Unmeasured Confounding in Observational Studies with Multiple Treatment Arms

Abstract

Comparing emergency department mortality across different levels of trauma care (nontrauma centers, level I and II centers) is important in evaluating regionalized care. Patient population characteristics differ across different levels of trauma care and it is essential to adjust for baseline covariates to make valid comparisons. Propensity score matching has been established as a more robust method to infer causal relationship in observational studies than conventional regression adjustment. We designed and implemented a three group matching methodology. First, we conducted optimal pair matching between the treatment group (nontrauma centers) and the first control group (level I trauma centers); second, we conducted optimal pair matching between the nontrauma centers and the second control group (level II trauma centers); the final step was to link the two sets of matched pairs by the common treatment subjects to form matched triplets. We then implemented a sensitivity analysis with three treatment arms, Lu's imputation based method, to assess the impact due to potential unmeasured confounding. The results showed that if the most severe adult trauma patients treated in nontrauma centers were to be treated in level I or II trauma centers, the odds of emergency department death would be reduced dramatically (odds ratio [OR] = 0.61 and 0.49). There was no difference between being treated in level I or II trauma centers. The sensitivity analysis revealed that the positive association between being treated at level I or II trauma centers and the reduced odds of mortality would remain present even in the presence of strong unmeasured confounding.

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

“…Some facilities, like level 1 trauma centres, will be better equipped than others to handle certain types of patients, and that reality is not captured in this study. [28][29][30] Instead, we based our ML on patient demographics and injury characteristics so prehospital emergency medical services could use the prediction to guide patient with trauma field triage. Finally, the deidentified nature of the data used means our model can only analyse the outcomes of individual visits rather than the patients themselves.…”

Model Assisted Sensitivity Analyses for Hidden Bias with Binary Outcomes

“…Shi et al. () compares two different types of trauma center with the non‐trauma center through matching. Each matched set is a triplet with one patient from each type of hospitals.…”

“…Our approach can be easily extended to the matching design with multiple exposure levels, where there is little discussion on SA. Shi et al (2016) compares two different types of trauma center with the nontrauma center through matching. Each matched set is a triplet with one patient from each type of hospitals.…”

“…We considered trauma patients, aged 18–64, characterized by a severe trauma (injury severity score greater or equal to 25). Detailed description of the data can be found elsewhere (Shi et al., ).…”

Flexible template matching for observational study design

“…Other covariates include sex, age, comorbidity of chronic conditions, multiple injuries, median household income by zip code, expected primary payer, and urban‐rural designation for patient's county of residence. More detailed description of this dataset is provided in the work by Vickers et al, 18 Shi et al, 19 and Nattino et al 12 Out of the 21 855 patients included in the dataset, 16 541 (75.7%) patients were admitted to TC, and 5314 (24.3%) patients were admitted to NTC.…”