Biomedical subjects

Daniel Commenges

Publications and source records attributed to Daniel Commenges.

8 recordsLinked to original sources

PubMed2003

Maximum penalized likelihood estimation in a gamma-frailty model.

The shared frailty models allow for unobserved heterogeneity or for statistical dependence between observed survival data. The most commonly used estimation procedure in frailty models is the EM algorithm, but this approach yields a discrete estimator of the distribution and consequently does not allow direct estimation of the hazard function. We show how maximum penalized likelihood estimation can be applied to nonparametric estimation of a continuous hazard function in a shared gamma-frailty model withright-censored and left-truncated data. We examine the problem of obtaining variance estimators for regression coefficients, the frailty parameter and baseline hazard functions. Some simulations for the proposed estimation procedure are presented. A prospective cohort (Paquid) with grouped survival data serves to illustrate the method which was used to analyze the relationship between environmental factors and the risk of dementia.

Algorithms↗

PubMed2003

Cardiovascular mortality and calcium and magnesium in drinking water: an ecological study in elderly people.

BACKGROUND: Previous studies found relations between cardiovascular mortality and minerals in drinking water, but the major works considered water hardness or neglected the differences between adults and elderly. Drinking water is an important source of calcium in the elderly particularly because of increased needs and decreased consumption of dairy products. METHODS: We collected informations about all deaths (14,311) occurring in 69 parishes of the South-West of France during 7 years (1990-1996). We obtained the causes of deaths from a special service of INSERM for each death, with age at death and sex. The exposure value was supplied by administrative source (DDASS) and by measurement surveys. We use an extra-Poisson variation model to take into account the heterogeneity of the population of these parishes. RESULTS: A significant relationship was observed between calcium and cardiovascular mortality with a RR: 0.90 for non-cerebrovascular causes and RR: 0.86 for cerebrovascular (when calcium is higher than the second tercile: 94 mg/l). We found a protective effect of magnesium concentrations between 4 and 11 mg/l with a RR: 0.92 for non-cerebrovascular and RR: 0.77 for cerebrovascular mortality, as compared to concentrations lower than 4 mg/l. CONCLUSIONS: These findings strongly suggest a potential protective dose-effect relation between calcium in drinking water and cardiovascular causes. For magnesium, a U-shape effect is possible, especially for cerebrovascular mortality.

Aged↗

PubMed2002

Endometrial vascularity and ongoing pregnancy after IVF.

BACKGROUND: Embryo transfer is prone to failure. AIM: To investigate whether endometrial vascularity influences in vitro fertilization (IVF) outcome. METHODS: Total 144 patients receiving IVF (conventional or microinjection) were assessed with color and power Doppler on the day of embryo transfer: age, IVF type, number and quality of embryos, endometrial thickness and aspect, mean uterine PI, uterine notch, type of endometrial vascularity (peripheral or sub- and intra-endometrial), and pregnancy involving second trimester were recorded. RESULTS: 27 (18.7%) pregnancies were obtained. By univariate analysis, two parameters were significant: high frequency of uterine notch (P = 0.03) and peri-endometrial vascularity (P = 0.012) in the group of failures. Multivariate analysis by logistic regression clearly showed that the absence of sub- and intra-endometrial color signal decreased the chances of pregnancy eight-fold odds ratio (OR) = 0.14 [CI: 0.029-0.68]. CONCLUSION: In this limited series, the presence of sub- and intra-endometrial vascularity on the day of transfer seemed to be mandatory for obtaining an ongoing pregnancy.

Adult↗

PubMed2002

Penalized likelihood approach to estimate a smooth mean curve on longitudinal data.

This paper aims to propose a penalized likelihood approach to estimate a smooth mean curve for the evolution with time of a Gaussian variable taking into account the correlation structure of longitudinal data. The model is an extension of the mixed effects linear model including an unspecified function of time f(t). The estimator (circumflex)f(t) is defined as the solution of the maximization of the penalized likelihood and is approximated on a basis of cubic M-spline with a reduced number of knots. We present modifications of four criteria (cross-validation, generalized cross-validation, T of Rice, Akaike's criterion) to estimate the smoothing parameter when data are correlated; these four criteria gave very similar results in the simulation study. The simulation study showed also the superiority of the Bayesian confidence bands of the mean curve over the frequentist ones. We develop empirical Bayes estimates of subject-specific deviations. This approach was applied to study the progression of CD4+ lymphocyte counts in a cohort of HIV patients treated with protease inhibitors.

