THTugba Akkaya Hocagil

Tugba Akkaya Hocagil

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Tugba Akkaya Hocagil has a total of 60 co-authors. Here are the top co-authors by number of shared publications:

  • Louise M. Ryan

    School of Mathematical and Physical Sciences University of Technology Sydney Ultimo New South Wales Australia

    6 shared publications
  • Sandra W. Jacobson

    Department of Psychiatry and Behavioral Neurosciences Wayne State University Detroit Michigan USA

    6 shared publications
  • Joseph L. Jacobson

    Department of Psychiatry and Behavioral Neurosciences, School of Medicine Wayne State University Detroit Michigan USA

    5 shared publications
  • Richard J. Cook

    Department of Statistics and Actuarial Science University of Waterloo Waterloo Ontario Canada

    4 shared publications
  • Claire D. Coles

    Department of Psychiatry and Behavioral Sciences Emory University Atlanta, Georgia USA

    2 shared publications
  • Doğacan Sarısoy

    Department of Pediatrics, Division of Pediatric Rheumatology, Ankara University School of Medicine, Ankara, Turkey

    2 shared publications

Latest Publications

  • Open Access

    Two‐Stage Estimation for Causal Inference Involving Semi‐Continuous Exposures

    by Xiaoya Wang, Richard J. Cook, Yeying Zhu, Tugba Akkaya‐Hocagil, R. Colin Carter, Sandra W. Jacobson, Joseph L. Jacobson, Louise M. Ryan

    Methods for causal inference are well developed for binary and continuous exposures, but in many settings, the exposure has a substantial mass at zero—such exposures are called semi‐continuous. We propose a general causal framework for such semi‐continuous exposures, together with a two‐stage estimation strategy. A two‐part propensity structure is introduced for the semi‐continuous exposure, with one component for exposure status (exposed vs unexposed) and another for the exposure level among those exposed, and incorporates both into a marginal structural model that disentangles the effects of exposure status and dose. The two‐stage procedure sequentially targets the causal dose–response among exposed individuals and the causal effect of exposure status at a reference dose, allowing flexibility in the choice of propensity score methods in the second stage. We establish consistency and asymptotic normality for the resulting estimators, and characterize their limiting values under misspecification of the propensity score models. Simulation studies evaluate finite sample performance and robustness, and an application to a study of prenatal alcohol exposure and child cognition demonstrates how the proposed methods can be used to address a range of scientific questions about both exposure status and exposure intensity.

  • Open Access

    Beyond Age: Radiological and Inflammatory Factors Associated with Hospital Stay After VATS in Pediatric Complicated Pneumonia—A 10-Year Single-Center Experience

    by Ihsan Ozdemir, Emre Yasar Karbancioglu, Tugba Akkaya Hocagil, Gulnur Gollu, Ufuk Ates, Ergun Ergun, Ergin Ciftci, Halil Ozdemir, Gul Arga, Tanil Kendirli, et al.

    Background/Objectives: We aimed to describe the clinical, laboratory, and radiological characteristics of children with complicated pneumonia/empyema who underwent video-assisted thoracoscopic surgery (VATS) and to identify factors associated with length of hospital stay. Methods: We retrospectively reviewed 68 children who presented to the pediatric emergency department of a tertiary university hospital and underwent VATS between April 2015 and April 2025. Factors associated with length of stay were assessed using univariable and multivariable negative binomial regression. Results: Median age was 60 months (interquartile range 36–138), and 66% were male. On univariable analysis, C-reactive protein (CRP) was associated with length of stay (incidence rate ratio [IRR] = 1.03 per 100 mg/L; p = 0.034), whereas age was not (p = 0.542). In the multivariable model, CRP remained borderline associated (IRR = 1.03; 95% confidence interval [CI] 1.00–1.06; p = 0.049), whereas performance of computed tomography was not (p = 0.142). An atypical (gastrointestinal or pain-predominant) presentation occurred in 25%. Streptococcus pneumoniae was the most frequently detected bacterial pathogen, mainly by nasopharyngeal polymerase chain reaction. Conclusions: In this single-center, retrospective cohort, higher CRP at presentation, but not age, was modestly associated with a longer hospital stay; this small association needs confirmation in larger prospective studies, and most children (82%) were managed with VATS alone.

