Background. Tree models provide easily interpretable prognostic tool, but instable results. Two approaches to enhance the generalizability of the results are pruning and random survival forest (RSF). The aim of this study is to assess the generalizability of saturated tree (ST), pruned tree (PT), and RSF. Methods. Data of 607 patients was randomly divided into training and test set applying 10-fold cross-validation. Using training sets, all three models were applied. Using Log-Rank test, ST was constructed by searching for optimal cutoffs. PT was selected plotting error rate versus minimum sample size in terminal nodes. In construction of RSF, 1000 bootstrap samples were drawn from the training set. C-index and integrated Brier score (IBS) statistic were used to compare models. Results. ST provides the most overoptimized statistics. Mean difference between C-index in training and test set was 0.237. Corresponding figure in PT and RSF was 0.054 and 0.007. In terms of IBS, the difference was 0.136 in ST, 0.021 in PT, and 0.0003 in RSF. Conclusion. Pruning of tree and assessment of its performance of a test set partially improve the generalizability of decision trees. RSF provides results that are highly generalizable.

Objective: To investigate the impact of household food insecurity during the third trimester of pregnancy on the growth indicators of infants aged less than six months. Design: Retrospective longitudinal study. Setting: 137 healthcare centres (15 cities) in Khorasan Razavi province, Iran. Data were extracted from the Sina Electronic Health Record System (SinaEHR®). Participants: This study was conducted on 2,481 mother and infant dyads during November 2016-March 2019. The household food insecurity access scale (HFIAS; nine-item version) was used to measure food insecurity in the third trimester of pregnancy. Women who delivered singleton infants were included in the study, and anthropometric indices of infants were measured throughout the first sixth months of life. Results: Approximately 67% of the participants were food-secure, while 33% had varying degrees of food insecurity. The children born to the mothers in the food-insecure households were respectively 2.01, 3.03, and 3.83 times more likely to be stunted at birth (95% CI: 1.17-3.46), four months (95% CI: 1.21-7.61), and six months of age (95% CI: 1.37-10.68) compared to their counterparts in the food-secure households. However, there were no significant differences in mean birthweight, birth height, and head circumference at birth between the two groups. Conclusions: Household food insecurity during pregnancy is a risk factor for stunting in infants aged less than six months. Therefore, national nutrition programs could considerably support women in food-insecure households during and before pregnancy.

Background & Aims: The electronic health record software of Mashhad University of Medical Sciences is the Integrated Information system, called “SINA” which was launched in 2016 to enhance public health and manage mother and child information. This study aimed to identify the challenges of “SINA” focusing on the information management of pregnant women from the healthcare midwives’ perspective. Methods: This qualitative study was conducted on 14 healthcare midwives currently working in the health centers affiliated with Mashhad University of Medical Sciences who were selected through purposeful sampling. Data were collected through in-depthand semi-structured interviews. Data analysis was carried out through thematic content analysis. Results: The main theme challenges identified from the analyzed data were system management problems and software systems problems. The four subthemes were system flow and process management, structural problems, interface, and technical problems. Conclusion: Based on the healthcare midwives’ perspectives, Maternity care information and training of pregnant women were well documented in “SINA”. This study can advise the policymakers to succeed in the Practical applications of the “SINA” by enhancing its strong features and ability to handle its challenges.

Background: Bladder cancer is one of the most common cancers worldwide and also in Iran. Understanding of bladder cancer epidemiology is of great value for policymakers and assists in the prevention and early detection of the disease. Objectives: The aim of the present study was to report national and subnational incidence trends of bladder cancer in Iran between 2003 and 2015. Methods: This study investigated the age-standardized incidence rates (ASIRs) per annum of bladder cancer from 2003 to 2015 in Iran using the data from the cancer registration system of the Ministry of Health and Medical Education of Iran. The crude incidence rates were calculated by dividing incident cases by the country and province population, which were provided by the Statistic Center of Iran. Age standardization was performed using the WHO standard population, and ASIRs were compared by age, sex, and province. Results: The ASIR of bladder cancer increased from 8.35 in 2003 to 13.57 in 2015 in men. The ASIR of bladder cancer also showed a mild increase in females, 2.12 in 2003 versus 2.86 in 2015. The province of Yazd had the highest rate of bladder cancer in men, and West Azerbaijan had the highest rate for women (15.13 and 7.79 per 100,000), while Sistan va Baluchestan and Ilam had the lowest ASIRs for men and women (3.01 and 0.96 per 100,000, respectively). Conclusions: The increasing trend of bladder cancer incidence in Iran, despite the clear decreasing global trend accompanying to ongoing aging of the population highlights the diseases as a potential health problem in the upcoming years in Iran. Therefore, it is necessary for health organizations to implement effective research and control programs in the country to prevent further increases in disease burden.

Introduction: Falling is one of the most common problems of the elderly people with a multi factorial nature and frequent cases. This study aimed to determine the risk factors of falls in old people covered by Mashhad University of Medical Sciences. Methods: This retrospective case-control study was conducted on 15,600 elderly participants. Data were extracted from Sina Electronic Health Record System (SinaEHR®, Iran). The obtained data were analyzed using STATA through odds ratio formula. Results: Risk factors of falls in the elderly subjects included age, fear of falling, higher body mass index, diabetes, anemia, gastrointestinal problems, hypothyroidism, use of sedatives, and smoking, after adjusting the potentially confounding effects of other variables. Conclusion: The current study provides evidence that patient-related factor such as diabetes, anemia, hypothyroidism and smoking are associated with falls in the elderly. The results of the present research can be used by health policy-makers to reduce fall-related costs of the old people by focusing on care services and high-risk individuals.