If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USADepartment of Internal Medicine, Division of Infectious Diseases, University of Michigan Medical School, Ann Arbor, MI, USA
Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USADepartment of Internal Medicine, Division of Infectious Diseases, University of Michigan Medical School, Ann Arbor, MI, USA
Full vaccination of children is high in Bangladesh, although it varies by vaccine type.
•
Full vaccination coverage was lowest among non-locals in all regions.
•
The mothers’ access to care and autonomy in healthcare decision-making are associated with higher vaccination coverage in their children.
•
Increasing childhood vaccination coverage will be key to meeting national goals for disease elimination and to improve child health in Bangladesh.
Abstract
Objectives
Childhood vaccination in Bangladesh has improved, but there is room for improvement. This study estimated full immunization coverage in Bangladeshi children and characterized risk factors for incomplete immunization.
Methods
Using the 2014 Bangladesh Demographic and Health Survey (DHS), full vaccination of children aged 12 to 24 months was examined; this was defined as the receipt of one dose of bacillus Calmette–Guérin (BCG), three doses of pentavalent vaccine, three doses of oral polio vaccine (OPV), and one dose of measles-containing vaccine (MCV). Associations between full vaccination and selected risk factors were assessed by logistic regression.
Results
Overall, 83% of children were fully vaccinated. BCG had the highest completion (97%), followed by OPV (92%), pentavalent vaccine (91%), and MCV (85%). Full vaccination coverage ranged from 64.4% in Sylhet to 90.0% in Rangpur and was lowest among non-locals of all regions (78.4%). Children who were in the lowest wealth quintile, who had mothers without antenatal care visits, or who had mothers without autonomy in healthcare decision-making were less likely to be fully vaccinated.
Conclusions
Overall, full vaccination of children is high, but varies by vaccine type. Disparities still exist by wealth and by region. Maternal access to care and autonomy in healthcare decision-making are associated with higher vaccination coverage.
A strong national commitment to childhood vaccination has contributed substantially to Bangladesh’s success in reaching Millennium Development Goal 4 to reduce childhood mortality. The death rate in Bangladeshi children under the age of 5 years declined from 133 per 1000 live births in 1993 to 46 per 1000 in 2014 (
). Implementation of the World Health Organization (WHO) Expanded Programme on Immunization (EPI) in Bangladesh occurred in 1979, but had minimal impact until the government made a public commitment to improve childhood vaccine coverage in 1985 (
). Bangladesh has recently experienced robust economic growth, with the 20th highest average increase (6.3%) in gross domestic product (GDP) of any country globally since 2008 (
). This prolonged period of economic growth has correlated with improvements in childhood health outcomes, despite a poor perception of the public health system by its citizens (
Bangladesh has expended considerable public health effort targeting reductions in childhood vaccine-preventable diseases. Sustained high polio vaccination coverage has rapidly decreased incident polio, with the last reported case in 2006 (
). Three nationwide supplementary immunization activities (SIAs) from 2000 to 2016 immunized approximately 108.9 million children with measles-containing vaccine (MCV) (
). In 2014, the Bangladeshi government set a goal to eliminate measles by 2018, aligning with the WHO South East Asia Regional goal of realizing measles elimination by 2020 (
). Despite these successes, Bangladesh, which continues to be one of the most densely populated countries in the world with over 162 million residents (
National Institute of Population Research and Training (NIPORT), Mitra and Associates, and ICF International. 2016. Bangladesh Demographic and Health Survey 2014. Dhaka, Bangladesh, and Rockville, Maryland, USA: NIPORT, Mitra and Associates, and ICF International.
), placing them among the top 10 countries with the highest childhood mortality globally.
