Abstract
Background:
As stimulant use increases across the United States, emergency medical services (EMS) are crucial touchpoints in the health care system. To better measure the prevalence of stimulant use, misuse, and EMS incidents related to stimulant intoxication, definitions for stimulant-related incidents (SRIs) are needed.
Methods:
We used the Massachusetts Ambulance Trip Record Information System (MATRIS) from 2013 to 2020 to develop definitions of stimulant-related incidents. EMS runs reported to MATRIS were categorized based on stimulant-related words and symptoms. The three tiers were “any stimulant use” (class 1), “problematic stimulant use” (class 2), and “acute stimulant-related incidents” (class 3). A group of four reviewers studied over 650 cases in eight rounds to refine the search terms, achieving definitions with a correct characterization of over 80% of cases that the code selected.
Results:
SRI definitions were applied against all EMS runs within Massachusetts between 2013 and 2020 (n = 6,584,836 runs). Of these, 43,538 (0.7%) met the class 1 definition, 38,669 (0.6%) met the class 2 definition, and 19,157 (0.3%) met the class 3 definition. Incidents at all tiers of severity increased over time and were more likely to occur among younger adults and males. Race and ethnicity data indicated that Hispanic/Latinx and Black non-Hispanic/non-Latinx residents formed a disproportionately large percentage of SRIs relative to their total percentage of EMS runs.
Conclusions:
The prevalence of all three tiers of SRIs are increasing in Massachusetts, and this protocol provides a source of administrative data on stimulant use that complements sources such as hospital, treatment-based, and/or prescribing records.
Keywords: Stimulant, Amphetamine, Methamphetamine, Overdose, Emergency medical service
1. Introduction
Globally in 2019, the estimated prevalence of cocaine use was 0.4% and of amphetamine use was 0.5% in individuals ages 15–64 (“World Drug Report”, 2021). Dependence affects 16% of those with cocaine use and 11% of those with amphetamine use (Farrell et al., 2019). Internationally, the highest proportion of individuals with cocaine use and with amphetamine use were found in the regions of North America and Oceania (especially Australia and New Zealand) (“World Drug Report”, 2021, Farrell et al., 2019). Within the United States (US), the rate of stimulant misuse and use disorder is steadily increasing (Ciccarone, 2021). From 2015 through 2018, there were an estimated 1.6 million adults in the US who reported methamphetamine use in the past year, half of whom met the criteria for methamphetamine use disorder (Jones et al., 2020a). Even in regions of the country where stimulants such as methamphetamine have historically been uncommon, supply and use are steadily increasing (Vivolo-Kantor et al., 2020). Among persons with co-occurring substance use disorders, those who use stimulants are a particularly vulnerable subset of the population. Individuals with amphetamine-type stimulant use along with opioid use disorder (OUD) in general are younger, have higher rates of comorbid chronic diseases, are less likely to seek treatment for their substance use issues, and experience higher rates of homelessness, unemployment, and incarceration than those with only OUD (Fischer et al., 2021; Daniulaityte et al., 2020; Chawarski et al., 2020).
Though some individuals may intentionally co-use stimulants and opioids, many are exposed to synthetic opioids (e.g., fentanyl) unknowingly due to a contaminated stimulant supply (Jones et al., 2020b). As a result, stimulants are increasingly involved in fatal and non-fatal drug overdoses. Between 2013 and 2018, the rates of both cocaine-involved and psychostimulant-involved overdose deaths roughly tripled in the US (Hedegaard et al., 2020; Mattson et al., 2021). Among these deaths, there are substantial racial and ethnic disparities. In 2018, Black non-Hispanic/non-Latinx (nH/nL) individuals experienced the highest cocaine-involved overdose mortality rate while American Indian/Alaska Natives nH/nL experienced the highest psychostimulant-involved overdose mortality rate (Cano, 2021).
Our epidemiological understanding of stimulant use is limited because most data come from small cohort studies or administrative sources such as claims data, which are imperfect for a variety of reasons. First, much of the previous research identified patients using symptoms and their related International Class Diagnosis (ICD) codes from Emergency Department (ED) visits. Patients who present with symptoms such as acute cardiovascular complaints, those related to psychiatric, comorbid mental illnesses, or behavioral disturbances, and those that are traumatic in nature (including gunshot wounds, motor vehicle accidents, assaults, self-harm, and intentional overdoses) can be indications of stimulant use, but these diagnoses are non-specific and, therefore, not sufficiently probative (Richards et al., 2017; Chivaurah et al., 2019; Sibanda et al., 2019; Jones et al., 2018; Lee et al., 2009; Isoardi et al., 2019). Without the toxicology data that underlies these symptoms, misclassification is likely. Other data come from drug seizure records, which are also problematic, as seizures are considerably skewed by social, economic, and racial demographics, it is unknown how these seizures correlate with the overall consumption of illicit substances, and they are more likely to indicate drug availability rather than use (Beckett et al., 2006; Koch et al., 2016; Mitchell and Caudy, 2015). Even large surveys such as the National Survey on Drug Use and Health (NSDUH) provide an incomplete picture of stimulant use since it is a household survey that does not account for people who may be experiencing homelessness or incarceration, or those who are reluctant to disclose unlawful and stigmatized activities (Johnson and Fendrich, 2005; Reuter. et al., 2021). Additionally, unlike OUD, where an array of administrative records may help assess the prevalence of people enrolled in treatment, there are no FDA-approved medications for stimulant use disorder, and clinical treatment for the condition is less common than for OUD. Lastly, measures that rely on the number of patients seeking access to treatment or the use of harm reduction resources are subject to their availability and the volitional choices made by a person. Measuring fatal overdoses alone does not capture the predominantly nonfatal sequalae of substance use and can vary based on changes to the potency of the drug supply rather than the prevalence of consumption. To understand the scope of stimulant use disorder in the US, we need to integrate a broad array of data from various touchpoints in the health care system, social services, and systems of public administration.
