Mengtao Dai

Evaluating the Use of Existing Data Sources, Probabilistic Linkage, and Multiple Imputation to Build Population-based Injury Databases Across Phases of Trauma Care

OBJECTIVES The objective was to evaluate the process of using existing data sources, probabilistic linkage, and multiple imputation to create large population-based injury databases matched to outcomes. METHODS This was a retrospective cohort study of injured children and adults transported by 94 emergency medical systems (EMS) agencies to 122 hospitals in seven regions of the western United States over a 36-month period (2006 to 2008). All injured patients evaluated by EMS personnel within specific geographic catchment areas were included, regardless of field disposition or outcome. The authors performed probabilistic linkage of EMS records to four hospital and postdischarge data sources (emergency department [ED] data, patient discharge data, trauma registries, and vital statistics files) and then handled missing values using multiple imputation. The authors compare and evaluate matched records, match rates (proportion of matches among eligible patients), and injury outcomes within and across sites. RESULTS There were 381,719 injured patients evaluated by EMS personnel in the seven regions. Among transported patients, match rates ranged from 14.9% to 87.5% and were directly affected by the availability of hospital data sources and proportion of missing values for key linkage variables. For vital statistics records (1-year mortality), estimated match rates ranged from 88.0% to 98.7%. Use of multiple imputation (compared to complete case analysis) reduced bias for injury outcomes, although sample size, percentage missing, type of variable, and combined-site versus single-site imputation models all affected the resulting estimates and variance. CONCLUSIONS This project demonstrates the feasibility and describes the process of constructing population-based injury databases across multiple phases of care using existing data sources and commonly available analytic methods. Attention to key linkage variables and decisions for handling missing values can be used to increase match rates between data sources, minimize bias, and preserve sampling design.

Description of the 2012 NEMSIS Public-Release Research Dataset

Abstract Background. The National Emergency Medical Services Information System (NEMSIS) is a federally funded project designed to standardize emergency medical services (EMS) patient care reporting and facilitate state and national data repositories for the assessment of EMS systems of care. The purpose of this assessment is to characterize the annual NEMSIS 2012 Public-Release Research Dataset, detailing the strengths and limitations associated with use of these data for EMS quality assurance and/or research purposes. Methodology. Using descriptive statistics, we evaluated the dataset completeness (i.e., presence of missing/null values) and dataset content. To assess data generalizability, we compared age distributions to the 2010 National Hospital Ambulatory Medical Care Survey (NHAMCS). Subanalyses were conducted for cardiac arrest- and trauma-related EMS activations to describe their characteristics, treatments, and dispositions. The analysis focuses primarily on 9-1-1 EMS activations resulting in the treatment and transport of a patient, excepting the analysis conducted for out-of-hospital cardiac arrests (OHCA), which includes scene deaths. Results. In 2012, 19,831,189 EMS activations were reported to NEMSIS by 8,439 agencies located in 42 states and territories. Of the 10,733,925 9-1-1 EMS activations reportedly treating and transporting a patient, the majority were attended by a paid EMT-paramedic (82%) employed by a fire-based EMS agency (25%) working in an urban area (53%). 9-1-1 call centers most likely dispatched EMS for a “sick person” (14%), while providers most likely reported pain (26%) as the patients primary symptom and “traumatic injury” (13%) as the likely cause. NEMSIS data adequately characterize specific patient populations and demonstrate significant similarity to independent distributions of EMS patients (i.e., NHAMCS), yet missing data and use of null values remain prevalent. Conclusions. The annual NEMSIS Public-Release Research Dataset is a valuable resource for evaluating the U.S. EMS activation population and can be used to conduct in-depth descriptions of the care of specific populations. However, the utility of the data are limited until the number of null values can be diminished and reporting becomes universal.

