Mask adoption, symptom burden, and machine learning insights from Bangladesh during and after COVID-19

A comprehensive investigation of mask usage, recent symptom burden, and post-pandemic health behavior in Bangladesh was conducted using a cross-sectional survey framework that integrated descriptive statistics, inferential analysis, theoretical comparison, and machine-learning methods. Data from respondents across diverse demographic and residential strata were analyzed to examine mask-related attitudes, willingness to pay, willingness to purchase, age-specific vaccination patterns, and recent self-reported symptoms. Chi-square tests, linear regression, multivariable logistic regression, random-forest and extra-trees models, permutation feature-importance analysis, symptom co-occurrence analysis, and exploratory clustering were applied to identify behavioral and health-related determinants. Strongly favorable attitudes toward mask use were observed, yet substantial heterogeneity was found in economic valuation and product-specific acceptance. Willingness to pay increased markedly with perceived mask efficiency, while acceptance declined as price increased, indicating that both perceived benefit and affordability shaped adoption. Adjusted analysis showed that willingness to purchase higher-value masks was strongly associated with belief in mask effectiveness, urban context, higher education, and income, whereas machine-learning results indicated that areas of residence, worry, education, and preventive beliefs were among the most influential predictors. Comparative modeling further showed that mask products were not behaviorally interchangeable: one option demonstrated the broadest population appeal, another showed intermediate appeal in several adult and middle-income strata, and a third appeared more selective and prevention oriented. Recent symptom burden was found to be multidimensional, age-patterned, and unevenly distributed, with older adults showing higher constitutional and musculoskeletal complaints and younger respondents showing more febrile and upper-respiratory symptoms. Exploratory post-COVID-related analyses suggested structured symptom co-occurrence and phenotype-like groupings rather than isolated complaints. Overall, mask adoption and symptom burden were shown to be shaped by interacting demographic, socioeconomic, perceptual, and health-related factors, supporting the need for segmented communication, affordability-sensitive policy, and more context-specific preventive strategies in Bangladesh.

Authors

Institutions

Publication Details

Journal
Discover Public Health
Published
2026-11-14
DOI
https://doi.org/10.1186/s12982-026-03303-0
Primary Topic
COVID-19 Prevention and Impact
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Mask adoption, symptom burden, and machine learning insights from Bangladesh during and after COVID-19

K. M. Ariful Kabir, Abu Zobayer, Md. Fahimur Rahman Shuvo
Discover Public Health
COVID-19 Prevention and Impact
article

Mask adoption, symptom burden, and machine learning insights from Bangladesh during and after COVID-19

K. M. Ariful Kabir, Abu Zobayer, Md. Fahimur Rahman Shuvo
article en

Abstract

A comprehensive investigation of mask usage, recent symptom burden, and post-pandemic health behavior in Bangladesh was conducted using a cross-sectional survey framework that integrated descriptive statistics, inferential analysis, theoretical comparison, and machine-learning methods. Data from respondents across diverse demographic and residential strata were analyzed to examine mask-related attitudes, willingness to pay, willingness to purchase, age-specific vaccination patterns, and recent self-reported symptoms. Chi-square tests, linear regression, multivariable logistic regression, random-forest and extra-trees models, permutation feature-importance analysis, symptom co-occurrence analysis, and exploratory clustering were applied to identify behavioral and health-related determinants. Strongly favorable attitudes toward mask use were observed, yet substantial heterogeneity was found in economic valuation and product-specific acceptance. Willingness to pay increased markedly with perceived mask efficiency, while acceptance declined as price increased, indicating that both perceived benefit and affordability shaped adoption. Adjusted analysis showed that willingness to purchase higher-value masks was strongly associated with belief in mask effectiveness, urban context, higher education, and income, whereas machine-learning results indicated that areas of residence, worry, education, and preventive beliefs were among the most influential predictors. Comparative modeling further showed that mask products were not behaviorally interchangeable: one option demonstrated the broadest population appeal, another showed intermediate appeal in several adult and middle-income strata, and a third appeared more selective and prevention oriented. Recent symptom burden was found to be multidimensional, age-patterned, and unevenly distributed, with older adults showing higher constitutional and musculoskeletal complaints and younger respondents showing more febrile and upper-respiratory symptoms. Exploratory post-COVID-related analyses suggested structured symptom co-occurrence and phenotype-like groupings rather than isolated complaints. Overall, mask adoption and symptom burden were shown to be shaped by interacting demographic, socioeconomic, perceptual, and health-related factors, supporting the need for segmented communication, affordability-sensitive policy, and more context-specific preventive strategies in Bangladesh.

Discover Public HealthVol. 23(1)
Bangladesh University of Engineering and Technology (BD)
COVID-19 Prevention and Impact
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon