A Bayesian order-restricted model for hormonal dynamics during menstrual cycles of healthy women

Department of Mathematics and Statistics, University of Maryland Baltimore County, Baltimore, MD, 21250, USA.
Statistics in Medicine (Impact Factor: 1.83). 09/2012; 31(22):2428-40. DOI: 10.1002/sim.4419
Source: PubMed


We propose a Bayesian framework for analyzing multivariate linear mixed effect models with linear constraints on the fixed effect parameters. The procedure can incorporate both firm and soft restrictions on the parameters and Bayesian model selection for the random effects. The framework is used to analyze data from the BioCycle study. One of the main objectives of the BioCycle study is to investigate the association between markers of oxidative stress and hormone levels during menstrual cycles of healthy women. Contrary to the popular belief that ovarian hormones are negatively associated with level of F (2) -isoprostanes, a known marker for oxidative stress, our analysis finds a positive association between ovarian hormone levels and isoprostane levels. The positive association corroborates the findings from a previous analysis of the BioCycle data. Copyright © 2011 John Wiley & Sons, Ltd.

5 Reads

  • Statistics in Medicine 09/2012; 31(22):2457-60. DOI:10.1002/sim.5500 · 1.83 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We propose a general framework for performing full Bayesian analysis under linear inequality parameter constraints. The proposal is motivated by the BioCycle Study, a large cohort study of hormone levels of healthy women where certain well-established linear inequality constraints on the log-hormone levels should be accounted for in the statistical inferential procedure. Based on the Minkowski-Weyl decomposition of polyhedral regions, we propose a class of priors that are fully supported on the parameter space with linear inequality constraints, and we fit a Bayesian linear mixed model to the BioCycle data using such a prior. We observe positive associations between estrogen and progesterone levels and F2-isoprostanes, a marker for oxidative stress. These findings are of particular interest to reproductive epidemiologists.
    Journal of the American Statistical Association 12/2012; 107(500). DOI:10.1080/01621459.2012.712414 · 1.98 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: F2-isoprostanes (F2-IsoP) are oxidant stress biomarkers that are higher in HIV-infected women than men. We explored whether the effect of hemoglobin (Hgb), serum iron, or anemia on F2-IsoP is different between HIV-infected women and men. Plasma F2-IsoP were quantified by gas chromatography/mass spectrometry; clinical and laboratory data were collected at enrollment or from the medical record. Multivariable linear regression was used to assess associations between F2-IsoP and Hgb, anemia as a dichotomous variable, and serum iron with adjustment for age, sex, race, body mass index, CD4 lymphocyte count, self-reported current smoking status, and antiretroviral therapy. Compared with men, women had lower Hgb [median: 12.7 (interquartile range: 11.8-13.9) vs. 14.9 (13.7-15.8) g/dL, P < 0.001], lower iron levels [75 (47-97) vs. 90 (69-121) µg/dL, P = 0.004], more anemia (29% vs. 10%, P < 0.001), and higher levels of F2-IsoP [42 (32-62) vs. 36 (25-46) pg/mL, P < 0.001]. The relationship between iron and F2-IsoP differed significantly between men and women (interaction P = 0.02). Men had a 21% (95% confidence interval: 8 to 36) increase in F2-IsoP per interquartile increase in iron (P = 0.001), whereas no relationship was seen among women [-4% (-17 to 13, P = 0.65]. Although women have overall higher F2-IsoP than men, a relationship between circulating F2-IsoP and iron levels was observed in men but not in women with HIV infection. The association between female sex and higher F2-IsoP is not explained by iron or Hgb levels because the association persists when controlling for these factors. The role of iron in oxidant stress and sex-specific differences among HIV-infected individuals require further study.
    JAIDS Journal of Acquired Immune Deficiency Syndromes 12/2013; 64(4):367-73. DOI:10.1097/QAI.0b013e3182a60f36 · 4.56 Impact Factor
Show more

Similar Publications