Social capital, trust in health information, and acceptance of Measles-Rubella vaccination campaign in Tamil Nadu: A case-control study.

Background: Parents' decision about vaccination of children is influenced by social relationships and sources of information. The aim of this study was to assess the influence of social capital and trust in health information on the status of Measles-Rubella (MR) vaccination campaign in Tamil Nadu. Materials and
Methods: This was a case-control study carried out in Kancheepuram district in Tamil Nadu where the MR vaccination campaign offered by Government of Tamil Nadu had poor acceptance. Cases were parents of children who had refused the MR vaccine and controls were parents having children in the same age group who had accepted the vaccine. Data on social capital and trust in health information were collected by using social capital scale developed by the researchers and trust in the source of information was measured by using simple questions on the level of trust in the information source.
Results: Nonadministration of MR vaccine was high among young parents and parents of younger children. Vaccine acceptance was higher when it was offered at school (P < 0.000) and also among parents who trusted school teachers (P < 0.003) and other school children (P < 0.014) as source of information. MR vaccine acceptance was less among parents who trusted social media and WhatsApp information. Greater levels of health-related physical social capital led to greater vaccine hesitancy. Multivariate analysis revealed that greater the age of the child, better parental attitudes toward vaccination, poorer health-related physical social capital, and greater trust in health information provided by school teachers led to overall greater acceptance of the MR vaccine.
Conclusion: Strong homogeneous bonding social capital had a negative influence on MR vaccine acceptance. Schools and school teachers played a vital role in influencing parental decision to vaccinate.

Introduction

Vaccination is one of the most cost-effective, proven child survival interventions and prevents sufferings from avoidable sickness, disability, and death.[12] After elimination of small pox, polio, maternal and neonatal tetanus, and recently yaws in India, the current goal is to eliminate measles and control congenital rubella syndrome (CRS) by 2020.[3] Achieving this is possible if immunization is carried out in a campaign mode initially and switched over to a program mode. Because Pune-based serum Institute of India is the only indigenous manufacturer of the measles–rubella (MR) vaccine, it is being introduced in phases in the country. In the first phase, Tamil Nadu, Karnataka, Puducherry, Goa, and Lakshadweep are covered under the MR campaign.[3]

The Department of Public Health of Tamil Nadu targeted to provide MR vaccine to 1.76 crore children between February 6, 2017 and February 28, 2017.[4] However, the department was able to cover only 77 lakh children due to hesitation of parents about the vaccine after rumors that spread through mobile phone and social media.[4]

New modes of communication rapidly spread the information and affect public opinion and confidence on vaccines, which leads to hesitancy, delay, and refusal in accepting vaccination.[5] Parents mentioned rumors and fear of side effects as major reasons for not administering MR vaccine for their children in Tamil Nadu.[4] Previous studies carried out in other countries have also highlighted the importance of rumors in creating vaccine hesitancy.[678] Some studies have analyzed the influence of antivaccination movement in social networks on vaccination decisions and public trust in vaccines.[910] A study which analyzed YouTube videos on human papilloma virus vaccine showed that most videos were negative in attitude and were liked more by viewers than positive ones.[11] Amplification of misinformation affects perception and behaviors of people in the society and it is termed as “social amplification of risk.”[12] Such social amplification of risks of MR vaccine is likely to have affected the vaccine coverage in Tamil Nadu.

Generally, social capital is defined as social relationship between people that enables protective outcomes.[13] The values, norms, trust, and reciprocity within a society are regarded as social capital and are seen as resources held between individuals within the networks and are subjectively verifiable.[1415] Based on the extent of homogeneity and relationships within the group, social capital is also classified as bonding within strongly knit homogeneous groups, such as families, bridging, with heterogeneous groups, such as people from other communities, and linking with systems and people in a power hierarchy.

