Intensive longitudinal data are becoming increasingly common in substance use research and HIV/AIDS research. Substance use researchers employ technological innovations such as hand-held computers, cell phones and web-based assessment to collect data in which many observations are collected closely spaced at time. These data collection approaches have revolutionized the process of scientific research and knowledge discovery in the substance use area. Additionally, many HIV/AIDS studies have become mature longitudinal studies that have collected many waves of data and have amassed many observations over the course of years of data collection. Thus, mature longitudinal data with many waves may be viewed as intensive Tongitudinal data as well. In theory, intensive longitudinal data can provide answers to important questions in drug abuse research and HIV/AIDS research. However, there are no appropriate statistical procedures that can be applied for such data to address important scientific questions, such as: What is the relationship between mood and urge to smoke over the period from quitting smoking to the first relapse? What factors predict first instance of unsafe sex after a diagnosis of HIV? In this project, we propose new models for joint analysis of intensive longitudinal data and time-to-event data. These models possess many valuable features which make them the most appropriate to use for testing key hypotheses in drug use research and HIV/AIDS research using intensive longitudinal data. The proposed new models allow effects to vary over time and changing across individual subjects, and keep the structure of random error process very flexible. We further propose estimation procedures for the proposed new models. We plan to develop software to implement the proposed procedure and assess the performance of the proposed procedures by extensive Monte Carlo simulations. We further plan to apply the new procedures we have developed to address important questions in substance use and HIV/AIDS research areas using intensive longitudinal data on HIV/AIDS and on tobacco, alcohol and marijuana.
Tobacco, alcohol and marijuana are the most widely used substances within the US and have been linked to a myriad of both short and long-term consequences. Findings in this project will help public health researchers to undertand the natural history of HIV infection, including the progression of HIV infection to AIDS, and the etiology of drug use and dependence and further inform drug use and treatment efforts.
Showing the most recent 10 out of 443 publications
