Machine learning approach for predicting cardiovascular disease in Bangladesh: evidence from a cross-sectional study in 2023.

BACKGROUND: Cardiovascular disorders (CVDs) are the leading cause of death worldwide. Lower- and middle-income countries (LMICs), such as Bangladesh, are also affected by several types of CVDs, such as heart failure and stroke. The leading cause of death in Bangladesh has recently switched from severe infections and parasitic illnesses to CVDs. MATERIALS AND METHODS: The study dataset comprised a random sample of 391 CVD patients' medical records collected between August 2022 and April 2023 using simple random sampling. Moreover, 260 data points were collected from individuals with no CVD problems for comparison purposes. Crosstabs and chi-square tests were used to determine the association between CVD and the explanatory variables. Logistic regression, Naïve Bayes classifier, Decision Tree, AdaBoost classifier, Random Forest, Bagging Tree, and Ensemble learning classifiers were used to predict CVD. The performance evaluations encompassed accuracy, sensitivity, specificity, and area under the receiver operator characteristic (AU-ROC) curve. RESULTS: Random Forest had the highest precision among the five techniques considered. The precision rates for the mentioned classifiers are as follows: Logistic Regression (93.67%), Naïve Bayes (94.87%), Decision Tree (96.1%), AdaBoost (94.94%), Random Forest (96.15%), and Bagging Tree (94.87%). The Random Forest classifier maintains the highest balance between correct and incorrect predictions. With 98.04% accuracy, the Random Forest classifier achieved the best precision (96.15%), robust recall (100%), and high F1 score (97.7%). In contrast, the Logistic Regression model achieved the lowest accuracy of 95.42%. Remarkably, the Random Forest classifier achieved the highest AUC value (0.989). CONCLUSION: This research mainly focused on identifying factors that are critical in impacting patients with CVD and predicting CVD risk. It is strongly advised that the Random Forest technique be implemented in a system for predicting cardiac diseases. This research may change clinical practice by providing doctors with a new instrument to determine a patient's CVD prognosis.

Impact of climate change on dengue incidence in Singapore: time-series seasonal analysis.

This study aimed to identify the meteorological factors that contribute to dengue epidemics. The monthly incidence of dengue was used as the outcome variable, while maximum temperature, humidity, precipitation, and sunshine hours were used as independent variables. The results showed a consistent increase in monthly dengue cases from 2013 to 2021, with seasonal patterns observed in stationary time-series data. The ARIMA (2, 1, 3) × seasonal (0, 1, 2)12 model was used based on its lowest Akaike Information Criterion (AIC) values. The analysis revealed that a 1-unit increase in rainfall was positively correlated with a small 0.062-unit increase in dengue cases, whereas a 1-unit increase in humidity was negatively associated, leading to a substantial reduction of approximately 16.34 cases. This study highlights the importance of incorporating weather data into national dengue prevention programs to enhance public awareness and to promote recommended safety measures.

Corrigendum to "Impact of COVID-19 pandemic on the mental health of children in Bangladesh: A cross-sectional study" [Child. Youth Serv. Rev. 117 (2020) 105277].

[This corrects the article DOI: 10.1016/j.childyouth.2020.105277.].

Regional education and wealth-related inequalities in malnutrition among women in Bangladesh.

OBJECTIVES: This paper examines the associations of socio-economic and demographic correlates with malnutrition among women and investigates education and wealth-related inequalities in malnutrition among women by region. DESIGN: We utilise a two-level mixed-effects logistic regression model to evaluate the associations and employ the concentration, Wagstaff and Erreygers's correction indices to measure socio-economic inequalities in malnutrition among women. SETTING: Bangladesh Demographic and Health Survey data. PARTICIPANTS: Non-pregnant women aged 15-49 years. RESULTS: We find evidence of a significant cluster effect in the data. Women's age, marital status, total children ever born, education level, husband's/partner's education level, residence and wealth index appear to be significantly associated with women underweight and overweight/obesity status. Underweight status is higher among less-educated women and women from poor households, whereas overweight/obesity is more concentrated among higher educated women and women from wealthy households. The southwestern region of the country demonstrates lower education and wealth-related inequalities in malnutrition among women. In contrast, the central and the northeastern areas apparently experience the highest education and wealth-related inequalities in malnutrition among women. The regional differences in predicted probabilities of being underweight shrink at higher education level and the richest quintile, whereas the differences in overweight/obese diminish at the primary education level and lower quintile households. CONCLUSIONS: Our findings strengthen the evidence base for effective regional policy interventions to mitigate education and wealth-related inequalities in malnutrition among women. There is a need for developing regional awareness programmes and establishing regional monitoring cells to ensure proper health and nutrition facilities in underprivileged regions.

