Speed kills? Quantifying the association between police traffic stops, types of stops, and motor vehicle collisions.

BACKGROUND: Motor vehicle crashes (MVCs) are a leading cause of traumatic death and injury. Police traffic stops (PTS) are a common approach to enforcing motor vehicle laws intended to prevent MVCs. However, it is unclear which types of PTS are most effective. This study examined the relationship of PTS subtypes among municipal police patrols on non-interstate roads and MVCs and MVC-related deaths. METHODS: PTS subtype data were characterized from six North Carolina cities: Charlotte, Durham, Fayetteville, Greensboro, Raleigh, and Winston-Salem. The primary outcomes of this study were yearly non-interstate MVC and MVC-related death rates per 100 population. The data were analyzed as balanced time-series cross-sectional data. The statistical analysis accounted for time-dependent and city-dependent confounding. We used a two-way fixed effects model to analyze the relationship between PTS and MVC or MVC-related deaths. We also utilized the difference in difference (DID) analysis to analyze if the reduction of PTS following a 2012 policing administrative change in Fayetteville had an association with MVC or MVC-related deaths. RESULTS: We found no significant overall association between non-interstate PTS and MVCs (Coeff: -0.00006; p = 0.43) or MVC-related deaths (Coeff: -0.00011; p = 0.15). Panel regression suggested no significant relationship between MVCs and MVC-related deaths and PTS related to driving while impaired (p = 0.36), safe movement violation (p = 0.43), or seatbelt violations (p = 0.17). However, speed limit violations (Coeff: -0.00025; p = 0.032) and stop-light/sign violations (Coeff: -0.00147; p = 0.017) related to PTS significantly reduced MVC-related deaths. The DID regression showed no significant impact on MVCs (p = 0.924) or MVC-related deaths (0.706) before and after the police reform period. CONCLUSIONS: The evidence regarding the absence of an overall association and any association with most PTS subtypes suggest that PTS are not effective for MVC death prevention. Policymakers may proceed with exploring modifications to policing efforts without detriments to public safety as defined by MVC and MVC-related deaths. LEVEL OF EVIDENCE: Retrospective epidemiological study, level IV.

Computational risk model for predicting 2-year malignancy of pulmonary nodules using demographic and radiographic characteristics.

OBJECTIVES: To determine whether discriminatory performance of a computational risk model in classifying pulmonary lesion malignancy using demographic, radiographic, and clinical characteristics is superior to the opinion of experienced providers. We hypothesized that computational risk models would outperform providers. METHODS: Outcome of malignancy was obtained from selected patients enrolled in the NAVIGATE trial (NCT02410837). Five predictive risk models were developed using an 80:20 train-test split: univariable logistic regression model based solely on provider opinion, multivariable logistic regression model, random forest classifier, extreme gradient boosting model, and artificial neural network. Area under the receiver operating characteristic curve achieved during testing of the predictive models was compared to that of prebiopsy provider opinion baseline using the DeLong test with 10,000 bootstrapped iterations. RESULTS: The cohort included 984 patients, 735 (74.7%) of which were diagnosed with malignancy. Factors associated with malignancy from multivariable logistic regression included age, history of cancer, largest lesion size, lung zone, and positron-emission tomography positivity. Testing area under the receiver operating characteristic curve were 0.830 for provider opinion baseline, 0.770 for provider opinion univariable logistic regression, 0.659 for multivariable logistic regression model, 0.743 for random forest classifier, 0.740 for extreme gradient boosting, and 0.679 for artificial neural network. Provider opinion baseline was determined to be the best predictive classification system. CONCLUSIONS: Computational models predicting malignancy of pulmonary lesions using clinical, demographic, and radiographic characteristics are inferior to provider opinion. This study questions the ability of these models to provide additional insight into patient care. Expert clinician evaluation of pulmonary lesion malignancy is paramount.