Immunization Coverage Among Slum Children: A Case Study of Rajshahi City Corporation, Bangladesh
..........................................................................................................................

Md. Rafiqul Islam (1), Md. Mahfuzar Rahman (2), and Md. Mosfequr Rahman (3)

1. Dr. Md. Rafiqul Islam
Associate Professor and Chairman
Department of Population Science & Human Resource Development
University of Rajshahi-6205, Bangladesh.
E-mail: rafique_pops@yahoo.com

2. Md. Mahfuzar Rahman
Research Fellow
Department of Population Science & Human Resource Development
University of Rajshahi-6205, Bangladesh.

3. Md. Mosfequr Rahman
Lecturer
Department of Population Science & Human Resource Development
University of Rajshahi-6205, Bangladesh.
E-mail: mosfeque @ gmail.com

Correspondence to:
Dr. Md. Rafiqul Islam
Associate Professor and Chairman
Department of Population Science & Human Resource Development
University of Rajshahi-6205, Bangladesh.
E-mail: rafique_pops@yahoo.com

..........................................................................................................................

ABSTRACT The study attempted to identify important effects of some selected variables in complete child immunization coverage. The data for the study were collected in 2006 from the slum areas of Rajshahi City Corporation, Bangladesh. With regard to immunization coverage for the children under age five who were still alive at the time of the survey, the figure for full immunization was higher (92.3%) in the higher ages (24+ months) than the age 12-23 months (89.5%). Application of logistic regression model suggests that demographic and socio-economic factors are associated with the chance of child immunization. Place of delivery and exposure to mass media has highly significant effects on child immunization. The results show that the partial immunization coverage among the children is gradually decreasing when the age of the child increases. Similarly, the prevalence of the under weight and obese among the children increased in spite of being fully immunized.

Key words: Immunization coverage, Slum child, Logistic regression model and Expanded Program on Immunization (EPI).
...........................................................................................................................

Bangladesh, situated in South Asia, emerged as a unitary and independent country on December 16, 1971. It is a country of 1,47, 570 square kilometers and around 147 million people (World Population Data Sheet, 2006) with the highest population density (839 per km2) in the world (U.S. Department of State, 2004). The people of Bangladesh are mostly poor and it is well known for its rapid population growth. Just over 25% of its total population lives in urban areas and the rest (75%) lives in rural areas (Slums of Urban Bangladesh: Mapping and Census, 2005). Life expectancy at birth is 61 years for male and 62 years for female (World Population Data Sheet, 2006).

The World Health Organization (WHO) launched the Expanded Program on Immunization (EPI) in 1974. The program focused on tackling major childhood diseases: measles, tuberculosis, pertussis (whooping-cough), diphtheria, tetanus and poliomyelitis, aimed at universal immunization of children against all the above- mentioned diseases by 1990. Under the EPI, a child is likely to receive one dose of BCG for protection against tuberculosis, three doses of DPT (diphtheria, pertussis and tetanus), three doses of OPV for poliomyelitis protection and one dose of measles vaccine by his/her first birthday.

Children are the future assets of a country. Women are the heart of development; they rear and bear the children. The study conducted in Ludhina slums has found that there is a significant relationship between education as well as poverty with the acceptance of complete immunization (Panda et al., 1993). De Partha and Bhattacharya (2002) have shown that those mothers, who have gone for prenatal care and delivery care after giving birth, are most likely to immunize their child. Different studies have shown that the adoption/practices of immunization of pregnant women and children is positively and significantly correlated with the educational status and income of the mothers/parents (Kaur and Narwal, 1988; Srivastava and Saksena, 1988; Viswnathan and Rohed, 1990). This is due to the linkage between education and income with awareness and knowledge of all types of vaccination and also motivation of the people (Roy et al., 1988).

A survey conducted in the slums of Indore provides evidence that access to essential services, such as delivery and immunization, is different across different categories of slums: complete immunization was 34%, 45% and 49% for the most vulnerable, moderately vulnerable and others, while the percentage of home deliveries were 69, 50 and 38 for those respectively (Environmental Health Project-India, 2004).

The EPI is considered a successful story in Bangladesh because of its remarkable progress during the past 20 years. Immunization was started in Bangladesh in 1979 with the partnership effort between the Government of Bangladesh and the Non Government Organizations (NGOs) under the active initiative of WHO and UNICEF to combat six vaccine preventable diseases with the objective of reducing morbidity, mortality and disabilities occurring due to these diseases, by making free vaccination services available to all the eligible children. The EPI provides almost universal access to immunization services as measured by the percentage of children under the age of one receiving BCG, which has increased a mere 2% in 1985 to cover 95% during the year from 1994-2003. However, the percentage of children under the age of one receiving all doses of vaccine at the right time and interval has been maintained only between 50-63% during the same period due to high dropout rate and invalid doses and 64% of the 12-23 months old children were nationally fully immunized with valid doses of all antigens by 12 months of age (EPI Coverage Evaluation Survey, 2005). A study conducted in Bangladesh (Jamil, K Bhuiya A. et al., 1999) showed that children living in communities where outreach clinics were further than 2 miles away were 30% less likely to be immunized than children living in communities where outreach clinics were within 2 miles.