Body mass index of women in Bangladesh: comparing Multiple Linear Regression and Quantile Regression.

This study explored the association between socio-demographic factors and the body mass index (BMI) of women of reproductive age (15-49 years) in Bangladesh. Data from the 2014 Bangladesh Demographic and Health Survey (BDHS-14) were analysed using Multiple Linear Regression (MLR) and Quantile Regression (QR) analyses. The study sample comprised 15,636 non-pregnant women aged 15-49. The mean BMI of the women was 22.35±4.12 kg/m2. Over half (56.75%) had a BMI in the normal range (18

Impact of COVID-19 pandemic on the mental health of children in Bangladesh: A cross-sectional study.

COVID-19 pandemic poses a significant mental health threat among children in Bangladesh. This study aims to explore the impact of COVID-19 on the mental health of children during the lockdown in Bangladesh. An online cross-sectional study was conducted from 25th April to 9th May 2020 among 384 parents having at least one child aged between 5-15 years using non-probability sampling. K-means clustering used to group children according to mental health score and confirmatory factor analysis (CFA) performed to identify the relationship among the parental behavior and child mental health, and also these associations were assessed through chi-square test. Children were classified into four groups where 43% of child had subthreshold mental disturbances (mean Major Depressive Disorder (MDD)-10; 2.8), 30.5% had mild (mean MDD-10; 8.9), 19.3% suffered moderately (mean MDD-10; 15.9), and 7.2% of child suffered from severe disturbances (mean MDD-10; 25.2). The higher percentage of mental health disturbances of children with the higher education level of parents, relative infected by COVID-19 (yes), parents still need to go the workplace (yes), and parent's abnormal behavior but lower to their counterparts. This paper demonstrates large proportions of children are suffering from mental health disturbances in Bangladesh during the period of lockdown. Implementation of psychological intervention strategies and improvement in house-hold financial conditions, literacy of parents, taking care of children, and job security may help in improving the psychological/mental status of children and the authors believe that the findings will be beneficial to accelerate the rate of achieving the Sustainable Development Goal (SDG) linked to health status in Bangladesh.

Association between the salinity level with miscarriage and unintended pregnancy in Bangladesh: Impact of salinity level on miscarriage and unintended pregnancy in Bangladesh.

Miscarriage is a significant public health concern worldwide, particularly in developing nations like Bangladesh. Moreover, people in coastal areas are more affected by miscarriage as compared to other areas. Increasing sea levels and salinity is the main reason for this discrepancy. This study aimed to investigate the association between different salinity levels (S1, S2, S3, S4, and S5) and miscarriage and unintended pregnancy. The outcome variables are pregnancy-related outcomes (miscarriage, unintended pregnancy), and the independent variables are different salinity levels. A frequency table and correlation analysis were done to find the descriptive scenarios of miscarriage, unintended pregnancy, and salinity levels. We found 621 miscarriage patients and 2271 unintended pregnant patients in our study. Furthermore, the Poisson regression model was used to observe the incidence of miscarriage and unintended pregnancy for different salinity levels. A higher amount of miscarriage and unintended pregnancy rate was found in Dhaka and Khulna, while these rates were lower in Barisal and Chittagong. However, the salinity levels were highest in Barisal and Khulna. Both miscarriage and unintended pregnancy are highly and negatively correlated with salinity levels. The Poisson regression model shows that the salinity levels s1-s5 are strongly associated with miscarriage. Lower and moderate levels of salinity are strongly associated with miscarriage than higher levels of salinity. Again, the average number of miscarriages decreases with the salinity levels. Likewise, unintended pregnancy was also negatively associated with salinity levels. However, it only reported a significant association with lower and moderate salinity levels, and higher salinity levels did not affect unintended pregnancy. Taking initiatives for raising awareness from government and non-government organizations, setting up deep tube water pumps extensively, and properly treating coastal areas women during pregnancy would be the ideal next step to reduce the miscarriage and unintended pregnancy rate in coastal zones in Bangladesh.

