Stephen Privitera

Stephen Privitera

Auto-generated profile, based on publicly available information

Stephen Privitera

Global Commercialization, Bayer AG, Berlin, Germany

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 December 2023 in Pharmaceutical Statistics
Verified
Journal Article
Open access
SPStephen Privitera
Stephen Privitera
HSHooman Sedghamiz
Hooman Sedghamiz
AHAlexander Hartenstein
Alexander Hartenstein
TVTatsiana Vaitsiakhovich
Tatsiana Vaitsiakhovich
FKFrank Kleinjung
Frank Kleinjung
Matching reduces confounding bias in comparing the outcomes of nonrandomized patient populations by removing systematic differences between them. Under very basic assumptions, propensity score (PS) matching can be shown to eliminate bias entirely in estimating the average treatment effect on the treated. In practice, misspecification of the PS model leads to deviations from theory and matching quality is ultimately judged by the observed post‐matching balance in baseline covariates. Since covariate balance is the ultimate arbiter of successful matching, we argue for an approach to matching in which the success criterion is explicitly specified and describe an evolutionary algorithm to directly optimize an arbitrary metric of covariate balance. We demonstrate the performance of the proposed method using a simulated dataset of 275,000 patients and 10 matching covariates. We further apply the method to match 250 patients from a recently completed clinical trial to a pool of more than 160,000 patients identified from electronic health records on 101 covariates. In all cases, we find that the proposed method outperforms PS matching as measured by the specified balance criterion. We additionally find that the evolutionary approach can perform comparably to another popular direct optimization technique based on linear integer programming, while having the additional advantage of supporting arbitrary balance metrics. We demonstrate how the chosen balance metric impacts the statistical properties of the resulting matched populations, emphasizing the potential impact of using nonlinear balance functions in constructing an external control arm. We release our implementation of the considered algorithms in Python.

16 May 2025 in Cardiology and Therapy
Verified
Journal Article
Open access
PDF file
TVTatsiana Vaitsiakhovich
Tatsiana Vaitsiakhovich
AHAlexander Hartenstein
Alexander Hartenstein
SPStephen Privitera
Stephen Privitera
MPManesh R. Patel
Manesh R. Patel
JPJonathan P. Piccini
Jonathan P. Piccini
FKFrank Kleinjung
Frank Kleinjung
Introduction The aim of this study was to assess the applicability of an external control arm (ECA) approach in the clinical development of the oral factor XIa inhibitor asundexian for stroke prevention in patients with atrial fibrillation (AF), using prospectively collected data from the phase 2 PACIFIC-AF trial (NCT04218266) and real-world individual-level data from patients with AF treated with apixaban in the Optum® de-identified Electronic Health Record data set (Optum® EHR) 2013–2019. Methods To build ECAs, real-world patients meeting trial eligibility criteria were matched to patients enrolled in PACIFIC-AF. The primary outcome was the composite of International Society on Thrombosis and Haemostasis–defined major bleeding or clinically relevant non-major bleeding. Event rates were compared between PACIFIC-AF and ECAs at 85 days of trial duration and projected up to 2 years. Results Overall, 160,153 real-world patients met PACIFIC-AF eligibility criteria and were matched to patients from the PACIFIC-AF apixaban arm on 101 variables, with matching ratios of 1:10, 1:5, and 1:1. At day 85, the number of events for the primary outcome was 92 (3.68%) in the 1:10 ECA (2500 patients) and 6 (2.40%) in the PACIFIC-AF apixaban arm (250 patients), with incidence rates of 16.67 (90% confidence interval [CI] 13.92–19.63) and 11.10 (90% CI 4.83–19.45) per 100 person-years, respectively. Conclusions ECAs matching the PACIFIC-AF apixaban arm could be built from EHRs with concordant event rates for key trial endpoints. The ECA approach enabled the determination of event rates for treatment duration up to 2 years, thereby informing the asundexian pivotal phase 3 trial in AF.