BMC Medical Research Methodology最新文献

{"title":"Is there a limit to limitations? A proposal for structured limitations reporting in scientific manuscripts.","authors":"Christos G Tsagkaris, Irem Sevik","doi":"10.1186/s12874-026-02992-0","DOIUrl":"10.1186/s12874-026-02992-0","url":null,"abstract":"<p><p>Reporting studies limitations is essential for transparent interpretation. While reporting guidelines such as CONSORT, STROBE, PRISMA mandate disclosure of limitations, no widely adopted framework standardizes how these are reported in a structured and comparable format. In practice, limitations sections often extend over several paragraphs reiterating predictable design-related constraints. This limits comparability, contributes to inefficiency and may overshadow genuine study-specific considerations. We propose a semi-structured checklist combining standardized limitation domains with a concise narrative component. This approach aims to ensure minimum systematic disclosure while preserving critical interpretation. Structured reporting may improve clarity, comparability, and enable meta-research on limitations across studies. Consensus-based refinement and pilot implementation are warranted to evaluate feasibility and impact.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":"26 1","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13523189/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148849832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A simulation-based comparison of Boruta, LASSO, and Elastic Net for variable selection in logistic regression, with an ovarian cancer miRNA application.","authors":"Reza Arabi Belaghi, Hulya Yurekli, Farzaneh Hamidi, Neda Gilani","doi":"10.1186/s12874-026-02939-5","DOIUrl":"10.1186/s12874-026-02939-5","url":null,"abstract":"<p><strong>Background: </strong>Variable selection is a central challenge in logistic regression, particularly in high-dimensional biomedical applications where correlated predictors and limited sample sizes complicate reliable identification of relevant variables. This study aims to systematically compare three widely used variable selection approaches - Boruta, LASSO, and Elastic Net - under a range of data-generating conditions and to illustrate their performance using an ovarian cancer miRNA dataset.</p><p><strong>Methods: </strong>We conducted a simulation study across 36 logistic regression scenarios varying in sample size, dimensionality, predictor correlation, and effect magnitude. Performance was evaluated using true-positive and false-positive selection rates. In addition, all three methods were applied to a real-world serum miRNA expression dataset, and discriminative performance was assessed using the area under the receiver operating characteristic curve (AUC).</p><p><strong>Results: </strong>Boruta, Elastic Net, and LASSO exhibited distinct variable selection behaviors across simulation scenarios. Boruta maintained strong true-positive recovery while controlling false positives in most settings, particularly when predictors were highly correlated. Elastic Net consistently achieved high sensitivity but produced comparatively large false-positive rates. LASSO showed the most conservative behavior, recovering fewer true predictors while maintaining low false-positive rates across nearly all scenarios. In the ovarian cancer miRNA application, all three methods achieved similarly strong test-set AUC performance, despite marked differences in the size of the selected biomarker panels.</p><p><strong>Conclusions: </strong>The results demonstrate clear trade-offs among the three methods. Boruta offers a favorable balance between sensitivity and specificity in highly correlated settings. Elastic Net prioritizes sensitivity at the cost of increased false discoveries, whereas LASSO provides stricter false-positive control with reduced sensitivity. These findings offer practical guidance for selecting variable selection methods in logistic regression, particularly for high-dimensional biomedical applications.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":"26 1","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13531732/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148863657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reproducibility, transparency, and open science: a cultural evolution is needed to move from ideals we celebrate to implementation.","authors":"Livia Puljak, Vladimir Ruf, Josip Šimić, Melissa K Sharp, Ivan Buljan","doi":"10.1186/s12874-026-02964-4","DOIUrl":"10.1186/s12874-026-02964-4","url":null,"abstract":"","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":"26 1","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13499322/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148788180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing patient-reported sociodemographic measures in health care - a national example of the development and evaluation of a practical standardized instrument for collecting patient data in studies.","authors":"Sigrid Boczor, Heike Hansen, Thomas Kloppe, Claudia Mews, Cathleen Muche-Borowski, Anja Rakebrandt, Ingmar Schäfer, Nadine