Emergency Medical Services Data Research Articles

Variations in the California Emergency Medical Services Response to Opioid Use Disorder.

IntroductionOpioids contributed to over 300,000 deaths in the United States in the past 10 years. Most research on drug use occurs in clinics or hospitals; few studies have evaluated the impact of opioid use on emergency medical services (EMS) or the EMS response to opioid use disorder (OUD). This study describes the perceived burden of disease, data collection, and interventions in California local EMS agencies (LEMSA).MethodsWe surveyed medical directors of all 33 California LEMSAs with 25 multiple-choice and free-answer questions. Results were collected in RedCap and downloaded into Excel (Microsoft Corporation, Redmond WA). This study was exempt from review by the Alameda Health System - Highland Hospital Institutional Review Board.ResultsOf the 33 California LEMSAs, 100% responded, all indicating that OUD significantly affects their patients. Most (91%) had specific protocols directing care of those patients and repeat naloxone dosing. After naloxone administration, none permitted release to law enforcement custody, 6% permitted patient refusal of care, and 45% directed base hospital contact for refusal of care. Few protocols directed screening or treatment of OUD or withdrawal symptoms. Regular data collection occurred in 76% of LEMSAs, with only 48% linking EMS data with hospital or coroner outcomes. In only 30% did the medical director oversee regular quality improvement meetings. Of respondents, 64% were aware of public health agency-based outreach programs and 42% were aware of emergency department BRIDGE programs (Medication Assisted Treatment and immediate referral). Only 9% oversaw naloxone kit distribution (all under the medical director), and 6% had EMS-based outreach programs. In almost all (94%), law enforcement officers carried naloxone and administered it anywhere from a few times a year to greater than 200 in one LEMSA.ConclusionThis study represents an important description of EMS medical directors’ approaches to the impact of OUD as well as trends in protocols and interventions to treat and prevent overdoses. Through this study, we can better understand the variable response to patients with OUD across California.

Design of a home-based intervention for Houston-area African-American adults with asthma: Methods and lessons learned from a pragmatic randomized trial

A growing body of evidence demonstrates that home-based, multicomponent interventions can effectively reduce exposures to asthma triggers and decrease asthma symptoms. However, few of these studies have targeted adults. To address this and other research gaps, we designed and implemented a pragmatic randomized clinical trial, the Houston Home-based Integrated Intervention Targeting Better Asthma Control (HIITBAC) for African Americans, to assess the effectiveness of a home-based intervention to improve asthma control and quality of life in African-American adults—a population disproportionately affected by asthma. The primary goals were to help participants reduce allergens and irritants in their homes and better manage their disease through knowledge, improved medication use, and behavior change. HIITBAC had two groups: clinic-only and home-visit groups. Both groups received enhanced clinical care, but the home-visit group also received a detailed home assessment and four additional home visits spaced over roughly one year. We recruited 263 participants. Of these, 152 (57.8%) were recruited through electronic health record data, 51 (19.4%) through Emergency Medical Services data, and 60 (22.8%) through other efforts (e.g., emergency departments, community events, outreach). Seventy participants (26.6%) were lost to follow up, substantially more in the home-visit than in the clinic-only group. We describe the HIITBAC methodology and cohort, discuss lessons learned about recruitment and retention, and highlight adaptations we implemented to address these lessons.

Public Health Surveillance for the Prevention of Pesticide-Related Illness in Illinois.

There is no pesticide related illness (PRI) surveillance program in Illinois. This study examines the quality of state-based data sources for their ability to inform public health surveillance on PRI. We estimated the counts of PRI by probabilistic data linkage of hospital discharge, emergency department, and poison center databases from 2010 to 2015. We characterized identified PRI cases. We identified 3867 unique cases of PRI and 6269 asymptomatic pesticide exposures. Out of the 3867 PRI cases, there were 1319 emergency department visits and 321 hospitalizations. We identified 13 deaths and 1640 major or moderate effects from PRI. Over half of ingestion related exposures occurred in children aged 0 to 4 years. Workers' compensation and Emergency Medical Service data were unusable. An effective public health surveillance on PRI requires reliable state data sources and cost-effective methods of assembling data from multiple sources.

