M. Iftakhar Alam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
M. Iftakhar Alam has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
Mohaimen Mansur
Affiliation Institute of Statistical Research and Training, University of Dhaka, Dhaka, Bangladesh
5 shared publicationsBarbara Bogacka
Queen Mary University of London
3 shared publicationsD. Stephen Coad
Queen Mary University of London
3 shared publicationsNigar Sultana
Institute of Statistical Research and Training, University of Dhaka, Dhaka, Bangladesh
3 shared publications- Ahmed Hossain
College of Health Sciences, University of Sharjah, Sharjah, United Arab Emirates
2 shared publications - Hasinur Rahaman Khan
Institute of Statistical Research and Training, University of Dhaka, Dhaka, Bangladesh
2 shared publications
Latest Publications
- Open Access
- 20 March 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (3) , e0342297
Effect of maternal education on the literacy-numeracy development of children: A propensity-score matched analysis
This study examined the causal relationship between maternal education and children’s literacy-numeracy development in Bangladesh using nationally representative data from the Bangladesh Multiple Indicator Cluster Survey. The analysis focused on 9,454 children aged 3–4 years. To strengthen causal inference, we applied propensity score matching, which reduces selection bias by balancing observed covariates between groups. From the analysis, we found that the average treatment effect of the treated was almost 18.4%, which means after matching, the probability of literacy-numeracy development in children was 18.4% higher among the mothers who had higher level of education compared to the same mothers had they not higher level of education. Subgroup analyses further revealed heterogeneity by place of residence: the effect of maternal education was stronger in urban areas (26.2%) than in rural areas (16.3%), suggesting that contextual factors may amplify or constrain the benefits of maternal education. These findings provide an evidence of the critical role of maternal education in early childhood development.
- 20 March 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (3) , e0342297
Open AccessEffect of maternal education on the literacy-numeracy development of children: A propensity-score matched analysis
This study examined the causal relationship between maternal education and children’s literacy-numeracy development in Bangladesh using nationally representative data from the Bangladesh Multiple Indicator Cluster Survey. The analysis focused on 9,454 children aged 3–4 years. To strengthen causal inference, we applied propensity score matching, which reduces selection bias by balancing observed covariates between groups. From the analysis, we found that the average treatment effect of the treated was almost 18.4%, which means after matching, the probability of literacy-numeracy development in children was 18.4% higher among the mothers who had higher level of education compared to the same mothers had they not higher level of education. Subgroup analyses further revealed heterogeneity by place of residence: the effect of maternal education was stronger in urban areas (26.2%) than in rural areas (16.3%), suggesting that contextual factors may amplify or constrain the benefits of maternal education. These findings provide an evidence of the critical role of maternal education in early childhood development.
- Open Access
- 20 February 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (2) , e0343489
Association of anxiety and depression with cognitive challenges in children in Bangladesh
This study primarily investigated the association of anxiety or depression (AD) with communication, learning, remembering, or concentrating (CLRC) difficulties in children aged 5-17 years in Bangladesh. Data were taken from the 2019 Bangladesh Multiple Indicator Cluster Survey (MICS). Random intercept binary logistic regression model was used to assess the association between CLRC and AD, after controlling sex, age, division, maternal education, maternal functional difficulty, and wealth index. Children with AD difficulties had significantly higher odds of CLRC (OR = 7.82; 95% CI: 6.29–9.73). Males were 30% more likely to have CLRC than females (OR = 1.30; 95% CI: 1.12–1.51). Compared with children living in Barishal-the reference administrative region-children residing in several other divisions had significantly lower odds of experiencing CLRC. Children of mothers with secondary education had reduced odds of CLRC (OR = 0.80), while those whose mothers had functional difficulties had higher odds (OR = 3.00). To conclude, AD difficulties is strongly associated with CLRC, alongside other factors like sex, geographic location, maternal education, and wealth. Addressing these factors is crucial for reducing CLRC difficulties in children.