Janis Pohontsch, Vivien Böttcher, Thorben W Fründt, Anna Sophie Hoffmann, Jan Dietrich Philipp Köster, Martin Scherer","doi":"10.1186/s12874-026-02977-z","DOIUrl":"https://doi.org/10.1186/s12874-026-02977-z","url":null,"abstract":"<p><strong>Background: </strong>The development of patient-oriented health care is continuous subject to old and new challenges. To better identify patient needs, studies are necessary and standardized data collection is essential to describe, analyze and compare data. Associated demographic variables remain a topic of discussion in health services research. Common approaches either differ in relevant sociodemographic aspects, or their standardization is too detailed to be feasibly used in studies with a clinical or health services research focus. Our aim was to develop and evaluate a comprehensive and practical instrument for collecting sociodemographic patient data in studies.</p><p><strong>Methods: </strong>A variant (interdisciplinary expert panel) of the nominal group technique was used as a consensus method, in addition, a focus group was held and a 2-phase pretest according to Prüfer and Rexroth (method: \"Think Aloud\") was used for evaluation. In the first phase we conducted 6 interviews with potential patients. The main phase was planned to enroll at least 200 patients to test the instrument with a quantitative sample as recommended for testing patient-reported outcome measures. It took place in eight departments of the University Medical Center Hamburg-Eppendorf. Exploratory analyses were conducted using SPSS. The interdisciplinary expert panel revised the instrument.</p><p><strong>Results: </strong>In the main phase, 290 (51%) of 574 approached patients participated [median (minimum; maximum) age: 40 (19; 86) years; 55% female]. Of these respondents 66%/ 31%/ 3% considered the length of the questionnaire to be appropriate/ too long/ too short; 96% found it understandable. Seven questions on living and income situation (commented at least three times) showed a need for clarification and modification. Non-responders optionally provided reasons for decline of which 7 clusters could be formed (predominantly: time). Final standardized variables: biological sex; year/country of birth; marital status/biological children; living/household type/persons/age; highest school/ qualification/current profession; household income/specialties/persons; country of birth mother/father; nationalities/residence status/native languages/German skills.</p><p><strong>Conclusions: </strong>Patient involvement is helpful in the development of patient survey instruments. Our instrument is based on the patients' perspective and understanding of these measurements. The aim is to further strengthen patient-centered health care. The revised instrument will be used in studies. The expert group will continue to identify and implement socio-demographic developments.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":"26 1","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13491617/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148788157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Validating methods for inferring co-occurring diseases: a flexible framework for simulating synthetic data.","authors":"Hannah Marchi, Sophie Schmiegel, Tamara Schamberger, Christiane Fuchs","doi":"10.1186/s12874-026-02966-2","DOIUrl":"https://doi.org/10.1186/s12874-026-02966-2","url":null,"abstract":"<p><strong>Background: </strong>The validation of methods is an integral part of statistical research, defining conditions under which methods yield reliable results. Empirical validation requires a solid data basis to control and manage relevant characteristics like sample size, dimensionality, and underlying dependency structures. Real-world data often fails to meet these requirements, particularly in medical contexts where privacy regulations restrict availability. For this reason, synthetic data is an effective alternative for method validation. However, generating synthetic data is demanding when it must precisely mirror complex dependence structures while simultaneously controlling specific target characteristics.</p><p><strong>Methods: </strong>We address the medical context of co-occurring diseases, where symptoms may overlap or conflict. We propose a four-step framework to generate synthetic data for the simulation-based validation of statistical methods. The framework involves: (I) generating patient covariates; (II) connecting this information to predictors for single or joint disease occurrence; (III) transforming predictors into disease probabilities or scores; and (IV) converting these into disease occurrences. Each step offers several alternatives for modeling the overall dependence structure. We apply our framework to a case study of pain-causing diseases which share certain similarities in their clinical presentations, and which can occur either individually or jointly. By employing five combinations of methodological alternatives, we evaluate the approaches' ability to achieve target characteristics and demonstrate their specific strengths and weaknesses.