In the last two decades, many researchers have studied the relationship between social capital and individual health and health behaviors.[1617] While bonding social capital increases physical and emotional support for adapting healthy behaviors, bridging provides assets and more credible information and linking capital increases compliance to suggested behaviors such as administering vaccines.[1819] People accept health protective action if they trust the information sources such as interpersonal networks, media, social networks, and the government.[20212223] Higher social capital is related to high immunization coverage when communities spread affirmative messages about vaccine that helps to build trust and increases utilization of vaccination.[22] Though the role of social capital in vaccine acceptance has been well described, there is dearth of literature on the effect of social capital on vaccine hesitancy and vaccine acceptance in the Indian context.

The present study aims to examine how different forms of social capital and trust in health information was associated with parents' acceptance of MR vaccination for their children in rural Tamil Nadu.

Materials and Methods

Study setting

Tamil Nadu is one of the states in South India, with a population of 7.2 crore (72 million). The state was ranked sixth in India according to the Human Development Index (HDI) in the year 2011. The state also has very good health indicators and is regarded as one of the best performing states as far as health care delivery is concerned. This study was carried out in Tamil Nadu where 68% of children in the age group of 12–23 months receive full immunization in comparison to the national coverage rate of 62%.[23]

The study was conducted in Kancheepuram district in Tamil Nadu, which is a high-focus district because only about 50% of the children are fully covered by vaccination.[24] During the MR vaccine drive, the acceptance to start with was only 70% in the initial campaign period, which increased to 85% by March 2017.[25] The Government of India has identified 201 high-focus districts which have 50% of all unvaccinated or partially vaccinated children in the country based on the Universal Immunization Program coverage data. Kancheepuram is one among the identified district in Tamil Nadu.[26]

Study design and sampling

The study was conducted using a case–control design. Because this study focused on the hesitancy to accept MR vaccine, cases were children in the age group of 9 months to 15 years, whose parents had refused the MR vaccine during the campaign and controls were children in the same age group whose parents had accepted the MR vaccine. The vaccination status was self-reported. No matching was performed during sampling.

The sample size was calculated for comparison of two proportions using prevalence of negative attitudes regarding vaccinations among the controls to be 84%, for an expected odds ratio of 5, with 95% confidence and 80% power.[27] Negative attitudes toward vaccines were taken into consideration instead of positive attitudes as the study focuses on vaccine hesitancy and refusal. The required sample size was 92 cases and controls and it was rounded off to 100 in each.

The study was carried out in the Kattankulathur block of the Kancheepuram district. Four villages were selected randomly and door-to-door enumeration of families with children between 9 months to 15 years of age was performed. Equal number of cases and controls were selected from each village from the enumeration data. The number of cases and controls were proportional to the size of the village.

Study instruments

Based on the domains of social capital identified by the researchers in a previous study, a social capital scale was developed.[28] As defined previously, social capital is the sum of all social relationships which support them. Physical social capital—the physical support provided by social relationships—and informational social capital—which provides information about the vaccines—are the most influential domains in determining vaccine acceptance. Therefore, these two domains were selected for measurement. Based on who was the social asset, the social capital was categorized as bonding capital when it was immediate family members; bridging capital when it was neighbors, friends, relatives; and linking when it was doctors and health workers.[14] The first item in the scale captured information capital pertaining to the social asset who provided health-related information. Physical social capital was measured by information about who and how many helped to take care of their children and helped with daily life responsibilities and helped them access health care facilities and vaccination in particular. The social capital scale was tested for face and content validity by a panel of experts and validated. Trust in the source of information about the MR vaccination was measured using simple questions on the level the respondent trusted the information source for different types of information about the MR vaccine such as place of vaccine availability, benefits of vaccine, potential risks, and age group who is eligible for the vaccine. Basic demographic characteristics of the parents were also obtained. Attitude of parents toward vaccination and vaccine hesitancy was measured using a standard vaccine hesitancy scale.[29] Data were collected between March and April 2017, which was the period during and immediately following the campaign. All the data were collected by the first author of this paper personally. She consulted the other two authors during the design of the questionnaire for purposes of standardization of questionnaire administration.

