Simulator Pre-Screening of Underprepared Drivers Prior to Licensing On-Road Examination: Clustering of Virtual Driving Test Time Series Data.

BACKGROUND: A large Midwestern state commissioned a virtual driving test (VDT) to assess driving skills preparedness before the on-road examination (ORE). Since July 2017, a pilot deployment of the VDT in state licensing centers (VDT pilot) has collected both VDT and ORE data from new license applicants with the aim of creating a scoring algorithm that could predict those who were underprepared. OBJECTIVE: Leveraging data collected from the VDT pilot, this study aimed to develop and conduct an initial evaluation of a novel machine learning (ML)-based classifier using limited domain knowledge and minimal feature engineering to reliably predict applicant pass/fail on the ORE. Such methods, if proven useful, could be applicable to the classification of other time series data collected within medical and other settings. METHODS: We analyzed an initial dataset that comprised 4308 drivers who completed both the VDT and the ORE, in which 1096 (25.4%) drivers went on to fail the ORE. We studied 2 different approaches to constructing feature sets to use as input to ML algorithms: the standard method of reducing the time series data to a set of manually defined variables that summarize driving behavior and a novel approach using time series clustering. We then fed these representations into different ML algorithms to compare their ability to predict a driver's ORE outcome (pass/fail). RESULTS: The new method using time series clustering performed similarly compared with the standard method in terms of overall accuracy for predicting pass or fail outcome (76.1% vs 76.2%) and area under the curve (0.656 vs 0.682). However, the time series clustering slightly outperformed the standard method in differentially predicting failure on the ORE. The novel clustering method yielded a risk ratio for failure of 3.07 (95% CI 2.75-3.43), whereas the standard variables method yielded a risk ratio for failure of 2.68 (95% CI 2.41-2.99). In addition, the time series clustering method with logistic regression produced the lowest ratio of false alarms (those who were predicted to fail but went on to pass the ORE; 27.2%). CONCLUSIONS: Our results provide initial evidence that the clustering method is useful for feature construction in classification tasks involving time series data when resources are limited to create multiple, domain-relevant variables.

Measuring Health Information Dissemination and Identifying Target Interest Communities on Twitter: Methods Development and Case Study of the @SafetyMD Network.

BACKGROUND: Little is known about the ability of individual stakeholder groups to achieve health information dissemination goals through Twitter. OBJECTIVE: This study aimed to develop and apply methods for the systematic evaluation and optimization of health information dissemination by stakeholders through Twitter. METHODS: Tweet content from 1790 followers of @SafetyMD (July-November 2012) was examined. User emphasis, a new indicator of Twitter information dissemination, was defined and applied to retweets across two levels of retweeters originating from @SafetyMD. User interest clusters were identified based on principal component analysis (PCA) and hierarchical cluster analysis (HCA) of a random sample of 170 followers. RESULTS: User emphasis of keywords remained across levels but decreased by 9.5 percentage points. PCA and HCA identified 12 statistically unique clusters of followers within the @SafetyMD Twitter network. CONCLUSIONS: This study is one of the first to develop methods for use by stakeholders to evaluate and optimize their use of Twitter to disseminate health information. Our new methods provide preliminary evidence that individual stakeholders can evaluate the effectiveness of health information dissemination and create content-specific clusters for more specific targeted messaging.

Traumatic Stress, Depression, and Recovery: Child and Parent Responses After Emergency Medical Care for Unintentional Injury.

OBJECTIVE: To assess psychological symptoms in injured children (aged 8-17 years) and their parents after emergency department (ED) care to examine the relationship between posttraumatic stress and depression symptoms, co-occurrence of symptoms within families, and the relationship of these symptoms to parent-reported overall recovery. METHODS: Children and parents (n = 263 child-parent dyads) were enrolled during ED treatment for unintentional injury. Approximately 5 months later, children and parents (n = 178 dyads) completed standardized measures of posttraumatic stress and depression symptoms and parents reported on child overall recovery. RESULTS: Follow-up assessments found significant posttraumatic stress symptoms in 15% of children and 5% of parents, significant depression symptoms in 13% of children and 16% of parents, and problematic overall recovery in 17% of children. For both children and parents, posttraumatic stress and depression symptom severity were strongly associated. Child and parent symptoms were only modestly associated with each other, and there were few families in which both child and parent had significant posttraumatic stress or depression. Parent symptoms, but not child symptoms, were inversely associated with children's overall recovery. CONCLUSIONS: For about 1 in 6 children and parents, unintentional injury treated in the ED can be associated with negative psychological sequelae and suboptimal recovery. Within families, child and parent responses may differ; their relative association with overall recovery deserves additional research. To promote emotional recovery, ED clinicians should be aware of the potential psychological impact of unintentional injury, provide timely evidence-based anticipatory guidance, and communicate these concerns to primary care clinicians.

Acute stress disorder and posttraumatic stress disorder in parents of injured children.