</p><p><strong>Results: </strong>Matching the data-generating process with the estimation method allows for the successful recovery of input information, such as coefficients and correlations. Target properties like disease prevalence and associations are achieved to varying degrees depending on the methods used.</p><p><strong>Conclusions: </strong>While the proposed theory-driven framework is broadly applicable beyond the specific medical use case, it relies on careful, domain-informed parameter curation to generate meaningful synthetic datasets. Its flexible, adjustable input settings enable researchers to tailor data generation to their precise methodological requirements, providing a controlled basis for simulation-based validation without implying direct clinical inference.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":"26 1","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13488252/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148788168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Applications of causal and structural equation modeling in epidemiology: a systematic and critical review.","authors":"Scholastique Midokpè Merveille Essetcheou, Houétchénou Gislain Fortuné Dovonou, Souand Peace Gloria Tahi, Sèton Calmette Ariane Houetohossou, Valère Kolawolé Salako, Marcel Tadogbè Donou Hounsode, Romain Glèlè Kakaï","doi":"10.1186/s12874-026-02944-8","DOIUrl":"https://doi.org/10.1186/s12874-026-02944-8","url":null,"abstract":"<p><strong>Background: </strong>Structural equation modeling (SEM) and causal modeling (CM) are powerful statistical approaches for identifying complex interrelationships among variables. However, their application in epidemiology remains limited and under-documented, especially in infectious disease research, which requires integrated analytical frameworks for effective control.</p><p><strong>Methods: </strong>To examine how SEM and CM have been applied, their methodological characteristics, and reporting practices, a systematic and critical review was conducted following PRISMA guidelines. The search covered studies published between 1987 and 2025 across PubMed, Scopus, Web of Science, ScienceDirect, SpringerLink, Google Scholar, and the Directory of Open Access Journals. After rigorous screening, 458 articles were thoroughly evaluated.</p><p><strong>Results: </strong>Most studies focused on neuropsychiatric (32.1%) and chronic (30.1%) conditions, with few addressing infectious diseases (24.0%), primarily malaria, tuberculosis, and HIV, particularly in low-income countries where context-specific evidence is urgently needed to inform targeted interventions. SEM studies predominantly used maximum likelihood estimation (57.7%) and large samples ([Formula: see text] observations in 85%) with CB-SEM remaining the dominant approach across all sample size categories. In contrast, CM studies showed substantial variability in sample sizes across approaches, ranging from fewer than 100 to over 200 observations (coefficient of variation [Formula: see text]), with no consistent sample size threshold across methods. Methodological reporting was often incomplete, notably regarding study design (17.4%), measurement validity (15.8%), and model fit criteria (5.2%), reducing transparency and reproducibility.</p><p><strong>Conclusion: </strong>Overall, broader application of SEM and CM to infectious diseases, combined with improved methodological transparency, could substantially strengthen causal inference and guide evidence-based disease control strategies. Moreover, integrating longitudinal study designs would further enhance the robustness and interpretability of causal findings.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":" ","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148547618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic review and meta synthesis of barriers and facilitators to Black women's participation in healthcare research.","authors":"S S Hall, G Clancy, S Patel, N Thorpe, B Majmudar, C Merriman, S Agyapong, J Usman, C Henshall","doi":"10.1186/s12874-026-02953-7","DOIUrl":"https://doi.org/10.1186/s12874-026-02953-7","url":null,"abstract":"<p><strong>Background: </strong>Black women are under-represented in clinical research, contributing to persistent health inequities and undermining the validity and generalisability of research findings. Understanding facilitators and barriers to their research engagement is essential to addressing these challenges. This study aimed to identify and synthesise evidence on barriers and facilitators which influence Black women's participation in healthcare research, to inform inclusive recruitment strategies and equitable research designs.</p><p><strong>Methods: </strong>This systematic review was conducted in accordance with PRISMA guidelines and registered with PROSPERO (CRD42024587308). Findings were narratively synthesised using thematic analysis and discussed iteratively through Patient and Public Involvement. Five databases were searched (CINAHL, PsycInfo, Embase, MIDIRS, MEDLINE) for studies published since 2010. Papers in any language were screened for inclusion. Patient and Public Involvement members informed review objectives and data interpretation. Primary research studies reporting on Black women's experiences, attitudes, and facilitators/barriers to participating in any type of healthcare research. Studies were included if ≥ 50% (to ensure primary representation) of the sample recruited were Black women.