A Real-Time Automated Machine Learning Algorithm for Predicting Mortality in Trauma Patients: Survey Says it's Ready for Prime-Time.

BACKGROUND: Though artificial intelligence ("AI") has been increasingly applied to patient care, many of these predictive models are retrospective and not readily available for real-time decision-making. This survey-based study aims to evaluate implementation of a new, validated mortality risk calculator (Parkland Trauma Index of Mortality, "PTIM") embedded in our electronic healthrecord ("EHR") that calculates hourly predictions of mortality with high sensitivity and specificity. METHODS: This is a prospective, survey-based study performed at a level 1 trauma center. An anonymous survey was sent to surgical providers and regarding PTIM implementation. The PTIM score evaluates 23 variables including Glasgow Coma Score (GCS), vital signs, and laboratory data. RESULTS: Of the 40 completed surveys, 35 reported using PTIM in decision-making. Prior to reviewing PTIM, providers identified perceived top 3 predictors of mortality, including GCS (22/38, 58%), age (18/35, 47%), and maximum heart rate (17/35, 45%). Most providers reported the PTIM assisted their treatment decisions (27/35, 77%) and timing of operative intervention (23/35, 66%). Many providers agreed that PTIM integrated into rounds and patient assessment (22/36, 61%) and that it improved efficiency in assessing patients' potential mortality (21/36, 58%). CONCLUSIONS: Artificial intelligence algorithms are mostly retrospective and lag in real-time prediction of mortality. To our knowledge, this is the first real-time, automated algorithm predicting mortality in trauma patients. In this small survey-based study, we found PTIM assists in decision-making, timing of intervention, and improves accuracy in assessing mortality. Next steps include evaluating the short- and long-term impact on patient outcomes.

Burn and Trauma Burden and Screening for Interpersonal Violence During the COVID-19 Pandemic.

INTRODUCTION: Nationwide shelter-in-place (SIP) orders during the pandemic have had long-lasting effects, including increased rates of domestic violence and interpersonal violence. Screening for violence varies by institution, which tool is used, and when. Given increases in burn and trauma admissions over the course of the pandemic, we sought to examine trends at our institution during this time period to better guide care and anticipate system-level effects. METHODS: We performed a retrospective cohort study of pediatric burn and adult burn and trauma patients at our level 1 trauma/burn center between March-May 2019 and March-May 2020. Home safety screening was performed by nursing staff using a 1-part screening questionnaire. Patients presenting before March 15, 2020, were defined as "pre-SIP; " between March 16-May 19, 2020, were "during SIP; " and those after May 19, 2020, were designated as "post-SIP." Descriptive and chi-square statistics were used. Demographic, injury patterns, and screening information were collected. RESULTS: Blunt trauma comprised 60% of injuries, followed by burns (30%) then penetrating injury (7%). Over the entire time period analyzed, 1822 patients had documented home safety screening; ∼2% of patients screened reported a safety concern pre-SIP, compared to 3% of patients during SIP. There were higher rates of burns and penetrating injury during SIP compared to other periods (P ≤ 0.0001). Home safety screening rates were 94%-95% pre- and during SIP, but dropped to 85% post-SIP (P < 0.0001). Home safety concerns were reported almost 2% of the time pre-SIP and 3% during SIP (P = 0.016). CONCLUSIONS: We noted an increase in trauma and burns during and after SIP orders, consistent with the experiences of other institutions. Implementation of a nurse-driven screening process demonstrated high compliance with appropriate referrals. The burden of burn and traumatic injury remains significant, highlighting a need for continued psychosocial screening and the provision of psychosocial support resources in the acute trauma setting.

Medical management of early pregnancy loss is cost-effective compared with office uterine aspiration.

BACKGROUND: Early pregnancy loss, also referred to as miscarriage, is common, affecting approximately 1 million people in the United States annually. Early pregnancy loss can be treated with expectant management, medications, or surgical procedures-strategies that differ in patient experience, effectiveness, and cost. One of the medications used for early pregnancy loss treatment, mifepristone, is uniquely regulated by the Food and Drug Administration. OBJECTIVE: This study aimed to compare the cost-effectiveness from the healthcare sector perspective of medical management of early pregnancy loss, using the standard of care medication regimen of mifepristone and misoprostol, with that of office uterine aspiration. STUDY DESIGN: We developed a decision analytical model to compare the cost-effectiveness of early pregnancy loss treatment with medical management with that of office uterine aspiration. Data on medical management came from the Pregnancy Failure Regimens randomized clinical trial, and data on uterine aspiration came from the published literature. The analysis was from the healthcare sector perspective with a 30-day time horizon. Costs were in 2018 US dollars. Effectiveness was measured in quality-adjust life-years gained and the rate of complete gestational sac expulsion with no additional interventions. Our primary outcome was the incremental cost per quality-adjust life-year gained. Sensitivity analysis was performed to identify the key uncertainties. RESULTS: Mean per-person costs were higher for uterine aspiration than for medical management ($828 [95% confidence interval, $789-$868] vs $661 [95% confidence interval, $556-$766]; P=.004). Uterine aspiration more frequently led to complete gestational sac expulsion than medical management (97.3% vs 83.8%; P=.0001); however, estimated quality-adjust life-years were higher for medical management than for uterine aspiration (0.082 [95% confidence interval, 0.8148-0.08248] vs 0.079 [95% confidence interval, 0.0789-0.0791]; P<.0001). Medical management dominated uterine aspiration, with lower costs and higher confidence interval. The probability that medical management is cost-effective relative to uterine aspiration is 97.5% for all willingness-to-pay values of ≥$5600/quality-adjust life-year. Sensitivity analysis did not identify any thresholds that would substantially change outcomes. CONCLUSION: Although office-based uterine aspiration more often results in treatment completion without further intervention, medical management with mifepristone pretreatment costs less and yields similar quality-adjust life-years, making it an attractive alternative. Our findings provided evidence that increasing access to mifepristone and eliminating unnecessary restrictions will improve early pregnancy care.

