Iranian Journal of

---

Background and Objectives: Missing data exist in many studies, e.g. in regression models, and they decrease the model's efficacy. Many methods have been suggested for handling incomplete data: they have generally focused on missing outcome values. But covariate values can also be missing.
Materials and Methods: In this paper we study the missing imputation by the EM algorithm and auxiliary variable and compare the result with case-complete analysis in a logistic regression model dealing with factors that influence the choice of the delivery method.
Our data came from a cross-sectional study of factors associated with the choice of the delivery method in pregnant women. The sample size in this cross-sectional study was 365 and the data were collected through interviews, using questionnaires covering several demographic variables, delivery history, attitude, and some social factors. We used standard deviations to compare the efficiency of the two methods.
Results: The results show that maximum likelihood analysis by EM algorithm is more effective than case-complete analysis.
The problem of missing data is common in surveys and it causes bias and decreased model efficacy. Here we show that the EM algorithm for imputation in logistic regression with missing values for a discrete covariate is more effective than case-complete analysis.
Conclusion: On the other hand if missing values occur for a continuous covariate then we have to use other methods or change the variable into a discrete one.

---

Background and objective: Disease or mortality mapping are statistical methods aimed at providing precise estimates of rates across geographical maps. The aim of this research is to improve the precision of relative risk (RR) estimates of infant mortality (IM) for different rural areas, using empirical and full Bayesian methods.
Methods: Infant mortality data were extracted from the vital horoscope (Zij-Hayati) for years 2001 and 2006 across rural areas of Iran. Maximum Likelihood, Empirical Bayes with Poisson-Gamma model and full Bayesian models were used. Mont Carlo Markov Chain method was used for latter models. Deviance information criterion (DIC) was computed to check the models fittings. R, WinBUGS and Arc GIS software were employed.
Results: Based on the full Bayesian method, the highest RR of infant mortality was 1.73 (95%CI: 1.58-1.88) in year 2001 and 1.62 (95%CI: 1.50-1.75) in 2006 which belonged to Sistan-va-Blouchestan area in comparison to the whole country. In 2001, the rural areas of Birjand (1.45), Kordistan (1.23) and Khorasan (1.21) and in 2006, Birjand (1.42), Zanjan (1.39), Kordistan (1.36), Ardebil (1.32), Zabol (1.28), West Azerbaijan (1.18) and finally Golestan (1.14) had significant RR of IM (all p<0.05). The lowest RR of infant mortality for year 2001 were belong to rural areas of Tehran University (0.56) and for year 2006 to former Iran University (0.52).
Conclusion: To estimate the mortality map parameters, the full Bayesian method is preferred compared to empirical Bayes and maximum likelihood.

---

Background & Objectives: Studying several linked markers provides more information on locating disease genes locus by using genetic association analysis.  The aims of this study were to introduce Multimarker Family Base Association Tests (FBAT-MM) and its Linear Combination (FBAT-LC) in multimarker genetic association analysis and to examine the association of selected microsatellites with HDL-C in an Iranian population.
Methods: One hundred twenty five (125) families having at least one member with metabolic syndrome and at least two members with low HDL-C were selected from participants of the Tehran Lipid and Glucose Study (TLGS). Multimarker genetic association of HDL-C level with some microsatellites in the chromosomes 8, 11, 12, and 16 were examined using FBAT-MM and FBAT-LC methods.
Results: The families consisted of 563 individuals (269 males and 294 females). FBAT-MM showed significant genetic association only between HDL-C and three microsatellites in Chromosome 11 (P<0.05). The microsatellite D11S1304 was found as the significant factor for multimarker genetic association.
Conclusion: FBAT-MM and FBAT-LC did not show shortcomings such as excessive conservatism and low power which are, usually, observed in other multimarker methods.  Finding microsatellites associated with HDL-C level can provide background for further researches on the role of predisposing genes in metabolic syndrome.

Normal