Subgroup identification for differential treatment effect : model-based recursive partitioning approach.

Cardinal Scholar

Show simple item record

dc.contributor.advisor Begum, Munni, 1970-
dc.contributor.author Parh, Md Yasin Ali
dc.date.accessioned 2020-10-27T13:06:36Z
dc.date.available 2020-10-27T13:06:36Z
dc.date.issued 2020-07-18
dc.identifier.uri http://cardinalscholar.bsu.edu/handle/123456789/202442
dc.description Access to thesis permanently restricted to Ball State community only. en_US
dc.description.abstract The goal of this study is twofold: i) identification of risk factors associated with three chronic diseases and with mortality under three scenarios, and (ii) identification of subgroups with differential treatment effects. Multivariate analysis is performed to identify the risk factors associated with chronic diseases, hypertension, diabetes, and dyslipidemia. Both multivariate parametric and semi-parametric regression models are applied to identify risk factors for time to death from all-cause mortality, from cardiovascular diseases (CVD) and from cancer. For subgroup identification, we applied a model-based recursive partitioning approach. This method fits a local model in each subgroup of the population rather than fitting one global model for the whole population. The method starts with a model for the overall effect of treatment and checks whether the overall impact of the treatment is equally applicable for all individuals under the study based on parameter instability of M fluctuation test over a set of partitioning variables. The procedure produces a segmented model with a differential effect of treatment corresponding to each subgroup. The subgroups are linked to predictive factors learned by the recursive partitioning approach. The methods are applied to the data from the Ball State Adult Fitness Program Longitudinal Lifestyle Study, where we considered the level of cardiorespiratory fitness (CRF) as a treatment variable. The overall results indicate that CRF is inversely associated with chronic diseases and mortality under three scenarios. The predictive factors that are selected for both chronic diseases and mortality scenarios in subgroup analysis are related to diseases and mortality scenarios. The subgroup-specific results of chronic diseases indicate that for each subgroup, the chance of chronic diseases increases with low CRF. The subgroup-specific results of all-cause mortality demonstrate that the risk of death for younger (age <= 49 years) is higher if their CRF is low. A similar scenario is observed for the subgroupspecific result of mortality due to CVD. en_US
dc.description.sponsorship Department of Mathematical Sciences
dc.subject.lcsh Hypertension -- Risk factors -- Mathematical models
dc.subject.lcsh Diabetes -- Risk factors -- Mathematical models
dc.subject.lcsh Hyperlipidemia -- Risk factors -- Mathematical models
dc.subject.lcsh Hypertension -- Mortality -- Mathematical models
dc.subject.lcsh Diabetes -- Mortality -- Mathematical models
dc.subject.lcsh Hyperlipidemia -- Mortality -- Mathematical models
dc.subject.lcsh Cardiopulmonary fitness
dc.title Subgroup identification for differential treatment effect : model-based recursive partitioning approach. en_US
dc.description.degree Thesis (M.S.) en_US


Files in this item

This item appears in the following Collection(s)

  • Master's Theses [5510]
    Master's theses submitted to the Graduate School by Ball State University master's degree candidates in partial fulfillment of degree requirements.

Show simple item record

Search Cardinal Scholar


Browse

My Account