Generalizing treatment effects with incomplete covariates: Identifying assumptions and multiple imputation algorithms.

We focus on the problem of generalizing a causal effect estimated on a randomized controlled trial (RCT) to a target population described by a set of covariates from observational data. Available methods such as inverse propensity sampling weighting are not designed to handle missing values, which are however common in both data sources. In addition to coupling the assumptions for causal effect identifiability and for the missing values mechanism and to defining appropriate estimation strategies, one difficulty is to consider the specific structure of the data with two sources and treatment and outcome only available in the RCT. We propose three multiple imputation strategies to handle missing values when generalizing treatment effects, each handling the multisource structure of the problem differently (separate imputation, joint imputation with fixed effect, joint imputation ignoring source information). As an alternative to multiple imputation, we also propose a direct estimation approach that treats incomplete covariates as semidiscrete variables. The multiple imputation strategies and the latter alternative rely on different sets of assumptions concerning the impact of missing values on identifiability. We discuss these assumptions and assess the methods through an extensive simulation study. This work is motivated by the analysis of a large registry of over 20,000 major trauma patients and an RCT studying the effect of tranexamic acid administration on mortality in major trauma patients admitted to intensive care units. The analysis illustrates how the missing values handling can impact the conclusion about the effect generalized from the RCT to the target population.

Association between in-ICU red blood cells transfusion and 1-year mortality in ICU survivors.

BACKGROUND: Impact of in-ICU transfusion on long-term outcomes remains unknown. The purpose of this study was to assess in critical-care survivors the association between in-ICU red blood cells transfusion and 1-year mortality. METHODS: FROG-ICU, a multicenter European study enrolling all-comers critical care patients was analyzed (n = 1551). Association between red blood cells transfusion administered in intensive care unit and 1-year mortality in critical care survivors was analyzed using an augmented inverse probability of treatment weighting-augmented inverse probability of censoring weighting method to control confounders. RESULTS: Among the 1551 ICU-survivors, 42% received at least one unit of red blood cells while in intensive care unit. Patients in the transfusion group had greater severity scores than those in the no-transfusion group. According to unweighted analysis, 1-year post-critical care mortality was greater in the transfusion group compared to the no-transfusion group (hazard ratio (HR) 1.78, 95% CI 1.45-2.16). Weighted analyses including 40 confounders, showed that transfusion remained associated with a higher risk of long-term mortality (HR 1.21, 95% CI 1.06-1.46). CONCLUSIONS: Our results suggest a high incidence of in-ICU RBC transfusion and that in-ICU transfusion is associated with a higher 1-year mortality among in-ICU survivors. Trial registration ( NCT01367093 ; Registered 6 June 2011).

Regulatory T Cells in Melanoma Revisited by a Computational Clustering of FOXP3+ T Cell Subpopulations.

CD4(+) T cells that express the transcription factor FOXP3 (FOXP3(+) T cells) are commonly regarded as immunosuppressive regulatory T cells (Tregs). FOXP3(+) T cells are reported to be increased in tumor-bearing patients or animals and are considered to suppress antitumor immunity, but the evidence is often contradictory. In addition, accumulating evidence indicates that FOXP3 is induced by antigenic stimulation and that some non-Treg FOXP3(+) T cells, especially memory-phenotype FOXP3(low) cells, produce proinflammatory cytokines. Accordingly, the subclassification of FOXP3(+) T cells is fundamental for revealing the significance of FOXP3(+) T cells in tumor immunity, but the arbitrariness and complexity of manual gating have complicated the issue. In this article, we report a computational method to automatically identify and classify FOXP3(+) T cells into subsets using clustering algorithms. By analyzing flow cytometric data of melanoma patients, the proposed method showed that the FOXP3(+) subpopulation that had relatively high FOXP3, CD45RO, and CD25 expressions was increased in melanoma patients, whereas manual gating did not produce significant results on the FOXP3(+) subpopulations. Interestingly, the computationally identified FOXP3(+) subpopulation included not only classical FOXP3(high) Tregs, but also memory-phenotype FOXP3(low) cells by manual gating. Furthermore, the proposed method successfully analyzed an independent data set, showing that the same FOXP3(+) subpopulation was increased in melanoma patients, validating the method. Collectively, the proposed method successfully captured an important feature of melanoma without relying on the existing criteria of FOXP3(+) T cells, revealing a hidden association between the T cell profile and melanoma, and providing new insights into FOXP3(+) T cells and Tregs.

To be or not to be, when synthetic data meet clinical pharmacology: A focused study on pharmacogenetics.

The use of synthetic data in pharmacology research has gained significant attention due to its potential to address privacy concerns and promote open science. In this study, we implemented and compared three synthetic data generation methods, CT-GAN, TVAE, and a simplified implementation of Avatar, for a previously published pharmacogenetic dataset of 253 patients with one measurement per patient (non-longitudinal). The aim of this study was to evaluate the performance of these methods in terms of data utility and privacy trade off. Our results showed that CT-GAN and Avatar used with k = 10 (number of patients used to create the local model of generation) had the best overall performance in terms of data utility and privacy preservation. However, the TVAE method showed a relatively lower level of performance in these aspects. In terms of Hazard ratio estimation, Avatar with k = 10 produced HR estimates closest to the original data, whereas CT-GAN slightly underestimated the HR and TVAE showed the most significant deviation from the original HR. We also investigated the effect of applying the algorithms multiple times to improve results stability in terms of HR estimation. Our findings suggested that this approach could be beneficial, especially in the case of small datasets, to achieve more reliable and robust results. In conclusion, our study provides valuable insights into the performance of CT-GAN, TVAE, and Avatar methods for synthetic data generation in pharmacogenetic research. The application to other type of data and analyses (data driven) used in pharmacology should be further investigated.

