Women’s understanding of immunization card and its correlation with vaccination behaviors

ABSTRACT

Background: Despite the proven effectiveness of vaccinations, vaccination uptake is limited in Nigeria. According to the Multiple Indicator Cluster Survey (MICS), one of the main barriers is the lack of accurate knowledge of the vaccination schedule. This study evaluates caregivers’ knowledge of the vaccination schedule and their ability to read the immunization card.

Methods: The study evaluated the knowledge of caregivers in 11 settlements in the Jada local government area of Adamawa State in September 2019. The change in knowledge among caregivers before and after referring to the immunization card was evaluated using a simple statistical hypothesis testing (chi-square test). We also used logistic regression analysis to evaluate the determinants of vaccination knowledge, as well as the correlation between knowledge and actual vaccination behaviors.

Results: More than half of the women had correct knowledge of the vaccination schedule for critical vaccines. However, the knowledge of the caregivers did not improve after referring to the immunization card which contained the information. Caregivers who brought their children to the clinic for vaccination recently were more likely to know the vaccination schedule correctly. Accurate knowledge was highly correlated with the actual vaccination behaviors.

Conclusion: Reference to the immunization card did not improve the knowledge of vaccination schedule, especially among the less-educated population. To increase the demand for vaccinations, one potential policy is to target the uneducated population and help them increase their knowledge.

KEYWORDS:

• Immunization card
• knowledge
• northern Nigeria

Introduction

Vaccination is one of the most cost-effective interventions existent today.^1,2 Ozawa et al. (2017) estimated that vaccinations would avert over 200 million deaths and save 350 million dollars in cost of illness between 2001 and 2020 in developing countries.³

However, the vaccination rate remains low in Nigeria. According to the Nigeria Demographic and Health Survey conducted in 2018, only 31% of children aged 12–23 months have received all basic vaccinations.⁴

Based on the Nigeria Multiple Indicator Cluster Survey (MICS) conducted in 2016/2017, one of the main reasons for incomplete vaccination is misunderstanding – caregivers think their children had completed their vaccination schedule – which accounts for 44.8% of the reasons for incomplete vaccination.⁵

This finding from MICS data indicates that a gap exists between what caregivers know about the vaccination schedule and the actual one. This knowledge gap can be a critical demand-side barrier among caregivers, which prevents them from visiting the clinic for children’s vaccination at appropriate timings.⁵

In the literature, several studies have explored the potential barriers for the low vaccination coverage both from the demand side and the supply side. Meta-analysis studies have consistently found that educational intervention to improve vaccination knowledge among mothers increases vaccination uptake.^6,7

Caregivers can obtain the information on vaccination schedule through various sources. One of the most important information sources is an immunization card. According to the World Health Organization (WHO), the immunization card usually contains critical information, notably the due date of the next immunizations and the immunization schedule specific to the locality.⁸ The role of health workers in informing the caregivers about the vaccination schedule is critical. Health workers are expected to inform the caregivers of the next vaccination schedule and to fill the information on the due date of the next vaccinations in the immunization card.⁸

Despite the potentially critical role that immunization cards have, little attention has been paid to caregivers’ ability to comprehend the contents of the immunization card. Usman et al. (2009, 2011) are the exceptions; recognizing the shortcomings of the design of the immunization card in Pakistan, they redesigned the immunization card and observed the increase in vaccination rate after the introduction of the redesigned card.^9,10

This study evaluates caregivers’ knowledge of vaccination schedule and their ability to read the immunization card. We also evaluate if vaccination knowledge is correlated with the actual vaccination behaviors.

Methods

Study population

This study was conducted in 11 settlements in the Jada local government area of Adamawa State in the northeastern region of Nigeria in September 2019. Convenience sampling was employed to identify the 11 settlements. All the 11 settlements were under the catchment area of one health clinic, Jada I PHCC (Primary Health Care Clinic). We selected settlements that were large so that we could find as many eligible respondents as possible. In each settlement, one woman from each household was selected. Our interviewers visited all the households in each settlement to identify the eligible women. The eligibility criteria was that a mother should have a child who was 8 months old or younger at the time of the baseline survey. If one household had more than one mother with such children, interviewers randomly selected one of them per household.