One such source is the data generated by emergency medical service (EMS) responses. In contrast to the issues outline above, the substance-use related incidents that generate an ambulance response and an accompanying Massachusetts Ambulance Trip Record Information System (MATRIS) report occur with relatively high frequency, often occur regardless of the patient’s preferences or intentions, and are not dependent on the availability of a particular intervention as a proxy measure for substance use disorder (e.g., a county without a syringe exchange program has no means to use needle consumption as an indication of OUD prevalence, but calls for OUD-related emergency medical services in that county will continue to occur regardless). For many people who use stimulants, EMS is often an initial point of contact with the health care system, or results from police encounters where officers summon medical attention (Yatsco et al., 2020). As such, protocols that accurately identify stimulant-related incidents in EMS records can provide important insights to the changes in the prevalence and severity of stimulant use in a community.
Since 2016, Massachusetts has used MATRIS to track opioid overdose incidents in EMS data (“MA opioid-related”, 2021). While MATRIS was not specifically designed to track opioid overdose incidents, analysts used existing fields and created criteria for text searches to develop a definition that accurately identifies opioid-related ambulance trips (Bettano et al., 2021). This definition was used in a capture and recapture analysis to estimate the statewide prevalence of OUD (Barocas et al., 2018), a technique that was subsequently replicated in British Columbia (Min et al., 2020). Given the success of using EMS data to identify opioid-related incidents, we hypothesized that a similar methodology could be used to identify stimulant-related ambulance trips. Our objectives were to 1) develop a series of definitions for a range of stimulant-related incidents (SRI) for use in MATRIS data to identify individuals with stimulant use, misuse, and EMS incidents related to stimulant use and 2) characterize the population with SRI in Massachusetts from 2013 to 2020.
2. Methods
2.1. Data source and inclusion/exclusion criteria
We used MATRIS, a statewide database comprised of all ambulance responses or runs by licensed ambulance services in Massachusetts for our analysis. All emergency response runs from 2013 to 2020 entered in MATRIS were considered for inclusion (n = 7,114,600). We excluded runs with an indication of dialysis, a section 12 hold (i.e., an involuntary psychiatric commitment), transport only (i.e., signifying it is not an emergency response run), and those where the incident state or county were not in Massachusetts (n = 529,764). There were 6,584,836 total runs remaining.
2.2. Data elements
The structure and fields of MATRIS reports are based on the National Emergency Medical Service Information System (NEMSIS) standard (“Administrative Requirement”, 2019). Data collected during ambulance responses include dates and times, incident and transport location information, patient demographics and vital signs, medications administered, treatments/procedures performed, and additional narrative information pertinent to the scene and the patient’s care. We developed a tiered case definition classification scheme. EMS runs reported to MATRIS were categorized into the SRI tiers based on criteria of key words in the chief, secondary, or other patient complaint, the EMS provider’s primary, secondary, or other impression, the narrative report of the incident, and the patient’s age. Currently EMS provider impression fields are ICD – 10 codes, while all other fields used for the search besides the age of the patient were free text, meaning there were no guides or constraints regarding what data EMS providers could enter. We included search terms that are a collection of both stimulant-related words (prescribed and illicit) and stimulant-related symptoms (cardiac, psychiatric, behavioral, and substance use-related). The terms were coded into SAS, and the runs were analyzed in SAS Studio (Enterprise Edition 3.8) and Joinpoint (version 4.8.0.1). The text searches were case insensitive, treating upper and lowercase letters as being the same. This project was reviewed by Massachusetts’ Institutional Review Board (IRB) and determined to be exempt from IRB oversight.
2.3. Case definition classification process
A group of four reviewers including an infectious disease physician and addiction researcher (JAB), a former municipal chief of police and current public health researcher (BdP), and two state epidemiologists (AB and DB) each independently reviewed over 650 records in eight rounds to refine the search terms comprising the three definition classes in an iterative review process. After each round of review, if there was not complete agreement between the four reviewers on a case, they worked to consensus on how the case should be categorized. When definitions failed at accurately identifying stimulant-related runs, the authors tried to identify ways to improve the algorithm: these included requiring additional search terms to be found in conjunction with a less precise term, by supplementing the exclusion terms to remove false positives, and by adding in age criteria to increase the likelihood that a selected case would be correct. If a search term could not be improved enough to prevent a significant number of false positives (more than 30% of the sample), that search term would be removed. Additionally, if the review uncovered any new inclusion or exclusion terms for a given class, these were incorporated into the coding. Cases related to these changes were selected for assessment in the following round to determine if the new terms were incorporated satisfactorily and if any additional adjustments were required.
2.4. Analyses
After they were finalized, the case definitions were applied to all MATRIS reports between 2013 and 2020 to determine if they met the criteria for stimulant-related incidents. The results were summarized in descriptive statistics by sex (male, female, missing), age category (0–10, 11–14, 15–24, 25–34, 35–44, 45–54, 55–64, 65 +, missing), race/ethnicity (American Indian/Alaska Native & Other Race, nH/nL; Asian & Pacific Islander, nH/nL; Black, nH/nL; Hispanic/Latinx; White, nH/nL; missing), by year when the run occurred, and if the run was opioid-related. To understand if there were time trends in the occurrence of the stimulant-related runs, Joinpoint software from the National Cancer Institute was used to fit models for the yearly rates of SRIs by class per 1000 EMS Runs. Tests of significance were conducted by the Joinpoint software using a Monte Carlo Permutation method with an alpha of 0.05 for statistical significance.
