HOME /

Presentations

Paradoxical collider effect in the analysis of non-communicable disease epidemiological data: a reproducible illustration and web application, at Oviedo, Spain, Friday, September 6, 2019:
xxvii_see19_LUQUE_colider.pdf
Classical epidemiology has focused on the control of confounding, but it is only recently that epidemiologists have started to focus on the bias produced by colliders. A collider for a certain pair of variables (e.g. an outcome Y and an exposure A) is a third variable (C) that is caused by both. In a directed acyclic graph (DAG), a collider is the variable in the middle of an inverted fork (i.e. the variable C in A → C ← Y). Controlling for, or conditioning an analysis on a collider (i.e. through stratification or regression) can introduce a spurious association between its causes. This... Read more about Paradoxical collider effect in the analysis of non-communicable disease epidemiological data: a reproducible illustration and web application
Ensemble Learning Targeted Maximum Likelihood Estimation for Stata Users, at Pompeu Fabra University, Barcelona, Spain (Spanish Stata Users Meeting, 2018), Wednesday, October 24, 2018:
malf_eltmle_sum_bcn_2018.pdf

eltmle is a Stata program implementing the targeted maximum likelihood estimation (TMLE) for the ATE for a binary or continuous outcome and binary treatment. eltmle includes the use of a super-learner called from the SuperLearner package v.2.0-21 (Polley E., et al. 2011). Modern Epidemiology has been able to identify significant limitations of classic epidemiological methods, like outcome regression analysis, when estimating causal quantities such as the average treatment effect (ATE) for observational data. For...

Read more about Ensemble Learning Targeted Maximum Likelihood Estimation for Stata Users
Cross-validated Area Under the ROC curve for Stata users: cvauroc, at Pompeu Fabra University, Barcelona, Spain (Spanish Stata Users Meeting, 2018), Wednesday, October 24, 2018:
ma-luque-stata-ugm-bcn-auroc-18.pdf

Receiver operating characteristic (ROC) analysis is used for comparing predictive models, both in model selection and model evaluation. This method is often applied in clinical medicine and social science to assess the trade-off between model sensitivity and specificity. After fitting a binary logistic regression model with a set of independent variables, the predictive performance of this set of variables - as assessed by the area under the curve (AUC) from a ROC curve - must be estimated for a sample (the 'test' sample) that is independent of the sample used...

Read more about Cross-validated Area Under the ROC curve for Stata users: cvauroc
Pattern of comorbidities among Colorectal Cancer Patients and impact on treatment and short-term survival, at Copenhange, Denmark (European Network of Cancer Registries: Conference), Friday, September 28, 2018:
maluque_encr18_comcor.pdf

 

Background: Colorectal cancer (CRC) is the most frequently diagnosed cancer in Spain in both sexes with 41,441 new cases in 2015. There is little evidence regarding the pattern and impact of comorbidities on time from cancer diagnosis to surgical treatment and short-term mortality among CRC patients in Spain.

...

Read more about Pattern of comorbidities among Colorectal Cancer Patients and impact on treatment and short-term survival
Clinical Epidemiology in the Era of the Big Data Revolution: New Opportunities , at NDMORS, Medical Science Division, Oxford University, Oxford, UK, Thursday, November 2, 2017:
oxfordtalk.pdf
Science is moving towards a new data-driven paradigm. The big data revolution offers new opportunities for the development of Patient Centered Epidemiologic Methods (PCEM) under the Comparatives Effectiveness Research (CER) framework for clinical and applied epidemiologists and statisticians.
IMPACT EVALUATION IN PUBLIC HEALTH: A COUNTERFACTUAL FRAMEWORK, at Baltimore, US, Wednesday, July 23, 2014:
jhsph-ie-2014_-v10.pdf

Talk given in the Department of International Health at the Johns Hopkins School of Public Health

Summary:

When a random clinical trial is not feasible, the evaluation of the effectiveness of a health intervention should not be prevented. Providing that high quality administrative data are available, we should plan an evaluation using the assumptions of the counterfactual framework (quasi-experimentation or causal inference).

Pages