The vaccination status among these children is the focus of our study. Although unlikely, in case women had more than one child aged 8 months or younger, we asked the vaccination status and behaviors of the youngest child.

We chose the age range of 8 months or younger to simplify the procedure and identify the eligible respondents. The study aimed to identify children who had not received all the vaccinations to evaluate how our program influenced their caregivers’ decision for vaccination. Focusing on children aged 8 months or younger simplifies the selection process because all of such children still need to receive measles and yellow fever vaccines at the age of 9 months.

Our sample included 534 women from 11 settlements.

Procedures

The questionnaire was administered by the interviewers to all women who agreed to participate (refusal rate: 0.37%). We intensively trained 8 nurses to become interviewers prior to the administration of the questionnaire.

The questionnaire had four sections, one of which was to evaluate women’s knowledge of child vaccination. In this section, we first asked a series of questions to evaluate their knowledge of vaccination, such as the number of recommended types of vaccination and the timing of critical vaccinations, without having them look at the immunization card. We then asked them to look at the immunization card, which contained all the information necessary to answer the questions around the child’s vaccination schedule. After the women looked at the immunization card, we asked them the same sets of questions to evaluate if they know how to read the immunization card. Interviewers were trained not to tell the women how to read the immunization card so that we can assess their understanding ability.

We asked four identical questions before and after showing the immunization card: (1) the number of recommended types of vaccinations for children under 12 months, (2) the minimum number of clinic visits necessary to achieve full vaccination, (3) the timing of the first vaccination in months, and (4) the timing of the measles vaccination in months.

According to the immunization card, there are 11 types of vaccinations in 12 months: BCG, HepB, OPV, penta, PCV, rota, IPV, measles, vitamin A, yellow fever, and meningitis. Because the rotavirus vaccine had not been introduced at the time of the survey, 10 types of vaccinations is also considered a correct answer. The minimum number of clinic visits for full vaccination is 6: at birth, 6 weeks, 10 weeks, 14 weeks, 6 months, and 9 months. Because vitamin A at 6 months is not practiced in this region, we also consider 5 as a correct answer. The timing of the first vaccination is at 0 month (at birth), and the timing of the measles vaccination is at 9 months. Because the immunization card also indicates that the second measles vaccination is scheduled at 15 months, the answer 15 is also considered correct.

We additionally asked about the timing of the pentavalent vaccination only after having the respondents look at the immunization card. We asked this question once under the assumption that most of the respondents would not know the answer without looking at the immunization card.

We also collected the sociodemographic information and vaccination status of the two youngest children. The vaccination status was collected based on the records at the immunization card. If the caregivers did not have the immunization card, then we relied on the caregivers’ self-report. In case the caregivers could not recall whether a child received a certain vaccine or not, we recorded the vaccination status as unknown.

Outcomes

The primary objective of this study is to evaluate women’s level of understanding of the immunization card, and its primary outcome is the change in their understanding level of the types of child vaccinations recommended and the vaccination schedule before and after looking at the immunization card.

As secondary objectives, we evaluate the determinants of the correct knowledge of the vaccination schedule, as well as the correlation between the knowledge and the actual vaccination behaviors.

Statistical analysis

Changes in the understanding level of the vaccination types and schedule among caregivers before and after showing them the immunization card are evaluated using chi-square tests for univariate frequency distribution. In particular, we tested if the proportion of women who correctly answered the question about vaccination changed before and after the immunization card provision.

To evaluate the determinants of correct knowledge, we used multivariate logistic regression. In this regression analysis, we included the individual-level factors of mothers, such as age, education level, number of children, work status, and past vaccination experience of older children. We adopted the conceptual framework introduced in Ababu et al. (2017) to identify the potential determinants.¹¹ They employed the modified steps of the Behavioral Change Model to evaluate the determinants of vaccination behaviors.¹¹

To evaluate the correlation between the knowledge and the actual vaccination behaviors, we also used multivariate logistic regression. In the regression, we included indicators that show accurate knowledge of the vaccination as the main independent variable. We also included the sociodemographic characteristics of caregivers, which were identical to the ones included in the previous regression for the determinants of correct knowledge.

Results

Descriptive statistics

Out of 534 women, we have complete information on the baseline characteristics of 515 women who had a child aged 8 months old or younger.