What Are We Missing?: Evaluating an Intimate Partner Violence Screening Program in a Pediatric Emergency Department.

OBJECTIVES: Our study sought to explore and assess pediatric emergency department (ED) health care providers' knowledge, attitudes, and behaviors surrounding an existing intimate partner violence (IPV) screening program 4 years after initial implementation. METHODS: We used anonymous electronic surveys and telephone interviews to obtain provider perspectives using a mixed-methods analysis. We used χ2 tests to analyze the quantitative survey results, and an unstructured qualitative approach to analyze the telephone interviews. RESULTS: We analyzed 141 survey responses, which correlated to a response rate of about 35% of all the providers reached, and 20 telephone interviews. Our results demonstrate that pediatric ED providers have some knowledge of our existing caregiver IPV screening program in the pediatric ED and universally endorse routine caregiver IPV screening, which both are suggestive of postimplementation cultural shifts. However, reported provider behaviors still indicate selective/targeted screening. For example, many providers reported screening males and nontraditional caregivers less often compared with female caregivers. Reported barriers potentially explaining such screening habits mirror those in existing literature: patient acuity, time, multiple caregivers being present, and more. CONCLUSIONS: Our study indicates that more research must be done to assess root causes of provider barriers to IPV screening in pediatric ED settings because trainings and a long-standing program do not seem to be changing screening practices. Addressing these issues may lead to truly sustainable and effective IPV screening programs in pediatric ED settings.

Cost-effectiveness of Mifepristone Pretreatment for the Medical Management of Nonviable Early Pregnancy: Secondary Analysis of a Randomized Clinical Trial.

Importance: Early pregnancy loss (EPL) is the most common complication of pregnancy. A multicenter randomized clinical trial compared 2 strategies for medical management and found that mifepristone pretreatment is 25% more effective than the standard of care, misoprostol alone. The cost of mifepristone may be a barrier to implementation of the regimen. Objective: To assess the cost-effectiveness of medical management of EPL with mifepristone pretreatment plus misoprostol vs misoprostol alone in the United States. Design, Setting, and Participants: This preplanned. prospective economic evaluation was performed concurrently with a randomized clinical trial in 3 US sites from May 1, 2014, through April 30, 2017. Participants included 300 women with anembryonic gestation or embryonic or fetal demise. Cost-effectiveness was computed from the health care sector and societal perspectives, with a 30-day time horizon. Data were analyzed from July 1, 2018, to July 3, 2019. Interventions: Mifepristone pretreatment plus misoprostol administration vs misoprostol alone. Main Outcomes and Measures: Costs in 2018 US dollars, effectiveness in quality-adjusted life-years (QALYs), and treatment efficacy. Incremental cost-effectiveness ratios (ICERs) of mifepristone and misoprostol vs misoprostol alone were calculated, and cost-effectiveness acceptability curves were generated. Results: Among the 300 women included in the randomized clinical trial (mean [SD] age, 30.4 [6.2] years), mean costs were similar for groups receiving mifepristone pretreatment and misoprostol alone from the health care sector perspective ($696.75 [95% CI, $591.88-$801.62] vs $690.88 [95% CI, $562.38-$819.38]; P = .94) and the societal perspective ($3846.30 [95% CI, $2783.01-$4909.58] vs $4845.62 [95% CI, $3186.84-$6504.41]; P = .32). The mifepristone pretreatment group had higher QALYs (0.0820 [95% CI, 0.0815-0.0825] vs 0.0806 [95% CI, 0.0800-0.0812]; P = .001) and a higher completion rate after first treatment (83.8% vs 67.1%; P < .001) than the group receiving misoprostol alone. From the health care sector perspective, mifepristone pretreatment was cost-effective relative to misoprostol alone with an ICER of $4225.43 (95% CI, -$195 053.30 to $367 625.10) per QALY gained. From the societal perspective, mifepristone pretreatment dominated misoprostol alone (95% CI, -$5 111 629 to $1 801 384). The probabilities that mifepristone pretreatment was cost-effective compared with misoprostol alone at a willingness-to-pay of $150 000 per QALY gained from the health care sector and societal perspectives were approximately 90% and 80%, respectively. Conclusions and Relevance: This study found that medical management of EPL with mifepristone pretreatment was cost-effective when compared with misoprostol alone. Trial Registration: ClinicalTrials.gov Identifier: NCT02012491.