While Bangladesh has made significant gains in childhood vaccination coverage, greater focus on series completion of all recommended EPI vaccines (Table 1) is needed to achieve further gains in decreasing childhood morbidity and mortality (
). Sustainable Development Goal 3 (United Nations Sustainable Development Goals (SDGs)) addresses the goal of universal health coverage (target 3.8), with a recommended indicator metric of attaining and sustaining 90% national vaccine coverage and 80% in every district with all vaccines in national programs (
The United Nations Children’s Fund (UNICEF) and the WHO estimated that individual vaccine coverage in Bangladesh in 2015 was 98% for bacillus Calmette–Guérin (BCG), 94% for three doses of pentavalent vaccine (diphtheria–tetanus–pertussis, Haemophilus influenzae type b, and hepatitis B virus; DTP–Hib–Hep B), 94% for three doses of oral polio virus vaccine (OPV), and 88% for the first dose of measles-containing vaccine (MCV) (
). These composite estimates were based on administrative coverage data aggregated from health service providers, estimated coverage data from national authorities, coverage from population-based household surveys, and relevant research reported in the published and grey literature (
). The administrative estimates are often unreliable due to incomplete or inaccurate primary recording of vaccinations. Therefore, it is important to bolster administrative surveys with more accurate examination of vaccination coverage using national surveys. The Demographic and Health Survey (DHS) is a nationally representative, standardized survey used globally, containing validated questions. It provides more robust data and allows factors associated with vaccination coverage to be studied (
). The Bangladesh DHS in 2014 is the most recent publicly available, nationally representative survey of vaccination coverage in Bangladesh.
Ongoing assessment of childhood vaccination through the DHS program is key to providing the essential information needed to inform and guide future program development through the identification of gaps in coverage, including differences based on residence or socio-demographic group. This study used the most recently available data from the 2014 DHS to calculate full immunization coverage in Bangladeshi children and to characterize risk factors for incomplete immunization, such as socioeconomic status and decision-making autonomy.
Methods
Study population
The DHS is an internationally recognized program that has conducted nationally representative surveys in over 90 developing countries. It includes population, health, and nutrition indicators. Seven DHS were conducted in Bangladesh between 1993 and 2014. At the time of the 2014 DHS, Bangladesh was divided into seven administrative regions: Barisal, Chittagong, Dhaka, Khulna, Rajshahi, Rangpur, and Sylhet. After completion of the 2014 DHS, an eighth region, Mymensingh, was divided from the Dhaka administrative region in 2015. Complete details of the DHS survey have been reported elsewhere (
National Institute of Population Research and Training (NIPORT), Mitra and Associates, and ICF International. 2016. Bangladesh Demographic and Health Survey 2014. Dhaka, Bangladesh, and Rockville, Maryland, USA: NIPORT, Mitra and Associates, and ICF International.
Briefly, the DHS survey enrolled individuals from all seven administrative regions of Bangladesh using a two-stage stratified sample of households. First, 600 enumeration areas were selected based on probability proportional to size. Then, within each enumeration area, a systematic sample of 30 households on average was selected, resulting in a total of 17 989 residential households. The data were collected by 20 interview teams consisting of five female interviewers during a 4-month period in 2014. The Women’s Questionnaire module of the DHS collected information from ever-married women aged 15–49 years, including background characteristics, reproductive behavior, perinatal care, children’s health, aspects of women’s empowerment, husband’s background, and other topics. From the selected households, 17 683 ever-married women were interviewed, resulting in data on a total of 43 772 children of all ages.
Measures
The main outcome for this study was full vaccination of children 12–24 months of age. Full vaccination was defined as having received all eight EPI-recommended vaccine doses (one dose of BCG, three doses of pentavalent vaccine, three doses of OPV, and one dose of MCV), in accordance with the WHO definition of full vaccination coverage (
Demographic, socioeconomic, and clinical data were abstracted from the DHS. Individuals who were not de jure residents at the location where they were interviewed were considered to be ‘non-locals’. A variable was derived to represent a respondent having autonomy in her own care; this variable was created by combining the categories of ‘respondent alone’ and ‘respondent and husband/partner’ for the question “Who is the person who usually decides on the respondent’s healthcare?”