Abstract WP282: Use of Emergent Ambulance Transport With Lights and Siren Leads to Significant Time Savings for Inter-Facility Transfers of Patients With Large Vessel Occlusion

Introduction: When emergency medical services (EMS) responds to a field call for a patient experiencing stroke symptoms, the response is emergent, necessitating the use of lights and siren (L&S) to allow for expedited transport to a stroke-capable center. With the expanded use of endovascular therapy (EVT) for large-vessel occlusion (LVO), many stroke-capable centers are transferring larger numbers of LVO patients to EVT-capable centers for definitive care. Interestingly, many EMS systems do not respond to or transport inter-facility transfers emergently with L&S. This can potentially lead to delays in care and worse clinical outcomes. Given increased scrutiny surrounding the safety and utility of L&S transport among EMS providers, we investigated the difference between emergent and non-emergent transfer of confirmed LVO stroke patients from two institutions with varied distances and traffic patterns. Methods: A retrospective analysis was performed of 127 consecutive inter-facility transfers for LVO from two facilities, Hospital A (38.5 miles) and Hospital B (5.5 miles), to the University of Michigan Comprehensive Stroke Center over 3 years and 2 years respectively. Transfers by helicopter (17/127; 13.4%) and those without available EMS data (9/127; 7.1%) were excluded. Final review included 50 cases from Hospital A and 51 from Hospital B. Run times and use of L&S during transport were collected. A t-test was used to examine whether the observed differences in transport times were statistically significant. Results: Of the 50 transfers from Hospital A, 22 were transported without L&S use and 28 with L&S. The mean transport time was 44 minutes versus 35 minutes, respectively. From Hospital B, there were 14 transfers transported without L&S use and 37 with L&S. The mean transport time was 15 minutes versus 9 minutes, respectively. For both samples, p-value was <0.01. Conclusion: Despite a small sample size, this analysis demonstrates significant time savings using L&S during inter-facility transfer of stroke patients with confirmed LVO. While inherent risk is associated with the use of L&S during EMS transport, judicious use for confirmed time-sensitive indications seems warranted.

Abstract TP247: A Comparison of Time to Treatment Between an Emergency Department-Focused Stroke Protocol and Mobile Stroke Units

Background: San Francisco is a relatively compact city with a population of 884,000 and nine stroke centers within a 47 square mile area. Emergency medical services (EMS) transport distances and times are short and there are currently no mobile stroke units (MSUs). Methods: We evaluated EMS activation to CT (EMS-CT) and EMS activation to thrombolysis (EMS-TPA) times for acute stroke in the first two years after implementation of an emergency department (ED)-focused, direct EMS-to-CT protocol entitled “Mission Protocol” (MP) at a safety net hospital in San Francisco and compared our performance to published reports from MSUs. EMS times were abstracted from ambulance records. Geometric means were calculated for MP data and pooled means were similarly calculated from published MSU data. Results: From July 2017 through June 2019, a total of 423 patients with suspected stroke were evaluated under the MP, and 166 of these patients were ultimately diagnosed with ischemic stroke. EMS and treatment time data were available for 129 of these patients with 60 patients (47%) receiving IV-tPA with mean EMS-CT and EMS-TPA times of 41 minutes (95% CI 39-43) and 64 minutes (95% CI 58-70), respectively. The pooled estimates for MSUs suggested a mean EMS-CT time of 35 minutes (95% CI 27-45) and a mean EMS-TPA time of 48 minutes (95% CI 39-60). MSUs achieved faster EMS-CT and EMS-TPA times (p<0.0001 for both). Conclusions: In moderate-sized, urban settings with high population densities, our protocol was able to achieve EMS activation to treatment times that were comparable but ultimately slightly slower than published performance data from MSUs.

