[Analyzing the methodologies to estimate age at natural menopause and its effective factors]

Menopause, cessation of menstruation, is a public event for all women that occur between the ages of 45-55 [or even sooner]. A variety of researches done in Iran and the rest of the world show age at menopause can depend on numerous factors such as race and genetics, socioeconomics, history of fertility, physical activity, nutrition, sexual behaviors, diseases and etc. However, there are inconsistencies between results from different researches which can be related to methodology and methods of analysis. Present research has been done to analyze methodologies which estimate age at natural menopause and its associated factors by considering their power and weakness points. The cohort and cross-sectional methodologies to estimate age at natural menopause are introduced as well as appropriate statistical techniques to determine effective factors. Iranian and the some foreign papers, which their main object or one of the principle objects was estimating age at menopause, are introduced and analyzed. The results show that prediction and estimation of age at menopause are more complicated than it seems. Nearly all Iranian and some foreign surveys are poor in methodology and methods of analysis which decreases the efficacy and correctness of their findings

[Sample size determination for matched studies]

One faced to matched data when the main object for a study is investigation of the effect of a given factor on an outcome, such that the samples are matched on the some factors which are known as confounders. The response can be according to quantitative or qualitative scales. Matched case-control, before-after and crossover studies are used frequently in medicine. The significant tests are required to sufficient samples. In this paper, formulas for determination of sample size in matched studies are introduced by considering the scale of response and extended for matched case-control studies with multiple controls and illustrated using practical examples. Using statistical techniques, formulas for determination of sample size are introduced according to the scale of response. The practical examples interpreted in order to clarify the formulas and their applications. One can increase the power and the accuracy of the statistical tests for matched studies, using the introduced formulas

Analyzing receiver operating characteristic curves to compare medical diagnostic tests

One exposes with diagnosis problems when she or he does an experiment or modeling to predict and allocate objects or persons to certain groups. For example in medicine in order to discriminate diabetes or cancers [level 2 of prevention] different criterions or indices can used. The simplest status is allocating objects to two possible categories, therefore one can measure a test variable in ordinal or continuous scale and regarding an appropriate cut-off in range of test variable and sensitivity, specificity and value of loss function, he or she can determine objects for each category. A suitable and single value index to evaluate test variable is A, area under receiver operating characteristic [ROC] curve. Since probably there are several test variables that measured on a unique sample, so there are natural correlations between A's. When one wants to compare and select the best test [s] among them, ignoring of these correlations can lead to confused results. We have detailed a method to compute A's and their variance-covariance matrix and introduced an adequate statistical test to compare them also using a set of simulated data have showed effectiveness of correlations on statistical results. For applied purposes we have prepared a software package using Delphi5. Based on simulated data for two indices we found: A[1] = 0.660, SE[A[1]] = 0.054, A[2] -0.49, SE [A[2]] - 0.06. By ignoring correlations between A[1], A[2] we computed Z =2.1, it leads to reject equality of As in a alpha = 0.05 level, otherwise by regarding correlation, Z =1.92 and equality will accept. Ignoring correlation between As can lead to incorrect results

[Comparing predictions of type of pregnancy using artificial neural networks and multinomial logistic regression]

Some medical and epidemiological surveys have been designed to predict a nominal response variable with several levels. Type of pregnancy is one of them such that it has four possible states: wanted, wife unwanted, husband unwanted and couple unwanted. In this paper we have predicted type of pregnancy as influencing factors on it, using two different models and compared them together. Regarding type of pregnancy with several levels, we developed a multinomial logistic regression and a neural network based on data and compared their results using three statistical indices: sensitivity, Specificity and kappa. Based on three indices, neural network showed a better fit and prediction on data in comparison to multinomial logistic regression. When relations between variables are complex, one can use neural networks instead of multinomial logistic regression to predict nominal response variables with several levels in order to gain more accurate predictions