death. In addition, nearly one-fifth of overdose decedents had an EMS encounter with recognized drug or alcohol use. This suggests that EMS has an opportunity to reduce overdose mortality by not only intervening after overdose, but also after recognition of drug use. We recommend future research to explore the feasibility and efficacy of interventions by EMS to include Screening, Brief Intervention and Referral to Treatment (SBIRT), EMS follow-up after encounters with suspected drug use, transport to drug treatment facilities, and harm reduction counseling. In addition, we recommend standardization of documentation surrounding drug and alcohol use among EMS agencies to aid in future research efforts and development of interventions.