Predictors of Need for First and Recurrent Emergency Medical Service Transport to Emergency Department after Dialysis Initiation

Background: Dialysis patients are frequently transported to the emergency department (ED) by Emergency Medical Services (EMS) due to acute and severe illness. However, little is known about predictors of first and recurrent transport to the ED (EMS-ED), based on characteristics at the time of dialysis initiation.Methods: We analyzed a cohort of adult (≥18 years) patients affiliated with a large quaternary care center who initiated chronic dialysis from 2009 to 2013 (last follow-up: 2015). Data on patient characteristics at the time of dialysis initiation were linked to regional EMS data. Candidate predictors of first and recurrent EMS-ED transport included comorbid conditions, dialysis characteristics and frailty severity (using the first version of the clinical frailty scale score; CFS). Time to first EMS-ED was analyzed using a multivariable sub-hazards regression model accounting for competing events (transplantation or death). Time to recurrent EMS-ED was analyzed using the Anderson-Gill counting approach, accounting for competing risks.Results: A total of 455 patients were included in the study, 243 (53%) had 1+ EMS-ED events, 90 (20%) never required an EMS-ED at last follow-up, and 69 (15%) and 53 (12%) experienced transplant or death as their first event, respectively. The mean age of the cohort was 62 ± 15 years, 89% were Caucasian, and 35% were female sex. Patients were highly comorbid and 97/381 (25.5%) with available data on frailty severity had a CFS score of ≥5, inclusive of CFS scores ranging from mildly to severely frail. After adjustment, a CFS score of ≥5 (relative to 1–2) was associated with a > 2-fold increase in the risk of first EMS-ED (subdistribution relative hazard; SHR 2.28, 95% confidence interval; CI 1.30–3.98). A history of peripheral vascular disease (SHR 1.43, 95% CI 1.00–2.03) and rheumatologic disease (SHR 1.84, 95% CI 1.00–3.38) was also associated with first EMS-ED. Frailty severity was the only factor associated with recurrent EMS-ED.Conclusion: Patients are at a high risk of EMS-ED after dialysis initiation. Frailty severity (at the time of dialysis initiation) is a strong predictor of first and recurrent EMS-ED and this may be important to guide informed decision making and resource planning for dialysis patients who require EMS.

The Detection of Opioid Misuse and Heroin Use From Paramedic Response Documentation: Machine Learning for Improved Surveillance.

BackgroundTimely, precise, and localized surveillance of nonfatal events is needed to improve response and prevention of opioid-related problems in an evolving opioid crisis in the United States. Records of naloxone administration found in prehospital emergency medical services (EMS) data have helped estimate opioid overdose incidence, including nonhospital, field-treated cases. However, as naloxone is often used by EMS personnel in unconsciousness of unknown cause, attributing naloxone administration to opioid misuse and heroin use (OM) may misclassify events. Better methods are needed to identify OM.ObjectiveThis study aimed to develop and test a natural language processing method that would improve identification of potential OM from paramedic documentation.MethodsFirst, we searched Denver Health paramedic trip reports from August 2017 to April 2018 for keywords naloxone, heroin, and both combined, and we reviewed narratives of identified reports to determine whether they constituted true cases of OM. Then, we used this human classification as reference standard and trained 4 machine learning models (random forest, k-nearest neighbors, support vector machines, and L1-regularized logistic regression). We selected the algorithm that produced the highest area under the receiver operating curve (AUC) for model assessment. Finally, we compared positive predictive value (PPV) of the highest performing machine learning algorithm with PPV of searches of keywords naloxone, heroin, and combination of both in the binary classification of OM in unseen September 2018 data.ResultsIn total, 54,359 trip reports were filed from August 2017 to April 2018. Approximately 1.09% (594/54,359) indicated naloxone administration. Among trip reports with reviewer agreement regarding OM in the narrative, 57.6% (292/516) were considered to include information revealing OM. Approximately 1.63% (884/54,359) of all trip reports mentioned heroin in the narrative. Among trip reports with reviewer agreement, 95.5% (784/821) were considered to include information revealing OM. Combined results accounted for 2.39% (1298/54,359) of trip reports. Among trip reports with reviewer agreement, 77.79% (907/1166) were considered to include information consistent with OM. The reference standard used to train and test machine learning models included details of 1166 trip reports. L1-regularized logistic regression was the highest performing algorithm (AUC=0.94; 95% CI 0.91-0.97) in identifying OM. Tested on 5983 unseen reports from September 2018, the keyword naloxone inaccurately identified and underestimated probable OM trip report cases (63 cases; PPV=0.68). The keyword heroin yielded more cases with improved performance (129 cases; PPV=0.99). Combined keyword and L1-regularized logistic regression classifier further improved performance (146 cases; PPV=0.99).ConclusionsA machine learning application enhanced the effectiveness of finding OM among documented paramedic field responses. This approach to refining OM surveillance may lead to improved first-responder and public health responses toward prevention of overdoses and other opioid-related problems in US communities.