  • Cytomegalovirus Infections Following Pediatric Hematopoietic Stem Cell Transplantation: An 18‐Year Single‐Center Experience

    by Doğacan Sarısoy, Elif Ünal İnCe, Dilber Talia İlEri, Ergin Çiftçi, Zeynep Ceren Karahan, Hasan Fatih Çakmaklı, Gül Arga, Tuğba Akkaya Hocagil, Candan İsLamoğlu, Halil Özdemir, et al.

    Aim: The aim of this study was to evaluate the risk factors for CMV reactivation and the effects of CMV on HSCT outcomes in patients managed using a standard prophylaxis and preemptive treatment approach in settings where letermovir prophylaxis could not be used. Methods: Patients who underwent HSCT between January 2006 and December 2023 in our pediatric hematology department were included in this retrospective study. Results: Among 217 patients included, 96 (44.4%) underwent HSCT for malignant diseases. The donor was a matched‐sibling for 65.4%, and 86.1% had myeloablative conditioning regimen. All patients received acyclovir prophylaxis (1500 mg/m 2 /day) for the first 30 days. The prevalences of acute and chronic GVHD were 28.1% and 18%, respectively. In a median (IQR) follow‐up of 49 (26.5–93.3) months, 60 (27.6%) patients experienced a total of 92 CMV reactivations. Patient group that experienced CMV reactivations was characterized by higher rates of malignant diseases, higher rates of mismatched donors, lower rates of bone marrow as stem cell source, higher acute and chronic GVHD prevalences, and higher grades of acute GVHD. In the group that did not experience CMV reactivation, 2‐year overall survival rates were higher and treatment‐related mortality rates were lower. This association remained consistent when evaluated among non‐malignant patients. However, no significant difference was observed in the malignant diseases. Conclusion: After an effective prophylaxis, CMV DNA monitoring combined with preemptive treatment have an important role in CMV management. However, the survival rates among patients experiencing CMV reactivation remain lower. image


  • Open Access

    Clinical predictors of relapse and severe disease phenotype in children with non-systemic juvenile idiopathic arthritis

    by Doğacan Sarısoy, Fatma Aydın, Özen Taş, Onur Bahçeci, Betül Öksüz Aydın, Elif Erorhan, Tuğba Akkaya Hocagil, Zeynep Birsin Özçakar

    The aim of this study was to identify the predictors of relapse and severe disease in non-systemic juvenile idiopathic arthritis (JIA), a heterogeneous childhood disease. Patients with JIA were grouped based on relapse status, and those with ≥ 2 relapses requiring biologics were classified as severe disease phenotype. A total of 142 patients (63.4% female) were included in the study. Seventy-three patients (51.4%) experienced at least one relapse after achieving remission, who were significantly characterized by female gender, younger age at diagnosis, positive ANA test, and longer disease duration. Ankle, elbow, metacarpophalangeal (MCP), and temporomandibular joint (TMJ) involvement was more prevalent in patients who had experienced at least one relapse. Longer disease duration, higher number of joints involved at the time of diagnosis, and MCP involvement were found as independent risk factors for relapse. Twenty patients (14%) were grouped as having a severe disease phenotype, characterized by a younger age at the time of diagnosis, longer disease duration, and a higher number of joints involved throughout the disease course. Younger age at diagnosis, ankle involvement, and TMJ involvement were found to be independent risk factors for a severe disease phenotype. Conclusion : In our study, longer disease duration, MCP joint involvement, and a higher number of joints involved at the time of diagnosis were found to be associated with relapse in non-systemic JIA patients. On the other hand, younger age at diagnosis, ankle and TMJ involvement were associated with severe disease phenotype. What is Known:• JIA is a chronic joint disease characterized by relapses and remissions, and approximately half of patients experience at least one relapse after remission. What is New:• Multiple relapses and bDMARD requirement may define a severe disease course. Younger age, ankle, and TMJ involvement may predict severe disease course.• Relapses may occur with longer disease duration; not every relapse implies a severe disease course.


