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 Objectives: Logistic regression is one of the most widely used generalized linear models for analysis of the relationships between one or more explanatory variables and a categorical response. Strong correlations among explanatory variables (multicollinearity) reduce the efficiency of model to a considerable degree. In this study we used latent variables to reduce the effects of multicollinearity in the analysis of a case-control study.
Methods: Our data came from a case-control study in which 300 women with breast cancer were compared to 300 controls. Five highly correlated quantitative variables were selected to assess the effect of multicollinearity. First, an ordinary logistic regression model was fitted to the data. Then, to remove the effect of multicollinearity, two latent variables were generated using factor analysis and principal components analysis methods. Parameters of logistic regression were estimated using these latent as explanatory variables. We used the estimated standard errors of the parameters to compare the efficiency of models.
Results: The logistic regression based on five primary variables produced unusual odds ratio estimates for age at first pregnancy (OR=67960, 95%CI: 10184-453503) and for total length of breast feeding (OR=0). On the other hand, the parameters estimated for logistic regression on latent variables generated by both factor analysis and principal components analysis were statistically significant (P<0.003). The standard errors were smaller than with ordinary logistic regression on original variables. The factors and components generated by the two methods explained at least 85% of the total variance.
Conclusions: This research showed that the standard errors of the estimated parameters in logistic regression based on latent variables were considerably smaller than that of model for original variables. Therefore models including latent variables could be more efficient when there is multicollinearity among the risk factors for breast cancer.

---

Background & Objectives: To determine the prevalence of HBV infection in the Golestan Province (southeastern part of the Caspian littoral, Iran).
Methods: A single cluster study was conducted in 2005, based on a sample of households, representative of the population aged 25-65 years in Golestan. All participants were invited for face-to-face interviews to gather demographic data. Blood samples were drawn and analyzed for serum markers of HBV infection such as HBsAg and HBcAb by the ELISA method. Factors associated with hepatitis B seroprevalence were analyzed using SPSS13 and STATA /8.
Results: A total of 1850 subjects were screened. The age- & sex-standardized prevalence for HBsAg positivity was 9.7% (95%CI=0.07-0.11). Rates were higher in males than in females (10.8% vs. 8.6%) (OR=1.28 95% CI=0.9-1.7). HBV seroprevalence in unmarried individuals was significantly higher than in those who were married (OR=2.13 95%CI=1.29-3.5). HBsAg(+) status was more frequent in urban areas (OR=1.46 95% CI=0.9-2.3). Thirty-six percent of population was HBcAb positive. HBcAb(+) prevalence was significantly higher in females (OR=1.46 95% CI=1.19-1.8) and married people (OR=1.58 95%CI=1.02-2.45), and also in urban areas (OR=1.34 95% CI=1.09-1.6).
Conclusions: This study shows that the prevalence of HBsAg(+) status in the Iranian province of Golestan is at a level regarded as "high" by the World Health Organization. It is higher than reported by pervious studies in Iran so it is important- especially for health providers and policy makers- to recognize risk factors and design appropriate prevention programs.

---

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: Oral disease is a significant burden to all countries of the world. Since there is little know about this in Iran we decided to identify of oral mucosal disease in patients referred to Oral Medicine Center affiliated to Tabriz Dental School.
Methods: A consecutive sample of admitted patients to OMC between April to June 2007 were included in this study. We employed a standard questionnaire in order to
Results: Eight hundred two patients were completed the requested questionnaire. Seventy percent was female. The mean age of the study sample was 32.68 (SD=12.25) years. Approximately 70% of subjects had oral lesion, %19.2 had normal mucosa and %21.60 of patients had oral mucosal lesions that need to follow and control. The most common lesions were ankiloglossia (%29.7) then fissural tongue (%25.7) and coated tongue (%23. 7).
Conclusions: Many patients attending the center were unaware of oral lesions in their soft oral tissue which needed to follow up. This emphasizes that examination of soft oral tissue should be considered by health policy makers in oral health agenda at national level.

---

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

---

Background and Objectives: Diabetic patients are always at risk of hypertension. In this paper, the main goal was to design a native cost sensitive model for the diagnosis of hypertension among diabetics considering the prior probabilities.

Methods: In this paper, we tried to design a cost sensitive model for the diagnosis of hypertension in diabetic patients, considering the distribution of the disease in the general population. Among the data mining algorithms, Decision Tree, Artificial Neural Network, K-Nearest Neighbors, Support Vector Machine, and Logistic Regression were used. The data set belonged to Azarbayjan-e-Sharqi, Iran.

Results: For people with diabetes, a systolic blood pressure more than 130 mm Hg increased the risk of hypertension. In the non-cost-sensitive scenario, Youden's index was around 68%. On the other hand, in the cost-sensitive scenario, the highest Youden's index (47.11%) was for Neural Network. However, in the cost-sensitive scenario, the value of the imposed cost was important, and Decision Tree and Logistic Regression show better performances.

Conclusion: When diagnosing a disease, the cost of miss-classifications and also prior probabilities are the most important factors rather than only minimizing the error of classification on the data set.

---

Background and Objectives: In June 2018 , 537 residents of a rural area in Khuzestan Province presented to the regional Comprehensive Health Service Center for gastroenteritis symptoms. This study was designed to determine the extent and cause of the outbreak.
 
Methods: A case-control study was performed after random selection of the case and control groups (80 cases and 88 controls). Clinical and water samples were analyzed for parasitic, bacterial and viral pathogens in local, provincial and national laboratories. Odds ratios with corresponding 95% confidence intervals were used to assess the relationship between disease and exposure.
 
Results: The odds ratio of rural plumbing water consumption as a risk factor was 3.3 (95% CI: 1.7-6.2). Using in vitro methods, Shigella sonnei was isolated in clinical samples and enterohemorrhagic Escherichia coli and enteropathogenic Escherichia coli were isolated from both clinical samples and water samples taken from the intake basin of water supply facilities.
 
Conclusion: Consumption of rural plumbing water Since 26 June 2018 as well as the water stored in domestic tanks at certain water-shut-off times is associated with gastroenteritis outbreak. To prevent similar outbreaks, continuous chlorination of drinking water during distribution through rural pipelines should be done. To prevent secondary outbreaks after the epidemic phase, educating and informing people about personal hygiene is essential.