Improving early identification of the high-risk elderly trauma patient by emergency medical services

STUDY OBJECTIVE We sought to (1) define the high-risk elderly trauma patient based on prognostic differences associated with different injury patterns and (2) derive alternative field trauma triage guidelines that mesh with national field triage guidelines to improve identification of high-risk elderly patients. METHODS This was a retrospective cohort study of injured adults ≥65 years transported by 94 EMS agencies to 122 hospitals in 7 regions from 1/1/2006 through 12/31/2008. We tracked current field triage practices by EMS, patient demographics, out-of-hospital physiology, procedures and mechanism of injury. Outcomes included Injury Severity Score≥16 and specific anatomic patterns of serious injury using abbreviated injury scale score ≥3 and surgical interventions. In-hospital mortality was used as a measure of prognosis for different injury patterns. RESULTS 33,298 injured elderly patients were transported by EMS, including 4.5% with ISS≥16, 4.8% with serious brain injury, 3.4% with serious chest injury, 1.6% with serious abdominal-pelvic injury and 29.2% with serious extremity injury. In-hospital mortality ranged from 18.7% (95% CI 16.7-20.7) for ISS≥16 to 2.9% (95% CI 2.6-3.3) for serious extremity injury. The alternative triage guidelines (any positive criterion from the current guidelines, GCS≤14 or abnormal vital signs) outperformed current field triage practices for identifying patients with ISS≥16: sensitivity (92.1% [95% CI 89.6-94.1%] vs. 75.9% [95% CI 72.3-79.2%]), specificity (41.5% [95% CI 40.6-42.4%] vs. 77.8% [95% CI 77.1-78.5%]). Sensitivity decreased for individual injury patterns, but was higher than current triage practices. CONCLUSIONS High-risk elderly trauma patients can be defined by ISS≥16 or specific non-extremity injury patterns. The field triage guidelines could be improved to better identify high-risk elderly trauma patients by EMS, with a reduction in triage specificity.

Pediatric Out-of-Hospital Critical Procedures in the United States.

Objective: Regular clinical application is important for maintenance of difficult resuscitation skills. Although emergency medical services must provide life-saving care for critically ill and injured children, the frequency with which these procedures are performed is unknown. We sought to characterize critical pediatric procedures performed by emergency medical service personnel in the United States. Design: We performed a retrospective, descriptive study of emergency medical service responses. Setting and Patients: We included patients less than 18 years old in the 2011 National Emergency Medical Services Information Systems national data set. We identified emergency medical service cases receiving critical procedures, including intubation, cricothyroidotomy, cardiac pacing, cardioversion, defibrillation, needle decompression, pericardiocentesis, and intraosseous or central venous catheter placement. Interventions: None. Measurements and Main Results: We analyzed the data to determine the number and prevalence of procedures, success rates, and factors associated with success. Of the 14,371,941 emergency medical service responses, 865,591 (6.8%) involved children. Emergency medical service responses to pediatric patients most often involved traumatic injuries (35.7%) or respiratory complications (13.2%). Emergency medical service performed a total of 616,913 procedures on 246,016 pediatric cases. Critical procedures were infrequently performed (n = 11,026, 10 per 1,000 pediatric cases). The most common critical procedures performed were intubation (n = 3,599, 6.7 per 1,000 pediatric cases) and intraosseous access (n = 2,618, 5 per 1,000 pediatric cases). Overall, 81% of critical procedures were successful. Increasing age and interfacility transfers were associated with greater odds of procedural success (p < 0.01). Conclusion: Despite the broad range of pediatric conditions seen in the prehospital setting, pediatric critical procedures are infrequently performed. These data highlight factors that are associated with successful completion of critical pediatric procedures.

Procedures Performed by Emergency Medical Services in the United States.

Abstract Emergency medical services (EMS) must provide a wide range of care for patients in the out-of-hospital setting. Although previous work has detailed that EMS providers rarely perform certain procedures, (e.g., endotracheal intubation) there are limited data detailing the frequency of procedures across the breadth of EMS providers’ scope of practice. We sought to characterize procedures performed by EMS in the United States. We conducted an analysis of the 2011 National Emergency Medical Services Information System (NEMSIS) research data set, encompassing EMS emergency response data from 40 states and two territories. From these data, we report the number and incidence of EMS procedures. We also characterize procedures performed. There were 14,371,941 submitted EMS responses, of which 7,680,559 had complete information on procedures performed on adults. Of these, 4,206,360 EMS responses had procedures performed totaling 11,407,396 procedures. The most common procedures performed were peripheral venous access (28.4%), cardiac monitoring (16.1%) pulse oximetry (13.5%), and blood glucose analysis (10.4%). Procedures were performed most often in patients with traumatic injury (20.0%) followed by chest pain/discomfort (14.0%). Critical procedures (cardioversion, defibrillation, endotracheal intubation, etc.) were infrequently performed (n = 277,785, 2.4%). These data highlight the frequency with which EMS providers perform procedures across the United States. This may help to guide future EMS training and education efforts by highlighting the relative frequency and infrequency of specific procedures.