Hypertension, diabetes, and hypercholesterolemia in Bangladesh: Evaluating role of physical activity from cross-sectional STEPS 2018 survey.

Background and Aims: The objective of this study was to assess current condition of three noncommunicable diseases (NCDs)-diabetes, hypertension, and hypercholesterolemia and vulnerable cohorts among adults in Bangladesh. Methods: The STEPwise Approach to the NCD Risk Factor Surveillance survey of 2018 was analyzed to evaluate the association between NCDs and sociodemographic factors, levels of physical activity, and behavioral measurements. Complex survey weight-adjusted logistic models were fitted. Results: From 6875 samples, prevalence of diabetes, hypertension and hypercholesterolemia were 27.3%, 9.8%, and 30.2%, respectively. The prevalence of the three NCDs were higher among older respondents. Low occupational activity with obese-BMI status people was more prone to NCDs. Lower prevalence of hypercholesterolemia was observed in Rangpur, Rajshahi, and Mymensingh, whereas lesser diabetes in Khulna region. Likelihood of NCDs were higher among highly educated, urban residents with low physical activity. Lesser physical activity, and lack of weight maintenance were found to be key factors in higher rates of NCDs in Bangladesh. Conclusion: Greater education and less strenuous profession led to a higher chance of NCDs. Overall, physical activity and maintenance of weight seem to be factors driving higher rates of NCDs in Bangladesh. As a preventive measure against NCDs, an active lifestyle is to be encouraged, particularly to the most vulnerable cohort.

Addictive behavior and mental health of adolescents aged 11-17 years in Bangladesh in 2014: A cross-sectional study.

Background and Aims: This study aimed to evaluate the factors associated with addictive behavior and mental health in adolescents aged 11-17 years in Bangladesh. Methods: This study analyzed data from the Bangladesh Global School-based Health Survey (GSHS) conducted in 2014. Adolescents aged 11-17 years studying government schools were considered as respondents for this survey. A two-phase group sample design was utilized to deliver illustrative information of all understudies in grades (classes) 7-10 in Bangladesh. Bivariate analyses followed by a weight-adjusted multiple logistic regression was fitted to a sample of size 2989 adolescents. Results: One in ten and one in four adolescents had different substance addictions and some forms of mental health conditions, respectively. Sex of participants (adjusted odds ratio [AOR]: 4.49; confidence interval [CI]:2.28-8.84), being bullied (AOR: 3.08; CI:1.46-6.49), use of tobacco among parents (AOR: 5.82; CI:3.16-10.75), parental understanding of adolescents' problems (AOR: 0.45; CI:0.23-0.82), and food affordability (AOR: 1.24; CI:1.09-1.42) were associated with addictive behaviors of adolescents. Bullied males with nonempathetic parents were found to be more vulnerable to addictive behaviors and bullied females showed higher tendencies to mental health issues. Conclusion: Considering the elevated prevalence of reported mental health concerns, identifying vulnerable groups and formulating intervention-oriented policies engaging youths can pave the way towards achieving robust health and well-being for them in Bangladesh.

Contraceptive uses among married women in Bangladesh: a systematic review and meta-analyses.