Statistical analysis
The distributions of various demographic, socioeconomic, and clinical factors were examined using descriptive statistics. The associations between these factors and full vaccination were assessed using logistic regression with output estimates of odds ratios (OR) and 95% confidence intervals (CI) for under-vaccinated compared to fully vaccinated children. All descriptive and analytical statistics followed standard survey procedures, employing the individual weight from the Women’s Questionnaire. Additionally, the analysis included cluster and sample strata statements, based on the study design, to estimate standard errors. Significance was assessed at an α level of 0.05, and all analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC, USA).
Ethical approval
This study was limited to previously collected, publically available, secondary data, and therefore was not under the purview of the institutional review board at the University of Michigan.
Results
The total child sample size included in the DHS was 43 772. This study was limited to data collected about the children who were 12–24 months of age at the time of the survey, resulting in a final sample of 1701 children. The general characteristics of the study population are depicted in Table 2. In summary, the sample was almost evenly divided by sex and among wealth quintiles, and the majority of the study population was rural (62.4%). Many mothers had attained secondary education (48.7%) and had had one to three antenatal visits (46.1%), and most births had occurred at home (59.9%). Households were generally large (33.5% had seven or more household members), although the majority of children (68.1%) came from a home with no more than one child under the age of 5 years. All administrative regions were represented, ranging from 19.0% of participants in Chittagong to 11.5% in Khulna. Those whose mothers identified as Muslim comprised 92.5% of the study population, followed by Hindus (6.6%), Buddhists (0.7%), and Christians (0.2%).
Table 2Vaccination status of children ages 1–2 years by demographic characteristics—Bangladesh Demographic and Health Survey, 2014.
Vaccination coverage is shown in Table 2. Overall, 83.0% of the children were fully vaccinated, and vaccination rates were similar in males (83.1%) and females (82.8%). Full vaccination coverage increased progressively from the poorest to the richest wealth quintile and with greater maternal education, as did coverage for each vaccine type individually. Full vaccination was highest among urban residents (85.0%), followed by rural (82.7%) and non-local residents (78.4%). Full vaccination coverage was higher for children born outside the home (87.7% vs. 79.8%) and increased with more antenatal visits, as did coverage for each vaccine type individually. In general, larger households had lower vaccination coverage than smaller households.
By administrative region, Sylhet had considerably lower full vaccination (64.4%) than all other regions: Barisal (81.7%), Khulna (84.2%), Chittagong (85.8%), Rajshahi (86.7%), Dhaka (87.4%), and Rangpur (90.0%) (Figure 1). By individual vaccine type, nationwide coverage was highest for BCG vaccine (97.1%), followed by the third dose of OPV (91.8%), third dose of pentavalent vaccine (91.0%), and first dose of MCV (84.8%). This pattern held true across all administrative regions, with Sylhet having the lowest vaccine coverage for all vaccine series.
Figure 1Proportion of children ages 1–2 years who were fully vaccinated, by geographic region—Bangladesh Demographic and Health Survey, 2014.
The fully adjusted associations between complete vaccination and socio-demographic characteristics are shown in Table 3. Those falling in the poorest wealth quintile were more likely not to be fully vaccinated (73.0%) than those in the richest wealth quintile (91.1%) (OR 0.367, 95% CI 0.195–0.688); no other significant differences by wealth quintiles were observed. The number of antenatal visits, an oft-used proxy for access to care, was significantly associated with full vaccination; those without any visits had substantially lower full vaccination (71.6%) compared to those with four or more visits (88.5%) (95% CI 0.275–0.93). Women with more autonomy in healthcare decisions were more likely to have children who were fully vaccinated (86.1%) than those without autonomy (78.8%) (95% CI 1.079–2.317).
Table 3Adjusted associations between full childhood vaccination and selected socio-demographic characteristics—Bangladesh Demographic and Health Survey, 2014.
