According to the WHO, by 2008, measles vaccination coverage was 83% globally among children aged 12–23 months old. Nearly 700 million children aged between 9 months to 14 years living in high-risk countries, were vaccinated against the disease between 2000 to 2008, and globally measles deaths declined by 78% during this period, indicating that the global measles vaccination campaign successfully averted over 3.6 million deaths [1-3]. The lowest rates were in the South-East Asian (75%) and African (73%) regions. In low-income countries, 76% of children aged 12–23 months had received measles vaccination [3].

However, measles continues to be a major public health problem among children in developing nations, with measles deaths predominantly occurring among children aged below five years. Children living in developing countries with low income and poor health infrastructure are at the highest risk of measles-related morbidity and mortality. Most of the world’s measles-related deaths occur in nations with the lowest measles vaccine uptake [4,5].

In 2008 two-thirds of the 22.7 million children who missed receiving measles vaccine in lived in India, Nigeria, China, Democratic Republic of Congo, Pakistan, Ethiopia and Indonesia [4]. In India although the measles vaccination uptake increased to 70% in 2008, measles mortality is high, with about 300 children estimated to be dying from measles-related complications every day.

The major factors influencing measles and other childhood vaccinations operate at five different levels; (1) Intra-personal – Individual child’s characteristics; (2) Inter-personal – parental and household factors; (3) Community – community characteristics and service delivery factors; (4) Institutional: international coordination of vaccination efforts (5) Public policy – quality, coverage and enforcement of policies [6]. This study focuses on the interpersonal factors influencing vaccination, Our findings have implications for the other four levels which influence childhood vaccination.

Variables such as maternal education and occupation and household’s economic status, have consistently been shown to influence measles vaccination at an interpersonal level in developing countries [7-9]. However, much of this research has focused on the role of maternal education status, despite the patriarchal nature of most nations with low measles vaccination uptake. The few studies on the impact of paternal education on vaccination uptake were largely based on small samples in limited settings. For example, a study based on a 2006 survey of slum children in Bangladesh showed that paternal occupation significantly influenced vaccination uptake, with children with fathers in a business or service occupation being 1.059 times and 1.107 times more likely to vaccinate their children, relative to children whose fathers were labourers [9]. A recent study from Pakistan found that while maternal education status was influential in improving child nutritional outcomes, father’s education was a more important factor in one-off type health decisions such as receiving vaccination, in comparison to day-today decisions related to nutrition [10].

A study based on data from the 1992 Indian National Family and Health Survey found that, in rural areas, children with literate fathers were more likely to be vaccinated, even if the mothers are illiterate [11]. The positive influence of high paternal education status on measles vaccination uptake in developed nations was also demonstrated in a state-wide study of 1003 children aged 2 in Ohio, United States [12].

The Demographic and Health Survey was considered the best available gold standard for comparison of immunisation data. Investigations of the quality of DHS methodologies concluded that it is nationally representative and relatively free of systematic bias [13].

Table ​Table11 describes the variables used in the analysis. maternal and paternal education were included as categorical variables indicating the highest level of education attained by the child’s mother and father, that is, no education (reference category), primary, secondary or higher education.

Maternal and paternal occupation variables were included to further investigate differences in immunisation coverage across socio-economic groups. Occupation variables were defined as ‘not working’ (reference category), ‘agricultural self-employed’, ‘manual labour’ and ‘professional/clerical/services’. Household economic status is measured using the DHS wealth index quintiles, ranging from poorest (reference category) to richest.

In India when both parents had no education, the proportion of children who were vaccinated for measles was 33.9%, and this increased to 43.9% even when the mother had no education, but the father had Secondary schooling and above. Similarly, for Indonesia when both parents had primary schooling, 55.1% of children were vaccinated, increasing to 75.6% when the mother had primary schooling, but father’s education was Secondary schooling and above. For Nigeria, when both parents had no education, 15.8% of children were vaccinated, and this figure nearly doubled when the father had secondary schooling, but the mother had no education.

In Table ​Table22 the significance of differences in measles vaccination coverage between paternal education categories is tested, while controlling for maternal education level.

A striking finding was that, in all six countries studied, the proportion of children who were vaccinated for measles increased as father’s education increased, regardless of mother’s education levels. When both parents had no education, children were significantly less likely to be immunised against measles (p-value<0.05).