Acute stress disorder (ASD) and posttraumatic stress disorder (PTSD) were examined in 334 parents of children with traffic-related injuries. In the first month after their child's injury, 12% of parents had ASD and another 25% had partial ASD. Among 251 parents assessed again approximately 6 months postinjury, 8% had PTSD and another 7% had partial PTSD. The ASD and PTSD severity were associated (r = .54), but ASD status was not a sensitive predictor of later PTSD. Independent predictors of ASD severity included prior trauma exposure, peritrauma exposure and perceptions of the child's pain and life threat, and child ASD severity. Independent predictors of PTSD severity included prior trauma exposure, parent ASD severity, and parent-rated child physical health at follow-up.

Risk factors for death among older child and teenaged motor vehicle passengers.

OBJECTIVE: To identify and prioritize risk factors for passenger death. DESIGN: Retrospective cohort captured in 2 databases: US census of fatal crashes (Fatality Analysis Reporting System) and US-representative sample of police-reported crashes (National Automotive Sampling System Crash Data System). SETTING: Crashes in 50 states and the District of Columbia from January 1, 2000, to December 31, 2005. PARTICIPANTS: A total of 45 560 passengers aged 8 to 17 years in crashes, representing 2 545 168 passengers (weighted). Main Exposures Age and sex of driver, vehicle and road type, restraint use, seating position, day of week, month, time of day, and speed limit. Outcome Measure Passenger fatality. RESULTS: Approximately 424 000 passengers (aged 8-17 years) annually were in tow-away crashes and experienced a fatality rate of 3.9 per 1000. Just more than three-quarters as many passengers (aged 8-17 years) were in crashes with 16- to 19-year-old drivers as with all adult drivers (those aged > or =25 years), with double the passenger fatality rate. For drivers aged 16 to 17 years, passenger fatalities doubled between passenger ages 11 to 12 years and increased to a peak at a passenger age of 16 years. Of crashes with child passenger fatality, 21.2% involved alcohol. In crashes with drivers younger than 18 years, the greatest risk factors for death for passengers aged 8 to 17 years were drivers younger than 16 years, restraint nonuse, and high-speed roads (> or =55 and 45-54 mph). CONCLUSIONS: While passenger risks because of restraint nonuse require continued attention, interventions must be developed to address risk because of travel on high-speed roads and riding with drivers younger than 16 years. For safety, child passengers and their parents should monitor these characteristics of the drive and the driver. Anticipatory guidance about passenger risks should begin by the age of 11 years.

Translational research in child and adolescent transportation safety.

Motor vehicle crash (MVC)-related injury is the leading cause of death among youths age 1 to 19 years. Advances in MVC and injury prevention depend on interrelated combinations of technology, policy, and education, and research on these topics can inform policy and lead to improvement in safety. This article presents two examples of translational research in transportation injury prevention. In the first example of child passenger protection, the authors describe a program of research designed to reduce the gap between the theoretical and practical effectiveness of safety methods. In the second example of novice teen drivers, the authors examine translational research related to two complementary approaches to reducing the exposure of novice teen drivers to high-risk driving conditions-graduated driver licensing policy and parental management of novice teen drivers. The examples suggest the utility of systematic programs of research designed to improve the translation into practice of MVC and injury prevention technology and policy.

Heart rate and posttraumatic stress in injured children.

BACKGROUND: Elevated, acute heart rate has been related to later posttraumatic stress disorder (PTSD) development in injured adults, but this has not been examined in children and adolescents. Better understanding of the relationship between acute physiological arousal and later child PTSD could help elucidate the etiology of posttrauma responses in children and might identify useful markers for PTSD risk. OBJECTIVE: To evaluate the relationship between heart rate assessed in the emergency department (ED) during normal clinical care and later PTSD outcome in traumatically injured children. DESIGN: Prospective cohort study assessed heart rate at ED triage and PTSD an average of 6 months' postinjury. SETTING: Large, urban pediatric academic medical center in the northeastern United States. PARTICIPANTS: One hundred ninety children and adolescents (aged 8-17 years) hospitalized for traffic-related injury.Main Outcome Measure Clinician-Administered PTSD Scale for Children and Adolescents. RESULTS: The group of children who developed partial or full PTSD had a higher mean +/- SD heart rate at ED triage than those who did not go on to have PTSD (109.6 +/- 22.3 vs 99.7 +/- 18.0 beats per minute). Children with an elevated heart rate (defined as >/=2 SDs higher than the normal resting heart rate for their age and sex) at ED triage were more likely to meet criteria for partial or full PTSD at follow-up, even after adjusting for age, sex, and injury (adjusted odds ratio, 2.4 [95% confidence interval, 1.1-5.4]). CONCLUSION: These results suggest an association between early physiological arousal and the development or persistence of PTSD symptoms in injured children and point to the importance of better understanding the interplay between physiological and psychological functioning after a traumatic stressor.

Predicting child PTSD: the relationship between acute stress disorder and PTSD in injured children.