BACKGROUND: Although Bangladesh's economy has shown significant improvement over the past two decades, the high population growth rate has hindered development efforts. This study aimed to review the prevalence of different contraceptive methods used among women of reproductive age in Bangladesh, which could help control the population growth rate. METHODS: We conducted an extensive literature review and meta-analysis of 82 articles, identifying 20 articles for analysis. The analyses included heterogeneity and publication bias in published papers. RESULTS: The pooled prevalence of various contraceptive methods was as follows: condom use, 7.13%; Female Sterilization, 8.09%; injectables, 12.76%; intrauterine devices (IUDs), 3.76%; male sterilization, 2.34%; periodic abstinence, 6.71%; pills, 33.21%; and withdrawal, 3.27%. Modern contraceptive methods accounted for 62.91% of usage, while traditional methods constituted 8.79%. On average, only 1.95% of women opted for the implant method. The overall prevalence of contraceptive method usage was 59.48%, with 60.59% in urban areas and 54.54% in rural areas. We found statistically significant heterogeneity for all contraceptive methods used by women in Bangladesh. The funnel plot and Egger's test showed no publication bias for any of the contraceptive methods, except condoms (Z = 2.34, P = 0.0194). The contraceptive methods used by women in rural areas also showed publication bias (Z = -3.04, P = 0.0024). CONCLUSION: A renewed commitment from government bodies and independent organizations is needed to implement and monitor family planning strategies to ensure adherence to and provision of the most appropriate contraceptive method for couples.

Prevalence of HIV and syphilis and their co-infection among men having sex with men in Asia: A systematic review and meta-analysis.

Background: Studies found that the group of men who have sex with men (MSM) is at a very high level of risk of HIV and sexually transmitted infections (STIs) in Asian regions due to multiple reasons. Although the prevalence of HIV among general people in Asia is considered low, the prevalence of HIV and Syphilis among MSM in this region was found very high and usually, it goes unnoticed. This study aimed to inspect the prevalence of and trends in HIV, Syphilis, and their co-infection among MSM in Asia. Methods: A systematic search was performed on January 5, 2021, in PubMed, Web of Science, and Google Scholar databases. To evaluate the heterogeneity, Q-tests, and I 2 were used. To explore the publication bias, Eggers' test and funnel plot were used. The random-effect model and subgroup analysis were performed due to the significant heterogeneity. Results: A total of 2872 articles were identified, and 66 articles were included in the final analysis. The overall prevalence of HIV and Syphilis among MSM was estimated considering 69 estimates from 66 studies whereas 19 estimates of co-infection were found in 17 studies. The pooled HIV prevalence was 8.48% (CI: 7.01-9.95) and the pooled Syphilis prevalence was 9.86% (CI: 8.30-11.41) with significant heterogeneity and publication bias. The pooled prevalence of HIV and Syphilis co-infection was 2.99% (CI: 1.70-4.27) with significant heterogeneity and no publication bias. The HIV, Syphilis, and HIV-Syphilis co-infection prevalence estimates exhibited an upward trend during 2002-2017. Conclusions: HIV, Syphilis, and their co-infection are quite prevalent among MSM in the Asia-Pacific region. Integrated and intensified intervention strategies, HIV testing, and improved access to antiretroviral treatment as well as increased awareness are needed to reduce HIV, Syphilis, and their co-infection among the discussed vulnerable group.

Association of climate factors with dengue incidence in Bangladesh, Dhaka City: A count regression approach.

Background: In Bangladesh, particularly in Dhaka city, dengue fever is a major factor in serious sickness and hospitalization. The weather influences the temporal and geographical spread of the vector-borne disease dengue in Dhaka. As a result, rainfall and ambient temperature are considered macro factors influencing dengue since they have a direct impact on Aedes aegypti population density, which changes seasonally dependent on these critical variables. This study aimed to clarify the relationship between climatic variables and the incidence of dengue disease. Methods: A total of 2253 dengue and climate data were used for this study. Maximum and minimum temperature (°C), humidity (grams of water vapor per kilogram of air g.kg-1), rainfall (mm), sunshine hour (in (average) hours per day), and wind speed (knots (kt)) in Dhaka were considered as the independent variables for this study which trigger the dengue incidence in Dhaka city, Bangladesh. Missing values were imputed using multiple imputation techniques. Descriptive and correlation analyses were performed for each variable and stationary tests were observed using Dicky Fuller test. However, initially, the Poisson model, zero-inflated regression model, and negative binomial model were fitted for this problem. Finally, the negative binomial model is considered the final model for this study based on minimum AIC values. Results: The mean of maximum and minimum temperature, wind speed, sunshine hour, and rainfall showed some fluctuations over the years. However, a mean number of dengue cases reported a higher incidence in recent years. Maximum and minimum temperature, humidity, and wind speed were positively correlated with dengue cases. However, rainfall and sunshine hours were negatively associated with dengue cases. The findings showed that factors such as maximum temperature, minimum temperature, humidity, and windspeed are crucial in the transmission cycles of dengue disease. On the other hand, dengue cases decreased with higher levels of rainfall. Conclusion: The findings of this study will be helpful for policymakers to develop a climate-based warning system in Bangladesh.

