What predicts complete immunisation among 18-month to 24-month-old children in the urban slum area of Hlaingthayar Township, Yangon Region, Myanmar? A cross-sectional study Authors: Zayar Lynn, Wai Wai Han Published: 2024 ABSTRACT AbstractABSTRACTIntroductionIn Myanmar, there are significant disparities in the delivery of immunisation services and barriers to service accessibility, particularly in border regions, conflict-affected areas, physically and geographically hard-to-reach areas, urban slums and migratory populations. The study aimed to explore the predictors for complete immunisation among 18-month to 24-month-old children in the urban slum area of Hlaingthayar Township, Yangon Region, Myanmar.MethodsA cross-sectional study was conducted from 21 to 25 July 2018. A total of 298 mothers of 18-month to 24-month-old children were recruited from the urban slum areas under five randomly selected wards by applying the cluster sampling method. Multiple binary logistic regression was performed to explore the predictors for complete immunisation. A significance level of 0.05 was considered statistically significant.ResultsOut of a total of 298 children, 120 (40.3%) children were completely immunised, and the immunisation coverages ranged from 52.7% for the second dose of measles-rubella vaccine to 81.9% for the first dose of oral polio vaccine. Multivariate logistic regression model revealed that mothers having university and above education (AOR=2.65, 95% CI 1.07 to 6.56), mothers with permanent residency (AOR=2.19, 95% CI 1.16 to 4.15), primiparous mothers (AOR=1.76, 95% CI 1.05 to 2.95), mothers with medium knowledge (AOR=2.38, 95% CI 1.07 to 5.30) and high knowledge about childhood immunisation (AOR=5.89, 95% CI 2.23 to 15.60), and mothers with positive perception of childhood immunisation (AOR=2.26, 95% CI 1.33 to 3.85) were the significant predictors for complete immunisation among children.ConclusionThe proportion of complete immunisation was low, and associated with maternal education, residency, parity, knowledge and perception of childhood immunisation. Having a comprehensive understanding of these predictors will facilitate the government and other relevant stakeholders to establish effective strategies promoting access to immunisation services among the urban slum communities in Myanmar.Keywords: Vaccination, Public Health, Community Health, Cross-Sectional Studies, Public Health Practice WHAT IS ALREADY KNOWN ON THIS TOPIC In Myanmar, there is still the existence of inequalities in immunisation service delivery and difficulties in service accessibility especially in border areas, conflict-affected areas, physically and geographically hard-to-reach areas and among urban slums and migrant populations. In Myanmar, there is still the existence of inequalities in immunisation service delivery and difficulties in service accessibility especially in border areas, conflict-affected areas, physically and geographically hard-to-reach areas and among urban slums and migrant populations. WHAT THIS STUDY ADDS The proportion of complete immunisation was low, and the immunisation coverages were below the national targets. The proportion of complete immunisation was low, and the immunisation coverages were below the national targets. Maternal education, residency, parity, knowledge and perception of childhood immunisation were significant predictors for complete immunisation among children. Maternal education, residency, parity, knowledge and perception of childhood immunisation were significant predictors for complete immunisation among children. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY The government and other relevant stakeholders should establish effective strategies to promote access to immunisation services among the urban slum inhabitants, thereby reducing the outbreaks of vaccine preventable diseases and improving the health outcomes of newborns and under-5 children in Myanmar. The government and other relevant stakeholders should establish effective strategies to promote access to immunisation services among the urban slum inhabitants, thereby reducing the outbreaks of vaccine preventable diseases and improving the health outcomes of newborns and under-5 children in Myanmar. Introduction Immunisation has been proven as a successful strategy for minimising morbidity and mortality of newborns and under-5 children from vaccine preventable diseases (VPD). Globally, there are 3.5 to 5 million deaths that can be avoided by immunisation every year.1 Complete immunisation can prevent severe sickness and prolonged disability from VPD and can reduce 24% of mortality in under-5 children.2 Although global vaccination coverage for infants who received the recommended three doses of diphtheria–tetanus–pertussis vaccine was about 84% (approximately 108 million infants) in 2023, there were significant gap between the countries of different economic levels, with low-income countries lagging behind.3 Nearly 21 million infants were still missing routine immunisations in the world, of which 6.5 million infants were partially vaccinated and 14.5 million infants were never vaccinated in 2023.3 In August 2020, the Immunization