OBJECTIVE: To examine the prevalence of acute stress disorder (ASD) and posttraumatic stress disorder (PTSD) in injured children and to evaluate the utility of ASD as a predictor of PTSD. METHOD: Children hospitalized for injuries sustained in a traffic crash were enrolled in a prospective study. ASD was assessed in 243 children within 1 month after injury, and PTSD was assessed in 177 of these children 3 or more months after injury. The relationship between ASD and PTSD was examined via correlations between symptom severity scores and calculation of sensitivity, specificity, and positive and negative predictive values for categorical prediction of PTSD from ASD or subsets of ASD symptoms. RESULTS: Eight percent of children met the symptom criteria for ASD and another 14% had subsyndromal ASD; 6% met the symptom criteria for PTSD and another 11% had subsyndromal PTSD. ASD and PTSD symptom severity were associated. Sensitivity was low for prediction of child PTSD from child ASD. Subsyndromal ASD was a more effective predictor of PTSD. CONCLUSIONS: A substantial minority of injured children are affected by traumatic stress disorders. ASD in children may not be an optimal categorical predictor of PTSD. With increasing attention to early posttrauma services for children, empirically valid assessment/triage models deserve further study.

Acute stress disorder symptoms in children and their parents after pediatric traffic injury.

OBJECTIVE: The American Academy of Pediatrics highlights the important role of pediatricians in recognizing adverse child responses to tragic events, such as traffic crashes. One challenge in effectively identifying children and their parents with troubling psychological responses to trauma is that little is known about the normal range of acute psychological responses in children and their parents in the immediate aftermath of traumatic events, making identification of adverse child responses difficult. Within the first month after a traumatic event, individuals may display reexperiencing, avoidance, and hyperarousal symptoms as well as dissociation (eg, feelings of unreality or emotional numbing). The presence of these responses, collectively known as acute stress disorder (ASD), alerts providers to those who may be at risk for ongoing difficulties. For beginning to develop an evidence base to guide pediatric care providers in addressing acute traumatic responses, the aim of the current investigation was to describe systematically the range and type of symptoms of ASD in children and their parents after pediatric traffic injury. METHODS: A prospective cohort study was conducted of traffic-injured children, who were 5 to 17 years of age and admitted to the hospital for treatment of injuries from traffic crashes, and their parents. All children who met eligibility criteria between July 1999 and May 2000 were invited to participate in the study. After consent/assent was obtained, children and their custodial parents (or guardians) were interviewed within 1 month after injury via a structured assessment to determine the circumstances of the crash and the presence of ASD symptoms. Relevant demographic and clinical information (eg, age, race, gender, date of injury) was abstracted from the medical records of subjects. A survey instrument to assess the presence of ASD symptoms was completed by both the child and his or her guardian. All children completed the Child Acute Stress Questionnaire, and all parents completed the Stanford Acute Stress Reaction Questionnaire. Responses were scored for the presence of dissociation, reexperiencing, avoidance, and/or hyperarousal symptoms as well as broad distress (symptoms present in every category). RESULTS: Symptoms of ASD were commonly observed in the children and parents. Eighty-eight percent of children and 83% of parents reported having at least 1 clinically significant symptom; this affected 90% of the families. Broad distress was observed for a large minority: 28% of children and 23% of parents. No statistically significant association was found between child broad distress and either child age (r = -0.12) or child injury severity score (r = -0.05). chi(2) analyses revealed no significant association between broad distress and child gender, child race, or mechanism of injury. No statistically significant association was found between parent broad distress and child age (r = -0.06) or child injury severity score (r = 0.09). chi(2) analyses revealed no significant association between parent broad distress and child gender or parent presence at the crash scene. Associations were found between parent broad distress and race in that fewer white parents reported broad distress. In addition, mechanism of injury was associated with parent broad distress: more parents reported broad distress when their children were involved in pedestrian-motor vehicle crashes, and fewer parents reported broad distress when their children were injured in a bicycle fall. CONCLUSIONS: Pediatric care providers can expect to see some ASD symptoms in most children and parents in the immediate aftermath of traffic-related injury. Brief education is appropriate to explain that these symptoms are normal reactions that are likely to resolve. If symptoms persist for >1 month or are particularly distressing in their intensity, then referral for psychological care may be necessary for treatment of posttraumatic stress disorder. Given the high prevalence of pediatric traffic crashes and the underdiagnosis of posttraumatic stress disorder, probing for recent crash exposure might be appropriate during routine child health maintenance. The following are recommendations for pediatricians: 1) routinely call the family several days and 1 to 2 weeks after a traffic injury and ask about behavioral symptoms and family function; 2) make use of the ongoing physician-patient relationship to explore symptom presence and intensity and any functional impairment in the injured child; a brief office visit with the child and parents could serve this purpose; 3) be sure to explore the effect that the child's injury has had on the family; remember that the parent's experience posttraumatic stress symptoms after pediatric traffic-related injuries and these symptoms may limit the parent's ability to support the child; 4) provide supportive care and give families the opportunity to discuss the crash and their current feelings; do not force families to talk about the crash; 5) although any child in a traffic crash or his or her parent is at risk for posttraumatic symptomatology, regardless of injury severity, particular attention should be paid to the parents of child pedestrians who are struck by motor vehicles. These parents experience posttraumatic symptoms more commonly than parents of children in other traffic crashes.