Validating the Use of ICD-9 Code Mapping to Generate Injury Severity Scores.

The Injury Severity Score (ISS) is a measure of injury severity widely used for research and quality assurance in trauma. Calculation of ISS requires chart abstraction, so it is often unavailable for patients cared for in nontrauma centers. Whether ISS can be accurately calculated from International Classification of Diseases, Ninth Revision (ICD-9) codes remains unclear. Our objective was to compare ISS derived from ICD-9 codes with those coded by trauma registrars. This was a retrospective study of patients entered into 9 U.S. trauma registries from January 2006 through December 2008. Two computer programs, ICDPIC and ICDMAP, were used to derive ISS from the ICD-9 codes in the registries. We compared derived ISS with ISS hand-coded by trained coders. There were 24,804 cases with a mortality rate of 3.9%. The median ISS derived by both ICDPIC (ISS-ICDPIC) and ICDMAP (ISS-ICDMAP) was 8 (interquartile range [IQR] = 4–13). The median ISS in the registry (ISS-registry) was 9 (IQR = 4–14). The median difference between either of the derived scores and ISS-registry was zero. However, the mean ISS derived by ICD-9 code mapping was lower than the hand-coded ISS in the registries (1.7 lower for ICDPIC, 95% CI [1.7, 1.8], Bland–Altman limits of agreement = −10.5 to 13.9; 1.8 lower for ICDMAP, 95% CI [1.7, 1.9], limits of agreement = −9.6 to 13.3). ICD-9-derived ISS slightly underestimated ISS compared with hand-coded scores. The 2 methods showed moderate to substantial agreement. Although hand-coded scores should be used when possible, ICD-9-derived scores may be useful in quality assurance and research when hand-coded scores are unavailable.

Prehospital Pain Management: Disparity By Age and Race

Abstract Importance: Historically, pain management in the prehospital setting, specifically pediatric pain management, has been inadequate despite many EMS (emergency medical services) transports related to traumatic injury with pain noted as a symptom. The National Emergency Services Information System (NEMSIS) database offers the largest national repository of prehospital data, and can be used to assess current patterns of EMS pain management across the country. Objectives: To analyze prehospital management of pain using NEMSIS data, and to assess if variables such as patient age and/or race/ethnicity are associated with disparity in pain treatment. Design/Setting/Participants: A retrospective descriptive study over a three-year period (2012–2014) of the NEMSIS database for patients evaluated for three potentially painful medical impressions (fracture, burn, penetrating injury) to assess the presence of documented pain as a symptom, and if patients received treatment with analgesic medications. Results were analyzed according to type of pain medication given, age categories, and race/ethnicity of the patients. Main outcomes: Percentage of EMS transports documenting the three painful impressions that had pain documented as a symptom, received any of the six pain medications, and the disparity in documentation and treatment by age and race/ethnicity. Results: There were 276,925 EMS records in the NEMSIS database that met inclusion criteria. Pain was listed as a primary or associated symptom for 29.5% of patients, and the youngest children (0–3 years) were least likely to have pain documented as a symptom (14.6%). Only 15.6% of all activations documented the receipt of prehospital pain medications. Children (<15 years) received pain medication 14.8% [95% CI 14.33, 15.34] of the time versus adults (≥15 years) 15.6% [95% CI 15.48, 15.76, p = 0.004]. Morphine and fentanyl were the most commonly administered medications to all age groups. Black patients were less likely to receive pain medication than other racial groups. Conclusions: Documentation of pain as a symptom and pain treatment continue to be infrequent in the prehospital setting in all age groups, especially young children. There appears to be a racial disparity with Black patients less often treated with analgesics. The broad incorporation of national NEMSIS data suggests that these inadequacies are a widespread challenge deserving further attention.