- 20 February 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (2) , e0343489
Open AccessAssociation of anxiety and depression with cognitive challenges in children in Bangladesh
This study primarily investigated the association of anxiety or depression (AD) with communication, learning, remembering, or concentrating (CLRC) difficulties in children aged 5-17 years in Bangladesh. Data were taken from the 2019 Bangladesh Multiple Indicator Cluster Survey (MICS). Random intercept binary logistic regression model was used to assess the association between CLRC and AD, after controlling sex, age, division, maternal education, maternal functional difficulty, and wealth index. Children with AD difficulties had significantly higher odds of CLRC (OR = 7.82; 95% CI: 6.29–9.73). Males were 30% more likely to have CLRC than females (OR = 1.30; 95% CI: 1.12–1.51). Compared with children living in Barishal-the reference administrative region-children residing in several other divisions had significantly lower odds of experiencing CLRC. Children of mothers with secondary education had reduced odds of CLRC (OR = 0.80), while those whose mothers had functional difficulties had higher odds (OR = 3.00). To conclude, AD difficulties is strongly associated with CLRC, alongside other factors like sex, geographic location, maternal education, and wealth. Addressing these factors is crucial for reducing CLRC difficulties in children.
- Open Access
- 6 December 2025
- Published by Springer Nature in Discover Social Science and Health
- Vol. 6 (1)
Determinants of inequities in access to healthcare among transgender people in Bangladesh
- 6 December 2025
- Published by Springer Nature in Discover Social Science and Health
- Vol. 6 (1)
Open AccessDeterminants of inequities in access to healthcare among transgender people in Bangladesh
- 21 October 2025
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 96 (4) , 964-978
A combined criterion for optimizing doses in phase I clinical trials
- 21 October 2025
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 96 (4) , 964-978
A combined criterion for optimizing doses in phase I clinical trials
- 10 July 2024
- Published by Taylor & Francis in Journal of Biopharmaceutical Statistics
- Vol. 36 (2) , 151-176
A constrained optimum adaptive design for dose finding in early phase clinical trials
Recently, interest has grown in the development of dose-finding methods that consider both toxicity and efficacy as endpoints. Along with responses on these, the incorporation of pharmacokinetic (PK) data can be beneficial in terms of patients' safety and can also increase the efficiency of the design for finding the best dose for the next phase. In this paper, the maximum concentration () is used as the PK measure guiding the dose selection. The ethically attractive approach, which is based on the probability of efficacy, is used as a dose optimisation criterion. At each stage of an adaptive trial, that dose is selected for which the criterion is maximised, subject to the constraints imposed on the and the probability of toxicity. The inter-patient variability of the PK model parameters is considered, and population -optimal sampling time points for measuring the concentration of a drug in the blood are calculated. The method is illustrated with a one-compartment PK model with first-order absorption, with the parameters being assumed to be random. The Cox model for bivariate binary responses is employed to model the dose-response outcomes. The results of a simulation study for several plausible dose-response scenarios show a significant gain in the efficiency of the design, as well as a reduction in the proportion of toxic responses.- 10 July 2024
- Published by Taylor & Francis in Journal of Biopharmaceutical Statistics
- Vol. 36 (2) , 151-176
A constrained optimum adaptive design for dose finding in early phase clinical trials
Recently, interest has grown in the development of dose-finding methods that consider both toxicity and efficacy as endpoints. Along with responses on these, the incorporation of pharmacokinetic (PK) data can be beneficial in terms of patients' safety and can also increase the efficiency of the design for finding the best dose for the next phase. In this paper, the maximum concentration () is used as the PK measure guiding the dose selection. The ethically attractive approach, which is based on the probability of efficacy, is used as a dose optimisation criterion. At each stage of an adaptive trial, that dose is selected for which the criterion is maximised, subject to the constraints imposed on the and the probability of toxicity. The inter-patient variability of the PK model parameters is considered, and population -optimal sampling time points for measuring the concentration of a drug in the blood are calculated. The method is illustrated with a one-compartment PK model with first-order absorption, with the parameters being assumed to be random. The Cox model for bivariate binary responses is employed to model the dose-response outcomes. The results of a simulation study for several plausible dose-response scenarios show a significant gain in the efficiency of the design, as well as a reduction in the proportion of toxic responses.- Open Access
- 29 April 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (4) , e0302059
Prevalence and determinants of discrimination or harassment of women: Analysis of cross-sectional data from Bangladesh
This study aimed to explore the prevalence and determinants of discrimination against or harassment among Bangladeshi women. The nationally representative cross-sectional data of the Multiple Indicator Cluster Survey 2019 were employed in the study. Women aged 15-49 were asked whether they felt discrimination or harassment due to seven potential reasons in the last twelve months before the survey. The outcome was a binary variable indicating whether a woman has experienced discrimination or harassment for any of the seven reasons. Responses were obtained from 64378 women and analysed through bivariate and multivariate procedures. The data had a hierarchical structure since women were nested within the clusters. A mixed-effects logistic regression model was used to analyse the data appropriately. A strong association between discrimination or harassment was seen with the functional difficulties of women. Notably, women with functional difficulties were 1.629 times more likely to experience discrimination or harassment than women without such difficulties. Other significant factors were the respondents’ area, age, education, current marital status, wealth index quintile and ethnicity. The study concludes that education, marital status, functional difficulties and wealth index are the most important determinants of discrimination or harassment of women.