Agenda 2030 was set up by the World Health Assembly for the decade 2021–2030 with ambitious and overarching strategies by reducing 50% of zero-dose children, achieving 90% of global vaccination coverage, and limiting drop-out between the first dose of pentavalent vaccine and other vaccines to less than 5%.2 4 In Myanmar, immunisation coverages were still unsatisfactory and below the national targets, which are targeted to achieve at least 95% for the Bacillus Calmette Guerin (BCG) vaccine and the first dose of measles-rubella vaccine (MR-1), and at least 90% for other types of vaccines by 2021.5 The latest nationwide Demographic and Health Survey in Myanmar reported that only 55% of children aged 12–23 months completed all basic vaccines, 37% of children were partially vaccinated, and 8% of children were unvaccinated in 2015.6 From 2015 to 2017, outbreaks of Japanese B encephalitis, diphtheria and measles occurred in Myanmar, even in the areas with high immunisation rates.7 Additionally, there are significant disparities in the delivery of immunisation services and barriers to service accessibility, particularly in border regions, conflicted-affected areas, physically and geographically hard-to-reach areas, urban slums and migratory populations.5 Myanmar has the highest poverty rate among South-East Asian countries.8 Seventy per cent of Myanmar’s population resides in rural regions, where poverty is more common, accounting for around 30% of the country’s population.9 Therefore, people from rural areas have migrated to urban areas to seek higher income and better job opportunities. This, in turn, causes rapid and unplanned poorly managed urbanisation leading towards expansion of urban slum areas in major cities of Myanmar.9 10 In Myanmar, the population living in slum areas was reported as 56.1% of the total urban population in 2018.11 The vulnerable characteristics of slum areas can be described as being the first entry point for the poor, being hubs of rising income and ownerships of assets, having safety concerns for women because of the high incidence of violence, having negative impacts on health and child mortality, low education and abundant child labour, exclusion in administrative and healthcare registries, political and social exclusion as illegal settlements, and being prone to natural disasters and climate conditions.10 As the Yangon Region is the most socioeconomically developed region of the country, a significant proportion of rural-to-urban migration takes place in the region, especially in the slum area of Hlaingthayar Township with the highest number of slum dwellers of the country, which experiences increased congestion and limited access to basic health services.12 13 People living in urban slum areas have been reported as the highest occurrence of communicable diseases because of poor living conditions, poor health-seeking knowledge and practice, low utilisation of public healthcare services, and significant disparities also exist especially in health service coverage, including childhood immunisation.9 10 In 2017, the slum community of the township reported 185 deaths of under-5 children, which was the highest among Yangon Region.14 Moreover, more than half of the measles cases were detected in the slum community of the township during the occurrence of the measles outbreak in Yangon Region.15 Therefore, the study aimed to explore the predictors for complete immunisation among 18-month to 24-month-old children in the urban slum area of Hlaingthayar Township. The findings of the study would provide evidence for the policymakers for further strategic implementation of the Expanded Programme on Immunization (EPI) among pockets of low immunisation coverage areas in Myanmar as well as countries with low resource settings. The study will make a substantial contribution to the literature on the predictors of complete immunisation in the urban slum settings of low-income countries and will act as a roadmap to achieve the Immunization Agenda 2030. Improving immunisation coverage will reduce the risks of outbreaks of VPDs, thereby minimising preventable deaths of newborns and children under 5 years of age. Methods Study design and setting This was a cross-sectional study conducted from 21 to 25 July 2018. The study was purposively selected to be conducted in the urban slum area of Hlaingthayar Township, which is the most crowded and the largest urban slum area in the country.16 The township was first established as an industrial zone in 1989 and then, residential areas were further expanded by the military government in 1993. After the Nargis Cyclone disaster, the township experienced a sudden increase in population due to the invasion of migrants from rural to urban areas residing in the slum area with informal settlements in the township. The availability of empty land spaces and a wide variety of employment in the industries are the main reasons for choosing Hlaingthayar Township as their location. Nowadays, the township has a total of 181 informal slum settlements, which consist of approximately 24 865 households with an estimated population of 124,325.17 Study population and eligibility criteria