This research is important because it will investigate other research and creates a vast research field to improve programs on mother-child health and achieve 100% immunization coverage in Bangladesh. Therefore, the main aim and objective of this study is to identify the factors, which are associated with the immunization coverage.

The data of this study were collected in 2006 from 8 different slums of 3 wards of Rajshahi City Corporation, Bangladesh. These data were collected through personal interview method from 700 married women in the childbearing ages (15-49 years) who had at least one child under five years of age at the time of interview, taking into consideration that the selection should be consistent with our objectives. The 700 households that were selected from 8 different slums of 3 wards are presented in the following table.

Various alternative statistical tools exist for analyzing the extent of immunization coverage over time. This paper reports the results from multivariate logistic regression estimation. Logistic regression models were used to determine the relative effects of various characteristics on child immunization coverage. The dependent variable used in this model is given below:

Y=1, if the children are fully immunized (3 dose of Polio and DPT each, one dose of BCG and Measles) and Y=0, otherwise.

Independent variables used in the model are presented in Table 1.

Note: * Exposure to mass media (a composite index has been computed for this purpose, based upon two factors - whether she watches T.V. and listens to Radio every week).

According to the Expanded Program on Immunization a child who received BCG, Measles and three doses of DPT and Polio each, is considered as fully immunized. In Rajshahi City Corporation, 79.0 percent of children aged 12-23 months old are fully immunized with valid doses of all antigens by 12 months of age (EPI Coverage Evaluation Survey, 2005). This is also the highest valid fully immunized rate among the other City Corporations in Bangladesh. The antigen-specific valid coverage rate is 99.4 percent for BCG, 98.9 percent for OPV-1, 98.6 percent for OPV-2, 94.4 percent for OPV-3, 98.9 percent for DPT-1, 98.0 percent for DPT-2, 86.0 percent for DPT-3 and 86.1 percent for measles.

Source: National Immunization Mission, GOB.

4. RESULTS AND DISCUSSIONS

In spite of various socio-economic and cultural constraints of immunization services Bangladesh has achieved mentionable success in immunization coverage (Sarker, 1998). Table 2 shows that with regard to immunization coverage for the children under age five who were still alive at the time of the survey, the figure for full immunization was higher (92.3%) in the higher ages (24+ months) than the age 12-23 months (89.5%). Table 2 also shows the incidence of partial immunization was high (97.2%) for the child aged 0-11 months. The partial immunization is gradually decreasing when the age of the child increases. Table 2 also indicates that a mentionable number of the children (5.6%) were not being immunized at all.

Table 2. Percentage of the Children with Immunization Coverage, Bangladesh

Receiving Immunization	Children’s Age at Interview (month)
	0-11	12-23	24+	All Ages
Full	0%	89.5%	92.3%	86.9%
Partial	97.2%	3.9%	2.3%	7.6%
Not at All	2.8%	6.6%	5.4%	5.6%

Although a high under weight and obese prevalence (77.9%) was seen among the children (immunized and not immunized together) as compared to the not malnourished (22.1%), a marked difference is seen in the prevalence between these two groups of children (Table 3). Table 3 shows that among the immunized children, only 22.8% are not malnourished which is much lower than the under weight and obese children (77.2%). Again among the not immunized children, this figure is also approximately the same.

Table 3. Prevalence of Nutrition (having BMI 18.5-24.9) among Fully Immunized and not Immunized Children Aged 12-23 Months, Bangladesh 2006

Nutritional Status*	Vaccinated	Not Vaccinated	Total
Under Weight and Obese	125 (77.2%)	16 (84.2%)	141(77.9%)
Not Malnourished	37 (22.8%)	3 (15.8%)	40 (22.1%)
Total	162 (100%)	19 (100%)	181 (100%)

Note: * = Body Mass Index (BMI) of 18.5-24.5= Not Malnourished (Normal Weight)
Body Mass Index (BMI) less and above 18.5-24.5=Under Weight and Obese respectively.

Logistic regression analysis can go some way towards identifying those variables, which are truly related to child immunization coverage. The category with the relative odds of 1.00 represents the reference category for that categorical variable. In this section, Table 4 presents the estimate of logistic coefficients, standard error of these estimates, Wald chi-square, significant probability and the relative odds calculated for each category of the categorical variables. Here four independent variables statistically and significantly affected immunization coverage. These variables were the place of delivery, mother's education, family's monthly income and exposure to mass media of the mothers.

The low odds ratio for the uptake of immunization at higher age of child indicates that the probability of being immunized of the children is low in the higher ages. Our results reveal that the child of age 24 months and above is 0.231 times less likely to be completely immunized than the child in age group 0-23. From this it may be concluded that the parents of the children do not properly follow the immunization schedules by EPI guidelines with the increase of age of child. From the results of logistic regression analysis, it appears that the high birth order has a negative effect on full immunization coverage of children relative to the reference category. The results found that the child of birth order 2+ is 0.987 times less likely to be fully immunized than the single birth order child. This clearly shows the negligence by the mother regarding child immunization at higher birth order.