Table 1 presents the demographic characteristics of the respondents. On average, 26.4% of the respondents were 20 years old or younger, 28.9% were between 21 and 25 years old, 25.6% were between 26 and 30 years old, and the remaining was 31 years old or older. The majority of the respondents (94.6%) were Muslim (table not shown). Almost half (42.1%) had never received any formal education, while about one-third (34.1%) received secondary or higher education. On average, they had 3.9 children, and 1.5 of them were 5 years old or younger. Note that the respondents had at least one child because the eligibility criteria was to have a child who was 8 months old or younger. The household size was 6.9, and the average household earning in the previous month of the baseline survey was 17,904 naira (approximately 50 US dollars). A little less than half (42.5%) of respondents had paid work in the previous month.

Table 1. Summary statistics

	Number	%	Mean	Std. Dev
Total N = 515	(1)	(2)	(3)	(4)
Age
20 or below	136	26.4%
21–25	149	28.9%
26–30	132	25.6%
31–35	64	12.4%
36 or above	34	6.6%
Education
None	217	42.1%
Some primary	122	23.7%
Secondary and higher	176	34.2%
Households
Number of children			3.89	2.62
Number of children under 5			1.47	0.80
Number of household members			6.89	4.20
Monthly household earning			17904.3	17388.2
Has Paid work	219	42.5%
Previous vaccination experience for second youngest children in the past 5 years	76	14.8%

The sample is from 515 women who had a child of 6 months old or younger.

“Previous vaccination,” the last variable in Table 1, indicated if respondents had experiences in having their children receive vaccinations. We considered that women had previous vaccination experience if they brought their second to the youngest child for vaccination. Out of the total sample, about 15% (14.8) of the respondents had such experience of bringing their second to the youngest child for vaccination at least once. This proportion is low because many women did not have children aged 5 years old or younger other than the child of 8 months or younger. Out of 515 women, 118 (22.9%) had their second to the youngest children under 5 years old. Among 118 of them, 78 (66.1%) had experiences in taking them to vaccination (table not shown).

Understanding of immunization card

Table 2 Panel A presents the change in the understanding level before and after the respondents looked at the immunization card.

Table 2. Increase in vaccination knowledge after immunization card provision

	# Women with Correct Answer			Difference
	Timing	N	%	(chi-square test)
Panel A: Variables with both timing (Before & After)
#Vaccine types	Before	113	22.3%	0.02
	After	137	27.0%
#Clinic Visits for vaccination	Before	286	56.2%	0.53
	After	293	57.6%
The first vaccine timing	Before	388	75.2%	0.62
	After	383	74.2%
Measles vaccine timing	Before	273	54.6%	0.00
	After	328	65.6%
Panel B: Variables with one timing (After)
#times for Pentavalent vaccine	After	272	54.4%
Timing of 1st pentavalent	After	217	43.4%
Timing of 2nd pentavalent	After	98	19.6%
Timing of 3rd pentavalent	After	95	19.0%

The sample is from 515 women who had a child of 6 months old or younger.

Without women looking at the immunization card, 113 (22.3%) had the correct answer on the number of recommended types of vaccination. More than half (56.2% or 286 women) had correct knowledge of the number of minimum clinic visits to achieve the full vaccination. Almost three out of four (75.2% or 388 women) knew the timing of the first vaccination, while over a half (54.6% or 273 women) knew the timing of the measles vaccination.

As described in the Procedures subsection of the Methods section, women were instructed to look at the immunization card after we captured the baseline knowledge. Having looked at the immunization card, the proportion of respondents answering correctly about the number of recommended vaccinations increased by 4.7 percentage points to 27.0%. On the other hand, their knowledge of the number of minimum clinic visits for the full vaccination as well as of the first vaccination timing did not improve significantly.

We observed a strong improvement in the knowledge of the timing of the measles vaccination; the proportion of respondents answering correctly increased by 11.0 percentage points to 65.6%.

The timing of the pentavalent vaccination was asked only after having respondents look at the immunization card. Over half of the respondents (54.4%) correctly answered that children were supposed to receive pentavalent vaccines three times. We also asked the timing of each pentavalent vaccination; 43.4% of the respondents answered correctly that the first pentavalent should be received at 6 weeks, while less than twenty percent of them had a correct answer for the second and third schedule (10 and 14 weeks), 19.6% and 19.0%, respectively.