3. Results
3.1. Case definitions and review results
The classes start with an all-inclusive categorization of stimulant incidents (class 1) and then resolve to greater levels of acuity as the classes progress. We defined class 1 cases as “any stimulant use,” which included EMS incidents where the reason for the call was an acute stimulant-related issue such as intoxication or its associated behaviors, the person had a documented or claimed history of stimulant misuse or use disorder, or the person had a known prescription for stimulants. The stimulant substances could be anything from those that are prescribed to substances that are uncontrolled (e.g., caffeine) to illicit substances. This initial group was deliberately broad and inclusive to form the base of the stimulant-related incidents collection; classes 2 and 3 were created as subsets of class 1. We defined class 2 as “problematic stimulant use,” which included EMS incidents in which there was an acute health or behavioral issue associated with illicit stimulant consumption (including misuse of prescription stimulants) or when EMS responded to a person with a history of stimulant misuse or use disorder. Class 3 is a subset of class 2 (and class 1). We defined class 3 as “acute stimulant-related incidents,” which included only EMS incidents in which prescribed or illicit stimulants were implicated as either the primary reason for the call or closely related to the health or behavioral issues observed at the call regardless of the reason for the initial response. For classes 2 and 3, we excluded any incidents in individuals ages younger than 14 as the rate of false positives was too high. Appendix 1 details the class definitions and a complete listing of their search inclusion and exclusion terms. A run that meets one or more of the criteria in a class is identified as meeting the definition for that class of SRI.
During the final review rounds for each class, we tested codes by compiling a random sample of cases meeting the class definition. For class 1, of the 100 cases that the code selected as meeting the definition, the reviewers agreed that 90% (n = 90) were cases that accurately indicated “any stimulant use.” For class 2, of the 98 cases that were selected by the code, the reviewers agreed that 87.8% (n = 86) were instances of “problematic stimulant use.” For class 3, of the 88 cases that were identified using the developed coding, the reviewers agreed that 81.8% (n = 72) were instances of “acute stimulant-related incidents.”
Of the 6,584,836 runs between 2013 and 2020, 43,538 runs (0.7%) met the definition of “any stimulant use” (class 1), 38,669 (0.6%) met the definition of “problematic stimulant use” (class 2), and 19,157 (0.3%) met the definition of an “acute stimulant-related incident” (class 3) (Table 1).
Table 1.
Demographics of all EMS runs and Stimulant-Related Incident runs Classes 1–3 in MA, 2013–2020.
| All Runs (%) | Class 1 SRIs (%) | Class 2 SRIs (%) | Class 3 SRIs (%) | |
|---|---|---|---|---|
| Total Runs | 6,584,836 (100%) | 43,538 (100%) | 38,669 (100%) | 19,157 (100%) |
| Sex: | ||||
| Male | 3,137,054 (47.6%) | 28,469 (65.4%) | 25,748 (66.6%) | 12,726 (66.4%) |
| Female | 3,353,532 (50.9%) | 14,860 (34.1%) | 12,812 (33.1%) | 6,390 (33.4%) |
| Missing | 94,250 (1.4%) | 209 (0.5%) | 109 (0.3%) | 41 (0.2%) |
| Age: | ||||
| 0–10 | 207,286 (3.2%) | 389 (0.9%) | N/A | N/A |
| 11–14 | 83,033 (1.3%) | 359 (0.8%) | N/A | N/A |
| 15–24 | 578,743 (8.8%) | 6,105 (14.0%) | 4,808 (12.4%) | 2,686 (14.0%) |
| 25–34 | 670,530 (10.2%) | 13,986 (32.1%) | 12,898 (33.4%) | 6,985 (36.5%) |
| 35–44 | 595,725 (9.1%) | 10,355 (23.8%) | 9,663 (25.0%) | 4,946 (25.8%) |
| 45–54 | 809,589 (12.3%) | 7,312 (16.8%) | 6,861 (17.7%) | 3,034 (15.8%) |
| 54–64 | 950,259 (14.4%) | 3,610 (8.3%) | 3,379 (8.7%) | 1,219 (6.4%) |
| ≥ 65 | 2,590,870 (39.4%) | 1,180 (2.7%) | 1,060 (2.7%) | 287 (1.5%) |
| Missing | 98,801 (1.5%) | 242 (0.6%) | 0 (0%) | 0 (0%) |
| Race: | ||||
| American Indian/Alaska Native & Other Race, nH/nL | 197,673 (3.0%) | 1,314 (3.0%) | 1,183 (3.1%) | 575 (3.0%) |
| Asian & Pacific Islander, nH/nL | 61,754 (0.9%) | 212 (0.5%) | 178 (0.5%) | 86 (0.5%) |
| Black, nH/nL | 261,159 (4.0%) | 2,719 (6.3%) | 2,560 (6.6%) | 999 (5.2%) |
| Hispanic/Latinx | 342,786 (5.2%) | 3,729 (8.6%) | 3,490 (9.0%) | 1,619 (8.5%) |
| White, nH/nL | 2,847,269 (43.2%) | 16,916 (38.9%) | 14,466 (37.4%) | 7,351 (38.4%) |
| Missing | 2,874,195 (43.7%) | 18,648 (42.8%) | 16,792 (43.4%) | 8,527 (44.5%) |
| Year of Run: 2013 | 691,930 (10.5%) | 2,662 (6.1%) | 2,369 (6.1%) | 1,039 (5.4%) |
| 2014 | 726,469 (11.0%) | 3,032 (7.0%) | 2,591 (6.7%) | 1,190 (6.2%) |
| 2015 | 796,442 (12.1%) | 3,546 (8.1%) | 3,043 (7.9%) | 1,385 (7.2%) |
| 2016 | 829,522 (12.6%) | 4,602 (10.6%) | 4,045 (10.5%) | 1,935 (10.1%) |
| 2017 | 896,799 (13.6%) | 6,593 (15.1%) | 5,889 (15.2%) | 2,892 (15.1%) |
| 2018 | 922,097 (14.0%) | 7,463 (17.1%) | 6,673 (17.3%) | 3,360 (17.5%) |
| 2019 | 895,945 (13.6%) | 7,734 (17.8%) | 6,928 (17.9%) | 3,383 (17.7%) |
| 2020 | 825,632 (12.5%) | 7,906 (18.2%) | 7,131 (18.4%) | 3,973 (20.7%) |
| Opioid-Related Incident: | ||||
| No | 6,430,413 (97.7%) | 32,789 (75.3%) | 28,274 (73.1%) | 12,785 (66.7%) |
| Yes | 154,423 (2.4%) | 10,749 (24.7%) | 10,395 (26.9%) | 6,372 (33.3%) |
(percentages may not add to 100% due to rounding)
N/A = code for Classes 2 and 3 SRIs requires patients to be older than 14
SRI = Stimulant-related incident; EMS = Emergency Medical Services; nH/nL = non-Hispanic/non-Latinx
3.2. Study population demographics and trends
Approximately half of all EMS runs were among female patients (47.6% male, 50.9% female), but for SRI runs, two thirds were male and one third were female. The total number of EMS runs increased by age category with older adults making up many of the runs: 39.4% of all runs were for patients ages 65 and older, whereas 23.4% of all runs were for patients ages 0–34. Among SRIs (in any class), the highest percent of runs were for patients between the ages of 25 and 44 (class 1, 55.9% of runs; class 2, 58.3% of runs; class 3, 62.3% of runs) (Table 1).