Generally, it is expected that those mothers who are already familiar with different kinds of health services, the likelihood of immunizing their children is higher than those who are not familiar. In institutional delivery, the children are given the polio and BCG just after birth, along with a vaccination card recording the vaccination schedule. The mothers are advised to immunize their children according to the given schedule. In this study, it is seen that mothers who gave birth in an institution (hospital or other health center) has a positive significant effect on child immunization. Here the odds of immunizing their children are 1.038 times higher than those who did not go for institutional delivery (reference category). Education widens the mental horizon of people. An educated woman has better knowledge of the availability of different kinds of health services and their necessity than their illiterate counterparts. The study indicates that mother's education has a significant effect on immunizing their children. The literate mothers have the higher odds (1.035) of immunizing their children as opposed to the illiterate mothers. This may be due to the higher acceptability of preventive health services by the educated mothers.

Husband's occupation also exerts significant impact for complete immunizing of children. In this study we see that husband's occupation (business and service) plays a positive significant role in immunization of children. From table 4, it is seen that having a business occupation is 1.059 times and service is 1.107 times more likely for taking up full immunization of their children, than that of labor group. It is expected that the children belonging to a higher income household should experience higher practices of immunization. It is presumed that the per capita consumption of health boosting goods and services for the children in higher income household is more than the children in the lower income households. The result shows that the family's monthly income is another important factor that has positive and significant effect on child immunization. It indicates that the mothers having household monthly income TK. 2001-2500 is 1.318 times and monthly income TK. 2501+ is 1.235 times more likely to take full immunization of their children than the women having household's income 2000 per month (reference category).

Mass media like radio and T.V. being an informal channel, can play an important role in disseminating information about the availability of different mother and child health services and their usefulness, which may lead to increase the coverage of complete immunization. The regression co-efficient corresponding to exposure to mass media is calculated and the co-efficient is statistically significant. The result provides that the mothers who are exposed to any mass media, the likelihood of immunizing their children is more (2.732 times) as compared to the mothers who are not exposed at all.

Table 4: Logistic Regression Estimation for the Effect of Some Selected Characteristics with Immunization Coverage as the Dependent Variable, Bangladesh 2006

Characteristics	Coefficient ( )	S.E. of estimates	Wald	Significance	Odds Ratio
Child’s age at interview (months) 0-23 ®	-	-			1.000
24+	-1.463	0.238	37.725	0.629	0.231
Birth order of the index child 1®	-	-			1.000
2+	-0.013	0.248	0.003	0.959	0.987
Place of delivery Not institutional ®	-	-			1.000
Institutional	0.038***	0.239	0.0.25	0.000	1.038
Mother’s education Illiterate ®	-	-			1.000
Literate	0.035**	0.259	0.018	0.042	1.035
Husband’s occupation Labor ®	-	-	0.059	0.971	1.000
Business	0.057	0.383	0.022	0.882	1.059
Service	0.101	0.423	0.058	0.810	1.107
Family’s monthly income 2000®	-	-	0.199		1.000
2001-2500	0.276**	1.109	0.062	0.016	1.318
2501+	0.211***	0.558	0.143	0.000	1.235
Exposure to mass media Not exposed ®	-	-			1.000
Exposed	1.005***	0.413	5.923	0.000	2.732
Constant	1.419	0.460	10.495	0.001	4.443

Note: ® = Reference category

***,**, and * indicate p<0.001 (highly significant),p<0.01 (significant) and p<0.05(less significant) respectively. 

This study indicates that though the percentage of immunization is very high yet a considerable percentage of children did not take any immunization in the slum areas of Rajshahi City Corporation. Out of all the selected variables that are included in the logistic regression analysis to build the model, place of delivery, mother's education, family's monthly income and exposure to mass media play the significant role in determining immunization coverage. The results show that complete immunization coverage was lower for children whose age is above 24 months and for children whose mother has more than one child. The children who were not being started on immunization according to the immunization schedules provided by EPI guidelines, may have a lower chance of being immunized at later stages and may show the negligence of the mother regarding child immunization at higher birth order. The analyses also indicated that the risks of complete child immunization were found to be considerably higher among the selected variables for those mothers who were exposed to mass media. Therefore, an effective policy and recommendations are needed to achieve 100% immunization coverage in particular, in Rajshahi City Corporation, Bangladesh. The specific recommendations are as follows:

i). Child immunization coverage may be increased by informing the parents about the dangerous effect of not properly following the immunization schedules provided by the EPI guidelines through various mass media like TV, radio, newspaper, billboard etc.

ii). Create awareness among mothers about the proper schedule of immunization and encourage them to go for immunization at correct age of child. This could be done through information, education and communication (IEC) campaigns.

iii). To encourage parents to have a small family. Therefore, it will increase immunization coverage among the children and the target to achieve 100% immunization coverage throughout the country level will succeed.

iv). Government and non-government organizations efforts during specified National Immunization Days of high dose capsule distribution program must be carried out.