Knowledge, Attitude and Practice of Mothers Towards Immunization of Infants in Health Centres at Addis Ababa, Ethiopia

: Background: In Ethiopia, a considerable proportion of childhood morbidity and mortality is due to vaccine preventable diseases. According to EDHS 2011, only 24% of children were fully vaccinated nationally. Mothers’ knowledge, attitude and practice are key tools to decrease dropout rates and to prevent or control infectious diseases. Thus, this study was aimed to assess Knowledge, Attitude and Practice of mothers regarding infant immunization and their associated factors in immunization clinic in health centers at Addis Ababa, Ethiopia. Methods


Background
World Health Organization (WHO) initiated the Expanded Program on Immunization (EPI) in May1974 with the objective to vaccinate children throughout the world [1]. A major goal for the World Health Organization is the global control of certain infectious diseases [2]. The main strategies for the prevention of infection are to eliminate or diminish the amount of infecting microorganism from circulation, to enhance the host immune response and to treat the infected host. These strategies are achieved by two of immunization types (active and passive) [2].
The millennium development project 2009 emphasizes that, reducing child morbidity in a nation ensures a healthy and robust generation contributing to society. According to this report, Sub-Saharan Africa is the lowest performing region in terms of MDG4-reducing child mortality and Ethiopia is one of the lowest performers in all MDGs [3]. According to the Ethiopia Ministry of health 2011annual health and health related indicator report, national infant and under five mortality rates were 59 and 88/1000 live births, respectively [4]. In this report, Addis Ababa City administration infant and under five mortality rates were 39 and 50/1000 live births respectively. Moreover, reports indicated that the major responsible causes for childhood deaths in Ethiopia are diarrheal diseases, VPDs and malnutrition [5]. Measles attributed to 4% of child and infant deaths in 2004 which was highest of the world [6].
In Ethiopia, Expanded Program on Immunization was started in1980 with the intention of increasing the immunization coverage by 10% annually and to reach 100% coverage in 1990. But the objective set in1980 was not met because of factors such as poor health infrastructure, low number of trained man power, high turnover of staff, low awareness in parents regarding infant immunization, and lack of donor funding [5].
According to the global immunization division CDC report, the global parental Attitude and Knowledge regarding immunization services was low and parents have negative beliefs about measles and vaccination programs [7].Thus, Parental attitudes and beliefs about vaccines are an import ant factor in predicting child's immunization status [8].
Keeping the point of view, this study was conducted to assess mothers' knowledge, attitude and practice regarding infants' immunization and their associated factors in Addis Ababa. Figure 1. Geographical location of the study site, Addis Ababa. Infants in Health Centers at Addis Ababa, Ethiopia

Study Area
The study was conducted in Addis Ababa, the capital city of Ethiopia. It lies at an altitude of 7,546 feet (2,300 meters) above sea level. According to the 2011 federal democratic republic of Ethiopia ministry of health annual health and health related report, during 2013 total population and number of infants was projected to be 3,101,896 and 68,242, respectively. This number (68,242) is approximately represents the total population of mothers with infants in 2013 in the city. In this report, the infant mortality rate of Addis Ababa city administration was 39 per 1000 live births [10]. According to EDHS 2011 report, in Addis Ababa 79% of infants were fully vaccinated. Taking this 79% in to account, still 14,331(21%) infants found in the capital city of Ethiopia were not fully vaccinated [4]. Addis Ababa has 10 sub cities (Fig. 1). In these sub cities, there are 43 hospitals (11 governmental and 32 privates) and 53 health centers which give vaccination services.

Study Design and Sample Size Determination
Cross-sectional study was conducted from March 1 st to April 1 st , 2013 among mothers with infants attending Addis Ababa city administration health centers. Sample size was determined using single population proportion formula n= Z 2 p (1-p) / d 2 , with the following assumptions: prevalence (p) of 50% in order to get large sample size [11], 95% confidence level, 5% margin of error, 10% for anticipated non-response rate, and 1.5 design effect. Accordingly, the minimum sample size (n) was found to be 634. The total sample size was allocated proportionally to the selected health centers. Multistage sampling technique was used for participant selection (Fig. 2).

Data Collection
The data was collected by 18 well trained nurses through face-to-face interview using interviewer-administered questionnaire. Close-ended with some open-ended questions were used to collect information on socio-demographic variables and KAP of mothers regarding infant immunization. The questionnaire was prepared in English version and then translated in to Amharic version for ease of understanding by the data collectors and respondents. Finally, it was translated back again in English version to make ease of data analysis.