Integrating the ambulance dispatching and relocation problems to maximize system’s preparedness

In the Emergency Medical Service (EMS) context, the decision-making process plays a very important role since some decisions highly impact patients’ health. This work focuses on the operational level by solving the dispatching and relocation ambulance problems. Dispatching decisions assign ambulances to emergencies and the relocation problem decides on the base to which available ambulances should be (re)assigned. To improve effectiveness and efficiency in the EMS response, integrated optimization approaches are proposed: a mathematical model and a pilot method heuristic. The aim is to maximize system’s coverage using a time-preparedness measure allowing relocations to any base.Experiments are performed using EMS data from Lisbon, Portugal, where solving these problems is still a handmade task. The optimization approaches are adapted to compare the proposed strategy with the current Portuguese EMS strategy which dispatches the closest available ambulance for each emergency and relocates ambulances to their home bases. Results highlight the potential of the mathematical model and of the proposed strategy to be applied in real-time contexts since key performance indicators (KPIs) are, in general, better than the ones obtained through the heuristic and the current Portuguese EMS strategy. The heuristic should be used when more emergencies occur in the same time period since a solution can be obtained almost immediately in contrast to the MIP usage.

Abstract 390: Automating a Cardiac Arrest Registry: Challenges and Opportunities

Introduction: A registry for out-of-hospital cardiac arrests facilitates cardiac arrest research, and emergency medical services (EMS) and community health quality improvement programs. Maintaining a cardiac arrest registry is labor intensive and costly, both of which could be reduced through automation. Objective: To determine the feasibility of automating data population of a cardiac arrest registry and to identify the components needed for and barriers to automation. Methods: This is a prospective feasibility study at the Dallas Fort-Worth (DFW) Center for Resuscitation Research. Our cardiac arrest registry was part of the multi-center Resuscitation Outcomes Consortium (ROC) Epistry from 2006 to 2016. Since 2016, we modified and maintained it separately from the ROC and over 20,000 cases have been enrolled. We identified the following critical elements necessary for automated data population of the registry: Create a registry dataset; create Business Associate Agreements with EMS agencies and hospitals; identify EMS and hospital electronic health record programs; develop a search strategy; search for appropriate cases in city servers where EMS patient care reports (PCRs) are stored and download data; move EMS data through our institutional firewall to a secure server; perform the same steps for hospital medical records; match EMS cases with hospital medical records, and automatically populate the registry with EMS and hospital data on a regular basis. Results: At the DFW site, EMS agencies use five different PCR programs and hospitals use two different electronic health record programs. A combination of 13 search terms captured 90% to 100% of appropriate cardiac arrest cases across six EMS agencies. Two different strategies can be used for streaming data across our firewall: A proprietary program entitled “Move-It” or a university generated file transfer protocol. We used a probabilistic logical approach for matching EMS cases with hospital cases. Conclusions: Automating a cardiac arrest registry is feasible. A cardiac arrest registry has many elements that must be considered to accomplish automation. The authors found that the most time consuming element is establishing Business Associate Agreements with all parties that provide data.

Abstract 238: Automating Medical Record Matching: A Key Component of an Automated Cardiac Arrest Registry