The most conclusive results were under the following two scenarios. First, when maternal education is primary or secondary schooling, the proportion of children receiving measles immunisation was significantly higher when paternal education is secondary or higher (p-value<0.05). Second, when maternal education was primary and paternal education is no education or primary, the proportion of children not receiving measles vaccination was significantly higher (p-value<0.05). This was also seen when maternal education level was secondary and paternal education as primary (p-value<0.05). These findings confirm paternal education status as an independent correlate of measles immunisation coverage.

To investigate the combined influence of these factors, the change in measles vaccination coverage with increasing paternal education within a particular maternal education level was analysed. The correlation between maternal and paternal education levels and measles vaccination was established in the logistic regression analysis. Table ​Table33 presents the results of a logistic regression analysis. These results show the estimated likelihood of the child being vaccinated for measles separately for each country and for disaggregated samples for different levels of maternal education. Specifically, we estimated a series of logistic regression models for each of our six countries, separately for the full sample; sample where the mothers had no schooling; sample where the mothers had only primary schooling; and sample where the mothers had secondary and higher levels of schooling. Variables including household wealth, rural/urban residence, dummy variables for province of residence, occupations of mother and father and child’s religion were included in all regressions.

Table ​Table3,3, summarises the coefficients for paternal education within different levels of maternal education. The coefficients presented in Table ​Table33 are for the influence of father’s education

The results of the current study comply with previous studies investigating factors associated with childhood vaccination coverage in both developing and developed countries. A study conducted in Istanbul found that the rate of non-vaccination was 2.3 times higher among children whose fathers had less than secondary education compared to those whose fathers had attained secondary education or higher [14]. Marks et al. investigated vaccination in the United States, finding that paternal education was a significant predictor of childhood vaccination coverage. This result was found to be independent of maternal education levels [12]. Breierova and Duflo showed that paternal and maternal education levels were equally important determinants of child mortality, and that this may reflect “assortative matching”, i.e., maternal and paternal education levels are correlated because more educated people are likely to choose more educated spouses [15]. Ours is the first nationally representative study to explore the impact on paternal education status on measles vaccination update in nations with the highest numbers of missed measles vaccinations.

The relative increases in female education may account, in part, for the substantial improvements in measles immunisation coverage. With global gender parity in education status now virtually assured based on MDG trends, paternal education is likely to play a critical role in efforts to improve childhood immunisation. This is particularly the case in all nations with the highest numbers of unimmunised children, where patrilineal norms predominate. For example, a study from northern Nigeria based on a 2007 dataset revealed that children whose fathers approve of BCG immunization are almost three times as likely to be immunized as those whose fathers disapprove [16]. Similar trends are presumed for other nations with large numbers of unimmunised children.

Addressing paternal encumbrances to measles vaccination entails interventions at three levels. First, improvement in male educational levels, particularly in rural regions of developing nations where the labour input of children exceeds potential gains from poor-quality secular education. MDGs are based on male educational level as a reference. However, as male educational levels are currently sub-optimal in many developing nations, averaging 77% for proportion progressing through secondary school in the six nations studied, according to latest World Bank estimates [17], achieving gender parity in formal education will still leave a large proportion of individuals illiterate, both male and female. Greater investments in creating educational opportunities for children in developing nations, and removal of encumbrances such as excessive school fees are urgently required [18]. Second, reduce educational, occupational, cultural, religious, socio-economic and political encumbrances to gender equality, such as to enhance the opinions of mothers on issues of child health within families. Third, adequate health communication strategies on the benefits of vaccination should be designed to appeal to both fathers and mothers. Community and religious leaders should be encouraged to advocate in favour of vaccination. It is acknowledged that removing the encumbrances on paternal education in nations performing poorly with regards to measles vaccination will not result in significant progress unless vaccination-related issues are concurrently addressed at other levels [6].

We have identified three limitations in this study. First, measles vaccination data relied heavily on mother’s recall, which is subject to bias. Maternal recall however, is considered a valid measure of child vaccination coverage in the absence of vaccination records in developing countries [19]. It is likely that many children in the sampled countries did not have a completed vaccination card, for example, only 21% of children the Indonesian sample produced a complete vaccination card. Second, the effects of contextual factors such as social influence and social learning on the decision to have children vaccinated were not considered [20]. It is possible that both parent education levels and the level of education of those in the community may influence the likelihood that children are vaccinated against measles. Finally, factors relating to the availability and accessibility of primary health infrastructure were not considered due to the unavailability of data.