The study recruited mothers of 18-month to 24-month-old children from the urban slum area of Hlaingthayar Township. The inclusion criteria were the mothers who had been living in the slum area for more than 1 month and who willingly provided consent to participate in the study. We excluded the mothers who were under 18 years old and had psychological problems or serious medical conditions. The study only collected the immunisation coverages of the vaccines provided by the EPI of the Ministry of Health and Sports and did not include other sources of vaccination taken from the outside market. Sample size The latest published Myanmar Demographic and Health Survey 2015–2016 reported that only 55% of children aged between 12 and 23 months received all basic vaccinations.6 Assuming that this proportion would be similar in the study area, the required sample size was calculated by using Cochran’s sample size calculation formula with infinite population proportion at 95% CI and 8% precision with the design effect of 2.0. The total sample size was 298 participants. n=the estimated sample size Zα22 = the confidence level at 95% (standard value of 1.96) p=the estimated proportion q= (1 – p) d2=the desired level of precision (ie, the margin of error) DEFF=design effect. Sampling and data collection The study applied a cluster sampling method for the selection of the participants. First, 5 out of 20 wards in the township were randomly selected by lottery method. After that, the urban slum areas under five wards were selected for data collection. For identifying the required participants, sampling frames which contained the list of mothers of 18-month to 24-month-old children were requested from the pre-existing data of the Township Administration Office. Then, all eligible participants were invited to participate in the study with the help of Local Authorities, Basic Health Staff and Auxiliary Midwives. The research team explained the study’s objective and procedure to the participants, and if the participants agreed to participate, written informed consents were obtained and face-to-face interviews were conducted using a semistructured questionnaire. The questionnaire is available as online supplemental file. Study variables and measurement The dependent variable was complete immunisation. According to the schedule of the EPI programme delivered by the Ministry of Health and Sports in Myanmar, a child has to receive BCG vaccine at birth or 2 months of age; the first dose of pentavalent vaccine (Penta-1), the first dose of Pneumococcal Conjugate Vaccine (PCV-1) and the first dose of Oral Polio Vaccine (OPV-1) at 2 months of age; the second dose of pentavalent vaccine (Penta-2), the second dose of Pneumococcal Conjugate Vaccine (PCV-2), the second dose of Oral Polio Vaccine (OPV-2) and Inactivated Polio Vaccine (IPV) at 4 months of age; the third dose of pentavalent vaccine (Penta-3), the third dose of Pneumococcal Conjugate Vaccine (PCV-3) and the third dose of Oral Polio Vaccine (OPV-3) at 6 months of age; the first dose of MR-1 and Japanese B encephalitis vaccine (JE) at 9 months of age; and the second dose of MR-2 at 18 months of age respectively. The complete immunisation coverage of the child can be calculated starting from 18 months of age, and the study selected 18-month to 24-month-old children as the denominator based on the previous studies conducted in Myanmar (Than et al, 2016)18 and Belgium (Vellinga et al, 2002)19 in order to minimise information bias and recall bias. A child is considered to be completely immunised if he or she has completed all these basic vaccinations for his or her age, or incompletely immunised if he or she has missed at least one dose of the recommended vaccines. The information on immunisation coverages was recorded from the immunisation cards during data collection. The independent variables contained maternal age, parental education and occupation, maternal residency, monthly family income, type of family, decision-making person for healthcare, parity of mother, sex of the child, knowledge and perception of mothers on immunisation. The survey questionnaire was developed by the researcher by reviewing and adopting local and international literatures.20 21 The questions for knowledge about immunisation comprised 23 items regarding the number of vaccines included in the EPI programme, names of VPDs, age of the child to start immunisation, number of sessions needed to complete immunisation, number of doses needed to complete for each vaccine, and age of the child to complete immunisation. The total knowledge score ranged from 0 to 23, and one mark was given for every correct response. The level of knowledge was further divided into three categories: Low knowledge (score less than one SD below the mean value), Medium knowledge (score within one SD of the mean value), and High knowledge (score more than one SD above the mean value). The questions for the perception of immunisation contained 12 items with Yes or No responses. The total perception score ranged from 0 to 12, and the level of perception was further categorised into Negative perception (score below the mean value), and Positive perception (score above the mean value). The content validity of the questionnaire was reviewed by three experts and internal consistency reliability was tested by conducting a pilot study with 30 participants before data collection in the study area, and these participants were excluded from the study. The Cronbach’s alpha coefficients of knowledge and perception of immunisation were 0.86 and 0.75 respectively. Data analysis Double data entries were performed using EpiData 3.1, and then statistical analysis was conducted using SPSS version 23. Descriptive statistics were used to summarise the categorical variables as frequencies and percentages and to summarise the continuous variables as means and SD. Bivariate analysis was conducted with simple logistic regression to explore the unadjusted association between each independent variable and dependent variable. Then, the variables with p-values less than 0.25 in bivariate analysis to prevent failure in identifying predictor variables known to be important in the model,22 and with no multicollinearity among other predictor variables were included in multivariate analysis with multiple binary logistic regression, and the backward elimination method was applied. Model fitness was checked using the Hosmer and Lemeshow goodness-of-fit chi-square test. Interaction effect analysis was conducted by generalised estimating equations. Unadjusted odds ratios (OR) and adjusted odds ratios (AOR) with a 95% CI were computed, and a significance level of 0.05 was considered statistically significant. Results Characteristics of the participants The study population included 298 mothers of 18-month to 24-month-old children. Most of the mothers were 25 years and older (80.2%), dependent (62.4%), had below high-school level education (91.3%), and migrated from other areas (80.9%). The majority of the fathers also had below high-school level education (93.6%) and were employed as daily workers (60.1%). Over half of the participants had a nuclear family (72.1%) with over 250 000 MMK of monthly family income (55.4%). The main decision-maker for the healthcare of the child in the family was the mothers (66.8%). Almost half of the participants were multiparous mothers (50.3%) with male children (50.7%). Only a few mothers had high knowledge about childhood immunisation (16.4%) and just over one-third of the mothers had a positive perception of childhood immunisation (39.3%). The mean scores were 7.1±3.51 for maternal knowledge about childhood immunisation, and 10.1±1.23 for maternal perception of childhood immunisation (table 1). Table 1Characteristics of the participants (n=298)VariablesNumberPercentAge of mother 18–24 years5919.8 ≥ 25 years23980.2Mother’s education High school and below27291.3 University and above268.7Father’s education High school and below27993.6 University and above196.4Mother’s occupation Daily worker165.4 Factory worker or general seller9632.2 Dependent18662.4Father’s occupation Daily worker17960.1 Factory worker or general seller11939.9Residence of mother Migrant24180.9 Permanent5719.1Monthly family income <250 000 MMK13344.6 ≥250 000 MMK16555.4Type of family Nuclear family21572.1 Extended family8327.9Decision-making person for healthcare Mother19966.8 Other family member9933.2Parity of mother Multiparous15050.3 Primiparous14849.7Sex of the child Male15150.7 Female14749.3Knowledge about immunisation Low knowledge (Score<3.59)4816.1 Medium knowledge (Score 3.59–10.61)20167.5 High knowledge (Score>10.61)4916.4 Mean±SD7.1±3.51 Range0–19Perception of immunisation Negative perception (Score<10.1)18160.7 Positive perception (Score>10.1)11739.3 Mean±SD10.1±1.23 Range6–12MMKMyanmar Kyat Characteristics of the participants (n=298) MMKMyanmar Kyat Immunisation status and immunisation coverage among 18-month to 24-month-old children In this study, the immunisation status of the children was reported as 40.3% for complete immunisation and 59.7% for incomplete immunisation (ie, missed at least one dose of vaccine). The highest immunisation coverage was 81.9% for the first dose of OPV, and the lowest immunisation coverage was 52.7% for the second dose of MR. A 23.2% dropout rate between the first dose of pentavalent vaccine (Penta-1) and the third dose of pentavalent vaccine (Penta-3) was observed (table 2). Table 2Immunisation status and immunisation coverage among 18-month to 24-month-old children (n=298)VariablesNumberPercentImmunisation status of children Complete immunisation12040.3 Incomplete immunisation17859.7Immunisation coverage for the recommended vaccines BCG23779.5 Penta-124180.9 Penta-220267.8 Penta-318562.1 PCV-122274.5 PCV-218261.1 PCV-317057.0 OPV-124481.9 OPV-221672.5 OPV-318963.4 IPV19766.1 MR-122575.5 MR-215752.7 JE22976.8BCG, Bacillus Calmette Guerin vaccine ; IPV, inactivated polio vaccine; JE, Japanese B encephalitis vaccineMR, measles-rubella vaccine; OPV, oral polio vaccine; PCV, pneumococcal conjugate vaccine; Penta, pentavalent vaccine Immunisation status and immunisation coverage among 18-month to 24-month-old children (n=298) BCG, Bacillus Calmette Guerin vaccine ; IPV, inactivated polio vaccine; JE, Japanese B encephalitis vaccineMR, measles-rubella vaccine; OPV, oral