Race and ethnicity data are under recorded in MATRIS, with nearly half of all runs lacking these demographic data, both in the overall universe of MATRIS records and among the three classes of SRIs. Out of the total runs with a known race/ethnicity, White nH/nL patients made up 76.7% of runs whereas among stimulant-related incidents with a known race/ethnicity, White nH/nL patients made up fewer than 70% of runs in any of the SRI classes. The opposite was found in Hispanic/Latinx and Black nH/nL residents, who both made up larger percentages of SRI class runs with a known race/ethnicity than they did of total EMS runs with a known race/ethnicity (Black nH/nL residents were 7.0% of total runs but more than 10% of SRI runs while Hispanic/Latinx residents were 9.2% of total runs but over 15% of SRI runs). Among all EMS runs, 2.4% were categorized as being opioid-related; in stimulant-related runs the percent that were also opioid-related was much higher. In class 1 and class 2 stimulant-related incidents, one out of every four runs were also opioid-related and among class 3 stimulant-related incidents, one of every three runs were also opioid-related (Table 1).
The total number of SRIs has steadily increased by year. A Joinpoint trend analysis found runs that included “any stimulant use” (class 1) increased 15.5% per year for 2013–2020, runs that met the definition of “problematic stimulant use” (class 2) increased 16.1% per year for 2013–2020, and runs that met the definition of an “acute stimulant-related incident” (class 3) increased 19.4% per year for 2013–2020; all these trends were statistically significant (p < 0.05) (Graph 1).
Graph 1.
Stimulant-Related Incidents (SRI) per 1000 EMS Runs, 2013–2020, Caption: (SRI = Stimulant-related incident; EMS = Emergency Medical Services) This graph depicts the yearly trends of the stimulant-related incidents per 1000 EMS runs in MA between 2013 and 2020. Joinpoint modeling found that class 1 SRIs increased 15.5% per year, class 2 SRIs increased 16.1% per year, and class 3 SRIs increased 19.4% per year.
4. Discussion and Conclusion
4.1. Discussion
As SRIs increase in Massachusetts and across the nation, it is vital to have public health surveillance systems in place that can monitor these trends. MATRIS is a source of administrative data where individuals interacting with the health care system reveal their stimulant use (either as the cause of the event or during an interaction for another issue) that may otherwise not be recorded in other health care system data sources such as hospital, treatment-based, or prescribing records. In this way, our identification and classification process can be used to enhance the accuracy of prevalence measures of stimulant use and to add depth and context to other data sources. Additionally, MATRIS has been successfully used by the Massachusetts Department of Public Health to track the prevalence of opioid-related incidents, proving to be an important source of both fatal and non-fatal overdose information (“MA opioid-related”, 2021). The similar protocol for stimulants presented here can reveal important features of the growing problem of stimulant misuse, such as the relative severity of stimulant-related incidents and the populations most impacted. In its reliance on the low threshold of EMS contact with a person whose recollection or observed condition meets the relevant definitions, it can be used to forecast the need for treatment resources in a given geography, and to assess if stimulant prescribing trends are associated with behavioral health crises. When used in conjunction with other related measures, our identification and classification process has the potential to be a useful tool in measuring the severity and any changes in the stimulant crisis, allocating resources as necessary, formulating an effective public health response, and gauging the success of its implementation. Since MATRIS uses the national NEMSIS standard, this work is replicable in other states.
4.2. Limitations
This research has several limitations. First, determining the involvement of a particular stimulant relies on either EMS identification or patient recall. EMS staff do not have the ability to test substances in the field and therefore must rely heavily on their own professional knowledge of symptoms and on patient reports. Patients may be unaware of the exact substance they consumed or may be reluctant to disclose use if they fear reprisal or stigma, especially if police, family members, or employers are present. Furthermore, substances may be coused (intentionally or unintentionally), further complicating symptom presentation and patient accounts. Additionally, the MATRIS dataset only covers events in which 9–1–1 has been contacted or EMS is summoned to a location by police. This represents an incomplete subset of all stimulant use and misuse in the population and therefore findings are not generalizable to the population level without further analysis. Next, within the MATRIS system, each run is given a unique identifier, but each patient is not. While runs contain patient information such as their full names, date of birth, and residential address these entries are subject to human error which complicates using them to trace individuals over time. Analyses with a unique patient identifier would allow research into the frequency and timing of when these SRIs occur within individuals. Lastly, the stimulant-related incident tier searches are textbased queries that were developed without the benefit of text-mining software. Misspellings and abbreviations can subvert text searches, while negation terms placed too far from their objects (i.e., “patient has never used any medications such as Ritalin”) may be missed and result in false positives; without further study we cannot determine if these instances are randomly distributed or if they present as patterns among reports from particular EMS providers or regions. Use of improved natural language processing algorithms and random case reviews conducted periodically may help revise search terms as needed and reduce these errors.