Operational Definitions
Good Knowledge:-Those mothers who answer correctly the knowledge questions and if they score the median value and above Poor Knowledge:-Those mothers who answer correctly the knowledge questions and if they score below the median value Positive Attitude:-Those mothers who answer correctly the attitude questions and if they score the median value and above Negative Attitude:-Those mothers who answer correctly the attitude questions and if they score below the median value Good Practice:-Those mothers who answer correctly the practice questions and if they score the median value and above Poor practice:-Those mothers who answer correctly the practice questions and if they score below the median value Fully vaccinated:-An infant who received all doses of the nine vaccines before he/she celebrates the first birth day (one BCG, three doses of Pentavalent, three doses of OPV, three doses of PCV and one dose measles vaccine) Partially vaccinated:-An infant who misses at least one doses of the nine vaccines.

Data Analysis
Data were coded, checked for completeness and cleaned for any inconsistencies. The data were then entered and analyzed using SPSS version 16.0. Descriptive statistics and Binary logistic regression were used and 5% level of precision was used for checking the association between dependent and independent variables. For ease of analysis, to measure knowledge and practice of mothers, eight questions were selected and scored one for each correctly answered and zero for the incorrectly one. The minimum, median and maximum values were calculated. On the other hand, the attitude statements constructed with five alternatives previously (agree, strongly agree, don't know, disagree and strongly disagree) were modified in to three responses i.e. "strongly agree and agree" responses of mothers were taken as having "agree", "don't know" for those neither agree nor disagree responses taken as it is, and "disagree and strongly disagree" responses were coded as "disagree" about vaccination of infants and vaccine preventable diseases [9,10]. Results of the total attitude questions were summed; the minimum, maximum and median values were calculated. Finally, results are presented using tables and graphs.

Ethical Clearance
Approval was secured from the Ethical Clearance Committee of Addis Ababa University. Then, official support letter from the university had been written to Addis Ababa city administration health bureau. Data collection was under taken after permission was obtained from Addis Ababa city administration health Bureau and from the selected health centers. Verbal informed consent was asked and taken from every study participant before asking any information or interview.

Socio-demographic Characteristics
A total of 626 mothers with infants were participated in this study. The mean age of respondents was 26.6 + 4.8 years. Five hundred seventy six (92%) of the study participants were married and only 5.6% were single. Regarding mothers occupation 60.1% were house wives.Socio-demographic characteristics of study participants are summarized on Table 1. Regarding their infants' 328 (52.4%) were females. Age of infants ranged from 1 day to 11 months. The mean age of infants was 4.3 ± 3.3 months. About one third, 189 (30.2%) infants found in the age group of 3-9 ( Fig. 3)

Knowledge of Respondents on Infant Vaccination and Vaccine Preventable Diseases
Of 626 respondents, 542(86.6%) have ever heard information about vaccination (Table 2). Only 23.8% of the respondents knew the objective of infant immunization (

Attitude of Respondents Towards Immunization
The attitude of mothers towards infants' immunization is summarized on table 3.

Practice of Study Participants Regarding Immunization of Infants
Regarding infant immunization practice of mothers, 594 (94.9%) was always bringing their infants for immunization according to the schedule given by health professionals (Table 4). From the total 626 respondents, only 32 (5.1%) missed their appointments for EPI for different reasons (Fig. 5)

Factors Affecting Attitude of Mothers About Infants' Immunization
Literate respondents who attend elementary school were about two times [AOR= 1.688, 95%CI: (1.053, 2.707) where as mothers who achieved higher education were two times [AOR=2.160, 95%CI: (1.208, 3.864)] had positive attitude towards infant immunization than illiterate respondents. Mothers who had infants aged from 2-3 months were two times [AOR=2.014, 95%CI: (1.044, 3.883)] significantly associated with positive attitude about infant immunization program than mothers having infants in the age group less than one month. Factors affecting Attitude of mothers about their infants' immunization are summarized on Table 6.

