Bo Lu

Individualized treatment rules (ITRs) tailor treatments to individuals based on their unique characteristics to optimize clinical outcomes and resource allocation. Current approaches use outcome modeling or propensity score weighting to control confounding in complex medical data. To avoid model misspecification and the impact of extreme weights, matched‐learning (M‐learning) was recently proposed for continuous outcomes. In this paper, we expand the existing M‐learning methodology to estimate optimal ITRs under right‐censored data, as time‐to‐event outcomes are common in medical research. We construct matched sets for individuals by comparing observed times and incorporate an inverse probability censoring weight into the value function to handle censored observations. Additionally, we consider a full matching design as a possible alternative to the matching with replacement in M‐learning. We demonstrate that the proposed value function is unbiased for the true value function without censoring. To gain insight into the empirical performance, we conduct an extensive simulation study that compares M‐learning with two matching designs and a weighed learning approach. Results are evaluated based on winning probabilities and estimated values. The simulation reveals that all methods are generally fine in the absence of unmeasured confounders, and different methods show somewhat different performances under various scenarios. But their performance drops substantially in the presence of unmeasured confounders. Finally, we apply these methods to estimate optimal ITRs for patients with atrial fibrillation (AF) complications from an electronic medical record database, where full matching design shows slightly better performance.

Leveraging external data information to supplement randomized clinical trials has been a popular topic in recent years, especially for medical device and drug discovery. In rare diseases, it is very challenging to recruit patients and run a large‐scale randomized trial. To take advantage of real‐world data from historical trials on the same disease, we can run a small hybrid trial and borrow historical controls to increase the power. But the borrowing needs to be conducted in a statistically principled manner. Bayesian power prior methods and propensity score adjustments have been discussed in the literature. In this paper, we propose a matching‐assisted power prior approach to better mitigate observed bias when incorporating external data. A subset of comparable external subjects is selected by groups through template matching, and different weights are assigned to these groups based on their similarity to the current study population. Power priors are then implemented to incorporate the information into Bayesian inference. Unlike conventional power prior methods, which discount all control patients similarly, matching pre‐selects good controls, hence improved the quality of external data being borrowed. We compare its performance with the existing propensity score‐integrated power prior approach through simulation studies and illustrate the implementation using data from a real acupuncture clinical trial.

Objective The aim of this study was to determine how the COVID-19 pandemic affected patient demographics, injury mechanisms, interhospital transfers and mortality of patients with traumatic brain injuries (TBIs) treated in US emergency departments (EDs). Design This cross-sectional study analysed 2016–2020 Nationwide Emergency Department Sample (NEDS) data. Setting US EDs contained in the NEDS. Participants Participants were patients with moderate and severe TBI who visited the ED. Primary and secondary outcome measures Probability sampling design and survey weights generated nationally representative estimates of ED visits by patient demographics, hospital characteristics and COVID-19 diagnosis and the top four leading TBI causes. To assess COVID-19 impact, we calculated the per cent change of estimated TBI ED visits and disposition outcomes from 2016 to 2019 and 2019 to 2020, and proportion and 95% CI of injury severity groups and admitted/transfer by hospital type and trauma centre level. A multivariable logistic regression model identified the mortality OR by patient demographics, injury severity, hospital characteristics and COVID-19 positive diagnosis. Results In 2020, there were 527 123 ED visits nationally for TBI, 4.3% higher than 2016 but 1.0% lower than 2019. Patients with TBI transferring to short-term hospitals and other facilities increased by 16.0% and 18.2%, respectively, from 2016 to 2019 and were 3.7% and 14.1% higher in 2020 than in 2019. An estimated 3317 patients with TBI died (in ED or later in hospital) in 2020, which is 9.8% higher than 2019. Firearm-related TBI proportion was 15.1% higher in 2020 than in 2019. Patients with TBI with injury severity scores 25–75 were significantly higher at nontrauma (29.4% vs 31.9 %) and level III trauma centres (34.9% vs 38.2%) in 2020 than in 2019. Patients with TBI treated at rural hospitals had significantly higher odds of mortality (OR=1.95, CI=1.58–2.40) than those at urban hospitals. Conclusions TBI patient mortality was higher at all US hospital types and almost all trauma centre levels in 2020 than in 2019. Patients with TBI treated at rural hospitals had a significantly higher mortality risk.

A concern surrounding marijuana legalization is that driving after marijuana use may become more prevalent. Survey data are valuable for estimating policy effects, however their observational nature and unequal sampling probabilities create challenges for causal inference. To estimate population‐level effects using survey data, we propose a matched design and implement sensitivity analyses to quantify how robust conclusions are to unmeasured confounding. Both theoretical justification and simulation studies are presented. We found no support that marijuana legalization increased tolerant behaviors and attitudes toward driving after marijuana use, and these conclusions seem moderately robust to unmeasured confounding.

Being able to predict motor vehicle crashes which are a major public health concern would greatly improve traffic safety. The prevalence of mobile sensing platforms now allows spatially and temporally rich driving data to be collected relatively easily. Research efforts have been devoted to predicting crashes from such data. This paper seeks in particular to assess the feasibility and performance of using aggregated fine geo-resolution vehicle telemetric data for crash risk prediction. We acquired vehicle telemetric data from Geotab Inc., which recorded the frequency of hard acceleration, hard braking, harsh cornering, and the average magnitude of those harsh events among its registered commercial vehicles for every 150 × 150 m² roadway segment within Columbus, Ohio between January and April 2018. We aggregated the data, obtained the crash history from the Ohio Police Accident Report, and leveraged three machine learning–based algorithms to predict the crash risk. The results suggest that aggregated vehicle telemetric data could provide acceptable predictions for crash risk at a roadway segment level. Our models’ predictive performances were further improved and maximized by including in the models both vehicle telemetric data and roadway geometric characteristics. Several factors, such as the aggregated count of hard accelerations and the presence of an intersection, were shown to be the factors that potentially made the greatest contribution to crash occurrence. We concluded that vehicle telemetric data could provide complementary and valuable information about crash likelihood monitoring, which may enable the police and city planners to implement proactive safety interventions. Yet the nature of traffic crashes is still complex and multi-dimensional.

PDF File - 501K, PF02341066 is well tolerated in female athymic nude mice.

Matching is a popular design for inferring causal effect with observational data. Unlike model‐based approaches, it is a nonparametric method to group treated and control subjects with similar characteristics together, hence to re‐create a randomization‐like scenario. The application of matched design for real world data may be limited by: (1) the causal estimand of interest; (2) the sample size of different treatment arms. We propose a flexible design of matching, based on the idea of template matching, to overcome these challenges. It first identifies the template group which is representative of the target population, then match subjects from the original data to this template group and make inference. We provide theoretical justification on how it unbiasedly estimates the average treatment effect using matched pairs and the average treatment effect on the treated when the treatment group has a bigger sample size. We also propose using the triplet matching algorithm to improve matching quality and devise a practical strategy to select the template size. One major advantage of matched design is that it allows both randomization‐based or model‐based inference, with the former being more robust. For the commonly used binary outcome in medical research, we adopt a randomization inference framework of attributable effects in matched data, which allows heterogeneous effects and can incorporate sensitivity analysis for unmeasured confounding. We apply our design and analytical strategy to a trauma care evaluation study.