Multilevel multinomial logistic regression model for identifying factors associated with anemia in children 6-59 months in northeastern states of India

Objective: To examine the factors influencing the occurrence of childhood anemia in North-East India.

Method: : A nationally representative systematic multistage stratified cross-sectional sample of singleton children aged 6 to 59 months from all states of India. Data consists of 10,136 children in the age group 6-59 months in eight northeastern states. The level of anemia was the outcome variable with four ordinal categories (severe, moderate, mild, and non-anemic). A two-level random intercept multivariate logistic regression model was considered with state of residence as the level-2 variable.

Results: About 53% children are anemic in the northeastern states of India. Tripura has the highest prevalence of anemia cases (74%) whereas the lowest percentage of anemia cases was in Manipur (42%). Multivariate analysis suggests that age at marriage (OR=1.13, 95% CI: 1.05, 1.21) and the number of children even born (OR=1.09, 95% CI: 1.03, 1.15) have significant effect on being at or below lower hemoglobin level (severely anemic). Furthermore, age of child (OR=0.92, 95% CI: 0.86-1.00) was a significant predictor, indicating that odds of severe anemia decreases if the child is 48 months or older.

Conclusions: The high prevalence of mild and moderate anemia demands due emphasis in the programmes and policies of the government so that the overall prevalence of anemia among children aged 6-59 months can be reduced. Comprehensive strategies need to focus more at state level followed by individual level for combating anemia.

How to cite

Sanku Dey and Enayetur Raheem (2016). Multilevel multinomial logistic regression model for identifying factors associated with anemia in children 6-59 months in northeastern states of India. Cogent Mathematics, Posted Online 1 March 2016

Identifying the factors influencing institutional and non-institutional delivery practices in slums of Shillong city

An attempt has been made in this paper to determine the factors which resulted in preference for Institutional and Non-Institutional deliveries in slum areas of Shillong city. Cross-sectional study was conducted in slums of Shillong City. From a total of 17605 slum dwellers distributed in different proportions in sixteen wards in Shillong city, a representative sample of 1300 slum households was selected from thirty one localities, using an appropriate statistical formula. From these 1300 households, 1417 women were identified as married women eligible for the study. Information of 1417 married women was analyzed to interpret the place of delivery of slum women in relation to respondent’s religion, education, occupation, family income, age at marriage etc. SAS/STAT software was used to analyze the data. The method is based on the well-known statistical technique of factor analysis by which we essentially find out the principal component of the group consisting of various indicators in descending order of their importance. Results from factor analysis show that the total number of ever born children, demand for male child, economic status and customs and religious practices influence both institutional and non-institutional deliveries.

How to cite

Sanku Dey, Enayetur Raheem, and Preenan Sarkar  (2015). Identifying the factors influencing institutional and non-institutional delivery practices in slums of Shillong city. Health and Population: Perspective and Issues 37(3&4):76-87 · December 2014 (Published online: April 2015)

Association of TSH level with first trimester pregnancy loss in anti-TPO antibody negative women in Bangladesh

Objective: To test if TSH level above 2.1 mlU/L is associated with first trimester pregnancy loss in anti-TPO antibody negative women in Bangladesh.

Study Design: An unmatched case-control study was conducted in Bangladesh. Patients were recruited following predefined inclusion and exclusion criteria. Clinical measures were taken as well as data on socioeconomic and physical characteristics were collected. Patients were grouped according to their TSH level—Group I with TSH ≤2.1 mlU/L and Group II with TSH > 2.1 mlU/L.

Results: We found relatively higher number of women in the case group (18) whose TSH level was above 2.1 mlU/L compared to 7 women in control group. In Group I 45.74% had lost pregnancy while 54.26% had continuing pregnancy during the first trimester. Among the Group II patients, 78% had miscarriage and 28% did not have miscarriage. The association between TSH level and first trimester pregnancy loss was statistically significant (p=.0196).In multivariate analysis, odds ratio for TSH level (OR 4.0, 95% CI: 1.44-11.16) indicates that odds of having miscarriage whose TSH level is above 2.1 mlU/L is 4 times compared to those with TSH level below 2.1 mlU/L after adjusting for the effects of age and BMI.

Conclusion: At a global level, the findings of this study provide evidence to the existing discussion on redefining the upper limit of TSH level that is related to first trimester pregnancy loss. At the local level, the results will have direct implication in facilitating management of future pregnancies particularly during the first trimester among Bangladeshi thyroid autoantibody negative women.