Introduction: An automated registry for out-of-hospital cardiac arrest can facilitate cardiac arrest research and emergency medical services (EMS) and community health quality improvement programs. A key element of a cardiac arrest registry is outcome, especially survival to hospital discharge, which is usually obtained from the hospital medical record. A crucial step in capturing survival data automatically is being able to reliably match EMS patient care reports with hospital medical records. Objective: To determine the accuracy of a probabilistic record-linkage approach to match EMS and hospital encounters and identify barriers of automation. Methods: This is a prospective feasibility study from the Dallas Fort-Worth Center for Resuscitation Research. Our cardiac arrest registry was part of the Resuscitation Outcomes Consortium (ROC) Epistry from 2006 to 2016. Since 2016, we maintained it separately from the ROC and have since enrolled over 20,000 cases. Record linkage was performed using demographic data of encounters from a single EMS agency within the registry and a community hospital in Dallas where these were transported from January 2017 to May 2018. An expectation maximization algorithm to aid in discerning matches and nonmatches was performed using the R package RecordLinkage identifying record pairs as a match, non-match, or possible match based on a pre-determined threshold. Match and Possible Match pairs were manually reviewed to assess the algorithm’s accuracy. Results: The EMS and hospital data sets had 67 and 6,050 unique encounters, respectively. The record linkage algorithm identified 40 EMS encounters as having a match (60%), 12 with a possible match (18%), and 15 with no identified match (22%). After manual review, the algorithm had a positive predictive value (PPV) of 100% (40/40) for the match group with a sensitivity of 60%. Conclusions: Applying probabilistic linkage to aid in automating a cardiac arrest registry was successful in linking over half of the EMS encounters with a PPV of 100%. When reviewing the unmatched encounters, many were adjacent to each other when sorted by time suggesting the poor sensitivity is more likely a reflection of the hospital’s data collection method as opposed to the linkage methodology.

WHAT CAN COUNTY-LEVEL EMERGENCY MEDICAL SERVICES DATA TEACH PUBLIC HEALTH ABOUT OLDER ADULT FALLS

Fall-related injuries in older adults contribute to an increasing number of deaths, hospitalizations and Emergency Medical Services (EMS) responses. In Snohomish County, the third largest county in Washington State, EMS agencies with electronic health records cover 96% of the county’s population. This EMS data contains unique information on falls, which can estimate costs and clarify intervention priorities. We analyzed 2018 data from EMS in Snohomish County. Fall incidents were summarized by count, frequency, and rate per 1,000 population. Costs for transferring patients to emergency departments (ED) were estimated using 2015 Snohomish Community Paramedic Analysis, and direct medical costs averted by implementing a single intervention were estimated based on prior research by Stevens and Lee (2018). There were total 38,910 incidents in older adults, of those 4,777 incidents were caused by falls (1606 in males and 2906 in females). The mean age (SD) was 81.0 (±8.9). The incidence rate was 45.6 per 1000 (55.8 in females and 30.6 in males). There were 573 repeated falls (12%). Most of the falls happened at home (54.85%), followed by assisted living and nursing homes (27.84%). 85.53% of the falls were transferred to ED, at an estimated cost of 3.15 million dollars. We calculated that one million dollars in medical cost could be averted by implementing home modifications delivered by an occupational therapist (OT). This research demonstrates the utility of EMS data for describing fall injury and determining interventions. Fall prevention programs should focus on preventing repeated falls and addressing home safety risks.

Racial differences in overdose events and polydrug detection in Indianapolis, Indiana

BackgroundWe examine racial disparities in drug overdose death rates by analyzing trends in fatal and nonfatal overdose outcomes in a large metropolitan area (Indianapolis, Indiana). MethodsDeath certificate and toxicology records for accidental drug overdose deaths from 2011 to 2018 were linked with emergency medical services (EMS) data. Bivariate comparisons examined differences in toxicology findings at the time of death as well as prior EMS events both overall and by indicator of non-fatal overdose. ResultsFrom 2011–2018, 2204 residents (29.4 per 100,000) died of drug overdose, 18.6% were Black (N = 410, 19.5 per 100,000) and 78.5% White (N = 1730, 35.2 per 100,000). In the year prior to death, 33.5% (N = 656) of decedents had an EMS event, 12.1% (N = 237) had an overdose event, and 9.4% (N = 185) had naloxone administered. Overdose complaint and naloxone administration were more likely to occur among White than Black patients. White decedents were more likely than Black decedents to have had naloxone administered in the year prior to death (10.1% vs. 6.8%, χ2 = 4.0, p < .05, Cramer’s V=.05). Toxicology data illustrate changing polydrug combinations, with Black decedents more likely to test positive for fentanyl-cocaine polydrug use in recent years. ConclusionsRecent racial disparities in overdose deaths are driven by a combination of fentanyl and cocaine, which disproportionally impacts African American drug users, but may be addressed through expanded harm reduction and community outreach services. Additionally, there is a need to assess the role of differing practices in overdose emergency service provision as a contributing factor to disparities.