Analysis

Respondents were asked to rate their vaccination hesitancy on a Likert scale of 1–5 corresponding to strongly disagree to strongly agree. The negative attitude items were reverse scored and then overall vaccine attitude and hesitancy score was calculated and standardized. Standardization was done by the formula (Observed Value – Minimum Value)/(Maximum Value – Minimum Value). Informational social capital was computed for bonding, bridging, and linking assets separately by multiplying (i) the number of people providing the support, (ii) their distance from the respondent's residence (living within 5 km scored as 2 and more than 5 km scored as 1), (iii) frequency of contact (scored as 5: daily, 4: weekly once, 3: once in two weeks, 2: monthly once, and 1: whenever needed), and (iv) the perceived closeness of the persons with the respondent (scored 1: not close, 2: close to an extent, and 3: very close). Physical social capital was computed for bonding and bridging assets by multiplying (i) number of people providing support and (ii) the extent to which it was helpful (scored as 1: mild, 2: moderate, and 3: high). After calculating overall score for informational and physical social capital, each was standardized and added to obtain the overall social capital score. To calculate the score of trust in information, sources of MR vaccine information and the extent of trust (1: no trust, 2: some trust, 3: high trust) on the sources mentioned were used. Total trust score was calculated by adding the scores given to different sources for the different types of information about the MR vaccine and then it was standardized. Standardization was done to bring the different values measured in various scales to a theoretically common scale. t-Test analysis was performed to compare vaccination attitude scores, health-related social capital scores, and scores of trust in health information among MR vaccinated and nonvaccinated children. A multiple logistic regression was performed to explore the factors that influenced MR vaccination acceptance. The data were entered in MS Excel spreadsheet and analysis was performed using Statistical Package for the Social Sciences (SPSS) version 21.

Ethical considerations

The study was reviewed by the Institutional Review Board and Ethical Committee of the School of Public Health, SRM University, Tamil Nadu, India. Verbal informed consent was obtained from all the participants. The confidentiality of the participant details was maintained throughout the study.

Results

Table 1 shows the demographic characteristics of the study participants. About 80% of the respondents among the cases as well as controls were mothers. About half the parents were in the age group of 30–39 years. There was a significant association between age of the child and MR vaccine administration (P < 0.000), which shows that younger the children, lesser the administration of MR vaccine. Also there was a significant association between acceptance of vaccination and availability of MR administration at school (P < 0.000). There was ubiquitous access to smartphones in the study sample. Table 2 depicts the respondents' attitude regarding vaccination and vaccine hesitancy.

Table 1

Characteristics of the study population

Characteristic	Categories	MR vaccine accepted (%)	MR vaccine not accepted (%)	P
Parent	Father	17 (17%)	20 (20%)	0.585
	Mother	83 (83%)	80 (80%)
Age group of parent (years)	20-29	31 (31%)	52 (52%)	0.009*
	30-39	54 (54%)	40 (40%)
	40-49	15 (15%)	8 (8%)
Education of respondent	1-12th std	80 (80%)	67 (67%)	0.089
	Graduate	10 (10%)	18 (18%)
	Postgraduate	3 (3%)	9 (9%)
	Others (vocational/diploma)	7 (7%)	6 (6%)
Monthly family income	Less than Rs. 25000	94 (94%)	87 (87%)	0.118
	Rs. 26000-50000	6 (6%)	10 (10%)
	Greater than Rs. 50000	0	3 (3%)
Gender of the child	Male	55 (55%)	61 (61%)	0.390
	Female	45 (45%)	39 (39%)
Age group of the child	1-7	41 (41%)	69 (69%)	0.000*
	8-14	52 (52%)	31 (31%)
	15 and above	7 (7%)	-
Whether MR vaccine is administered in the school?	Yes	92 (92%)	65 (65%)	0.000*
	No	0	12 (12%)
	Not applicable	8 (8%)	23 (23%)

