Dropout due to loss to follow-up or death is common in prospective longitudinal cohort studies, such as the Multicenter AIDS Cohort Study (MACS). Patients that drop out may be more likely to have disease progression, use drugs or engage in high risk behaviors. Over time, the cohort evolves to be biased towards healthier subjects with lower risks. As a result, when estimating the relationship between drug use and longitudinal outcomes, analyses must consider subject losses or the consequences of drug use will be underestimated. When the probability of dropout depends on the unobserved outcomes, even after conditioning on observable data, the missing data are missing not at random (MNAR) and therefore nonignorable. Despite the likelihood of nonignorable dropout, traditional methods, such as mixed- or random-effects models are frequently used. This may be partially due to the complexity level of existing statistical methods and the inability to implement methods using standard software. In addition, many investigators are na?ve to possible biases and the resulting loss of power. Mixture model methods account for the dropout mechanism by factoring the joint outcome-dropout distribution into the dropout-time distribution, f(u), and f(y|u), the distribution of the outcome given dropout. The resulting complete data distribution, f(y), is +f(y|u)dF(u). Misspecification of a parametric form of f(y|u) can lead to bias. Recently developed varying-coefficient mixture models can be used to semi-parametrically model the outcome-dropout relationship for a continuous outcome. The method is computationally stable, highly flexible and relatively simple to implement using standard software. A simple extension of this varying-coefficient approach to binary outcomes will be developed. Application of these methods to the MACS data will accurately determine the consequences of drug use on clinical, risk and prevention-oriented outcomes. Varying-coefficient methods that account for dropout will be applied to estimate the relationship between drug use and the clinical outcomes of longitudinal CD4+ T cell count and HIV-1 RNA in untreated subjects (continuous outcomes) and to viral suppression in HAART-treated subjects (a binary outcome). In addition, the influence of drug use on HIV-risk and prevention-oriented outcomes such as high risk sexual risk behavior, HAART- adherence and needle sharing among injection drug users will be examined.

Public Health Relevance

Individuals with disease progression, drug use and other risk behaviors are more likely to drop out of the Multicenter AIDS Cohort Study due to death or loss to follow-up, such that over time, remaining subjects are healthier with lower risks. Analyses exploring the consequences of drug use must therefore account for subject loss. We will utilize new statistical methods to more accurately determine the consequences of drug use on clinical, risk and prevention-oriented outcomes.

Agency
National Institute of Health (NIH)
Institute
National Institute on Drug Abuse (NIDA)
Type
Small Research Grants (R03)
Project #
5R03DA026743-02
Application #
7780090
Study Section
Special Emphasis Panel (ZRG1-AARR-F (53))
Program Officer
Comolli, Jean C
Project Start
2009-03-01
Project End
2012-02-28
Budget Start
2010-03-01
Budget End
2012-02-28
Support Year
2
Fiscal Year
2010
Total Cost
$153,044
Indirect Cost
Name
University of Colorado Denver
Department
Type
DUNS #
041096314
City
Aurora
State
CO
Country
United States
Zip Code
80045
Moore, Camille M; MaWhinney, Samantha; Forster, Jeri E et al. (2017) Accounting for dropout reason in longitudinal studies with nonignorable dropout. Stat Methods Med Res 26:1854-1866
Forster, Jeri E; MaWhinney, Samantha; Ball, Erika L et al. (2012) A varying-coefficient method for analyzing longitudinal clinical trials data with nonignorable dropout. Contemp Clin Trials 33:378-85