Determinants of knowledge

Table 3 presents the determinants of correct knowledge for the two questions we found significant improvement in their understanding after showing the immunization card: the number of recommended types of vaccinations and the timing of the measles vaccination. Restricting the sample to women who got the wrong answer before looking at the immunization card, we evaluated the determinants of correct knowledge with the immunization card. In other words, we observed whose knowledge was more likely to improve with the help of the immunization card.

Table 3. Determinants of correct answer response after shown immunization card

	Wrong answer BEFORE
	Correct Answers [Adjusted OR [CI95%]]
	# Vaccine Types	Measles vaccine timing
Sample:	(1)	(2)
Education (control: none)
Some primary	1.53	0.98
	[0.56,4.14]	[0.47,2.02]
Secondary and higher	3.55**	1.07
	[1.48,8.54]	[0.53,2.16]
Previous vaccination in 5 years	1.84	3.84**
	[0.73,4.64]	[1.42,10.39]
N	356	220

The sample is women who had a child of 6 months old or younger and who did not have the correct answer before looking at the immunization card. The coefficients are adjusted odds ratio based on multivariate logistic regression with confidence intervals in the bracket (95%). Covariates also include age category, number of household members, monthly household earning, number of children under 5, and whether women have paid work. ** significant at 5%.

We found that higher education level (secondary school and higher) was positively correlated with the correct answer for the number of recommended types of vaccinations, while the experience of vaccination for older children was the significant determinant of the correct answer for the timing of the measles vaccination.

Table 4 presents the predictors of correct knowledge of the vaccination schedule when they have access to the immunization card. We observe consistent trends; women with higher education and with experiences of their older children receiving vaccinations are more likely to have correct knowledge. In particular, having experiences of their older children’s vaccination increases the likelihood of having the correct knowledge by two to four times for most of the questions.

Table 4. Determinants of correct knowledge on vaccination

	Correct Answers [Adjusted OR [CI95%]]
	# Vaccine Types	# Clinic Visits for vaccination	The first vaccine timing	Measles vaccine timing	# of Pentavalent	Timing of 1st Pentavalent	Timing of 2nd Pentavalent	Timing of 3rd Pentavalent
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Education (control: none)
Some primary	0.74	1.57	1.13	1.30	1.59	1.06	0.99	0.81
	[0.43,1.29]	[0.97,2.54]	[0.67,1.89]	[0.79,2.15]	[0.98,2.57]	[0.65,1.73]	[0.53,1.84]	[0.43,1.54]
Secondary and higher	1.25	1.34	1.60	1.26	2.27**	1.95**	1.47	1.38
	[0.78,2.02]	[0.87,2.07]	[0.97,2.63]	[0.79,1.99]	[1.44,3.57]	[1.25,3.04]	[0.84,2.57]	[0.78,2.41]
Previous vaccination in 5 years	1.20	1.75	2.81**	2.21**	2.83**	2.50**	4.62**	4.84**
	[0.64,2.25]	[0.94,3.26]	[1.30,6.05]	[1.12,4.37]	[1.46,5.47]	[1.34,4.67]	[2.31,9.22]	[2.38,9.86]
N	490	493	500	487	483	483	483	483

The sample is from 515 women who had a child of 6 months old or younger. The coefficients are adjusted odds ratio based on multivariate logistic regression with confidence intervals in the bracket (95%). Covariates also include age category, number of household members, monthly household earning, number of children under 5, and whether women have paid work. **significant at 5%.

Correlation between knowledge and vaccination behavior

Table 5 presents the correlation between mothers’ vaccination knowledge and the actual vaccination status of their youngest children aged 8 months or younger. Generally, correct knowledge of the number of clinic visits required for full vaccination and the vaccination schedule is positively correlated with most of the vaccination behaviors. For example, having correct knowledge of the number of clinic visits to achieve the full vaccination increases the odds of vaccination completion at birth by 58% (column 2) and the odds of vaccination at 6 weeks by about 49.2% (column 6). On the other hand, this knowledge was not significantly correlated with vaccination at a later stage (10 and 14 weeks) (columns 10 and 14).

