Search

The main objective of this paper is to present a method of evaluating several tumor markers by using the generalized ROC criterion. This criterion finds the best linear combination of the tumor markers such that the area under the ROC curve is maximized. Confidence intervals for the generalized ROC criteria are also presented. This methodology is applied to 51 patients with advanced colorectal cancer for whom the ACE tumor markers were measured before and during chemotherapy treatment. Two populations were defined according to clinical response to chemotherapy. Each marker taken separately, whether on the raw scale or on the transformed scale, contained 0.5 in the confidence interval and was thus non significant. This was also true for both markers on the raw scale. However, the best linear combination on the logarithms of ACE before and at evaluation gave a significantly better area under the ROC curve. A weighted change in ACE measurements significantly distinguishes between responders and non responders in patients with advanced colorectal cancer. We propose that the methodology presented in this paper be used for the evaluation of several tumor markers.

Objective The central objective of this paper is to explore the dynamic interactions between 5 sets of variables, which are Sociodemographic Characteristics, Satisfaction with Life, Perceived Discrimination, Religiosity and Emotional Distress within Montreal's Tariqa Qadiriya Boudchichiya, a Muslim Sufi way whose origins are Moroccan and date back to the 18th century. Method As a method, we considered psychological distress as the dependent variable and performed univariate descriptive statistical analyzes, bivariate correlation analyzes (Pearson correlation), one-way ANOVA analyzes, and multivariate analyzes (linear regressions). Results Our results, although preliminary due to a relatively small sample (n = 56), allow us to put forward a new hypothesis suggesting that the intense spiritual practice that characterizes the Tariqa, would allow "a work of the self on self" through a set of "techniques of the self" (Foucault) that contribute to a certain emotional well-being, if not to mental health. Conclusion Our conclusion underlines the importance of investigating more in depth the possible contribution of religiosity to the subject's capacity to act on oneself in order to emerge as a spiritual, ethical and political subject.

This paper deals with the modeling of insulation material lifespan in a partial discharge regime. Accelerated aging tests are carried out to determine the lifespan of polyesterimide insulation films under different various stress conditions. The insulation lifespan logarithm is modeled as a function of different factors: the electrical stress logarithms and of an exponential form of the temperature. The model parameters, or so-called factor effects, is estimated on a training set. The significance of the factors is evaluated through the analysis of variance (ANOVA). In a first step, the design of experiment method (DoE) is considered. The associated lifespan model is linear with respect to the factors. This method is well-known for reducing the number of experiments while providing a good accuracy. In a second step, the response surface method (RSm) is considered. This method takes also into account some second order terms and thus possible interactions between the stress factors. Performance of the two methods are analyzed and compared on a test set.

Summary The treatment of Angle Class II/2 malocclusion requires an understanding not only of the dental and maxillofacial features but also of the changes occurring with age in the dental relationships. This paper attempts to determine certain dental relationships in patients with Angle II/2 malocclusion and any changes that might occur with age. The study material included 96 lateral cephalometric X-rays of patients with Angle II/2 malocclusion, the youngest of whom was 7 years old and the eldest 35. Six cephalometric measurements were performed on every cephalometric X-ray. Statistical processing of data included one-variable analyses and correlations of two variables. Statistical one-variable analysis used methods from descriptive statistics. Methods of inferential statistics have been applied in order to revel statistical significance. The evaluation of the findings of this research study concluded that, in patients with Angle II/2 malocclusion, maxillary central incisors presented pronounced palatal tipping, which exacerbated significantly with age, while mandibular incisors displayed reduced labial tipping and projection, which significantly reduced with age. Furthermore, it was found that there was pronounced incisor overjet, which increased with age, due to increased protrusion of mandibular incisors.

In this methodological paper, we present the elaboration of the experimental design, central stage of the research cycle. First, we focus on the hypothesis elaboration from the research question. Then, we detail the hypothesis operationalisation by the means of the choice of variables. Finally, the experimental controls and the factorial designs are presented.

