Predicting hospital admission from emergency department triage data for patients presenting with fall-related fractures.

PURPOSE: Predict in advance the need for hospitalization of adult patients for fall-related fractures based on information available at the time of triage to help decision-making at the emergency department (ED). METHODS: We developed machine learning models using routinely collected triage data at a regional hospital chain in Pennsylvania to predict admission to an inpatient unit. We considered all patients presenting to the ED for fall-related fractures. Patients who were 18 years or younger, who left the ED against medical advice, left the ED waiting room without being seen by a provider, and left the ED after initial diagnostics were excluded from the analysis. We compared models obtained using triage data (pre-model) with models developed using additional data obtained after physicians' diagnoses (post-model). RESULTS: Our results show good discriminatory power on predicting hospital admissions. Neural network models performed the best (AUC: pre-model = 0.938 [CI 0.920-0.956], post-model = 0.983 [0.974-0.992]). The logistic regression analysis provides additional insights into the data and the relationships between the variables. CONCLUSIONS: Using limited data available at the time of triage, we developed four machine learning models aimed at predicting hospitalization for patients presenting to the ED for fall-related fractures. All the four models were robust and performed well. Neural network method, however, performed the best for both pre- and post-models. Simple, parsimonious machine learning models can provide high accuracy for predicting hospital admission.

A comparison of cervical ripening modalities among overweight and obese nulliparous gravidas.

Objective: To determine if differences exist among nulliparous overweight and obese gravidas undergoing cervical ripening employing three different agents (dinoprostone, misoprostol, or cervical catheter).Methods: A retrospective cohort study of nulliparous overweight and obese women who underwent induction of labor at two south-central Pennsylvania hospitals between January 2014 and December 2017. Nulliparous gravidas, ≥37 weeks' gestational age, with singleton pregnancies in the vertex presentation, were included in the study. We employed the following definitions: (1) overweight: BMI 25.0-29.9 kg/m2; (2) class I obesity: BMI 30.0-34.9 kg/m2; (3) class II obesity: BMI 35.0-39.9 kg/m2; and (4) class III obesity: BMI >40.0 kg/m2. The primary outcome measure was the mean difference in induction-to-birth time. A subanalysis was performed to assess the effect of BMI on the primary outcome. Secondary outcome measures included mode of delivery, induction-to-second-stage-of-labor time, estimated blood loss, neonatal feeding type, neonatal Apgar scores, and neonatal admission to triage or intensive care unit (ICU) after delivery. A priori power calculation estimated that 156 patients would be needed using the medium effective size. Data analysis was performed using ANOVA for continuous variables and chi-square tests for categorical variables.Results: Among 192 nulliparous overweight and obese gravidas, 70 received dinoprostone, 72 were given misoprostol, and 50 had cervical ripening with cervical catheters. There were no significant differences in mean induction to birth times among overweight and obese women when comparing the three cervical ripening agents (dinoprostone 24.5 ± 15.2 versus misoprostol 28.7 ± 12.3 and catheters 25.1 ± 12.9 hours), (p = .145, 95% CI -8.7 to 0.2 and -5.5 to 4.3, respectively). Overweight nulliparous women had shorter mean induction to birth time (22.9 ± 11.4 versus 29.2 ± 15.8 hours) as compared to class II obese women, (p = .037, 95% CI -12.0 to -0.38). When overweight women were compared to class III obese women, shorter mean induction to birth time (22.9 ± 11.4 versus 30.9 ± 13.9 hours) was also found, (p = .005, 95% CI -13.4 to -2.4).Conclusion: Among nulliparous overweight and obese gravidas, neither dinoprostone, misoprostol, or cervical catheter significantly impacted the induction to birth time. There was a longer induction to birth time for class II and class III obese women when compared to overweight women. Additional studies are warranted to improve cervical ripening in nulliparous overweight and obese women.

A Hospital-Level Intervention to Improve Outcomes of Opioid Exposed Newborns.

PURPOSE: The purpose of this quality improvement project was to determine if non-pharmacologic strategies such as a rooming-in approach to care for newborns at risk of developing neonatal abstinence syndrome (NAS) would reduce total length of stay (LOS) and reduce the need for pharmacologic treatment. DESIGN AND METHODS: This was a quality improvement project utilizing a retrospective chart review. Records of newborns with in-utero methadone or buprenorphine exposure were reviewed who were born between January 2016-July 2017 and July 2017-August 2018 at Wellspan Health York Hospital. Starting in July 2017, newborns exposed to opioids who transitioned normally remained with their mothers for monitoring in the newborn nursery. Monitoring for withdrawal was continued on the pediatric floor after the mother's discharge from the post-partum floor. RESULTS: The primary outcome of total LOS was reduced from 14 days to 10.1 days (p = 0.014). The total length of pharmacologic treatment decreased from 15.68 days to 9.71 days (p = 0.023). CONCLUSIONS: A rooming-in approach to care including management on a pediatric floor can reduce total length of stay and the duration of pharmacologic treatment in newborns with NAS. Newborns with NAS can be safely managed in an inpatient pediatric floor. PRACTICE IMPLICATIONS: Implementing a rooming-in approach to care of newborns at risk of developing NAS can improve outcomes through a decreased length of hospital stay and decreased duration of pharmacologic treatment. This approach improves access to critical care services by safely monitoring newborns with NAS on an inpatient pediatric floor.