OR
95% CI
p-Value
Lower
Upper
Child’s sex
Female vs. male
1.020
0.717
1.452
0.9117
Wealth quintile
Poorest vs. richest
0.367
0.195
0.688
0.0088
Middle vs. richest
0.669
0.365
1.227
Poorer vs. richest
0.574
0.297
1.112
Richer vs. richest
0.856
0.476
1.539
Residence type
Non-local vs. urban
0.663
0.357
1.232
0.0095
Rural vs. urban
1.454
0.953
2.219
Mother’s level of education
Higher vs. secondary
1.863
0.736
4.717
0.0671
No education, preschool vs. secondary
0.609
0.360
1.029
Primary vs. secondary
0.696
0.463
1.046
Place of birth
Home vs. non-home
0.866
0.569
1.319
0.5026
Number of antenatal visits
0 visits vs. 4+ visits
0.506
0.275
0.930
0.0855
1-3 visits vs. 4+ visits
0.751
0.482
1.172
Number of people in the house
4 vs. ≤3
0.743
0.39
1.415
0.1492
5 vs. ≤3
1.074
0.574
2.009
6 vs. ≤3
1.324
0.662
2.648
7+ vs. ≤3
0.751
0.415
1.358
Number of children in the home
2 vs. ≤1
0.798
0.505
1.26
0.5481
3+ vs. ≤1
0.793
0.407
1.544
Person who usually decides on the respondent’s healthcare
Sex of the child, type of residence, mother’s level of education, place of birth, number of people in the household, and number of children in the home aged less than 5 years were all found to be non-significant factors in determining whether or not a child had received full vaccination.
Discussion
Vaccinations are a key component of the United Nations SDGs, which include ending preventable deaths of young children, reducing under-five mortality to 25 per 1000 live births, and ensuring access to vaccines for all (
). Using a nationally representative sample from the most recent 2014 DHS, a relatively high vaccination coverage was found in Bangladeshi children 1–2 years of age, but also significant disparities: vaccination coverage was relatively low among children whose mothers had non-local residency, had low wealth, and who did not play a role in the family’s healthcare decision-making. These findings can help guide future immunization programming with respect to devoting more resources to routine immunization services for specific socio-demographic groups, or in deciding whom and where to target SIAs to improve coverage. The significance of maternal healthcare autonomy as a predictor of the child’s vaccination status also points to broad, non-specific impacts of more comprehensive social change, such as implementing policies that promote gender equality, also addressed in the SDGs (
The present estimates of vaccination coverage in 2014 based on the DHS data were slightly lower than composite estimates based on country-specific data reported to the WHO and UNICEF (
). However, full vaccination coverage increased by 20% between the 2004 (61.7%) and 2014 DHS assessments. Previous studies have revealed a number of factors associated with full vaccination of children. A study using 2004 DHS data found that higher vaccination of children was associated with wealth, distance from a healthcare facility, mother’s age, mother’s education, the mother having permission to go to the hospital alone, parity, visits from a family planning/health worker, mass media, and child’s sex (
). In addition, full childhood vaccination has been found to be positively associated with urbanicity for all vaccinations and with exposure to media for measles vaccine (
Relative to other countries in the area, Bangladesh compares favorably in childhood vaccination coverage with other countries in the WHO South East Asia Region, including the neighboring country of India, which had a much lower 59% overall childhood vaccination coverage in 2014 (
). The strength of Bangladeshi health programs has been attributed to its pluralistic health system, which actively utilizes non-governmental organizations and women-focused, equity-oriented targeted programs (
). While Bangladesh has established a goal of measles elimination by 2018, achieving such will require maintaining ≥95% coverage with two doses of MCV, which has proven elusive, as reflected in the recent increase in measles cases from 240 in 2015 to 972 in 2016 (
). In fact, MCV had the lowest coverage of any of the EPI immunizations, whereas BCG, administered at birth, had the highest, with OPV and pentavalent vaccine administered at 6, 10, and 14 weeks falling somewhere in between. As is commonly observed with immunizations in many countries, progressively more children are lost to follow-up accompanied by a corresponding drop in coverage for vaccines administered later in the schedule.