Table 5. Correlation between correct knowledge and actual vaccination

	[Adjusted OR [CI95%]]
	Completed Vaccination at birth (BCG, OPV0, HepB)				Completed Vaccination at 6week (OPV1, Penta1, PCV1)				Completed Vaccination at 10week (OPV2, Penta2, PCV2)				Completed Vaccination at 14week (OPV3, Penta3, PCV3)
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)
Correct Answer
# Vaccine types	0.60**				1.02				1.15				1.37
	[0.38,0.95]				[0.67,1.57]				[0.75,1.75]				[0.88,2.12]
# Clinic Visits for vaccination		1.58**				1.49**				1.35				1.30
		[1.04,2.41]				[1.02,2.19]				[0.92,1.98]				[0.87,1.93]
The first vaccine timing			1.77**				1.72**				1.83**				1.80**
			[1.13,2.76]				[1.13,2.63]				[1.19,2.83]				[1.12,2.88]
Measles vaccine timing				1.96**				2.47**				2.03**				2.55**
				[1.28,3.01]				[1.65,3.70]				[1.35,3.06]				[1.62,4.03]
Education (control: none)
Some primary	1.62	1.60	1.70**	1.64	1.09	1.05	1.09	1.11	1.23	1.20	1.23	1.29	1.44	1.39	1.40	1.45
	[0.97,2.71]	[0.95,2.68]	[1.02,2.83]	[0.97,2.75]	[0.68,1.74]	[0.65,1.69]	[0.68,1.75]	[0.68,1.81]	[0.76,2.00]	[0.74,1.94]	[0.76,1.99]	[0.79,2.11]	[0.86,2.40]	[0.84,2.31]	[0.84,2.32]	[0.86,2.44]
Secondary and higher	3.18**	3.01**	3.25**	3.20**	2.18**	2.09**	2.21**	2.28**	1.45	1.41	1.46	1.54	1.61**	1.57	1.59	1.77**
	[1.88,5.40]	[1.78,5.09]	[1.92,5.50]	[1.88,5.44]	[1.38,3.44]	[1.33,3.29]	[1.40,3.48]	[1.43,3.64]	[0.93,2.26]	[0.90,2.20]	[0.94,2.27]	[0.98,2.42]	[1.01,2.57]	[0.99,2.50]	[1.00,2.52]	[1.10,2.85]
Previous vaccination in 5 years	2.72**	2.53**	2.33**	2.38**	1.91**	1.79	1.66	1.56	1.81	1.74	1.59	1.45	1.48	1.44	1.33	1.10
	[1.29,5.74]	[1.19,5.36]	[1.11,4.87]	[1.11,5.09]	[1.02,3.57]	[0.95,3.36]	[0.89,3.11]	[0.82,2.99]	[0.97,3.38]	[0.93,3.26]	[0.85,2.97]	[0.77,2.75]	[0.76,2.87]	[0.74,2.80]	[0.69,2.57]	[0.55,2.17]
N	489	492	499	486	483	486	492	479	478	481	488	474	474	477	484	470

The sample is from 515 women who had a child of 6 months old or younger. The coefficients are adjusted odds ratio based on multivariate logistic regression with confidence intervals in the bracket (95%). Covariates also include age category, number of household members, monthly household earning, number of children under 5, and whether women have paid work. **significant at 5%

Correct knowledge of the timing of the first vaccination and the measles vaccination are both important predictors for vaccination behaviors at any given time. Knowledge of the timing of the first vaccination (at birth) increases the odds of vaccination at any given time (at birth, 6 weeks, 10 weeks, and 14 weeks) by 70% to 80% (Table 5 columns 3, 7, 11, and 15). Knowledge of the timing of the measles vaccination almost doubles the odds of all the vaccinations (columns 4, 8, 12, and 16).

Discussions

This study evaluated women’s knowledge of the vaccination schedule and their ability to read the immunization card. We further evaluated if the vaccination knowledge was correlated with the actual vaccination behaviors.

First, we found that without referring to the immunization card, women’s knowledge of the vaccination schedule was relatively high. More than half of women had correct knowledge of the number of clinic visits necessary to achieve children’s full vaccination, and up to 75% of women knew when they were supposed to visit the clinic for the vaccination for the first time.