- 29 April 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (4) , e0302059
Open AccessPrevalence and determinants of discrimination or harassment of women: Analysis of cross-sectional data from Bangladesh
This study aimed to explore the prevalence and determinants of discrimination against or harassment among Bangladeshi women. The nationally representative cross-sectional data of the Multiple Indicator Cluster Survey 2019 were employed in the study. Women aged 15-49 were asked whether they felt discrimination or harassment due to seven potential reasons in the last twelve months before the survey. The outcome was a binary variable indicating whether a woman has experienced discrimination or harassment for any of the seven reasons. Responses were obtained from 64378 women and analysed through bivariate and multivariate procedures. The data had a hierarchical structure since women were nested within the clusters. A mixed-effects logistic regression model was used to analyse the data appropriately. A strong association between discrimination or harassment was seen with the functional difficulties of women. Notably, women with functional difficulties were 1.629 times more likely to experience discrimination or harassment than women without such difficulties. Other significant factors were the respondents’ area, age, education, current marital status, wealth index quintile and ethnicity. The study concludes that education, marital status, functional difficulties and wealth index are the most important determinants of discrimination or harassment of women.
- Open Access
- 21 March 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (3) , e0300403
Functional difficulty among young children in Bangladesh: An analysis of nationally representative data
Functional difficulty in children is a crucial public health problem still undervalued in developing countries. This study explored the socio-demographic factors and anthropometry associated with children’s functional difficulty in Bangladesh. Data for 2-4-year-old children, obtained from Multiple Indicator Cluster Survey 2019, were used in this study. The mixed-effects logistic regression model was used to analyse the data. Children whose mothers had functional difficulty were found to be 2.75 times more likely to have functional difficulty than children whose mothers had no functional difficulty (95% CI 1.63-4.63). Male children were more likely to experience functional difficulty than female children (OR = 1.48). Furthermore, stunting was found to be significantly associated with functional difficulty (OR = 1.50). The study also revealed that division and mother’s education, specifically, children with mothers having higher secondary + education, had significant association with the outcome variable. The findings provided a vital overview of child disability in a developing country.
- 21 March 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (3) , e0300403
Open AccessFunctional difficulty among young children in Bangladesh: An analysis of nationally representative data
Functional difficulty in children is a crucial public health problem still undervalued in developing countries. This study explored the socio-demographic factors and anthropometry associated with children’s functional difficulty in Bangladesh. Data for 2-4-year-old children, obtained from Multiple Indicator Cluster Survey 2019, were used in this study. The mixed-effects logistic regression model was used to analyse the data. Children whose mothers had functional difficulty were found to be 2.75 times more likely to have functional difficulty than children whose mothers had no functional difficulty (95% CI 1.63-4.63). Male children were more likely to experience functional difficulty than female children (OR = 1.48). Furthermore, stunting was found to be significantly associated with functional difficulty (OR = 1.50). The study also revealed that division and mother’s education, specifically, children with mothers having higher secondary + education, had significant association with the outcome variable. The findings provided a vital overview of child disability in a developing country.