polio vaccine; PCV, pneumococcal conjugate vaccine; Penta, pentavalent vaccine Bivariate and multivariate analysis of the predictors for complete immunisation In bivariate logistic regression analysis, the variables of mother’s education, father’s occupation, residence of mother, type of family, parity of mother, knowledge about immunisation and perception of immunisation showed associations with complete immunisation at p<0.25, and these variables were taken into the multivariate logistic regression analysis. After adjusting for the covariates, the variables of maternal education, residency, parity, knowledge and perception showed statistically significant associations with complete immunisation (p<0.05). The mothers with university and above education were 2.6 times more likely to receive complete immunisation for their children compared with the mothers with high school and below education (AOR=2.65, 95% CI 1.07 to 6.56, p=0.035). Compared with the mothers who migrated from other areas, the mothers who permanently reside in the area were 2.2 times more likely to take full doses of immunisation for their children (AOR=2.19, 95% CI 1.16 to 4.15, p=0.016). The odds of children receiving complete immunisation were 1.8 times higher among primiparous mothers than multiparous mothers (AOR=1.76, 95% CI 1.05 to 2.95, p=0.033). Similarly, the children whose mothers had medium knowledge and high knowledge about childhood immunisation were 2.4 times and 5.9 times more likely to complete immunisation compared with the children whose mothers had low knowledge (medium knowledge: AOR=2.38, 95% CI 1.07 to 5.30, p=0.034; high knowledge: AOR=5.89, 95% CI 2.23 to 15.60, p<0.001). Moreover, the mothers with a positive perception of childhood immunisation had a 2.3 times higher probability of taking complete immunisation for their children than the mothers with a negative perception (AOR=2.26, 95% CI 1.33 to 3.85, p=0.003) (table 3). Table 3Bivariate and multivariate logistic regression analysis of the predictors for complete immunisation (n=298)VariablesIncomplete (n=178)Complete (n=120)OR (95% CI)AOR (95% CI)P valuen (%)n (%)Mother’s education High school and below169 (62.1)103 (37.9)11 University and above9 (34.6)17 (65.4)3.10 (1.33 to 7.21)2.65 (1.07 to 6.56)0.035Father’s occupation Daily worker114 (63.7)65 (36.3)11 Factory worker or general seller64 (53.8)55 (46.2)1.51 (0.94 to 2.42)1.18 (0.70 to 2.00)0.538Residence of mother Migrant151 (62.7)90 (37.3)11 Permanent27 (47.4)30 (52.6)1.86 (1.04 to 3.34)2.19 (1.16 to 4.15)0.016Type of family Nuclear family134 (62.3)81 (37.7)11 Extended family44 (53.0)39 (47.0)1.47 (0.88 to 2.45)1.49 (0.84 to 2.63)0.174Parity of mother Multiparous98 (65.3)52 (34.7)11 Primiparous80 (54.1)68 (45.9)1.60 (1.01 to 2.55)1.76 (1.05 to 2.95)0.033Knowledge about immunisation Low knowledge39 (81.3)9 (18.8)11 Medium knowledge122 (60.7)79 (39.3)2.81 (1.29 to 6.11)2.38 (1.07 to 5.30)0.034 High knowledge17 (34.7)32 (65.3)8.16 (3.21 to 20.75)5.89 (2.23 to 15.60)<0.001Perception of immunisation Negative perception122 (67.4)59 (32.6)11 Positive perception56 (47.9)61 (52.1)2.25 (1.40 to 3.63)2.26 (1.33 to 3.85)0.003AOR, adjusted OR Bivariate and multivariate logistic regression analysis of the predictors for complete immunisation (n=298) AOR, adjusted OR Interaction effects between residence of mothers with other predictors on complete immunisation The study assessed the interaction effects between residence of mothers (migrant and permanent) with other predictors on complete immunisation and presented in online supplemental table S1. The results showed that there were significant interaction effects between father’s occupation, parity of mother, knowledge about immunisation and perception of immunisation with residence of mothers in the context of complete immunisation. The mothers whose husbands were employed as factory worker or general seller and permanently reside in the area were 2.5 times more likely to take full doses of immunisation for their children than the mothers whose husbands were employed as daily worker and migrated from other areas (OR=2.52, 95% CI 1.13 to 5.58, p=0.023). Compared with the multiparous mothers migrated from other areas, the primiparous mothers who permanently reside in the area were 3.2 times more likely to receive complete immunisation for their children (OR=3.25, 95% CI 1.39 to 7.59, p=0.007). The children whose mothers had medium knowledge and high knowledge about childhood immunisation and migrated from other areas were 3.7 times and 10.2 times more likely to complete immunisation compared with the children whose mothers had low knowledge and migrated from other areas (Medium knowledge × Migrant: OR=3.75, 95% CI 1.39 to 10.14, p=0.009; High knowledge × Migrant: OR=10.24, 95% CI 3.27 to 32.09, p<0.001). Similarly, the children whose mothers had medium knowledge and high knowledge about childhood immunisation and permanently reside in the area were 6.4 times and 25.6 times more likely to complete immunisation compared with the children whose mothers had low knowledge and migrated from other areas (Medium knowledge × Permanent: OR=6.40, 95% CI 2.03 to 20.15, p=0.002; High