  • Bridging Statistical Rigor and Clinical Usability: The CORMeta App for Meta-Analysis of Correlated Outcomes

    by Tugba Akkaya Hocagil, Richard J Cook, Louise M Ryan

    Background: In clinical research, multiple outcomes are often measured within the same cohort, leading to statistical dependencies that violate assumptions of traditional meta-analytic methods. While advanced models can accommodate such correlations, they typically require programming expertise, limiting accessibility for many physician-researchers. Objective: We present a user-friendly, interactive Shiny web application designed to perform meta-analyses of correlated outcomes, with particular relevance for cohort-based clinical datasets. Methods: The application implements a modified multivariate meta-analytic framework that accounts for the correlation structure of outcomes within cohorts. Users can upload their data, define correlation matrices, and filter observations by any variable (e.g., age, domain, exposure) without writing code. The application provides graphical output (forest plots) along with estimates of overall effect size, heterogeneity (τ²), and p-values. Results: A demonstration dataset on prenatal alcohol exposure and neurodevelopmental outcomes is simulated to illustrate the application’s functionality. The application automatically generates correlation matrices where needed, adjusts for intra-cohort dependencies, and produces interpretable results suitable for clinical research reports. Conclusion: This open-access application bridges the gap between complex statistical modeling and clinical usability. It enables physicians to conduct robust meta-analyses of correlated outcomes with ease, supporting evidence-based practice and local research initiatives. The tool is particularly valuable in multi-domain or multi-cohort studies where outcome correlation is non-negligible.

  • Open Access

    Evaluation of the Risk of Malignant Arrhythmia Through Electrocardiography Parameters in Patients with Urinary Stone Disease

    by Hilal Hocagil, Abdullah Cüneyt Hocagil, Tuğba Akkaya Hocagil

    Introduction: With a lifetime frequency of 10-15% and a recurrence rate of up to 50% within 10 years, urinary stone disease is a major cause of comorbidity. Recently, it has become widely recognized that urinary system stone disease is not only limited to kidney stone development but is also linked to cardiovascular disease and various other illnesses. Malignant arrhythmias (sudden cardiac death and ventricular arrhythmias) are consequences of coronary artery disease. Despite numerous studies demonstrating that urinary system stone disease is a risk factor for coronary artery disease, no electrocardiographic parameters have been evaluated in this patient group This prospective observational study aims to assess ventricular repolarization parameters using electrocardiography, which are risk indicators for malignant arrhythmias in patients with urinary stones disease. Materials and Methods The study included patients diagnosed with urinary stone disease and healthy volunteers. All patients underwent 12-lead electrocardiography. The electrocardiographys were evaluated for QTc interval, QTc dispersion, T peak-end interval, and Tp-e/QTc, and compared with the control group. Results: When comparing the QTc interval, QTc dispersion, T peak-end interval, and Tp-e/QTc parameters between the patient and control groups, no statistically significant differences were found. Conclusion: Contrary to existing knowledge, this study found that malignant arrhythmias in patients with urinary stone disease were not significantly different from the normal population.

  • Open Access

    Benchmark dose profiles for bivariate exposures

    by Tugba Akkaya Hocagil, Louise M. Ryan, Richard J. Cook, Khue‐Dung Dang, R. Colin Carter, Gale A. Richardson, Nancy L. Day, Claire D. Coles, Heather Carmichael Olson, Sandra W. Jacobson, et al.