The acceptance of COVID-19 vaccine at early stage of development and approval: A global systematic review and meta-analysis.

Background: Vaccination seems to be the most effective way to prevent and control the spread of COVID-19, a disease that has adversely impacted the lives of over 7 billion people across the globe. Vaccine hesitancy represents an important threat to combat infectious diseases worldwide. This study aims to inspect the COVID-19 vaccine acceptance rate worldwide and the regional variation of the acceptance rates among the general population and healthcare workers across different territories of the world. In addition, it compares the vaccine acceptance rates between the pre- and post-vaccine approval periods. Method: A comprehensive systematic review was conducted using PRISMA statements. After quality evaluation, the data from eligible studies were analyzed using the random effect model. Q-test and I 2 statistics were used to search for heterogeneity. The publication bias was assessed by using Egger's test and funnel plot. Results: The combined COVID-19 vaccine acceptance rate among the general population and healthcare workers (n = 1,581,562) was estimated at 62.79% (95% CI: 58.98-66.60). The acceptance rate substantially decreased from 66.29% (95% CI: 61.24-71.35) to 56.69% (95% CI: 48.68-64.71) among the general population from the pre-to post-vaccine approval periods but remained almost constant at 58.25% (95% CI: 46.52-69.97) among healthcare workers. The acceptance rates also varied in different regions of the world. The highest acceptance rate was found in the South-East Asia region at 70.18% (95% CI: 58.12-82.25) and the lowest was found in African Region at 39.51% (95% CI: 23.42-55.59). Conclusion: Low COVID-19 vaccine acceptance rate might be a massive barrier to controlling the pandemic. More research is needed to address the responsible factors influencing the low global rate of COVID-19 vaccine acceptance. Integrated global efforts are required to remove the barriers.

Impact of Prolonged COVID-19 Lockdown on Body Mass Index, Eating Habits, and Physical Activity of University Students in Bangladesh: A Web-Based Cross-Sectional Study.

Objectives: This current study aims to assess the prevalence and factors associated with body mass index (BMI), dietary patterns, and the extent of physical activities among university students following the prolonged coronavirus disease 2019 (COVID-19) lockdown in Bangladesh. Methods: A cross-sectional web-based survey was conducted between July 10 to August 10, 2021, through a pre-designed Google Form to collect the data from Bangladeshi university students (age: ≥18 years). Informed consent was electronically obtained from each participant, and a simple snowball technique was employed during the sampling. Frequency and percentage distribution, paired t-test, chi-square [χ2] test, and multinomial and binary logistic regression analyses were consecutively applied to analyze the collected data. Results: Among the total participants (n = 1,602), 45.1% were female and 55.6% were 22-25 years' age group students. The BMI (mean ± standard deviation, SD) during the COVID-19 lockdown was 23.52 ± 7.68 kg/m2, which was 22.77 ± 4.11 kg/m2 during the pre-lockdown period (mean difference = 0.753; p < 0.001). The multinomial logistic regression analysis found a significant impact of gender [male vs. female: adjusted relative risk ratio (RRR) = 1.448; 95% confidence interval (CI) = 1.022, 2.053; p = 0.037], age (years) (<22 vs. >25: RRR =0.389, 95% CI = 0.213,0.710; p = 0.002, and 22-25 vs. >25: RRR = 0.473, 95% CI = 0.290, 0.772; p = 0.003), monthly family income (BDT) (<25,000 vs. >50,000: RRR = 0.525, 95% CI = 0.334,0.826; p = 0.005), university type (public vs. private: RRR = 0.540, 95% CI = 0.369, 0.791; p = 0.002), eating larger meals/snacks (increased vs. unchanged: RRR = 2.401, 95% CI = 1.597, 3.610; p < 0.001 and decreased vs. unchanged: RRR = 1.893, 95% CI = 1.218, 2.942; p = 0.005), and verbally or physically abuse (yes vs. no: RRR = 1.438, 95% CI = 0.977, 2.116; p = 0.066) on obesity during COVID-19 pandemic. Besides, the female students and those who have constant eating habits, were more likely to be underweight. Additionally, the binary logistic regression analysis found that the students from private universities [others vs. private: adjusted odds ratio (AOR) = 0.461, 95% CI = 0.313, 0.680; p < 0.001], urban areas (urban vs. rural: AOR = 1.451, 95% CI = 1.165, 1.806; p = 0.001), wealthier families (<25,000 BDT vs. >50,000 BDT: AOR = 0.727, 95% CI = 0.540, 0.979; p = 0.036), and who were taking larger meals/snacks (increased vs. unchanged: AOR = 2.806, 95% CI = 2.190, 3.596; p < 0.001) and had conflicts/arguments with others (no vs. yes: AOR = 0.524, 95% CI = 0.418, 0.657; p < 0.001), were significantly more physically inactive. Finally, the level of education and smoking habits significantly influenced the eating habits of university students during the extended strict lockdown in Bangladesh. Conclusion: The current findings would be helpful tools and evidence for local and international public health experts and policymakers to reverse these worsening effects on students mediated by the prolonged lockdown. Several effective plans, programs, and combined attempts must be earnestly implemented to promote a smooth academic and daily life.