Despite a relatively high level of knowledge at the baseline, however, referring to the immunization card did not improve the knowledge level further for most of the questions. For example, the immunization card did not increase the proportion of women who answered correctly on the number of clinic visits for vaccination. A slightly less proportion of women (74.2%) answered correctly on the timing of the first vaccination after looking at the immunization card as compared to before (75.2%), although the difference was statistically insignificant. This result of the insignificant/marginal improvement of the knowledge with the help of the immunization card implies that the immunization card is not useful in providing information on the vaccination schedule among women. Determinant analysis emphasized the importance of higher education to be able to read the immunization card. Among women who had a wrong answer prior to looking at the immunization card, the odds of correctly answering the question of the number of vaccine types is 3.6 times higher among highly educated women.

The exception is the timing of the measles vaccination. While about 54.6% of women had correct knowledge of the timing of the measles vaccination, the access to the immunization card increased the proportion of women with the correct knowledge significantly by 12 percentage points to 65.6%. Determinant analysis revealed that, among women who had a wrong answer prior to looking at the immunization card, if a woman had experience in taking her child to the measles vaccination in the past 5 years, they were 3.8 times more likely to be able to answer correctly after looking at the immunization card. This analysis might imply that it is not the immunization card that helps women to have correct knowledge but that their experiences might have simply helped them remember the timing.

Women’s previous experiences of taking their children for vaccination were highly and consistently correlated with the correct knowledge of the vaccination schedule (Table 4). This positive correlation was stronger than the correlation between higher educational attainment and correct knowledge. Combining this result with the finding above, our findings strengthen the argument that the immunization card does not help improve women’s knowledge of the vaccination schedule, but mostly the learning from experience does.

Knowledge of vaccination schedule was strongly correlated with the actual vaccination behaviors, regardless of the timing of the vaccination (Table 5). This result confirms that improving caregivers’ knowledge of vaccination can potentially increase vaccination uptake. This finding is consistent with previous studies which emphasize the importance of knowledge on the actual vaccination uptake.¹² We also found that higher education (secondary school or more) was strongly correlated with the likelihood of the vaccination at all the timings, while the experience of vaccination in the past 5 years increased the current vaccination, but only at the early stage. Contrary to our finding on the determinants of knowledge, higher education is more important than the previous experience of vaccination.

Overall, this paper has an important contribution to the literature. The current immunization card does not help women understand the vaccination schedule. This finding has an important policy implication. We should consider redesigning the immunization card so that women can better comprehend it. If low literacy is the barrier to the understanding of the immunization card, organizing an educational health session might be able to improve the comprehension level of the immunization card.

Although the literature consistently emphasizes the usefulness of reminding caregivers about the next vaccination schedule,¹³ there can be various constraints. For example, setting up the automated reminder system can be expensive.¹⁴ One of the challenges that the rural population in Nigeria faces is the lack of phone network services.¹⁵ In such a situation, using health staffs to remind caregivers might be one of the few possible options. However, health staffs often do not adequately inform the next schedule.¹⁶ In such a resource-limited setting, it is equally important to empower women with health knowledge so that they can be actively involved with the vaccination activity instead of passively waiting for the reminders from health providers.

Limitations

This study has several limitations. First, because the sample is not representative of the country or region, we do not generalize the findings. Second, although this study reveals the important finding that the immunization card does not improve caregivers’ knowledge of the vaccination schedule, we cannot identify if it is because of the complicated design of the immunization card or because of the low literacy of the caregivers.

Conclusion

This study examined how the immunization card helps caregivers understand the vaccination schedule correctly among women who had a child aged 6 months or below in Jada local government area of Adamawa State in Nigeria. The reference to the immunization card did not improve their knowledge of the vaccination schedule for the most part. In particular, only the educated and experienced women increased their understanding of the vaccination schedule after looking at the immunization card. However, because knowledge of the vaccination schedule is a strong predictor of the actual vaccination behaviors, one potential policy to increase the demand for vaccinations is to increase their knowledge, especially of the uneducated population, by redesigning the immunization card or by other methods.

Disclosure of potential conflicts of interest

Authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Funding

The study received no funding support.