knowledge × Permanent: OR=25.60, 95% CI 4.17 to 157.00, p<0.001). Moreover, the mothers with a positive perception of childhood immunisation and migrated from other areas had a 2.3 times higher probability of taking complete immunisation for their children than the mothers with a negative perception and migrated from other areas (OR=2.31, 95% CI 1.35 to 3.94, p=0.002). Furthermore, the mothers with a positive perception of childhood immunisation and permanently reside in the area had a 5.2 times higher probability of taking complete immunisation for their children than the mothers with a negative perception and migrated from other areas (OR=5.21, 95% CI 1.86 to 14.63, p=0.002). Reasons for complete and incomplete immunisation The main reason for mothers receiving complete immunisation for their children was because of worry about getting diseases to the children (59.2%), followed by an awareness of the benefits of immunisation (33.3%) and a short distance from home to the immunisation place (24.2%). The main reason for mothers receiving incomplete immunisation for their children was due to a tight work schedule (36%), followed by migration to another area or hometown (27%), and minor illness of the child at the time of immunisation (14%) (table 4). Table 4Reasons for complete and incomplete immunisationVariablesNumberPercentReasons for complete immunisation (n=120)* Worry about getting diseases7159.2 Awareness of the benefits of immunisation4033.3 A short distance from home to the immunisation place2924.2 Midwife regularly comes for immunisation2016.7 Counselling from healthcare providers1815.0 Advice from family members/relatives/neighbours86.7Reasons for incomplete immunisation (n=178)* Tight working schedule6436.0 Migration to another area or hometown4827.0 Minor illness of the child2514.0 Fear of side effects of immunisation2212.4 Didn't know how to access healthcare services2111.8 Unaware of the need for immunisation169.0*The question has multiple responses. Reasons for complete and incomplete immunisation The question has multiple responses. Discussion Although the EPI in Myanmar has evolved rapidly with increasing commitment from the government and engagement from the international community, the country faces a complex social, cultural and political concern that leads to under-reported unimmunised children from hard-to-reach and underserved communities.23 The urban slum population is often neglected in the formal healthcare system since they mostly reside in informal settlements with temporary buildings and sometimes hidden, which would create difficulty in accessing basic healthcare services, including immunisation services. Therefore, the study aimed to explore the predictors for complete immunisation among 18-month to 24-month-old children in the urban slum area. In this study, the proportion of children receiving complete immunisation for all basic vaccines according to their respective age was only 40.3%, which is similar to the findings of the studies conducted in the urban slum areas of Dhaka, Bangladesh (43%)24; and Kampala, Uganda (41.8%).25 However, our study finding was lower than that reported in the urban slum areas of Mumbai City, India (73.1%)26; Kathmandu Valley, Nepal (71.4%)27; Islamabad, Pakistan (71.1%)28; Abidjan, Cote d’Ivoire (68.6%)29; Nairobi, Kenya (66.6%)30; East Java Province, Indonesia (55.4%)31; and Pernambuco, Brazil (52.1%).32 The discrepancy in the completeness of immunisation may be related to the utilisation of immunisation services by the urban slum inhabitants in different slum settings of the countries. Moreover, the proportion of complete immunisation in our study was lower than that of national coverage, which was reported as 55% in the latest published Myanmar Demographic and Health Survey 2015–20166. The possible explanation may be due to the difference in socioeconomic background of the study populations, since most of the mothers in our study were poor, whereas the participants in the MDHS 2015–2016 included a wide range of socioeconomic conditions. Our study also revealed high immunisation coverage for the vaccines administered at birth and 2 months of age (79.5% for BCG, 80.9% for penta-1, 74.5% for PCV-1 and 81.9% for OPV-1), which indicated good access to immunisation services; however, these coverages were still lower than the national immunisation targets. After that, the coverages were dropped for the subsequent vaccine doses delivered after 2 months of age by a dropout rate of 23.2% from the first to the third dose of the pentavalent vaccine, indicating a decrease in utilisation of immunisation services over time. This high level of dropout rate demonstrates that the EPI programme should prioritise motivating mothers to complete the immunisation schedule in addition to initiating immunisation. The main reason for incomplete immunisation in our study was due to the tight working schedule of mothers, and this finding is consistent with the studies conducted in the urban slums of Mandalay, Myanmar33; Mumbai City, India26; and Semarang City, Indonesia.34 Therefore, the healthcare workers should engage with the employers to allow the working mothers to bring their