</p><p><strong>Results: </strong>Seventy-two studies were included. Most were from America and spanned research areas relating to oncology, Human Immunodeficiency Virus (HIV), dementia, maternity, and general health. Five overarching themes identified how Black women's participation in healthcare research is shaped by a complexity of factors. These were mistrust, interpersonal experiences, altruism, self-interest, and low health/research literacy and awareness. Patient and Public Involvement consultations concurred that the findings aligned with a \"bridge and barriers\" analogy, illustrating how various factors can facilitate or prevent Black women from engaging with research opportunities. This provided a clear framework to guide culturally sensitive implementation for researchers, healthcare professionals, and policymakers.</p><p><strong>Conclusions: </strong>Inclusive strategies for recruiting Black women into research must focus on co-constructing a bridge between communities and research systems. This requires structural and systemic change in research delivery environments and ongoing collaborations with all stakeholders including Black women, healthcare system providers, and research teams.</p><p><strong>Trial registration: </strong>PROSPERO (CRD42024587308).</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":" ","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148547665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust power and sample size calculations in quasi-likelihood models: methods and practice.","authors":"Shijie Yuan, Amy Cochran, Paul Rathouz","doi":"10.1186/s12874-026-02940-y","DOIUrl":"https://doi.org/10.1186/s12874-026-02940-y","url":null,"abstract":"<p><strong>Background: </strong>Accurate power and sample size calculations are essential in study planning, yet they are often difficult to carry out for quasi-likelihood (QL) models. Traditional power and sample size (PSS) approaches often rely on restrictive distributional assumptions, limiting their applicability when responses have non-standard distributions, variance functions are misspecified, or when covariates exhibit complex dependence structures.</p><p><strong>Methods: </strong>We examine whether two effect size measures-2 Standard Deviations in the Linear Predictor (2SLiP) and Pseudo-Partial [Formula: see text] (P2R2)-originally developed for Wald tests involving generalized linear models, are effective at power and sample size calculations in the QL framework. Through extensive simulations across diverse outcome types, link functions, and variance structures, we assess their performance under Wald tests and explore whether they remain useful for score tests. To illustrate practical utility, we apply these effect size measures to survey data on frontline health care workers to quantify the association between perceived personal protective equipment (PPE) adequacy and burnout risk during the COVID-19 pandemic, adjusting for covariates.</p><p><strong>Results: </strong>We show that the two generalized linear model (GLM)-based effect sizes are fundamentally moment-based objects, and therefore extend directly to QL models. Across all simulation settings, both measures remained accurate, with sample size and power estimates within 3% and 2% of the target, respectively. In the case study, the estimated effect sizes for perceived PPE adequacy (2SLiP = 0.096 and P2R2 = 0.020) correspond to small but meaningful associations with burnout risk after adjustment for demographic and occupational covariates; both yielded accurate recommendations for sample size.</p><p><strong>Conclusions: </strong>Both 2SLiP and P2R2 offer robust alternatives for power and sample size calculation in QL settings. By requiring minimal distributional assumptions, these measures enhance the flexibility and reliability of power and sample size calculations for realistic study designs commonly encountered in medical and public health research.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":" ","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148547755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bayesian semi-parametric inference for joint modelling of childhood growth and appetite phenotypes.","authors":"Andrea Cremaschi, Beatrice Franzolini, Maria De Iorio, Mary Chong, Jia Ying Toh, Navin Michael, Varsha Gupta, Fabian Yap, Yung Seng Lee, Johan Erikkson, Anna Fogel","doi":"10.1186/s12874-026-02950-w","DOIUrl":"https://doi.org/10.1186/s12874-026-02950-w","url":null,"abstract":"<p><strong>Purpose: </strong>Early eating behaviours have been associated with later weight outcomes, yet their longitudinal development and relationship with growth trajectories remain insufficiently understood. This study aims to investigate how appetite-related behaviours in early childhood co-evolve with growth patterns and contribute to obesity risk.