Hepatitis B vaccine and risk of acute myocardial infarction among patients with diabetes

The objective of our study was to incorporate stricter probable nonfatal opioid overdose case criteria, and advanced epidemiologic approaches to more reliably detect local clustering in nonfatal opioid overdose activity in EMS runs data.Data were obtained using emsCharts for our study area in southwestern Pennsylvania from 2007 to 2018. Cases were identified as emergency medical service (EMS) responses where naloxone was administered, and improvement was noted in patient records between initial and final Glasgow Coma Score. A subsample of all-cause EMS responses sites were used as controls and exact matched to cases on sex and 10-year-age category. Clustering was assessed using difference in Ripley's K function for cases and controls and Kulldorff scan statistics.Difference in K functions indicated no significant difference in probable nonfatal overdose EMS runs across the study area compared to all-cause EMS runs. However, scan statistics did identify significant local clustering of probable nonfatal overdose EMS runs (maximum likelihood = 16.40, P = 0.0003).Results highlight relevance of EMS data to detect community-level overdose activity and promote reliable use through stricter case definition criteria and advanced methodological approaches. Techniques examined have the potential to improve targeted delivery of neighborhood-level public health response activities using a near real-time data source.

Identifying high-risk areas for nonfatal opioid overdose: a spatial case-control study using EMS run data

PurposeThe objective of our study was to incorporate stricter probable nonfatal opioid overdose case criteria, and advanced epidemiologic approaches to more reliably detect local clustering in nonfatal opioid overdose activity in EMS runs data. MethodsData were obtained using emsCharts for our study area in southwestern Pennsylvania from 2007 to 2018. Cases were identified as emergency medical service (EMS) responses where naloxone was administered, and improvement was noted in patient records between initial and final Glasgow Coma Score. A subsample of all-cause EMS responses sites were used as controls and exact matched to cases on sex and 10-year-age category. Clustering was assessed using difference in Ripley's K function for cases and controls and Kulldorff scan statistics. ResultsDifference in K functions indicated no significant difference in probable nonfatal overdose EMS runs across the study area compared to all-cause EMS runs. However, scan statistics did identify significant local clustering of probable nonfatal overdose EMS runs (maximum likelihood = 16.40, P = 0.0003). ConclusionsResults highlight relevance of EMS data to detect community-level overdose activity and promote reliable use through stricter case definition criteria and advanced methodological approaches. Techniques examined have the potential to improve targeted delivery of neighborhood-level public health response activities using a near real-time data source.

Approach to Onboarding Emergency Medical Services (EMS) Data Into a Syndromic Surveillance System