children to the immunisation place during working hours, or schedule immunisation appointments for the working mothers on their convenient dates. The study revealed that maternal education was one of the significant predictors for complete immunisation among children, and mothers with higher education were more likely to take complete immunisation for their children. The finding is consistent with the studies conducted in the urban slum areas of Kathmandu Valley, Nepal27; Abidjan, Cote d’Ivoire29; Nairobi, Kenya30; Pernambuco, Brazil32 and Varanasi, India.35 This may be related to the fact that because of the contribution of education, the mothers may change their attitudes, and beliefs, as well as more autonomy and decision-making, all of which may have the potential to positively influence the health-seeking behaviour of the mothers. This finding suggests that healthcare providers focus more on the children of low-educated mothers by encouraging immunisation using visual rather than written communication to reach this population efficiently. The residency of the mother was also found to be a predictor for immunisation completeness in the study, and the mothers with permanent residency in the area were more likely to receive full doses of immunisation for their children compared with the mothers who migrated from other areas. Our study finding is aligned with the studies conducted in urban slum areas of Mandalay, Myanmar33; Delhi, India36; and Beijing, China.37 This may be explained by the fact that migration is one of the disparities in immunisation coverage of children. Migrants are mostly registered in their places of origin, and not included in any health and administrative registries of place of destination. The legal status of migrants can impact their access to health services and thus may result in lower immunisation coverage.10 It was also supported by the study conducted in the delta region of Myanmar assessing the immunisation status of migrant children in which completeness of immunisation was four times lower among migrant children compared with that of children from the general population.38 Therefore, it is suggested to strengthen immunisation service delivery by ensuring equitable access to all target populations in the community. Maternal parity was another determinant for completeness of immunisation in children, and the study observed that the children of primiparous mothers had a higher chance of receiving complete immunisation. The finding is supported by the studies conducted in the urban slum areas of Nairobi, Kenya30; Varanasi, India35; and Western Area Urban district, Sierra Leone.39 If a mother has more children, she has more constraints on the available resources, particularly in the informal settlements where poverty is prevalent and has an impact on healthcare utilisation.40 Therefore, multiparous mothers should be given great attention and provided more intensive health education on the importance of childhood immunisation during immunisation sessions. Moreover, knowledge of mothers about childhood immunisation significantly determined immunisation completeness in children, and the children of mothers having higher knowledge about childhood immunisation were more likely to complete immunisation for all basic vaccines. The study finding is similar to the studies conducted in the urban slum areas of Kathmandu Valley, Nepal27; Abidjan, Cote d’Ivoire29; and Beijing, China37. This might be because the mothers are more motivated to take complete immunisation for their children since they are more knowledgeable about the benefits of childhood immunisation, VPDs and immunisation schedule. Our study also revealed that there was only 16.4% high knowledge of mothers about childhood immunisation, and therefore, it is recommended to conduct more frequent health literacy promotion activities on childhood immunisation in the urban slum community by considering innovative methods to fill the knowledge gaps of mothers. Additionally, the perception of mothers on childhood immunisation was significantly associated with complete immunisation in children. The mothers with a positive perception of childhood immunisation had a higher probability of taking full doses of immunisation for their children. Our study finding is consistent with the studies conducted in the hilly region of Myanmar18 and among Myanmar migrant children in Samut Sakhon Province of Thailand.41 If the mother has a positive perception of childhood immunisation, she may have a positive attitude and practice to immunise her child according to schedule. The study observed that just over one-third of the mothers (39.3%) had a positive perception of childhood immunisation. The finding suggests proper health education and health promotion interventions should be carried out as a routine measure among the urban slum inhabitants to improve their positive perception of childhood immunisation. The study identified interactions between father’s occupation, parity, knowledge about immunisation and perception of immunisation with residence of mothers, suggesting synergistic effects on completeness of immunisation. The findings indicate that migrant children