Benchmarking missing-values approaches for predictive models on health databases.

BACKGROUND: As databases grow larger, it becomes harder to fully control their collection, and they frequently come with missing values. These large databases are well suited to train machine learning models, e.g., for forecasting or to extract biomarkers in biomedical settings. Such predictive approaches can use discriminative-rather than generative-modeling and thus open the door to new missing-values strategies. Yet existing empirical evaluations of strategies to handle missing values have focused on inferential statistics. RESULTS: Here we conduct a systematic benchmark of missing-values strategies in predictive models with a focus on large health databases: 4 electronic health record datasets, 1 population brain imaging database, 1 health survey, and 2 intensive care surveys. Using gradient-boosted trees, we compare native support for missing values with simple and state-of-the-art imputation prior to learning. We investigate prediction accuracy and computational time. For prediction after imputation, we find that adding an indicator to express which values have been imputed is important, suggesting that the data are missing not at random. Elaborate missing-values imputation can improve prediction compared to simple strategies but requires longer computational time on large data. Learning trees that model missing values-with missing incorporated attribute-leads to robust, fast, and well-performing predictive modeling. CONCLUSIONS: Native support for missing values in supervised machine learning predicts better than state-of-the-art imputation with much less computational cost. When using imputation, it is important to add indicator columns expressing which values have been imputed.

Association of Early Norepinephrine Administration With 24-Hour Mortality Among Patients With Blunt Trauma and Hemorrhagic Shock.

Importance: Hemorrhagic shock is a common cause of preventable death after injury. Vasopressor administration for patients with blunt trauma and hemorrhagic shock is often discouraged. Objective: To evaluate the association of early norepinephrine administration with 24-hour mortality among patients with blunt trauma and hemorrhagic shock. Design, Setting, and Participants: This retrospective, multicenter, observational cohort study used data from 3 registries in the US and France on all consecutive patients with blunt trauma from January 1, 2013, to December 31, 2018. Patients were alive on admission with hemorrhagic shock, defined by prehospital or admission systolic blood pressure less than 100 mm Hg and evidence of hemorrhage (ie, prehospital or resuscitation room transfusion of packed red blood cells, receipt of emergency treatment for hemorrhage control, transfusion of >10 units of packed red blood cells in the first 24 hours, or death from hemorrhage). Blunt trauma was defined as any exposure to nonpenetrating kinetic energy, collision, or deceleration. Statistical analysis was performed from January 15, 2021, to February 22, 2022. Exposure: Continuous administration of norepinephrine in the prehospital environment or resuscitation room prior to hemorrhage control, according to European guidelines. Main Outcomes and Measures: The primary outcome was 24-hour mortality, and the secondary outcome was in-hospital mortality. The average treatment effect (ATE) of early norepinephrine administration on 24-hour mortality was estimated according to the Rubin causal model. Inverse propensity score weighting and the doubly robust approach with 5 distinct analytical strategies were used to determine the ATE. Results: A total of 52 568 patients were screened for inclusion, and 2164 patients (1508 men [70%]; mean [SD] age, 46 [19] years; median Injury Severity Score, 29 [IQR, 17-36]) presented with acute hemorrhage and were included. A total of 1497 patients (69.1%) required emergency hemorrhage control, 128 (5.9%) received a prehospital transfusion of packed red blood cells, and 543 (25.0%) received a massive transfusion. Norepinephrine was administered to 1498 patients (69.2%). The 24-hour mortality rate was 17.8% (385 of 2164), and the in-hospital mortality rate was 35.6% (770 of 2164). None of the 5 analytical strategies suggested any statistically significant association between norepinephrine administration and 24-hour mortality, with ATEs ranging from -4.6 (95% CI, -11.9 to 2.7) to 2.1 (95% CI, -2.1 to 6.3), or between norepinephrine administration and in-hospital mortality, with ATEs ranging from -1.3 (95% CI, -9.5 to 6.9) to 5.3 (95% CI, -2.1 to 12.8). Conclusions and Relevance: The findings of this study suggest that early norepinephrine infusion was not associated with 24-hour or in-hospital mortality among patients with blunt trauma and hemorrhagic shock. Randomized clinical trials that study the effect of early norepinephrine administration among patients with trauma and hypotension are warranted to further assess whether norepinephrine is safe for patients with hemorrhagic shock.

Measuring multivariate association and beyond.

Simple correlation coefficients between two variables have been generalized to measure association between two matrices in many ways. Coefficients such as the RV coefficient, the distance covariance (dCov) coefficient and kernel based coefficients are being used by different research communities. Scientists use these coefficients to test whether two random vectors are linked. Once it has been ascertained that there is such association through testing, then a next step, often ignored, is to explore and uncover the association's underlying patterns. This article provides a survey of various measures of dependence between random vectors and tests of independence and emphasizes the connections and differences between the various approaches. After providing definitions of the coefficients and associated tests, we present the recent improvements that enhance their statistical properties and ease of interpretation. We summarize multi-table approaches and provide scenarii where the indices can provide useful summaries of heterogeneous multi-block data. We illustrate these different strategies on several examples of real data and suggest directions for future research.