ObjectiveTo describe the strategy and process used by the Florida Department of Health (FDOH) Bureau of Epidemiology to onboard emergency medical services (EMS) data into FDOH’s syndromic surveillance system, the Electronic Surveillance System for the Early Notification of Community-based Epidemics (ESSENCE-FL).IntroductionSyndromic surveillance has become an integral component of public health surveillance efforts within the state of Florida. The near real-time nature of these data are critical during events such as the Zika virus outbreak in Florida in 2016 and in the aftermath of Hurricane Irma in 2017. Additionally, syndromic surveillance data are utilized to support daily reportable disease detection and other surveillance efforts. Although syndromic systems typically utilize emergency department (ED) visit data, ESSENCE-FL also includes data from non-traditional sources: urgent care center visit data, mortality data, reportable disease data, and Florida Poison Information Center Network (FPICN) data. Inclusion of these data sources within the same system enables the broad accessibility of the data to more than 400 users statewide, and allows for rapid visualization of multiple data sources in order to address public health needs. Currently, the ESSENCE-FL team is actively working to incorporate EMS data into ESSENCE-FL to further increase public health surveillance capacity and data visualization.MethodsThe ESSENCE-FL team worked collaboratively with various public health program stakeholders to bring EMS data, aggregated by the FDOH Bureau of Emergency Medical Oversight Emergency Medical Services Tracking and Reporting System (EMSTARS) team, into ESSENCE-FL. The ESSENCE-FL team met with the EMSTARS team to discuss use cases, demonstrate both systems, and to obtain project buy-in and support. Initial project meetings included review of ESSENCE-FL system support, user types (roles and access), as well as data security and compliance. An overall project timeline was established, and deliverables were added into system support contracts. Multiple stakeholders, across disciplines representing each key use case, reviewed the Florida version of the National Emergency Medical Services Information System (NEMSIS) version 3.4 data dictionary to identify program-specific data element needs. An element scoring spreadsheet was returned to the ESSENCE-FL team. These scores were aggregated and discordant scores were reviewed by the ESSENCE-FL team. A one-month extract of EMS data was reviewed to assess variable completeness and relevance. Monthly team meetings facilitated the final decisions on the data elements by leveraging lessons learned through onboarding other data sources, findings from the analysis of the one-month extract, stakeholder comments, and advice from other states known to be leveraging EMS data for public health surveillance.ResultsThrough a collaborative and broad approach with partners, the ESSENCE-FL team attained stakeholder buy-in and identified 81 data elements to be included in the EMS feed to ESSENCE-FL. The final list of data elements was determined to best support health surveillance of this population prior to presenting to the ED. The inclusion of the EMS data in ESSENCE-FL will increase the epidemiologic characterization and analysis of the opioid epidemic in Florida. Additional key use cases identified during this project included enhanced injury surveillance, enhanced occupational health surveillance, and characterization of potential differences between EMS and ED visits.ConclusionsThis comprehensive approach can be used by other jurisdictions considering adding EMS data to their syndromic surveillance systems. When considering onboarding a new data source into a surveillance system, it is important to work closely with stakeholders from disciplines representing each of the key use cases to broaden buy-in and support for the project. Through employing this comprehensive approach, syndromic surveillance systems can be better developed to include data that are widely utilizable to many different stakeholders in the public health community.

Enhancing TX SyS by integrating EMS and Poison Data for Opioid Overdose Surveillance

ObjectiveTo improve Texas Syndromic Surveillance by integrating data from the Texas Poison Center and Emergency Medical Services for opioid overdose surveillance.IntroductionIn recent years, the number of deaths from illicit and prescription opioids has increased significantly resulting in a national and local public health crisis. According to the Texas Center for Health Statistics, there were 1340 opioid related deaths in 2015.1 In 2005, by comparison, there were 913 opioid related deaths. Syndromic surveillance can be used to monitor overdose trends in near real-time and provide much needed information to public health officials. Texas Syndromic Surveillance (TxS2) is the statewide syndromic surveillance system hosted by the Texas Department of State Health Services (DSHS). To enhance the capabilities of TxS2 and to better understand the opioid epidemic, DSHS is integrating both Texas Poison Center (TPC) data and Emergency Medical Services (EMS) data into the system.Much of the data collected at public health organizations can be several years old by the time it is released for public use. As a result, there have been major efforts to integrate more real-time data sources for a variety of surveillance needs and during emergency response activities.MethodsGuided by the Oregon Public Health Division’s successful integration of poison data into Oregon ESSENCE, DSHS has followed a similar path.2 DSHS already receives TPC data from the Commission on State Emergency Communication (CSEC), hence copying and routing that data into TxS2 requires a Memorandum of Understanding (MOU) with CSEC, which is charged with administering the implementation of the Texas Poison Control Network.EMS records are currently received by the DSHS Office of Injury Prevention (OIP) via file upload and extracted from web services as an XML file. Regional and Local Health Operations, the division where the syndromic surveillance program is located, and OIP, are both sections within DSHS. Therefore, it is not necessary to have a formal MOU in place. Both parties would operate under the rules and regulations that are established for data under the Community Health Improvement Division.CSEC and EMS will push data extracts to a DSHS SFTP folder location for polling by Rhapsody in Amazon Web Services. The message data will be extracted and transformed into the ESSENCE database format. Data are received at least once every 24 hours.ResultsTxS2 will now include TPC and EMS data, giving system users the ability to analyze and overlay real-time data for opioid overdose surveillance in one application. The integration of these data sources in TxS2 can be used for both routine surveillance and for unexpected public health events. This effort has led to discussions on how different sections within DSHS can collaborate by using syndromic surveillance data, and has generated interest in incorporating additional data streams into TxS2 in the future.ConclusionsWhile this venture is still a work in progress, it is anticipated that adding TPC and EMS data to TxS2 will be beneficial in surveilling not just opioid overdoses but other conditions and illnesses, as well as capturing disaster related injuries.