Diabetes-Related Emergency Medical Service Activations in 23 States, United States 2015

Abstract Objective: The use of emergency medical services (EMS) for diabetes-related events is believed to be substantial but has not been quantified nationally despite the diverse acute complications associated with diabetes. We describe diabetes-related EMS activations in 2015 among people of all ages from 23 U.S. states. Methods: We used data from 23 states that reported ≥95% of their EMS activations to the U.S. National Emergency Medical Services Information System (NEMSIS) in 2015. A diabetes-related EMS activation was defined using coded EMS provider impressions of “diabetes symptoms” and coded complaints recorded by dispatch of “diabetic problem.” We described activations by type of location, urbanicity, U.S. Census Division, season, and time of day; and patient-events by age category, race/ethnicity, disposition, and treatment with glucose. Crude and age-adjusted diabetes-related EMS patient-level event rates were calculated for adults ≥18 years of age with diagnosed diabetes using the Behavioral Risk Factor Surveillance System to estimate the population denominator. Results: Of 10,324,031 relevant EMS records, 241,495 (2.3%) were diabetes-related activations, which involved over 235,000 hours of service. Most activations occurred in urban or suburban environ- ments (86.4%), in the home setting (73.5%), and were slightly more frequent in the summer months. Most patients (72.6%) were ≥45 years of age and over one-half (55.4%) were transported to the emergency department. The overall age-adjusted diabetes-related EMS event rate was 33.9 per 1,000 persons with diagnosed diabetes; rates were highest in patients 18–44 years of age, males, and non-Hispanic blacks and varied by U.S. Census Division. Conclusions: Diabetes results in a substantial burden on EMS resources. Collection of more detailed diabetes complication information in NEMSIS may help facilitate EMS resource planning and prevention strategies.

Emergency Medical Services Response to Mass Shooting and Active Shooter Incidents, United States, 2014–2015

Abstract Background: The purpose of the current study was to describe the injury patterns, EMS response and interventions to mass shooting (MS) and active shooter (AS) incidents. Methods: Retrospective analysis of 2014–2015 National Emergency Medical Services Information System (NEMSIS) data sets. Date, time, and location for MS incidents were obtained from the Gun Violence Archive and then correlated with NEMSIS data set records. AS incidents were identified through Federal Bureau of Investigation (FBI) data. A de-identified database was generated for final analysis. Results: A total of 608 MS incidents were identified, of which 19 were also classified as AS incidents. NEMSIS patient care data was available for 652 EMS activations representing 226 unique MS incidents. Thirty-four EMS responses to 5 unique AS incidents were similarly identified: 76% of victims were male and 80% of victims were African American. Dispatch complaint did not suggest shooting (potentially dangerous scene environment) in 15.9% of records. The most commonly reported incident locations for MS were Street/Highway (38.2%) and Home/Residence (32.4%). Location of wounds included extremities (49%), chest (12%), and head/neck (13%). Tourniquet use was documented in 6 victims. 35.9% of victims were transported to the closest facility. Conclusions: MS and AS incidents are prevalent in the United States. Despite the fact that extremity wounds were common, documented EMS tourniquet use was uncommon. While MS events are high risk for responders, dispatch information was lacking in almost 15% of records. Responding EMS agencies were diverse, emphasizing the need to ensure all EMS providers are prepared to respond to MS incidents.

Injury-Based Geographic Access to Trauma Centers.

OBJECTIVE Previous studies examining access to trauma care use patient residence as a proxy for location and need for services, which could result in a flawed understanding of access to trauma centers. The objective of this study was to examine the geographic access of the U.S. population to trauma centers based on trauma incident locations. METHODS We conducted a cross-sectional study using 9-1-1 emergency medical services activations associated with traumatic injury from the 2014 National Emergency Medical Services Information System and trauma centers participating in the 2014 American Hospital Association Annual Survey. The measures included the percentage of trauma incidents that could reach a trauma center within 60 minutes by ground ambulance, capacity-to-demand ratio for each trauma center, and overall trauma care accessibility ratio for each U.S. zip code. RESULTS A total of 92.9% of all trauma incidents could be transported to an existing trauma center within 60 minutes by ground ambulance, and 85.3% could be transported to a Level I or II trauma center within this time frame in the 32 study states. While 94.7% of trauma incidents in the Northeast area could be transported to a Level I or II trauma center within a 60-minute driving time, the capacity-to-demand ratios of trauma centers in this region were low, indicating high utilization of those trauma center resources. By using the accessibility measure, we found that some Midwestern and Southern states had higher amounts of accessible trauma center resources relative to the number of injuries than Northeastern states. CONCLUSIONS These findings suggest that greater access to trauma care and significant variations can be observed throughout the 32 study states when using trauma incident location rather than patient residence to calculate access to trauma care. The proposed capacity-to-demand ratio and accessibility ratio can be applied to many other needs assessments in health care.