whose fathers work as daily labourers, mothers with multiparity, low knowledge and negative perception of childhood immunisation were more likely to be non-compliance with immunisation and should be given special consideration by basic health staff to ensure better childhood immunisation coverage. The findings of the study will provide evidence for the policymakers and healthcare professionals for strategic planning of EPI among the urban slum communities and will act as a roadmap to achieve the Immunization Agenda 2030 in Myanmar. Since incomplete or non-compliance with immunisation is generally associated with the decline of immunity and increase in risk of epidemics to VPDs, there is a need for detail exploration on the underlying causes in order to develop tailored strategies and implement for ensuring that no child is left behind. The study recommended modification of the existing immunisation service delivery in the urban slum communities such as expansion of outreach programmes, mobile team visits and supplementary immunisation activities, regularly updating EPI microplanning by ensuring inclusion of migrant population, regularly monitoring about the dropout and non-compliance with immunisation, etc. Moreover, the study recommended the government and donor agencies to strengthen immunisation programme management, human resources, financing and logistic supplies for delivering immunisation services in the urban slum communities. In addition, given that mobile health (mHealth) technology has been increasing popularity nowadays, the study recommended to implement immunisation reminder system using mHealth intervention among the urban slum inhabitants. Strengths and limitations The strength of the study is that a representative sample of all eligible mothers of 18-month to 24-month-old children from randomly selected clusters could be interviewed using the cluster sampling method. The immunisation coverages of 18-month to 24-month-old children for all basic vaccinations were requested to be checked from the immunisation cards to minimise information bias and recall bias. However, the study had the following limitations. First of all, our study only investigated immunisation completeness and we failed to record the timeliness of vaccines. For becoming an effective childhood immunisation, both completion and timeliness were important since infants with delayed immunisation can be a threat to other children as they can transmit infections and also threaten to develop herd immunity that will increase the likelihood of an epidemic. Moreover, our study only assessed demand-side predictors for complete immunisation and we did not investigate the supply-side predictors such as vaccine availability, health manpower, cold chains, etc, which might also influence the uptake of immunisation in the study area. Furthermore, it was not possible to establish a causal relationship between the predictors and the outcome variable because of the cross-sectional design of this study. In addition, it is less likely to generalise the findings for the entire population in other areas of Myanmar since the study only recruited mothers living in the urban slum area of one township. Conclusion In conclusion, the proportion of complete immunisation among 18-month to 24-month-old children in the studied urban slum area was found to be suboptimal, and the immunisation coverages were lower than the national immunisation targets. Mothers’ high educational status, permanent residency, primiparity, higher immunisation knowledge and positive perception of immunisation were the significant predictors for complete immunisation. A thorough understanding of these predictors will enable the government and other relevant stakeholders to design and implement effective policies, initiatives and campaigns to promote access to immunisation services among the urban slum inhabitants, thereby reducing the outbreaks of VPDs and improving the health outcomes of newborns and under-5 children in Myanmar. supplementary material Acknowledgements We would like to express our sincere gratitude to the Township Medical Officer, Basic Health staff, Auxiliary Midwives, and local authorities from Hlaingthayar Township for their supportive arrangement during the data collection process. We are also grateful to our interviewers and study participants. Footnotes Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Provenance and peer review: Not commissioned; externally peer-reviewed. Patient consent for publication: Not applicable. Ethics approval: This study involves human participants and was approved by Defence Services Medical Research Centre Institutional Review Board (DSMRC-IRB) (Ref: IRB/2018/16). Participants gave informed consent to participate in the study before taking part. Data availability free text: The datasets used and analysed during the current study are available from the corresponding author upon reasonable request. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Data availability statement Data are available upon reasonable request. References Associated Data Supplementary Materials Data Availability Statement Data are available upon reasonable request.