Acquired Immunodeficiency Syndrome↗

PubMed2002

Recombinant follicle-stimulating hormone versus human menopausal gonadotropin in the late follicular phase during ovarian hyperstimulation for in vitro fertilization.

OBJECTIVE: To study the effect of exogenous LH in the late follicular phase on ongoing pregnancies and at the different stages of IVF-ET (stimulation, fertilization, and implantation) in patients with low endogenous LH. DESIGN: Retrospective cohort study with modeling of the different phases of IVF-ET. SETTING: IVF center of the teaching hospital in Bordeaux, France. PATIENT(S): Women undergoing IVF and ICSI treatment. INTERVENTION(S): One group received recombinant FSH alone (FSH group) and the other received recombinant FSH and hMG in the late follicular phase (i.e., when the largest follicle reached 14 mm) (FSH/hMG group). MAIN OUTCOME MEASURE(S): Ongoing pregnancy, number of oocytes, and number of embryos. RESULT(S): The FSH/hMG group had a higher probability of having at least one oocyte (odds ratio [OR] = 2.75 [1.11-6.80]), of having at least one embryo after oocyte retrieval (OR = 2.84 [1.33-6.07]), and of ongoing pregnancy after ET (OR = 2.04 [0.83-5.01]), and globally had a higher probability of ongoing pregnancy (OR = 2.83 [1.19-6.71]). CONCLUSION(S): In ovarian hyperstimulation for IVF-ET, LH supplementation in the late follicular phase of women with low endogenous LH is beneficial for ongoing pregnancy by increasing the rate of success of all stages of the treatment.

Adult↗

PubMed2002

Bivariate linear mixed models using SAS proc MIXED.

Bivariate linear mixed models are useful when analyzing longitudinal data of two associated markers. In this paper, we present a bivariate linear mixed model including random effects or first-order auto-regressive process and independent measurement error for both markers. Codes and tricks to fit these models using SAS Proc MIXED are provided. Limitations of this program are discussed and an example in the field of HIV infection is shown. Despite some limitations, SAS Proc MIXED is a useful tool that may be easily extendable to multivariate response in longitudinal studies.

Antiretroviral Therapy, Highly Active↗

PubMed2002

Correction of the p-value after multiple tests in a Cox proportional hazard model.

We consider a situation which is common in epidemiology, in which several transformations of an explanatory variable are tried in a Cox model and the most significant test is retained. The p-value should then be corrected to take account of the multiplicity of tests. Bonferroni method is often too conservative because the tests may be highly positively correlated. We propose an asymptotically exact correction of the p-value. The method uses the fact that the tests are asymptotically normal to compute numerically the distribution of the maximum of several tests. Counting processes theory is used to derive estimators of the correlations between tests. The method is illustrated by a simulation and an analysis of the relation between concentration of aluminum in drinking water and risk of dementia.

Aluminum↗

PubMed2002

A penalized likelihood approach for an illness-death model with interval-censored data: application to age-specific incidence of dementia.

We consider the problem of estimating the intensity functions for a continuous time 'illness-death' model with intermittently observed data. In such a case, it may happen that a subject becomes diseased between two visits and dies without being observed. Consequently, there is an uncertainty about the precise number of transitions. Estimating the intensity of transition from health to illness by survival analysis (treating death as censoring) is biased downwards. Furthermore, the dates of transitions between states are not known exactly. We propose to estimate the intensity functions by maximizing a penalized likelihood. The method yields smooth estimates without parametric assumptions. This is illustrated using data from a large cohort study on cerebral ageing. The age-specific incidence of dementia is estimated using an illness-death approach and a survival approach.

Journal Article↗