Summary : Should we use analysis of variance rather than regression analysis for experiences using continuous independent variables ? In this paper, we discuss the two possible statistical approaches when IV are continuous : analysis of variance and multiple linear regression. The statistical model common to these two approaches, and analysis of covariance is first briefly reminded. Using general arguments, followed by a psycholinguistic illustration, we discuss the advantages and drawbacks of using an experimental control of confoundings and random variability, which leads to analysis of variance, or a statistical control, which leads to multiple regression analysis. The problems to avoid for insuring the validity of the chosen approach are also underlined., Dans cet article, nous discutons des deux approches statistiques possibles lorsque les VI sont continues : l'analyse de variance et la régression linéaire multiple. Le modèle statistique commun à ces deux approches, ainsi qu'à l'analyse de covariance, est tout d'abord rappelé de manière succincte. A travers des arguments généraux, puis une illustration issue de la psycholinguistique, nous discutons ensuite des avantages et des inconvénients du recours à un contrôle expérimental des confusions d'effets et de la variabilité aléatoire, qui amène à privilégier l'analyse de variance, et à celui d'un contrôle statistique, qui conduit à la régression linéaire multiple. Les écueils à éviter pour assurer une certaine validité à l'approche privilégiée sont aussi présentés., Méot Alain, Bonin Patrick. Faut-il préférer l'analyse de variance à l'analyse de régression dans les expériences utilisant des VI continues ?. In: L'année psychologique. 2005 vol. 105, n°4. pp. 669-695.

Prognostic studies aim to stress and quantify the effect of some individual characteristics on the evolution of an illness. Statistically speaking, most of the time, a prognostic study relying on a censored endpoint can be summarized in building a Cox model. Even though they are often encountered in clinical research or in epidemiology, prognostic studies are rarely reliable and there is no real consensus on the way to perform them. The aim of this article is to give some directions regarding the analysis of such studies. For illustration purpose, this paper relies on commented results from a study in oncology.

This paper proposes an approach for calculating the normal confidence interval of a weighted summary measure which requires a particular continuous transformation for its variance estimation. By using the transformation properties and applying the delta method, the variance of transformed measure is easily expressed in terms of the transformed specific measure variances and the squared weights. The confidence limits of the summary measure are easily deduced by inverse transformation of those of transformed measure. The method is illustrated by applying it to some well known epidemiological measures. It seems appropriate for application in stratified analysis context where size allows normal approximation.

This study was designed to determine the prevalence of HIV infection and associated high-risk behavior in female sex workers in two areas of Bujumbura (Bwiza and Musanga), Burundi. Between June 15th and July 15th 1993, 320 female sex workers responded to an anonymous questionnaire and gave a blood sample on filter paper for detection of HIV antibodies. The mean age of the prostitutes was 24.6 (+/- 5.9) years (range 15 to 46). The HIV infection rate was 42.2% (135 of the 320). Univariate analysis showed that the proportion of HIV-infected female sex workers was higher in the Musaga area than in Bwiza (48.3% versus 34.3%; p = 0.012). The prevalence of HIV infection increased with age (24.3% for those aged between 15 and 19; 40.8% for those aged between 20 and 24 and more than 51.4% infected in the over 25 age-group; p0.001). The median number of sexual partners in the preceding month was 3 (range 1 to 50) and was not correlated with HIV infection (Wilcoxon's test: p = 0.516). HIV infection was highly correlated with a previous history of STDs (62.7% versus 32.7%; p0.001) and was also correlated with STD symptoms (87.5% versus 51.7%; p = 0.01). The only factors correlated with HIV infection in multivariate analysis were age (p = 0.008; Trend's test) and previous history of STDs (p0.001). Only 40.2% (127 of 316) of the female sex workers had used condoms with their clients during the preceding month. Intervention strategies to combat HIV transmission among individuals with high levels of sexual activity in Burundi should focus on the female sex workers and their clients, particularly the manual laborers, drivers, soldiers, prisoners and other men who regularly use prostitutes. The prevention of HIV infection in individuals of high-risk groups is the best strategy to reduce HIV transmission in the general population of developing countries.