EMS Heroin Overdoses with Refusal to Transport &amp; Impacts on ED Overdose Surveillance

ObjectiveThe aim of this project was to explore changing patterns in patient refusal to transport by emergency medical services for classified heroin overdoses and possible implications on heroin overdose surveillance in Kentucky.IntroductionAs a Centers for Disease Control and Prevention Enhanced State Opioid Overdose Surveillance (ESOOS) funded state, Kentucky started utilizing Emergency Medical Services (EMS) data to increase timeliness of state data on drug overdose events in late 2016. Using developed definitions of heroin overdose for EMS emergency runs, Kentucky analyzed the patterns of refused/transported EMS runs for both statewide and local jurisdictions. Changes in EMS transportation patterns of heroin overdoses can have a dramatic impact on other surveillance systems, such as emergency department (ED) claims data or syndromic surveillance (SyS) data.MethodsAs part of the ESOOS grant, Kentucky receives all emergency-only EMS runs monthly from Kentucky Board for Emergency Medical Services, Kentucky State Ambulance Reporting System data. Heroin cases were classified based on text and medications (Narcan) administered, with comparisons to historic data discussed elsewhere (Rock &amp; Singleton, 2018). Transportation classifications are based on EMS standard elements defining treatment with transportation vs refusal to transport to hospital and canceled runs were excluded. Initial analysis included trend analysis at state and local levels, as well as demographic comparisons of refusal vs transported heroin overdose encounters.ResultsStatewide trends in EMS heroin overdoses with refusal transport significantly increased from 5% (n=42) in 2016 quarter three to 22% (n=290) in 2018 quarter two (Fig 1). Initial demographic analysis does not show any significant difference between refusals/transported for age, gender, or race. However, there are significant differences among geographic regions in Kentucky with heroin encounter refusal proportion ranging from 3%-48% in 2018 quarter two. Specifically, one urban area (Fig 2) shows the change in proportion of refusal increasing from 15% (n=23) in 2016 quarter three to 47% (n=110) in 2018 quarter two. In this geographic area, combined refused/transported EMS heroin overdoses compared to traditional ED data demonstrates opposing heroin overdose patterns for the same local with EMS showing and increasing trend overtime and ED showing a decreasing trend (Fig 3).ConclusionsTraditional public health surveillance for heroin overdose has historically relied on ED billing data, though agencies are starting to use syndromic surveillance, too (Vivolo-Kantor et al., 2016). These systems share similar underlying ED data, albeit with different components, quality, and limitations. However, in terms of the overdose epidemic, both are limited to only heroin overdoses that result in ED hospital encounters. The recent drastic increase in refused transport can have significant impacts on heroin surveillance. Jurisdictions relying on SyS or ED data for monitoring overdose patterns and/or evaluating interventions may be significantly underestimating acute overdose occurrence in the population. This analysis highlights the importance of this preclinical data source in surveillance of the heroin epidemic.ReferencesRock, P. J., &amp; Singleton, M. D. (2018). Assessing Definitions of Heroin Overdose in ED &amp; EMS Data Using Hospital Billing Data, 10(1), 2579.Vivolo-Kantor, A. M., Seth, P., Gladden, ; R Matthew, Mattson, C. L., Baldwin, G. T., Kite-Powell, A., &amp; Coletta, M. A. (2016). Morbidity and Mortality Weekly Report Vital Signs: Trends in Emergency Department Visits for Suspected Opioid Overdoses — United States, 67(9), 279–285. Retrieved from https://www.cdc.gov/mmwr/volumes/67/wr/pdfs/mm6709e1-H.pdf

Development of Text-Based Algorithm for Opioid Overdose Identification in EMS Data

Development of Text-Based Algorithm for Opioid Overdose Identification in EMS Data

Optimization of Linkage between North Carolina EMS and ED Data: EMS Naloxone Cases

Optimization of Linkage between North Carolina EMS and ED Data: EMS Naloxone Cases

Validating Syndromic Data for Opioid Overdose Surveillance in Florida

Validating Syndromic Data for Opioid Overdose Surveillance in Florida