How to cite

Jahan Y, Raheem E, Akhteruzzaman M, Hussain MA, Kazal RK, et al. (2015) Association of TSH Level with First Trimester Pregnancy Loss in Anti-TPO Antibody Negative Women in Bangladesh. Med J Obstet Gynecol 3(3): 1061

Adaptation to climate change in coastal saline area of south-western region of Bangladesh

Abstract

An adaptation research under Food and Agriculture Organization (FAO) funded LACC-II project started at coastal saline area of Laudove, Dakope upzila in Khulna district of Bangladesh through on-farm research division of Bangladesh Agricultural Research Institute (BARI), Khulna Bangladesh during May-June 2008 to May-June 2009. It aimed to find out the appropriate adaptation measures against salinity problem. After selection of adaptation options through three focus group discussions (FGD), homestead vegetable production started immediately with saline/excess soil moisture tolerant vegetables for coastal area. In coastal area several homestead vegetables were successfully produced through scientific management like making ridge and furrowing of bed. There was large participation of women in all the activities of home gardening from land preparation to marketing at Laudove. Through utilization of different niches of homestead farm family succeeded to increase their vegetable consumption three to five folds more from the bench mark, though intake was below recommended level. Economically homestead vegetable production was quite lucrative. Also social relationship of the farm family was improved with neighbors and relatives through free distribution of vegetables. Short duration T. aman rice variety (cv. BINAdhan 4) was tested at Laudove for facilitating timely planting of rabi crops. It gave better yield and one month shortening of field duration was possible. Farmers kept most of the produced seed for next year cultivation. In post-rainy season (Rabi) different field crops were tested, such as relaying (for timely planting and avoiding of increased soil salinity) of mustard, wheat, cowpea and later on creeping crops like watermelon and sweet gourd. Among them cowpea, water melon and sweet gourd proved to be promising.

Keywords

Adaptation, Salinity, FGD, Vegetable and Short duration.

How to cite

Yusuf Ali, Shah AL-Emran, M. B. Islam, and E. Raheem (2014). Adaptation to climate change in coastal saline area of south-western region of Bangladesh. Int. J. Sustain. Agril. Tech. 10(4): 09-16, April 2014

Shrinkage and absolute penalty estimation in linear regression models

Abstract

In predicting a response variable using multiple linear regression model, several candidate models may be available which are subsets of the full model. Shrinkage estimators borrow information from the full model and provides a hybrid estimate of the regression parameters by shrinking the full model estimates toward the candidate submodel. The process introduces bias in the estimation but reduces the overall prediction error that offsets the bias. In this article, we give an overview of shrinkage estimators and their asymptotic properties. A real data example is given and a Monte Carlo simulation study is carried out to evaluate the performance of shrinkage estimators compared to the absolute penalty estimators such as least absolute shrinkage and selection operator (LASSO), adaptive LASSO and smoothly clipped absolute deviation (SCAD) based on prediction errors criterion in a multiple linear regression setup. WIREs Comput Stat 2012, 4:541–553. DOI: 10.1002/wics.1232

Keywords
shrinkage estimation; absolute penalty estimation; LASSO; adaptive LASSO; SCAD

How to cite

S. Ejaz Ahmed, and S. E. Raheem, (2012). Shrinkage and absolute penalty estimation in linear modelsWIREs Computational StatisticsVolume 4, Issue 6, pages 541–553, November/December 2012.

Absolute penalty and shrinkage estimation in partially linear models

Abstract
In the context of a partially linear regression model, shrinkage semiparametric estimation is considered based on the Stein-rule. In this framework, the coefficient vector is partitioned into two sub-vectors: the first sub-vector gives the coefficients of interest, i.e., main effects (for example, treatment effects), and the second sub-vector is for variables that may or may not need to be controlled. When estimating the first sub-vector, the best estimate may be obtained using either the full model that includes both sub-vectors, or the reduced model which leaves out the second sub-vector. It is demonstrated that shrinkage estimators which combine two semiparametric estimators computed for the full model and the reduced model outperform the semiparametric estimator for the full model. Using the semiparametric estimate for the reduced model is best when the second sub-vector is the null vector, but this estimator suffers seriously from bias otherwise. The relative dominance picture of suggested estimators is investigated. In particular, suitability of estimating the nonparametric component based on the B-spline basis function is explored. Further, the performance of the proposed estimators is compared with an absolute penalty estimator through Monte Carlo simulation. Lasso and adaptive lasso were implemented for simultaneous model selection and parameter estimation. A real data example is given to compare the proposed estimators with lasso and adaptive lasso estimators.

Keywords
Partially linear model; James–Stein estimator; Absolute penalty estimation; Lasso; Adaptive lasso; B-spline approximation; Semiparametric model; Monte Carlo simulation

How to cite

Raheem, S. E, Ahmed S. E., Doksum K. A. (2012). Absolute penalty and shrinkage estimation in partially linear models. Computational Statistics and Data Analysis. 56(4):874-891.