Table 2

Attitudes regarding vaccination

Statement	Cases versus controls	Strongly agree	Agree	Neither agree nor disagree	Disagree	Strongly disagree
Childhood vaccines are important for my child’s health	MR vaccine denied		1 (1%)	10 (10%)		89 (89%)
	MR vaccine given				10 (10%)	90 (90%)
Childhood vaccines are effective	MR vaccine denied		1 (1%)	2 (2%)	15 (15%)	82 (82%)
	MR vaccine given			2 (2%)	14 (14%)	84 (84%)
Having my child vaccinated is important for the health of the others in the community	MR vaccine denied		7 (7%)	9 (9%)	18 (18%)	66 (66%)
	MR vaccine given		1 (1%)	7 (7%)	22 (22%)	70 (70%)
All vaccines offered by the government are beneficial for the community	MR vaccine denied		1 (1%)	9 (9%)	20 (20%)	70 (70%)
	MR vaccine given		3 (3%)	1 (1%)	21 (21%)	75 (75%)
New vaccines carry more risk than older vaccines	MR vaccine denied	20 (20%)	37 (37%)	33 (33%)	8 (8%)	2 (2%)
	MR vaccine given	8 (8%)	31 (31%)	24 (24%)	33 (33%)	4 (4%)
The information I receive from the government about vaccines is trustworthy	MR vaccine denied	1 (1%)	13 (13%)	13 (13%)	38 (38%)	35 (35%)
	MR vaccine given		2 (2%)	7 (7%)	42 (42%)	49 (49%)
Getting vaccinated is a good way to protect my child from diseases	MR vaccine denied		2 (2%)	2 (2%)	39 (39%)	57 (57%)
	MR vaccine given			1 (1%)	32 (32%)	67 (67%)
Generally I do what my doctor or health care provider recommends regarding vaccine for my child	MR vaccine denied		3 (3%)	1 (1%)	27 (27%)	69 (69%)
	MR vaccine given	1 (1%)			27 (27%)	72 (72%)
I am concerned regarding serious adverse events of vaccines	MR vaccine denied	31 (31%)	32 (32%)	11 (11%)	19 (19%)	7 (7%0
	MR vaccine given	8 (8%)	28 (28%)	6 (6%)	39 (39%)	19 (19%)
My child does not need vaccines for diseases that are not common	MR vaccine denied	11 (11%)	21 (21%)	6 (6%)	35 (35%)	27 (27%)
	MR vaccine given	2 (2%)	13 (13%)	3 (3%)	38 (38%)	44 (44%)

Figure 1 shows the comparison of vaccination attitude and vaccine hesitancy scores among parents of MR vaccinated versus nonvaccinated children. It was seen that those who had accepted the MR vaccination had a more positive attitude toward vaccinations in general. Table 3 shows the comparison of social capital scores among parents who accepted the MR vaccination and those who did not. It was seen that there was a significant difference in linking information capital scores for parents who accepted MR vaccine (score = 0.011, SD = 0.03) and parents who did not accept the vaccine (score = 0.039, SD = 0.12); t (198) = −2.03, P = 0.043. Also, there was a significant differences in bonding physical social capital scores between parents who accepted MR vaccine (score = 0.084, SD = 0.12) and those who did not (score = 0.218, SD = 0.30; t (197) = −4.0, P = 0.000). Figure 2 shows the comparison of the overall health-related social capital score among those who accepted the vaccination and those who did not, and it is seen that those who did not accept the MR vaccine had a greater overall social capital.

Figure 1

Comparison of vaccination attitude and vaccine hesitancy scores among parents of MR vaccinated versus nonvaccinated children

Table 3

Health-related social capital among MR vaccinated and nonvaccinated families

Type of social capital	Category of social capital	Mean social capital score (SD)		P (unpaired t-test)
		MR vaccinated	MR nonvaccinated
Health-related informational social capital	Bonding	0.324	0.331	0.916
	Bridging	0.118	0.105	0.753
	Linking	0.011	0.039	0.043*
Health-related physical social support	Bonding	0.084	0.218	0.000*
	Bridging	0.041	0.090	0.123

*Statistically significant P<0.05, unpaired t-test

Figure 2

Comparison of the overall health-related social capital score and status of MR vaccine administration

The results of the level of trust in source of health information related to MR vaccine are shown in Table 4. It was seen that among the MR vaccine acceptors were parents who placed a greater trust on teachers and other school children as sources of vaccine-related information. Parents who did not accept the vaccine placed greater trust on WhatsApp and other social media information.