4.3. Conclusions
Between 2013 and 2020, stimulant-related incidents in EMS data were more likely to occur among younger adults and male patients in Massachusetts. Stimulant-related incidents became more common over time, and initial race and ethnicity data suggests that Hispanic/Latinx and Black nH/nL residents may be overrepresented among SRIs relative to their total percent of EMS runs. Future plans for analyses involve including the SRI class definitions in the Massachusetts Public Health Data Warehouse, a patient-level, multi-year dataset that enables researchers to trace unique patients over time to analyze population health priorities and trends. In the Public Health Data Warehouse, these SRIs can be linked with insurance claims, hospital visits, toxicology results, and prescribing records to form a more comprehensive picture of stimulant use and misuse in Massachusetts. Doing so will allow public health officials, clinicians, and policy makers to understand the scope of stimulant use that may not be identified using hospital-or claims-based records.
Role of the funding source
This work was supported by the National Institutes of Health: The National Institute on Drug Abuse (USA) [DP2DA051864 to JAB and T32DA013911 to BdP] and by funds made available from the Centers for Disease Control and Prevention (USA) under Grant 1 [NB01OT009316-01-00] and the National Highway Traffic Safety Administration (USA) under Grant Section 405-c [# 20.616]. None of the funders had a role in the design, conduct, analysis of the study or in the decision to submit the manuscript for publication.
Appendix 1.:
Case Definition for Massachusetts Ambulance Trip Record Information System (MATRIS) Suspected Stimulant-Related Incident (SRI)
| NEMSIS Variable (s) | Criteria | |
|---|---|---|
| LEVEL 1 - (any stimulant use indicator) Defined as 1) EMS incidents in which there is an acute stimulant related issue (e.g., withdrawal, intoxication) from either prescribed or illicit stimulants, unintentional or intentional OR 2) EMS responds to a person with a history of stimulant misuse or use disorder, OR 3) EMS responds to a person with a known current or previous prescription for stimulants | ||
| Any Complaint [E09.05, E09.08, eSituation.03, eSituation.04] Or Narrative Report (NR) [E13.01, eNarrative.01] |
Criteria 1: Inclusion terms: stimulant, cocaine, cocain, coaine, coke, amphet, cacaine Exclusion terms: not stimulant, no stimulant, denies stimulant, denied stimulant, denies any stimulant, denied any stimulant, denies cocaine, denied cocaine, denies any cocaine, denied any cocaine, not cocaine, no cocaine, not coke, no coke, diet coke, rum and coke, jack and coke, bottle of, can of, glass of, cup of, denies amphet, denied amphet, denies any amphet, denied any amphet, not amphet, no amphet, coke cola, coke plant, over the counter, over-the-counter, vanilla coke, drink of |
|
|
Criteria 2: Inclusion terms: meth Exclusion terms: denies meth, denied meth, denies any meth, denied any meth, meth mile, [oiam]meth, meth[aieou], amethia, methotr, methanol, indometh, betameth, dexameth, methocarb, prometh, not meth, no meth, Dextromethorphan, sulfameth, simethic, methado, methodo, something, samething, Methuen, metheun, method, methyl | ||
|
Criteria 3: Inclusion terms: Speedball, speed ball, smoked crack, smoke crack, smoking crack, smokes crack, use crack, on crack, used crack, crack pipe, took upper, under the influence of upper Exclusion terms: crackle | ||
|
Criteria 4: Inclusion terms: Adderall, addarall, adderol, dextroamph, methylph, Dexedrine, Ritalin, Concerta, Strattera, atomoxetine, Vyvanse, focalin, methylin, procentra, zenzedi, evekeo, metadate, adzenys, aptensio, cotempla, daytrana, dyanavel, mydayis, quillichew, quillivant, lisdex | ||
|
Criteria 5: Inclusion terms: MDMA, bath salt, Cathinone, ecstasy, ecstacy, molly Exclusion terms: rn molly, molly rn | ||
| Any Impression [E09.15, E09.16, eSituation.11, eSituation.12] |
Criteria 6: Inclusion terms: cocaine related, F14.10, F14.11, F14.20, F14.21, F15.10, F15.11, F15.20, F15.21, F16 |
|
|
LEVEL 2 (problematic stimulant use indicator) Defined as 1) EMS incidents in which there is an acute stimulant related issue (e.g., withdrawal, intoxication) from an illicit stimulant or misuse of prescription stimulants OR 2) EMS responds to a person with a history of stimulant misuse or use disorder. | ||
| Any Complaint [E09.05, E09.08, eSituation.03, eSituation.04] Or Narrative Report (NR) [E13.01, eNarrative.01] & Patient Age [E06.14, E06.15, ePatient.15, ePatient.16] |
Criteria 1: Inclusion terms: stimulant, cocaine, cocain, coaine, coke, amphet, cacaine Exclusion terms: not stimulant, no stimulant, denies stimulant, denied stimulant, denies any stimulant, denied any stimulant, denies cocaine, denied cocaine, denies any cocaine, denied any cocaine, not cocaine, no cocaine, not coke, no coke, diet coke, rum and coke, jack and coke, bottle of, can of, glass of, cup of, denies amphet, denied amphet, denies any amphet, denied any amphet, not amphet, no amphet, coke cola, coke plant, over the counter, over-the-counter, vanilla coke, drink of AND Patient age > 14 |