- 31 July 2023
- Published by Springer Nature in Journal of Public Health
- Vol. 33 (2) , 435-443
Skilled birth attendants and caesarean sections: a case in Bangladesh
- 31 July 2023
- Published by Springer Nature in Journal of Public Health
- Vol. 33 (2) , 435-443
Skilled birth attendants and caesarean sections: a case in Bangladesh
- 13 September 2022
- Published by Taylor & Francis in Sequential Analysis
- Vol. 41 (3) , 1-20
Patient-specific dose finding in seamless phase I/II clinical trials
This article incorporates a covariate to determine the optimum dose in a seamless phase I/II clinical trial. A binary covariate and its interaction effect are assumed to keep the method simple. Each patient’s outcome is assumed to be trinomial, and the continuation ratio model is utilized to model the dose–response data. The Bayesian approach estimates parameters of the dose–response model. Eight plausible dose–response scenarios are investigated to check the proposed methodology. A simulation study shows that covariate consideration can enhance the identification of the optimum dose when it is appropriate to do.
- 13 September 2022
- Published by Taylor & Francis in Sequential Analysis
- Vol. 41 (3) , 1-20
Patient-specific dose finding in seamless phase I/II clinical trials
This article incorporates a covariate to determine the optimum dose in a seamless phase I/II clinical trial. A binary covariate and its interaction effect are assumed to keep the method simple. Each patient’s outcome is assumed to be trinomial, and the continuation ratio model is utilized to model the dose–response data. The Bayesian approach estimates parameters of the dose–response model. Eight plausible dose–response scenarios are investigated to check the proposed methodology. A simulation study shows that covariate consideration can enhance the identification of the optimum dose when it is appropriate to do.
- 24 March 2022
- Published by SAGE Publications in Violence and Gender
- Vol. 9 (2) , 87-95
Attitudes of Women Toward Domestic Violence: What Matters the Most?
Domestic violence (DV) is a global public health concern as it has a profound negative impact on women's health. Investigating DV itself was not the concern of this study, it focused on the acceptance of DV by women in Bangladesh. The specific purpose was to explore the prevalence of women's attitudes toward DV and the associated factors. The nationally representative data of the Multiple Indicator Cluster Survey 2019 were employed in the study. The 15–49-year-old women were asked whether they agreed to be hit or beaten by their husbands for five potential reasons. The outcome variable was a binary variable indicating whether a woman agreed on any of the five reasons. Responses were obtained from 64,378 women and analyzed through bivariate and multivariate procedures. The data had a hierarchical structure since women were nested within the clusters. A mixed-effects logistic regression model was used for the appropriate analysis of the data. The results revealed that many covariates were highly associated with the outcome variable. Education, functional difficulties, ethnicity, and wealth index appeared as the major contributing factors for accepting DV. The study recommends emphasizing education, poverty alleviation, and exposure to mass media to strengthen women's awareness and capabilities to stand against DV.
- 24 March 2022
- Published by SAGE Publications in Violence and Gender
- Vol. 9 (2) , 87-95
Attitudes of Women Toward Domestic Violence: What Matters the Most?
Domestic violence (DV) is a global public health concern as it has a profound negative impact on women's health. Investigating DV itself was not the concern of this study, it focused on the acceptance of DV by women in Bangladesh. The specific purpose was to explore the prevalence of women's attitudes toward DV and the associated factors. The nationally representative data of the Multiple Indicator Cluster Survey 2019 were employed in the study. The 15–49-year-old women were asked whether they agreed to be hit or beaten by their husbands for five potential reasons. The outcome variable was a binary variable indicating whether a woman agreed on any of the five reasons. Responses were obtained from 64,378 women and analyzed through bivariate and multivariate procedures. The data had a hierarchical structure since women were nested within the clusters. A mixed-effects logistic regression model was used for the appropriate analysis of the data. The results revealed that many covariates were highly associated with the outcome variable. Education, functional difficulties, ethnicity, and wealth index appeared as the major contributing factors for accepting DV. The study recommends emphasizing education, poverty alleviation, and exposure to mass media to strengthen women's awareness and capabilities to stand against DV.