Table 4

Level of trust in source of health information related to MR vaccination

Source of health information related to MR vaccine	Overall trust score		P
	MR vaccinated	MR nonvaccinated
Primary Health Centre Medical Officer	0.055	0.028	0.210
Community Health Worker	0.206	0.200	0.870
Nutritional Centre Worker (Anganwadi Worker)	0.095	0.070	0.400
Neighbors	0.052	0.093	0.076
Friends	0.028	0.021	0.646
Relatives	0.036	0.046	0.604
School teacher	0.246	0.134	0.003*
Private medical practitioner	0.655	0.640	0.963
Television	0.327	0.275	0.229
WhatsApp information	0.045	0.085	0.036*
Other Social Media information	0.014	0.055	0.023*
Newspaper	0.051	0.076	0.327
School children	0.040	0.006	0.014*
Co-workers of the parents	0.034	0.038	0.839

*Statistically significant, P<0.05, unpaired t-test

Table 5 shows the results of multiple logistic regression analysis of factors influencing MR vaccine acceptance. It was observed that after adjusting for multiple variables shown in the table, age of the child, parental attitude toward vaccination, and trust in information provided by school teachers, all positively influenced the uptake of MR vaccination. Health-related physical social capital negatively influenced MR vaccine acceptance.

Table 5

Factors influencing MR vaccine acceptance

Factor influencing MR vaccine acceptance	Adjusted odds ratio (Exp of B)	95% Confidence interval	P
Age of parents	0.570	0.250-1.299	0.181
Age of the child	0.264*	0.098-0.712	0.009
Gender of the child	1.828	0.742-4.50	0.190
Education of parent	1.759	0.123-25.064	0.677
Monthly family income	3.904	0.650-23.43	0.136
Parental attitude toward vaccination	250*	17.2-1000	<0.001
Health-related informational social capital	1.121	0.648-1.912	0.897
Health-related physical social capital	0.068*	0.009-0.438	0.002
Trust in information provided by school teachers	16.12*	2.398-111.111	0.004
Trust in information provided by WhatsApp	0.858	0.018-40.48	0.938
Trust in information provided by other social media	3.44	0.011-1.13	0.675

Discussion

This study on social capital, trust in health information, and its influence on MR vaccination acceptance has thrown light on some important aspects of vaccine acceptance and vaccine hesitancy in a developing health care setting such as Tamil Nadu, India. It was seen that young parents and parents of younger children were more likely to refuse the vaccine. Vaccine acceptance was greater when the vaccine was offered at the school. Greater levels of linking informational social capital and bonding physical social capital lead to greater vaccine hesitancy. Parents with a more positive attitude toward vaccination did accept the MR vaccine to a greater extent. When the parents trusted the school teacher and other school children as sources of information, the vaccine acceptance was greater than when they trusted WhatsApp and other social media information. The multivariate analysis revealed that greater the age of the child, better parental attitudes toward vaccination, poorer health-related physical social capital, and greater trust in health information provided by school teachers led to overall greater acceptance in the MR vaccination.

Social capital and MR vaccine acceptance

Previous studies have shown that a greater social capital has led to greater acceptance of vaccination.[293031] These studies showed that greater the linking social capital in the form of trust in the health system, greater was the acceptance of the vaccine. In another study of bridging social capital in the form of women's self-help groups in the villages and their influence on maternal and child health care in India, it was found that greater bridging social capital led to better outcomes.[31] However, there is evidence to show that strong bonding ties can reduce the utilization of preventive care services due to perpetuation of traditional norms.[32]

In this study the main finding of negative relationship between physical social capital and the MR vaccine acceptance can be explained by the fact that health-related physical social capital is a predominantly bonding social tie and such strong homogeneous ties tend to perpetuate the locally held popular social norms and beliefs related to the MR vaccine. On the contrary, informational social capital is predominantly bridging and linking and it is associated with a more heterogeneous form of capital, allowing for exchange of positive and negative beliefs. This indicates that there is a greater role for linking social capital in influencing health-seeking behaviors, while strong bonding social capital can hamper the same.