|
|
Criteria 2: Inclusion terms: meth Exclusion terms: meth mile, [oiam]meth, meth[aieou], amethia, methotr, methanol, indometh, betameth, dexameth, methocarb, prometh, not meth, no meth, Dextromethorphan, sulfameth, simethic, methado, methodo, something, samething, Methuen, metheun, method, methyl AND Patient age > 14 | ||
|
Criteria 3: Inclusion terms: Speedball, speed ball, smoked crack, smoke crack, smoking crack, smokes crack, use crack, on crack, used crack, crack pipe, took upper, under the influence of upper Exclusion terms: crackle AND Patient age > 14 | ||
|
Criteria 4: Inclusion terms: MDMA, bath salt, Cathinone, ecstasy, ecstasy, molly Exclusion terms: rn molly, molly rn AND Patient age > 14 | ||
Criteria 5:INCLUDES AT LEAST ONE TERM from list a: Adderall, addarall, adderol, dextroamph, methylph, Dexedrine, Ritalin, Concerta, Strattera, atomoxetine, Vyvanse, focalin, methylin, procentra, zenzedi, evekeo, metadate, adzenys, aptensio, cotempla, daytrana, dyanavel, mydayis, quillichew, quillivant, lisdexAND ONE OF THE FOLLOWING TERMS from list b:
| ||
| AND NONE OF THE FOLLOWING TERMS from list c: adhd, as prescri, has a prescri, unrespons, uncons AND Patient age ≥ 14 * (snort is the only term that does not consider the exclusion terms in list c – i.e. if your search term from list b is “snort” then the exclusion terms in list c are not searched) |
||
|
Criteria 6: INCLUDES AT LEAST ONE TERM from list a: Adderall, adderol, dextroamph, methylph, Dexedrine, Ritalin, Concerta, Strattera, atomoxetine, Vyvanse, focalin, methylin, procentra, zenzedi, evekeo, metadate, adzenys, aptensio, cotempla, daytrana, dyanavel, mydayis, quillichew, quillivant, lisdex AND ONE OF THE FOLLOWING TERMS from list b: overdose, od, o.d AND NONE OF THE FOLLOWING TERMS from list c: unrespons, uncons AND Patient age > 14 |
||
| Any Impression [E09.15, E09.16, eSituation.11, eSituation.12] & Patient Age [E06.14, E06.15, ePatient.15, ePatient.16] |
Criteria 7: Inclusion terms: cocaine related, F14.10, F14.11, F14.20, F14.21, F15.10, F15.11, F15.20, F15.21, F16 AND Patient age > 14 |
|
|
LEVEL 3 (acute stimulant-related incident) Defined as an EMS incident in which either prescribed or illicit stimulants are implicated (e.g. withdrawal, intoxication, adverse effect) as either the primary reason for the call or as a related factor for the call. | ||
| Any Complaint [E09.05, E09.08, eSituation.03, eSituation.04] Or Narrative Report (NR) [E13.01, eNarrative.01] & Patient Age [E06.14, E06.15, ePatient.15, ePatient.16] |
Criteria 1 (searches complaint fields only): Inclusion terms: stimulant, cocaine, cocain, coaine, coke, amphet, cacaine Exclusion terms: not stimulant, no stimulant, denies stimulant, denied stimulant, denies any stimulant, denied any stimulant, denies cocaine, denied cocaine, denies any cocaine, denied any cocaine, not cocaine, no cocaine, not coke, no coke, diet coke, rum and coke, jack and coke, bottle of, can of, glass of, denies amphet, denied amphet, denies any amphet, denied any amphet, not amphet, no amphet, coke cola, coke plant, over the counter, over-the-counter, vanilla coke, drink of AND Patient age > 14 |
|
|
Criteria 2 (searches complaint fields only): Inclusion terms: meth Exclusion terms: meth mile, [oiam]meth, meth[aieou], amethia, methotr, methanol, indometh, betameth, dexameth, methocarb, prometh, not meth, no meth, Dextromethorphan, sulfameth, simethic, methado, methodo, something, samething, Methuen, metheun, method, methyl AND Patient age > 14 | ||
|
Criteria 3 (searches complaint fields only): Inclusion terms: Speedball, speed ball, smoked crack, smoke crack, smoking crack, smokes crack, use crack, on crack, used crack, crack pipe, took upper, under the influence of upper Exclusion terms: crackle AND Patient age > 14 Criteria 4 (searches complaint fields only): Inclusion terms: MDMA, bath salt, Cathinone, ecstasy, ecstasy, molly Exclusion terms: rn molly, molly rn AND Patient age > 14 | ||
|
Criteria 5: Inclusion terms: psycho, anxi, manic, parano, delus, aggress, violent, confused, crawling, public disturbance, public nuisance Exclusion terms: dog, no psycho, not psycho, denie[sd] pyscho, denie[sd] any psycho, no anxi, not anxi, denie[sd] anxi, denies [sd] any anxi, no manic, not manic, denie[sd] manic, denie[sd] any manic, no parano, not parano, denie[sd] parano, denie[sd] any parano, no delus, not delus, denie[sd] delus, denie[sd] any delus, no aggress, not aggress, no violent, not violent, no confused, not confused, no crawling, not crawling, no public, not a public AND Patient age > 14 | ||
|
Criteria 6 (searchers complaint field only): Inclusion terms: tachy, hypertensive, heart racing, heart pounding, arrythmia Exclusion terms: no tachy, not tachy, denie[sd] tachy, no hypertensive, not hypertensive, denie[sd] hypertensive, no heart racing, not heart racing, denie[sd] heart racing, no heart pounding, not heart pounding, denie[sd] heart pounding, no arrythmia, not arrythmia, denie[sd] arrythmia OR Inclusion terms: CVA AND Patient age < 46 Exclusion terms: no CVA, not CVA, denie[sd] CVA, AND Patient age > 14 | ||