The multiplication of multicentric clinical studies leads to a generalization of the use of a center effect in statistical analyses. In the simple situation where the dependent variable is quantitative and there is only a simple factor of interest, the formalisation of such a protocol corresponds to a two way ANOVA (with the factor of interest and the center factor) with generally a term of interaction. Even such a simple situation leads to interrogations, one of them being to determine whether the centre effect is fixed or random. In the first situation, the conclusion of the study will concern only the centres evaluated in the study, while in the second, the conclusion will concern the whole population from which the centers have been randomly selected. Such a generalizability has a cost: a loss of power in determining the effect of the factor of interest (the loss being as important as the interaction factor of interest x centre factor is large) and the necessity of sophisticated statistical software. This paper will focus on successively: the notion of ANOVA, of fixed or random effect (with the conceptual, formal and numerical differences that distinguish these two approaches). A numerical example is proposed.

Patient satisfaction is increasingly used to evaluate the performance of health services. Validated French-language instruments to measure satisfaction are currently lacking. This study was designed to validate a questionnaire of this kind and to identify factors associated with patient satisfaction.Mail survey of 1027 patients who consulted at 4 different ambulatory health care settings in Geneva, Switzerland. The participation rate was 81%. The questionnaire measured 7 dimensions of satisfaction using 16 items adapted from other sources.The questionnaire was easy to respond to (scores were available for 95 to 99% of respondents, depending on the scale). The internal consistency of the scales was satisfactory (Cronbach alpha between 0.65 and 0.82) for 5 of 6 multi-item scales; it was lower for the scale which measures satisfaction with access to care. Factor analysis identified two principal components corresponding roughly to the "process" and to the "organization" of care. Open comments also confirmed the validity of the multi-item scales. Several patient or visit characteristics were independently associated with the level of satisfaction: older patients, those who were born in Switzerland, who had a visit appointment, who consulted a specialist, and those who saw the same physician as at their previous visit were more satisfied than other patients.The brief satisfaction questionnaire described in this paper is easy to use, and its reliability and validity are good. Its use can be recommended in ambulatory health care settings. Several variables associated with the level of satisfaction were identified; they should be measured in satisfaction surveys to allow correct appraisal of the results.

Time series analysis of health indicator can contribute to the improvement of knowledge in epidemiology and public health. It is necessary to use models and methods appropriate to the time domain. A class of models proposed by BoxJenkins are presented together with the necessary tools for identifying such models. Two useful generalisations are also discussed. To study the link between two time series, regression models need to include autocorrelated error terms while to evaluate the effect of an intervention, the use of intervention models are appropriate. This paper presents basic concepts of such analyses in a simple way to encourage the use of those methods.

Brucellin INRA is a new allergen used for the screening of brucellosis. It shows, by a delayed-type skin hypersensitivity reaction, the sensitization induced in animals by a brucella infection. The biological potency of each batch must be compared to that of a standard allergen. The use of guinea-pigs necessary for this quality control and the statistical analysis of the results are described and discussed. The aim of this paper is to provide the end-user with a simple and accurate method, and to alert him to the difficulties and limits of such a test.

In a prevention trial, when the randomised units are groups of subjects, the first step in the analysis of the results is to check whether the groups differ, for the variables under study, ie, to test a possible group effect. If this effect is not significant, the results are analysed as if the subjects had been randomised. On the other hand, it this effect is significant, the comparison must be carried out between the groups and no more between the subjects. In the latter case, the loss of efficiency of the randomization of groups instead of the randomization of subjects can be computed. When a group effect is present and the number of subjects in each group differs, the analysis is considerably more complex. In the Paris cardiovascular prevention trial, 160 groups of young men, with variable numbers of subjects in each group have been randomised. The change in weight, blood cholesterol and cigarette consumption after two years of intervention are analysed in the present paper with the methodological principles mentioned above.