Child Marriage and Adolescent Motherhood: A Nationwide Vulnerability for Women in Bangladesh.

The persistently high prevalence of girl-child marriage and adolescent motherhood is a public health concern in Bangladesh. This study investigated the division-wise prevalence and the influence of education and religious affiliation on child marriage and adolescent motherhood among women in Bangladesh along with their consequences using 15,474 women aged 15-49 years from the Bangladesh Demographic and Health Survey 2017-18. Staggeringly, 82.5% women were married before 18, 43.1% were married before 15, and 61.8% gave birth before 18 years of age. Binary logistic regression models for the complex survey showed that girl-children with primary, secondary, and higher secondary or above education were 16% (95% CI: 0.69, 1.03), 32% (95% CI: 0.55, 0.84), and 87% (95% CI: 0.10, 0.17) less likely to get married <18 years of age, respectively, compared to the uneducated. Also, girl-children with secondary and higher education were 21 and 83% less likely to become adolescent mothers, respectively, than the uneducated. Women married during childhood (<18 years) and adolescent mothers were 36 and 55% less likely to continue studies after marriage, respectively, and expressed that they significantly preferred a late marriage. Policy interventions need to address culturally-laden social norms influenced by religious-related beliefs, especially in rural areas.

The global prevalence of depression, anxiety, stress, and, insomnia and its changes among health professionals during COVID-19 pandemic: A rapid systematic review and meta-analysis.

BACKGROUND: During the COVID-19 pandemic, the health professionals who are at the frontline of this crisis have been facing extreme psychological disorders. This research aims to provide an overall scenario of the prevalence of depression, anxiety, stress, as well as insomnia and to inspect the changes in these prevalence over time by analyzing the existing evidence during this COVID-19 pandemic. METHODS: A systematic search was performed on March 30, 2021, in PubMed, MEDLINE, Google Scholar databases, and Web of Science. To assess the heterogeneity, Q-test, I 2 statistics, and Meta regression and to search for the publication bias, Eggers's test and funnel plot were used. The random-effect model and subgroup analysis were performed due to the significant heterogeneity. RESULTS: Among eighty-three eligible studies in the final synthesis, 69 studies (n = 144649) assessed the depression prevalence of 37.12% (95% CI: 31.80-42.43), 75 studies (n = 147435) reported the anxiety prevalence of 41.42% (95% CI: 36.17-46.54), 41 studies (n = 82783) assessed the stress prevalence of 44.86% (95% CI: 36.98-52.74), 21 studies (n = 33370) enunciated the insomnia prevalence of 43.76% (95% CI: 35.83-51.68). The severity of the mental health problems among health professionals increased over the time during January 2020 to September 2020. LIMITATIONS: A significant level of heterogeneity was found among psychological measurement tools and across studies. CONCLUSIONS: Therefore, it is an emergency to develop psychological interventions that can protect the mental health of vulnerable groups like health professionals.