Triplet Matching for Estimating Causal Effects With Three Treatment Arms: A Comparative Study of Mortality by Trauma Center Level

Efficient algorithms for building representative matched pairs with enhanced generalizability

Many recent efforts center on assessing the ability of real-world evidence (RWE) generated from non-randomized, observational data to produce results compatible with those from randomized controlled trials (RCTs). One noticeable endeavor is the RCT DUPLICATE initiative. To better reconcile findings from an observational study and an RCT, or two observational studies based on different databases, it is desirable to eliminate differences between study populations. We outline an efficient, network-flow-based statistical matching algorithm that designs well-matched pairs from observational data that resemble the covariate distributions of a target population, for instance, the target-RCT-eligible population in the RCT DUPLICATE initiative studies or a generic population of scientific interest. We demonstrate the usefulness of the method by revisiting the inconsistency regarding a cardioprotective effect of the hormone replacement therapy (HRT) in the Women's Health Initiative (WHI) clinical trial and corresponding observational study. We found that the discrepancy between the trial and observational study persisted in a design that adjusted for the difference in study populations' cardiovascular risk profile, but seemed to disappear in a study design that further adjusted for the difference in HRT initiation age and previous estrogen-plus-progestin use. The proposed method is integrated into the R package match2C.

Flexible template matching for observational study design

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.

Machine learning for outcome predictions of patients with trauma during emergency department care

OBJECTIVES

To develop and evaluate a machine learning model for predicting patient with trauma mortality within the US emergency departments.

METHODS

This was a retrospective prognostic study using deidentified patient visit data from years 2007 to 2014 of the National Trauma Data Bank. The predictive model intelligence building process is designed based on patient demographics, vital signs, comorbid conditions, arrival mode and hospital transfer status. The mortality prediction model was evaluated on its sensitivity, specificity, area under receiver operating curve (AUC), positive and negative predictive value, and Matthews correlation coefficient.

RESULTS

Our final dataset consisted of 2 007 485 patient visits (36.45% female, mean age of 45), 8198 (0.4%) of which resulted in mortality. Our model achieved AUC and sensitivity-specificity gap of 0.86 (95% CI 0.85 to 0.87), 0.44 for children and 0.85 (95% CI 0.85 to 0.85), 0.44 for adults. The all ages model characteristics indicate it generalised, with an AUC and gap of 0.85 (95% CI 0.85 to 0.85), 0.45. Excluding fall injuries weakened the child model (AUC 0.85, 95% CI 0.84 to 0.86) but strengthened adult (AUC 0.87, 95% CI 0.87 to 0.87) and all ages (AUC 0.86, 95% CI 0.86 to 0.86) models.

CONCLUSIONS

Our machine learning model demonstrates similar performance to contemporary machine learning models without requiring restrictive criteria or extensive medical expertise. These results suggest that machine learning models for trauma outcome prediction can generalise to patients with trauma across the USA and may be able to provide decision support to medical providers in any healthcare setting.