Discussion
The present study has tried to identify the Knowledge, Attitude and Practice of mothers about immunization of infants in Addis Ababa city administration, Ethiopia. In this study, 542(86.6%) of mothers have heard information about infant immunization, and 261(48.2%) of them have accesses of information from health professionals. Similarly, study participants included in other study areas have got information primarily from health professionals [11,12]. This is due to the fact that vaccination information are usually given for mothers by health professional just before infants/ children taking vaccines with each immunization sessions or at the time of antenatal checkups.
About 90% of respondents correctly mentioned the time when infants should begin vaccination programs (just after birth) and should follow immunization sessions as per of the scheduled time (97.6%). This is similar to a study done in India; all mothers had the knowledge that immunization is important for the child and all of them knew that immunization is to be started at birth and should follow vaccination sessions [13]. In contrast, mothers studied in Ambo, Ethiopia, only 6.7% of respondents know the exact time when infants should begin immunization [11]. The discrepancy between the present study and Ambo district may be due to the difference in information, education and communication (IEC) or in terms of health service accessibility in comparison to this study since 73% of the respondents in Ambo were from rural areas. Similarly, 52.4% of mothers mentioned the time when infants should finish vaccination correctly (at nine months or before the first birth day). This is different from other study done in rural Nigeria, only (14.1%) of respondents mentioned vaccination against childhood killer diseases should be completed at the age of nine months [14]. The inconsistency might be due to the difference in study participants' educational status since about 70% of mothers included for study in rural Nigeria were illiterates, which is quite higher than the illiteracy status of the present study (18%). Despite the fact that greater than half of respondents in this study have mentioned correctly the time when infants should begin and finish vaccination, only 23.8% of mothers correctly knew the objective of immunization (i.e. 'to prevent specific and killer diseases').This is consistent with the response of mothers in rural Nigeria which indicated that only 20.1% [14] and contradicts with another finding in Enugu, Nigeria, 81.2% of respondents mentioned correctly the objective of immunization [15]. The inconsistency between the present participants and the later Nigerian participants may be due to educational status differences in respondents since only 47% of mothers in the present study and 90% of mothers in Nigeria attended secondary school up to higher education.
Over all, in this study 344(55.0%) of mothers had good knowledge and this is lower than the study finding from Nnamdi Azikiwe University hospital, Nigeria which revealed that, 215(70.0%) of mothers had good knowledge [16]. The inconsistency may be due to sample size difference or educational back grounds of the participants. On the other hand, a study conducted in Connecticut (north eastern United States) showed that both fresh and experienced mothers scored poor knowledge about vaccination [10]. The possible explanation about poor knowledge of mothers in developing and developed countries is different. For example, the major health problem in developed countries is not communicable disease and mothers may not focus on vaccine preventable diseases and may not develop good knowledge on communicable diseases, while poor knowledge in developing countries may be due to social, economical or educational reason.
As it is presented on table, 98.7% and 96.5% of respondents in this study had favorable attitude towards immunization of infants, and health personnel, respectively. This is similar to the study done in Kinshasa, Congo; showed that 93.8% of respondents had positive attitude about children immunization and 93.5% of them had friendly seen the health personnel [12]. Despite the fact that respondents had positive attitude towards the EPI program and vaccine providers in general, a large proportion 47.6% of mothers believed that infants took too many vaccines and 73.2% of mothers believed that vaccines are given for infants to prevent non serious (simple) diseases. This type of perception is similarly described by other scholars from Texas health science center at Fort Worth (America) with three consistent belief/attitudes of mothers such as vaccines could harm child, children receive too many vaccinations and vaccinations are given to prevent diseases that are not serious [17]. Basically this similarity should not be seen from the same angle of reference. For example, mothers found in developed countries may develop this type of belief due to they pay attention more about vaccines safety while mothers found in developing countries like Ethiopia may be due to lack of knowledge or social influences. Surprisingly, 49.4% of respondents worried that vaccination may make infants sick and 7.5% perceived that vaccination can cause death in infants. Similarly, 15.0% and 16.1% of respondents believed that vaccination had no any use (not work) and decreases infants' natural immunity respectively. This finding is in line with other cross-sectional study done in Connecticut (north eastern United States) [10].
As far as their practice is concerned on infant immunization, about 98.7% of mothers had and showed EPI card on the spot of immunization. Contradictory, only 29% of mothers in Ethiopian demography and health survey (EDHS) report of 2011 [18] and 41.8% of respondents in Ambo district [11] had showed EPI cards for their children, respectively. Moreover, immunization practice in the present study (94.9%) is also a beat greater than EDHS 2011 and Ethiopia ministry of health 2011 annual health and health related indicator reports which represents 79% and 84.4% of infants were fully vaccinated, respectively [18,4]. The variation is due to a study design difference (institution in this study and national based survey in comparison studies).Nearly 95% of respondents in this study always bring their infants for immunization according to the scheduled time. From the total 455 infants aged less than nine months, 89.5% of them were received vaccines appropriate to their age where as from 171 infants aged nine months and above, 93.2% of infants were fully vaccinated. This is consistent with a cross sectional study done in India, that 98% and 93% of children completely immunized and had been immunized on the schedule, time, respectively [13]. In the contrary, inconsistencies had been seen with other findings. For example, in Congo, mothers' immunization practice based on immunization card showed about 37% [12] and in Ambo, Ethiopia only 35.6% of infants completed all the recommended vaccines [11]. The discrepancy may be due to social or cultural reasons, and study setting differences.