|
Criteria 7: Inclusion terms: withdraw, poly[ -]sub, polysub, overdose Exclusion terms: alcohol withdraw, etoh withdraw, no withdraw, not withdraw, denie[sd] withdraw, no poly, not poly, denie[sd] poly, no overdose, not overdose, denie[sd] overdose AND Patient age > 14 | ||
|
Criteria 8: Inclusion terms: seiz, hyperthermia, hallucinat, de-escalat, taser, pepper spray, jitter, abscess, abcess Exclusion terms: seizure history, seizure hx, no seiz, not seiz, denie[sd] seiz, no hyperthermia, not hyperthermia, denie[sd] hyperthermia, no hallucinat, not hallucinat, denie[sd] hallucinat, no de-escalat, not de-escalat, no taser, no pepper spray, no jitter, not jitter, denie[sd] jitter, no abscess, not abscess, denie[sd] abscess AND Patient age > 14 | ||
| Any Impression [E09.15, E09.16, eSituation.11, eSituation.12] & Patient Age [E06.14, E06.15, ePatient.15, ePatient.16] |
Criteria 9: Inclusion terms: cocaine related, F14.10, F14.11, F14.20, F14.21, F15.10, F15.11, F15.20, F15.21, F16 AND Patient age > 14 |
|
|
Criteria 10: Inclusion terms: F20, F22, F29, F41, R41, F30, F60, R45.6, altered mental status AND Patient age > 14 | ||
|
Criteria 11: Inclusion terms: R00, I16, I49, cardiac arrhythmia/dysrhythmia, hypertension OR Inclusion terms: chest pain & Patient Age < 55 OR Inclusion terms I63 & Patient Age < 46 AND Patient age > 14 | ||
|
Criteria 12: Inclusion terms: F19.23, polysubstance overdose, psychoactive substance related diso AND Patient age > 14 | ||
|
Criteria 13: Inclusion terms: L02, R56, R50.9, R44.3, hallucinogen related disorders, hyperthermia, seizure, seizures with status epilepticus, seizures without status epilepticus AND Patient age > 14 | ||
Footnotes
CRediT authorship contribution statement
Amy Bettano: Conceptualization, Formal Analysis, Methodology, Investigation, Writing – Original Draft. Brandon del Pozo: Methodology, Investigation, Writing – Review & Editing. Dana Bernson: Methodology, Investigation, Writing – Review & Editing. Joshua A. Barocas: Conceptualization, Methodology, Investigation, Writing – Review & Editing.
Conflict of Interest
No conflict declared.
References
- Administrative requirement manual: Statewide EMS minimum data set. Massachusetts Office of Emergency Medical Services. Published February 2019. Accessed October 7, 2021. 〈https://www.mass.gov/files/documents/2018/12/17/ar-5-403-final.pdf〉 [Google Scholar]
- Barocas JA, White LF, Wang J, Walley AY, LaRochelle MR, Bernson D, Land T, Morgan JR, Samet JH, Linas BP, 2018. Estimated prevalence of opioid use disorder in Massachusetts, 2011–2015: A capture-recapture analysis. AJPH 108, 1675–1681. 10.2105/AJPH.2018.304673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Beckett K, Nyrop K, Pfingst L, 2006. Race, drugs, and policing: Understanding disparities in drug delivery arrests. Criminology 44, 105–137. 10.1111/j.1745-9125.2006.00044.x. [DOI] [Google Scholar]
- Bettano A, Jones K, Fillo KT, Ficks R, Bernson D, 2021. Opioid-related incident severity and emergency medical service naloxone administration by sex in Massachusetts, 2013–2019. Subst. Abus 20, 1–7. 10.1080/08897077.2021.1949661. [DOI] [PubMed] [Google Scholar]
- Cano M, 2021. Racial/ethnic differences in US drug overdose mortality, 2017–2018. Addict. Behav 112. 10.1016/j.addbeh.2020.106625. [DOI] [PubMed] [Google Scholar]
- Chawarski MC, Hawk K, Edelman EJ, O’Connor P, Owens P, Martel S, Coupet E, Whiteside L, Tsui JI, Rothman R, Cowan E, Richardson L, Lyons MS, Fiellin DA, D’Onofrio G, 2020. Use of amphetamine-type stimulants among emergency department patients with untreated opioid use disorder. Ann. Emerg. Med 76, 782–787. 10.1016/j.annemergmed.2020.06.046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chivaurah BM, Lienert D, Coates D, 2019. Amphetamine-type-substance-related presentations to the emergency department mental health team of a local health district in Australia. Austral Psychiatry 27, 369–373. 10.1177/1039856219848836. [DOI] [PubMed] [Google Scholar]
- Ciccarone D, 2021. The rise of illicit fentanyls, stimulants and the fourth wave of the opioid overdose crisis. Curr. Opin. Psychiatry 34, 344–350. 10.1097/YCO.0000000000000717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Daniulaityte R, Silverstein SM, Crawford TN, Martins SS, Zule W, Zaragoza AJ, Carlson RG, 2020. Methamphetamine use and its correlates among individuals with opioid use disorder in a Midwestern U.S. city. Subst. Use Misuse 55, 1781–1789. 10.1080/10826084.2020.1765805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Farrell M, Martin NK, Stockings E, Bòrquez A, Cepeda JA, Degenhardt L, Ali R, Tran LT, Rehm J, Torrens M, Shoptaw S, McKetin R, 2019. Responding to global stimulant use: challenges and opportunities. Lancet 394, 1652–1667. 