    While benchmark dose (BMD) methodology is well‐established for settings with a single exposure, these methods cannot easily handle multidimensional exposures with nonlinear effects. We propose a framework for BMD analysis to characterize the joint effect of a two‐dimensional exposure on a continuous outcome using a generalized additive model while adjusting for potential confounders via propensity scores. This leads to a dose–response surface which can be summarized in two dimensions by a contour plot in which combinations of exposures leading to the same expected effect are identified. In our motivating study of prenatal alcohol exposure, cognitive deficits in children are found to be associated with both the frequency of drinking as well as the amount of alcohol consumed on each drinking day during pregnancy. The general methodological framework is useful for a broad range of settings, including combinations of environmental stressors, such as chemical mixtures, and in explorations of the impact of dose rate rather than simply cumulative exposure on adverse outcomes.


  • Open Access

    Meta-analysis on studies with heterogeneous and partially observed covariates

    by Tugba Akkaya Hocagil, Hon Hwang, Joseph L. Jacobson, Sandra W. Jacobson, Louise M. Ryan

    Individual participant data meta-analysis is a commonly used alternative to the traditional aggregate data meta-analysis. It is popular because it avoids relying on published results and enables direct adjustment for relevant covariates. However, a practical challenge is that the studies being combined often vary in terms of the potential confounders that were measured. Furthermore, it will inevitably be the case that some individuals have missing values for some of those covariates. In this paper, we demonstrate how these challenges can be resolved using a propensity score approach, combined with multiple imputation, as a strategy to adjust for covariates in the context of individual participant data meta-analysis. To illustrate, we analyze data from the Bill and Melinda Gates Foundation–funded Healthy Birth, Growth, and Development Knowledge Integration project to investigate the relationship between physical growth rate in the first year of life and cognition measured later during childhood. We found that the overall effect of average growth velocity on cognitive outcome is slightly, but significantly, positive with an estimated effect size of 0.36 (95% CI 0.18, 0.55).


  • A computationally efficient sequential regression imputation algorithm for multilevel data

    by Tugba Akkaya Hocagil, Recai M. Yucel

    Due to the computational burden, especially in high-dimensional settings, sequential imputation may not be practical. In this paper, we adopt computationally advantageous methods by sampling the missing data from their perspective predictive distributions, which leads to significantly improved computation time in the class of variable-by-variable imputation algorithms. We assess the computational performance in a comprehensive simulation study. We then compare and contrast the performance of our algorithm with commonly used alternatives. The results show that our method has a significant advantage over the commonly used alternatives with respect to computational efficiency and inferential quality. Finally, we demonstrate our methods in a substantive problem aimed at investigating the effects of area-level behavioral, socioeconomic, and demographic characteristics on poor birth outcomes in New York State among singleton births.


  • Open Access

    Bayesian outcome selection modeling

    by Khue‐Dung Dang, Louise M. Ryan, Richard J. Cook, Tugba Akkaya Hocagil, Sandra W. Jacobson, Joseph L. Jacobson

    In psychiatric and social epidemiology studies, it is common to measure multiple different outcomes using a comprehensive battery of tests thought to be related to an underlying construct of interest. In the research that motivates our work, researchers wanted to assess the impact of in utero alcohol exposure on child cognition and neuropsychological development, which are evaluated using a range of different psychometric tests. Statistical analysis of the resulting multiple outcomes data can be challenging, because the outcomes measured on the same individual are not independent. Moreover, it is unclear, a priori, which outcomes are impacted by the exposure under study. While researchers will typically have some hypotheses about which outcomes are important, a framework is needed to help identify outcomes that are sensitive to the exposure and to quantify the associated treatment or exposure effects of interest. We propose such a framework using a modification of stochastic search variable selection, a popular Bayesian variable selection model and use it to quantify an overall effect of the exposure on the affected outcomes. The performance of the method is investigated empirically and an illustration is given through application using data from our motivating study.