Trust in source of vaccine information and MR vaccine acceptance

This study also found that the parents who accepted the MR vaccine trusted the information provided by school teachers. Introduction of new vaccine always gives rise to hesitance and doubts. Vaccine uptake has been shown to be increased by stronger relationship between school and the health system.[3334] To improve vaccine coverage, school teachers have to act as a bridge between parents and health care providers.[3536] A study carried out in sub-Saharan Africa showed that high vaccine acceptance was achieved by school-based approach and also teachers play a vital role in vaccine delivery.[373839] A study carried out in Vietnam found that advice of health workers and teachers motivated the parents to accept vaccination.[38] Success of school-based vaccine program depends on teacher's attitude and knowledge about vaccine.[39] Thus school teachers have a major role in influencing parental decision to accept vaccine along with health professionals in school-based vaccination programs.

In bivariate analysis it was found that greater trust in WhatsApp information and information from other social media led to reduced acceptance of MR vaccine. This ceased to be statistically significant in multivariable analysis. This is probably because age of the parent and age of the child are strong confounders. Lesser the age of the parent, greater the access to social media and misinformation spread through social media. It is important to note that trust in school teacher's information was a significant positive influence on MR vaccine acceptance irrespective of adjustment for multiple confounders.

Lessons from this study to guide future vaccination campaigns

Social capital plays a very important role in determining the vaccine acceptance in a community. This study clearly shows that strong physical social capital, which is predominantly a bonding social capital, leads to reduced vaccination acceptance. This is probably because of strongly held local beliefs and misinformation about the vaccine, which tends to persist within the group. Therefore, in such strongly bonded rural settings in India, it is important to engage with the community on a large-scale basis through village level and door-to-door information, education, and communication activities. Locally relevant methods of information dissemination such as street theatre, public talks, video shows, and interpersonal communication should be used to spread information. In areas where strong bridging networks such as women's self-help groups are available, they should be optimally utilized for dissemination of credible information about the vaccine.

The study also highlights the importance of trustworthy vaccine-related information as a factor influencing vaccine acceptance. The parents who accepted the vaccine trusted school teacher as a source of information, more than even the health care providers and the medical officers. Schools and school teachers should be engaged in a greater manner in spreading credible, trustworthy information about vaccination.

Another key message from this study is that it is the younger parents, with greater access to social media misinformation, who are likely to be more hesitant to the vaccine. Therefore, greater focus should be placed on appropriately educating younger parents on the importance of vaccination. Parents who trusted the social media were more likely to be hesitant to accept the vaccine. Therefore, this study points out that there is a need to regulate the misinformation that is spread on social media platforms and credible information sources and the state department of public health should take affirmative steps to counter the spread of such misinformation.

Strengths and limitations of this study

To the best of the authors' knowledge, this is the first study to assess the relationship between social capital and vaccine acceptance in a setting with developing health care systems. The study used a case–control design, which is very good in identifying risk factors for vaccine nonacceptance. The other strength of the study is its timely nature, as it captured the social dynamics surrounding the MR vaccination during and immediately following the time of the campaign. There are several limitations of this study. The selection of cases and controls were nonrandom and based on convenience. However, on examination of the characteristics of the included children, it is clear that there is comparability in terms of sociodemographic variables, including sex of the child, education of the parent, and monthly income. The social capital scale that was used in this study captured only two dimensions, physical social capital and informational social capital, which are relevant to vaccine decision-making. Other dimensions of social capital—such as emotional support—and membership in local networks—such as women's self-help groups—have been shown to influence health-related decision-making, but these were not measured. Finally, the small sample size warrants that the findings of the study be interpreted keeping in mind the limitations in confidence levels and power of the study.

Conclusion

Social capital and trust in sources of health information had a strong influence on parental decision to accept or refuse MR vaccine during the campaign in Tamil Nadu. This study suggests definitive directions for the scale-up and future campaigns in other parts of India. Future campaigns should engage widely with communities through schools and social networks in the communities to spread credible information about the vaccine.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.