An encouraging increase in full vaccination in 2014 was observed among the most vulnerable populations compared to 2004 (e.g., the poorest quintile or those who had no antenatal visits), although the difference in coverage between the richest and poorest quintiles (18%) and between those with 0 and ≥4 antenatal visits (17%) was comparable to those found in the 2004 DHS (
). That the absolute difference in the vaccination disparities between rich and poor and those with and without healthcare access has remained constant over a decade span points to the difficulty in reaching the most vulnerable members of society. Clearly, further gains in vaccination coverage among the poor and underserved in addressing this gap will rely on successfully reaching these populations and require commensurately greater resource expenditure (
Maternal autonomy in healthcare decision-making emerged as a significant predictor of the child’s immunization status. Children whose mothers were included in healthcare decisions had higher full vaccination coverage than children whose mothers were not. Autonomy is a multifaceted concept, including factors such as decision-making capacity, mobility outside the home, financial independence, and attitudes towards domestic violence. A study using data from the National Family Health Survey in India in 2005–2006 found small but statistically significant differences in the child’s vaccination status based on the mother’s decision-making power: incomplete vaccination was 49% among those with low decision-making power and 46.0% among those with high decision-making power (
). Because the mother is traditionally more involved with caring for children, having a greater say in making decisions in the household and specifically with child healthcare, can lead to healthier infants. For example, one study in Sylhet district, Bangladesh, randomized certain communities to receive home visits from community health workers to educate mothers about best care practices. Neonatal mortality was noticeably lower (29.2 per 1000) compared to the control arm (43.5 per 1000) (
Effect of community-based newborn-care intervention package implemented through two service-delivery strategies in Sylhet district, Bangladesh: a cluster-randomised controlled trial.
). Engaging with mothers and providing them the education and autonomy to seek out the best healthcare for their children is an important component of any program aimed at reducing neonatal and infant mortality.
With regard to the strengths and limitations of this study, the robust database from the DHS is a strength and ensured a sample representative of the entire country. The data were obtained from previously validated questionnaires. However, the cross-sectional nature of the DHS complicates the determination of causality and is a limitation of the study. Some information is based on parental recall, and individuals of different demographic groups may remember information about their child’s vaccination status differently.
In conclusion, this study found that childhood immunization coverage in Bangladesh was among the highest in the WHO South East Asia Region for several antigens, and coverage increased rapidly over the 10-year period 2004–2014. While this is encouraging and may not be surprising given Bangladesh’s rapidly growing economy, significant disparities in coverage by wealth status and by healthcare access remain. Geographical and demographic differences in vaccination may portend difficulties in eliminating preventable diseases within Bangladesh, one of the most densely populated countries in the world. Future success in increasing childhood vaccination coverage will require improvements in healthcare access to reach the poor and underserved populations. The success of vaccination programs would also benefit from mothers having greater autonomy in making healthcare decisions about their children and themselves.
Acknowledgements
We are grateful to the public health workers for their tireless efforts in obtaining data for the Demographic and Health Surveys Program.
References
Afzal Nasrin
Zainab Begum
Determinants and status of vaccination in Bangladesh.
Effect of community-based newborn-care intervention package implemented through two service-delivery strategies in Sylhet district, Bangladesh: a cluster-randomised controlled trial.
National Institute of Population Research and Training (NIPORT), Mitra and Associates, and ICF International. 2016. Bangladesh Demographic and Health Survey 2014. Dhaka, Bangladesh, and Rockville, Maryland, USA: NIPORT, Mitra and Associates, and ICF International.
30036-5&id=gr1.jpg){kind=link}