A class of semiparametric transormation frailty models with application to estimating treatment effects and heterogenity in aids clinical traials

Muitivariate failure time data are common in medical research; com¬monly used statistical models for such correlated failure-time data include frailty and marginal models. Both types of models most often assume pro¬portional hazards (Cox, 1972); but the Cox model may not fit the data well This article presents a class of linear transformation frailty models that in¬cludes, as a special case, the proportional hazards model with frailty. We then propose approximate procedures to derive the best linear unbiased es¬timates and predictors of the regression parameters and frailties. We apply the proposed methods to analyze results of a clinical trial of different dose levels of didansine (ddl) among HIV-infected patients who were intolerant of zidovudine (ZDV). These methods yield estimates of treatment effects and of frailties corresponding to patient groups defined by clinical history prior to entry into the trial.     

Combining genotype groups and recursive partitioning: an application to human immunodeficiency virus type 1 genetics data

Summary.  Understanding the relationship between human immunodeficiency virus type 1 genotypic markers of resistance and response to therapy presents an analytic challenge because of the high dimensionality of the viral genome and the complex interactions across sites on the genome. Recursive partitioning is a natural approach to classifying patients on the basis of covariates that capture information on the variability of a response variable and is therefore well suited to handling both of these features of the problem. In the human immunodeficiency virus genetics setting, we aim to classify patients on the basis of the genotypic characteristics of their infecting viral population so that the resulting groups explain the most variability in drug susceptibility phenotype. Recursive partitioning is also informative in identifying meaningful genotypic patterns that define the resulting classes. By combining dimension reduction techniques with recursive partitioning we can arrive at different classification schemes which are potentially more informative. Furthermore, the use of weighted distance metrics in the clustering algorithm allows us to update and potentially to improve on this classification scheme. An illustration of the relative benefits of the various approaches is provided by using 2559 protease sequences and the corresponding 50% inhibitory concentrations for Indinavir.     

Prevalence of same‐gender sexual behavior and HIV in a probability household survey in Mexican men

The prevalence of male same‐gender sexual behavior in Mexico City in relation to HIV transmission was studied. A household probability survey of 8,068 adult men was conducted in 1992–93 using the Mexican National Health Survey sampling frame. The response rate was 59%. Differences between respondents and nonrespondents indicated no evidence for significant bias. A random subsample of 1,116 individuals provided serum or saliva for HIV testing. An estimated 2.5% of men practiced same‐gender sex in their lifetime: 2.1% (95% CI: 1.7 ‐ 2.4%) reported bisexual behavior and 0.4% (95% CI: 0.3 ‐ 0.6%) reported exclusively homosexual behavior. Among bisexuals, 70% reported sex only with women in the previous year, 7% reported sex only with men, 13% reported sex with both, and 10% were sexually inactive. A condom was used by 46% in their last homosexual encounter. An estimated 0.1% of married men were homosexually active in the previous year. The HTV prevalence estimate was 0.2% in the sample. The rate was 4% among homosexual/bisexual men and 0.09% in heterosexual men (p < 0.0001). Estimates of homosexual behavior and HTV infection from this population‐based sample are lower than results from nonprobability studies. The low prevalence of condom use anticipates future growth of the epidemic in the homosexual population. Bisexual behavior appeared to be infrequent and transitory, particularly among married men.     

Strategies for Cohort Sampling Under the Cox Proportional Hazards Model,Application to an AIDS Clinical Trial

In some studies that relate covariates to times of failure it is not feasible to observe all covariates for all subjects. For example, some covariates may be too costly in terms of time, money, or effect on the subject to record for all subjects. This paper considers the relative efficiencies of several designs for sampling a portion of the cohort on which the costly covariates will be observed. Such designs typically measure all covariates for each failure and control for covariates of lesser interest. Control subjects are sampled either from risk sets at times of observed failures or from the entire cohort. A new design in which the sampling probability for each individual depends on the amount of information that the individual can contribute to estimated coefficients is shown to be superior to other sampling designs under certain conditions. Primary focus of our designs is on time-invariant covariates, but some methods easily generalize to the time-varying setting. Data from a study conducted by the AIDS Clinical Trials Group are used to illustrate the new sampling procedure and to explore the relative efficiency of several sampling schemes.