10.1016/S0140-6736(19)32230-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fischer B, O’Keefe-Markman C, Lee AMH, Daldegan-Bueno D, 2021. ‘Resurgent’, ‘twin’ or ‘silent’ epidemic? A select data overview and observations on increasing psycho-stimulant use and harms in North America. Subst. Abus.: Treat. Prev. Policy 16. 10.1186/s13011-021-00350-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hedegaard H, Spencer R, Garnett MF, 2020. Increase in drug overdose deaths involving cocaine: United States, 2009–2018. U.S. Dept. of Health and Human Services, Centers for Disease Control and Prevention Natl. Cent. Health Stat. Data Brief. 2020. Published October 2020. Accessed October 8, 2021〈https://www.cdc.gov/nchs/data/databriefs/db384-H.pdf〉. [Google Scholar]
- Isoardi KZ, Ayles SF, Harris K, Finch CJ, Page CB, 2019. Methamphetamine presentations to an emergency department: Management and complications. Emerg. Med. Australas 31, 593–599. 10.1111/1742-6723.13219. [DOI] [PubMed] [Google Scholar]
- Johnson T, Fendrich M, 2005. Modeling sources of self-report bias in a survey of drug use epidemiology. Ann. Epidemiol 15, 381–389. 10.1016/j.annepidem.2004.09.004. [DOI] [PubMed] [Google Scholar]
- Jones CM, Bekheet F, Park JN, Alexander GC, 2020a. The evolving overdose epidemic: Synthetic opioids and rising stimulant-related harms. Epidemiol. Rev 42, 154–166. 10.1093/epirev/mxaa011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jones CM, Compton WM, Mustaquim D, 2020b. Patterns and characteristics of methamphetamine use among adults — United States, 2015–2018. MWR Morb. Mortal. Wkly. Rep 69, 317–323. 10.15585/mmwr.mm6912a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jones R, Woods C, Usher K, 2018. Rates and features of methamphetamine-related presentations to emergency departments: An integrative literature review. J. Clin. Nurs 27, 2569–2582. 10.1111/jocn.14493. [DOI] [PubMed] [Google Scholar]
- Koch DW, Lee J, Lee K, 2016. Coloring the war on drugs: arrest disparities in black, brown, and white. Race Soc. Probl 8, 313–325. 10.1007/s12552016-9185-6. [DOI] [Google Scholar]
- Lee MO, Vivier PM, Diercks DB, 2009. Is the self-report of recent cocaine or methamphetamine use reliable in illicit stimulant drug users who present to the emergency department with chest pain? J. Emerg. Med 37, 237–241. 10.1016/j.jemermed.2008.05.024. [DOI] [PubMed] [Google Scholar]
- MA opioid-related EMS incidents 2013–2020. Massachusetts Department of Public Health. Published Nov 2021. Accessed November 12, 2021. 〈https://www.mass.gov/doc/emergency-medical-services-data-november-2021/download〉. [Google Scholar]
- Mattson CL, Tanz LJ, Quinn K, Kariisa M, Patel P, Davis NL, 2021. Trends and geographic patterns in drug and synthetic opioid overdose deaths-United States, 2013–2019. MMWR Morb. Mortal. Wkly. Rep 70, 202–207. 10.15585/mmwr.mm7006a4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Min JE, Pearce LA, Homayra F, Dale LM, Barocas JA, Irvine MA, Slaunwhite AK, McGowan G, Torban M, Nosyk B, 2020. Estimates of opioid use disorder prevalence from a regression-based multi-sample stratified capture-recapture analysis. Drug Alcohol Depend. 217. 10.1016/j.drugalcdep.2020.108337. [DOI] [PubMed] [Google Scholar]
- Mitchell O, Caudy MS, 2015. Examining racial disparities in drug arrests. Justice Q. 32, 288–313. 10.1080/07418825.2012.761721. [DOI] [Google Scholar]
- Reuter P, Caulkins JP, Midgette G, 2021. Heroin use cannot be measured adequately with a general population survey. Addiction 116, 2600–2609. 10.1111/add.15458. [DOI] [PubMed] [Google Scholar]
- Richards JR, Hamidi S, Grant CD, Wang CG, Tabish N, Turnipseed SD, Derlet RW, 2017. Methamphetamine use and emergency department utilization: 20 years later. J. Addict 2017, 1–8. 10.1155/2017/4050932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sibanda NC, Kornhaber R, Hunt GE, Morley K, Cleary M, 2019. Prevalence and risk factors of emergency department presentations with methamphetamine intoxication or dependence: a systematic review and meta-analysis. Issues Ment. Health Nurs 40, 567–578. 10.1080/01612840.2018.1553003. [DOI] [PubMed] [Google Scholar]
- Vivolo-Kantor AM, Hoots BE, Seth P, Jones CM, 2020. Recent trends and associated factors of amphetamine-type stimulant overdoses in emergency departments. Drug Alcohol Depend. 216, 1–13. 10.1016/j.drugalcdep.2020.108323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- World Drug Report 2021, Booklet 4: Drug market trends: Cocaine amphetamine-type stimulants. 2021. United Nations publication, Sales No. E.21.XI.8. 〈https://www.unodc.org/res/wdr2021/field/WDR21_Booklet_4.pdf〉. Published June 2021. Accessed March 8, 2022. [Google Scholar]
- Yatsco AJ, Champagne-Langabeer T, Holder TF, Stotts AL, Langabeer JR, 2020. Developing interagency collaboration to address the opioid epidemic: a scoping review of joint criminal justice and healthcare initiatives. Int. J. Drug Policy 83, 1–7. 10.1016/j.drugpo.2020.102849. [DOI] [PubMed] [Google Scholar]