</p><p><strong>Methods: </strong>We develop a Bayesian semi-parametric joint modelling framework to analyse repeated measures of growth indicators, such as body mass index, alongside questionnaire-based eating behaviour scores collected at multiple time points in children from the Singaporean GUSTO cohort (\"Growing Up in Singapore Towards Healthy Outcomes\"). The approach extends established models for ordinal questionnaire data to accommodate longitudinal observations and covariate effects, while growth trajectories are flexibly represented using spline-based regressions. Subject-specific demographic and clinical covariates are incorporated into both components of the model, allowing their effects on growth and eating behaviour to be assessed simultaneously. The two components are linked through a Bayesian nonparametric prior, specifically a Normalised Generalised Gamma Process, that enables data-driven identification of subgroups of children with similar developmental profiles.</p><p><strong>Results: </strong>The proposed framework captures the dynamic interplay between appetite phenotypes and growth trajectories over time, allowing the identification of clinically meaningful clusters characterised by distinct behavioural and growth patterns.</p><p><strong>Conclusion: </strong>The proposed integrated modelling strategy provides a nuanced understanding of how eating behaviours and growth co-develop in early life, offering new insights into mechanisms underlying childhood obesity risk and supporting the design of targeted early interventions.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":" ","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148535289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimating treatment effects from non-overlapping cohorts with application to antimicrobial resistance.","authors":"Avi Baraz, Daniel Nevo, Amos Cahan, Tal Brosh-Nissimov, Michal Chowers, Uri Obolski","doi":"10.1186/s12874-026-02947-5","DOIUrl":"https://doi.org/10.1186/s12874-026-02947-5","url":null,"abstract":"<p><strong>Background: </strong>The positivity assumption, also known as overlap, is one of the key assumptions for causal inference. However, in certain observational studies it does not hold. This work is motivated by such a scenario, where the goal was to compare the effect of ceftriaxone versus cefuroxime treatment on subsequent 3rd generation cephalosporin resistance within one year post-treatment in Gram-negative bacteria among hospitalized patients. Our data consisted of electronic medical records from 24,969 patients hospitalized between 2015-2025 in two hospitals with distinct policies: one prescribed cefuroxime, while the other prescribed ceftriaxone. These prescription policies created two non-overlapping patient cohorts, as each hospital treated patients with either cefuroxime or ceftriaxone. We developed and applied a statistical framework that leverages additional data, for which positivity holds, and clinically driven assumptions, to identify the average treatment effect among these non-overlapping cohorts, without removing any observations.</p><p><strong>Methods: </strong>Our framework is based on a logistic regression model for the conditional expected bacterial resistance outcome, and it can be applied to any generalized linear model. The model is identifiable under a clinically driven assumption that the effect modification of the hospital on cefazolin equals that of cefuroxime and ceftriaxone, when compared to penicillins. We rely on cefazolin and penicillins being prescribed in both hospitals, and on penicillins inducing negligible resistance to 3rd generation cephalosporins. We also implemented inverse probability of censoring weighting for censored outcomes, and utilized aminoglycoside resistance as a negative control outcome to assess sufficient confounding adjustment.</p><p><strong>Results: </strong>The estimated average treatment effect, on the risk difference scale, of ceftriaxone compared to cefuroxime was 5.83% (95% CI: -2.06%, 13.39%). The conditional odds ratio for ceftriaxone compared to cefuroxime, adjusted for confounders, was 1.31 (95% CI: 0.91, 1.89). No evidence of unmeasured confounding was found using the negative control outcome falsification test.</p><p><strong>Conclusions: </strong>We develop and apply a statistical framework applicable in settings of non-overlapping cohorts that violate the positivity assumption. Our framework allows identification of the average treatment effect thanks to parametric assumptions about the conditional expected outcome, additional data and clinically driven assumptions. The empirical findings, though not statistically significant, may point to a modest resistance increase with ceftriaxone compared to cefuroxime use; and suggest further research about the preferential use of cefuroxime is needed.</p>","PeriodicalId":9114,"journal":{"name":"BMC Medical Research Methodology","volume":" ","pages":""},"PeriodicalIF":3.7,"publicationDate":"2026-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148535303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}