Estimating Risks of Progressing to Aids when Covariates are Measured

SUMMARY For some time, there has been interest in quantifying the risks of progressing to Aids from infection with the human immunodeficiency virus (HIV) associated with changes in the immunological markers. Until recently, little attention has been paid to the considerable variability of these markers and the corresponding effect on estimates of relative risk. We estimate the risks of progression to Aids associated with immunological markers among a group of homosexual men infected with HIV who have been followed at quarterly intervals for 5 years. Errors in measurement in the markers are shown to bias these estimates. Smoothing techniques are compared for their effectiveness in reducing the amount of variation in the markers and the subsequent bias in the coefficients. A simulation study compares the effectiveness of various methods of smoothing in reducing the bias of the estimated coefficient.

[1]  J. Malone,et al.  Sources of variability in repeated T-helper lymphocyte counts from human immunodeficiency virus type 1-infected patients: total lymphocyte count fluctuations and diurnal cycle are important. , 1990, Journal of acquired immune deficiency syndromes.

[2]  J M Taylor,et al.  The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1. , 1990, The New England journal of medicine.

[3]  A. Madansky The fitting of straight lines when both variables are subject to error , 1959 .

[4]  Raymond J. Carroll,et al.  Covariate Measurement Error in Logistic Regression , 1985 .

[5]  R. Prentice,et al.  Further results on covariate measurement errors in cohort studies with time to response data. , 1989, Statistics in medicine.

[6]  R. Tibshirani,et al.  Generalized additive models for medical research , 1986, Statistical methods in medical research.

[7]  D.Sc. Joseph Berkson Are there Two Regressions , 1950 .

[8]  D. Richman,et al.  A pilot study of low-dose zidovudine in human immunodeficiency virus infection. , 1990, The New England journal of medicine.

[9]  T. Merigan,et al.  A controlled trial comparing continued zidovudine with didanosine in human immunodeficiency virus infection. The NIAID AIDS Clinical Trials Group. , 1992, The New England journal of medicine.

[10]  R. Prentice Covariate measurement errors and parameter estimation in a failure time regression model , 1982 .

[11]  J. Goedert,et al.  Natural history of human immunodeficiency virus infections in hemophiliacs: effects of T-cell subsets, platelet counts, and age. , 1987, Annals of internal medicine.

[12]  R Brookmeyer,et al.  Biases in prevalent cohorts. , 1987, Biometrics.

[13]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[14]  A. Phillips,et al.  p24 antigenaemia, CD4 lymphocyte counts and the development of AIDS. , 1991, AIDS.

[15]  J. Montaner,et al.  Progression to AIDS and predictors of AIDS in seroprevalent and seroincident cohorts of homosexual men. , 1989, AIDS.

[16]  N. Dubin,et al.  CD4% is the best predictor of development of AIDS in a cohort of HIV‐infected homosexual men , 1991, AIDS.

[17]  C. Soskolne,et al.  A prospective study of male sexual contacts of men with AIDS-related conditions (ARC) or AIDS: HTLV-III antibody, clinical, and immune function status at induction. , 1986, Canadian journal of public health = Revue canadienne de sante publique.

[18]  N. Jewell,et al.  Patterns of T lymphocyte changes with human immunodeficiency virus infection: from seroconversion to the development of AIDS. , 1989, Journal of acquired immune deficiency syndromes.

[19]  J. Goedert,et al.  Markers of risk in HIV-1. , 1990, The New England journal of medicine.

[20]  J. Phair,et al.  Predictors of the acquired immunodeficiency syndrome developing in a cohort of seropositive homosexual men. , 1987, The New England journal of medicine.