purpose

Traumatic injury is a leading cause of death and disability in the United States. This study aimed to: 1) to elicit the acceptability of technology-assisted injury outcome monitoring, 2) to test the feasibility of symptom assessment and extended patient communication using a web-based platform, and 3) to assess long-term symptom burden and recovery challenges in patients at particularly high-risk for poor outcomes.

methods

26 individuals (21-61 years), 12-36 months after serious injury, were recruited from a concluded cohort study which longitudinally followed 623 Black/African American men in Philadelphia for three months following hospitalization. Participants completed qualitative interviews which explored their: long-term symptomology, facilitators and barriers to recovery, and typical use of smartphone and communication technology. 25 participants then trialed a web-based remote monitoring intervention that automated 12 waves of text and email delivered assessments of: functional status, psychosocial well-being, workforce participation, pain, sleep, neighborhood environment, and healthcare access. Qualitative data were analyzed with content analysis in NVIVO. Survey data were analyzed descriptively in STATA.

results

Participants describe significant and persistent symptoms of psychological distress, physical pain, and altered workforce participation and financial stability 12-36 months after injury. Trial of remote monitoring identified participants’ preference for text-delivered surveys, and the feasibility of collecting outcome data across multiple domains. Survey responses reinforced participants’ qualitative perceptions of chronic pain burden, limited workforce participation, and disturbance in sleep quantity and quality.

conclusions

Technology-assisted remote monitoring may be leveraged for future interventions and research aimed at minimizing the long-term and public health consequences of traumatic injury.

background/purpose: Although analyses of crash data have advanced our understanding of motor vehicle crashes, the vast majority of studies are plagued by an inability to account for “driving exposure”— the extent to which drivers actually drive and are thus “at risk” for a crash. As a result, most studies have been unable to validly compare crash rates between driver groups or measure changes in rates within individuals over time. In addition, we do not have high-quality methods to estimate young drivers’ exposure to (i.e., frequency of engagement in) high-risk driving behaviors. To address these two critical gaps, we aimed to establish a cost-efficient and highly generalizable Quasi Induced Exposure (QIE)-based method that can: (1) identify specific teen driver groups who spend more time driving in high-risk conditions and (2) account for differences in time “at risk” when estimating teen driver crash risk.

methods/results: In this methodological presentation, we will describe our proposed approach to utilize QIE methods to measure state-wide and group-specific exposure to (i.e., frequency of engagement in) high risk behaviors. As an applied example, we analyzed data from the New Jersey Traffic Safety Outcomes data warehouse to determine the proportion of newly-licensed young drivers who drive older (i.e., less safe) vehicles.

conclusions: Our proposed approach can be used in other jurisdictions and with various driver populations (e.g., teen drivers, older drivers, with child passengers) to develop knowledge on how to best reduce crash risk. This, in turn, will provide practitioners with clear direction on the most efficacious interventions.

Background. This study profiles traumatic injury in New Jersey based the newly established statewide trauma registry. This paper addresses how trauma centers (TC) compare with non-trauma hospitals (NTH) with respect to numbers treated, severity of injury (SOI), and survival post-treatment. Secondarily, the study profiles the statewide distribution of trauma in terms of type and external mechanism of injury and demographic characteristics of injury victims. Methods. The data under study include all patients admitted to New Jersey Level I & II trauma centers, and patients presenting to non-trauma NJ hospitals during 2016-2017. Study endpoints include treatment site and survival; other covariates include: principle and secondary ICD10 diagnoses, trauma center-assigned injury severity score (ISS) and SOI evaluated via ICDPIC-R computer algorithm; E-codes; age, sex, race/ethnicity, residence, LOS, ICULOS, and discharge status. Descriptive statistics are presented for study variables; other methods include difference of means (T-tests) for ISS, LOS, ICULOS by treatment site; chi square statistics for contingency table analysis; and general GEE logistic regression analysis for binary outcomes (treatment at TC; survival) (Stata, V15). The models evaluated in this study include (1) trauma ctr treatment site-as predicted by ICDPICR ISS; (2) mortality- as predicted by ISS and treatment site, adjusted for age/sex; and (3) mortality as predicted by treatment site, race/ethnicity and interaction of ISS with age. Results: ICDPIC-R and ISS are statistically significantly correlated; ICDPIC-R appears to understate actual SOI. ICDPIC-R is a good predictor of trauma triage AND mortality. Conclusion: Findings hold policy significance for QI and trauma system evaluation.

Background: There is a need for more objective methods to assist in the determination of return to play (RTP) timelines. RTP is typically reported using proportions, but this can be inaccurate in instances where a lack of follow-up data excludes some athletes from analysis. Time to event analysis can provide more accurate estimates by accounting for all injured athletes.

Methods: Athletic trainers (ATs) participating in the NCAA Injury Surveillance Program (NCAA-ISP) reported data for 1,940,462 football athlete-exposures (AEs) during the 2009/10-2017/18 academic years. RTP was determined for new lower extremity (LE) injuries. Time lost from participation was calculated using injury and RTP dates. For athletes with no reported return date, time-loss was calculated using season end dates by division; these injuries were censored cases. Incidence proportions (IP) were calculated by division for non-time-loss (NTL), 1-6, 7-13, 14-29 and ≥30 days; RTP probabilities were calculated using the Life-Table Method.

Results: ATs reported a total of 8,152 new LE injuries. Few censored cases were reported for NTL injuries; the IP and RTP probabilities were equal for all divisions. Both metrics indicate an athlete has the highest chance of RTP in the first week. After the first week, the IP underestimates RTP by excluding censored cases, and fails to provide an accurate estimate for RTP timelines.

Conclusions: The inclusion of censored cases provides a more accurate estimate of RTP timelines, particularly in instances with a larger proportion of censored. In these instances, commonly reported statistics can underestimate RTP timelines by excluding these cases.

Background/Purpose

Emergency trauma response for high-acuity patients requires collaboration under stress within a multi-disciplinary team of providers. The current medical training model is not optimized for learning in a chaotic environment and among different professions and levels of learners ranging from students to fellows. This study utilized Virtual Reality (VR) technology and Cognitive load theory (CLT) to determine curricular needs for trauma providers.

Methods

This study combines principles of CLT with qualitative interviewing methods. CLT states that increased load on a learner’s working memory can compromise the learner’s acquisition of knowledge. Participants completed think-aloud interviews and a debrief questionnaire while watching a previously recorded trauma resuscitation through VR Oculus headsets. We conducted 25 interviews with first- and second-year residents across Emergency Medicine, Anesthesia, and General Surgery, as well as 3 faculty experts. Interview data were synthesized into emerging themes describing sources of cognitive load.

Results/Outcomes

Interviews revealed 6 sources of cognitive load for junior-level resident that impeded their ability to learn during a trauma: 1) Orienting to VR technology; 2) Orienting to the room layout; 3) Learning team members’ roles and responsibilities; 4) Managing the primary survey of the patient; 5) Learning how and when to be anticipatory; 6) Understanding clinical content.

Conclusions

The study revealed needs to include in a curriculum. VR technology aided in identifying sources of cognitive load among junior resident physicians while delivering trauma care. With this data we can develop and sequence the curriculum.