Treatment-Outcome Complex and Analysis of Observational Data

• Activity Number: 421
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Lev S. Sverdlov*+
• Companies: Independent Researcher
• Keywords: Treatment - Outcome Complex ; Observational Data ; Longitudinal ; Micro-experiment ; Intended and Unintended Effects ; Inferences

Abstract

The goal is to create a logical framework for the analysis of intended and uninintended effects of treatment using the concept of the treatment-outcome complex (TOC) for the analysis of longitudinal observational data. This approach assumes that at any moment during the treatment of chronic conditions we can in fact assess neither the severity of the disorder, nor the outcome of treatment. Rather we observe the momentary functional state of the patient under the influence of therapy and under a condition described by a set of observed and unobserved covariates as they stand at this moment (TOC). The process of treatment then can be described as a sequence of TOCs. The change in the outcome component of each sequential TOC depends on the change in treatment, i.e. each sequential pair of adjacent TOCs within an individual patient record can be interpreted as a micro-experiment. Various temporal patterns of outcomes can be analyzed using relevant aggregations of the TOCs with a possibility of measuring the effects of history and maturation. The approach is designed for assessing treatment effects in changing environments. The opportunities and limitations of the approach are