The present study also tried to identify predictors of maternal knowledge, attitude and practice towards infant immunizations using multivariate analysis. Accordingly, Infants in Health Centers at Addis Ababa, Ethiopia literate mothers attending grade 1-8 th were about two times (AOR= 1.781, 95% CI: 1.035, 3.065) more likely to have good knowledge of immunization than illiterate mothers. Similarly, other study findings conducted in United Arab Emirates and Enugu (Nigeria) mothers attending secondary or higher educational status was significantly associated with good knowledge and acceptance of immunization [15,9]. In addition to literacy status, mothers who had infants aged from 3-9 months and 9-12 months old were about two times more likely to have good knowledge of immunization than mothers who had infants below one month. This might be due to the fact that mothers could get information from health professionals and add knowledge of immunization with repeated vaccination visits than mothers who couldn't visit immunization clinics repeatedly. Mothers who gave births two times and above were also about two times more likely to have good knowledge of infant immunization than respondents who delivered only once. Similarly, a study conducted in Kinshasa (Democratic Republic of Congo), showed that mother's experience of an EPI-targeted disease emerged as significant predictors of knowledge of immunization [12]. Inconsistencies regarding birth order of infants were also seen between this study and other study findings which were conducted in Arab Emirates and Ambo district, Ethiopia, [19,11]. The inconsistency might be due to differences in study design, sample size or study participants' educational status.
In respect to respondents' attitude on immunization, literate respondents who attend elementary school or higher education were about two times more likely to have positive attitude towards infant immunization than illiterate mothers. This is consistent with results of different articles reviewed by global immunization division centers for disease control and prevention from 1999-2009 [7]. The possible explanation might be, literate mothers may have more opportunities to understand about vaccination and its importance than illiterate mothers and this may create favorable attitude towards infant immunization than illiterates. On the other hand, study conducted from Oromia zone of Amhara region of Ethiopia which showed no association between maternal education and favorable attitude towards immunization [20]. The discrepancy may be due to educational back ground difference between the two study respondents since only 18% of mothers in the present study and the majority80% of the study participants in Oromia zone Amhara region were illiterates, respectively. It may be also due to study period difference and could be associated with the efforts of health extension workers since the comparison study was conducted six years ago. Similar to knowledge association with infants' age, mothers who had infants aged from 2-3 months were two times more likely to develop positive attitude about infant immunization program than mothers who had infants aged 0-1 month. This may be due to frequent contact of immunization sessions.
Regarding immunization practice, those mothers who had infants with the age range of 1-2 months, 2-3 months and 9-12 months were four times significantly associated with mothers' immunization practices than those mothers who had infants aged 0-1month old. On the other hand, mothers' who had infants aged 3-9 months old were about three times, had infants' immunization practice than their counter parts, respectively.This may be due to BCG vaccines opened only one day per a week. So that mothers who delivered on non BCG vaccination days went to their home without vaccinating their neonates. Moreover, place of delivery was also significantly associated with infant immunization practices. In the present study, respondents who delivered in health institutions were twenty three times significantly associated to infant immunization practice than those mothers who delivered at home. Similarly, respondents who heard information about vaccination and vaccine preventable diseases were about two times significantly associated with infant immunization practice than who didn't heard information yet. This finding is similar with research findings conducted in Ambo, Ethiopia. Besides, Mothers who responded correctly for the time 'when infants should begin vaccination', who mentioned correctly 'the number of vaccination sessions needed' and 'when infants' should finish vaccination' were two times significantly associated with their practices than those mothers who responded wrongly. This is also consistent to other research findings [11]. This paper has its own strength because it is based on primary data and can be used as base-line information for intervention programs and further investigations. Moreover, the sample size was large enough and representative and the questionnaire was pre tested. On the other hand, as a limitation the study was a cross-sectional design and the associations observed may not be causal. Besides, it was institution based study and previously disappointed or dissatisfied mothers may not come to the same health institution again and the number of mothers having unfavorable attitude could be decreased.

Conclusions
In this study, only 55.0%, and 53.8 of respondents had good knowledge and attitude towards immunization of infants, respectively. Despite inadequate knowledge and attitude of mothers towards infant immunization, the majority 84.0% of mothers had good practice of infant immunization. From this point of view, it is possible to conclude that mothers' immunization practice was not really based on their knowledge and attitude regarding immunization of infants. Maternal education and birth order were significantly associated with good knowledge. Similarly, Mothers' education, infants' aged from 2-3 months was significantly associated with favorable attitude towards immunization of infants. Good infant immunization practice was significantly associated with mothers who have ever heard information about vaccination, who know correctly the time when infants should begin immunization, who know correctly the number of sessions needed, who know the time when infants should complete immunization and place of delivery.