Health inequalities in vaccination: A challenge for measles elimination (the case of outbreak in Ecuador, 2011–2012)

Introduction/background Low immunization coverage in specific population groups, together with high viral infectivity, has aggravated the measles outbreaks happening in the past decade. Ecuador, a country in northwestern South America, introduced routine measles vaccination more than three decades ago. As a result, the last cases of autochthonous measles happened in that country around 1996. However, a new measles outbreak occurred in 2011–12. This study aimed to quantify socioeconomic inequalities associated with measles immunization coverage at the population level. Methods An ecological study was performed using two datasets: the results of a measles immunization survey performed in Ecuador in 2011 and socioeconomic data from the 2010 census, aggregated by canton. The survey included 3,140,799 people aged 6 months to 14 years living in 220 cantons of Ecuador, in whom at least one dose of measles-containing vaccine received was inquired. Variables included were: previous measles immunization, Unsatisfied basic needs (UBNI) of urban cantons, percentage of the 15–17 year-old population in the canton attending school, percentage of the population self-identifying as indigenous or African-Ecuadorian in the canton, and employment rate. Multiple spatial regression was performed to identify socioeconomic inequalities associated with measles immunization coverage. Spatial autocorrelation was detected and conditional autoregressive analysis was performed for adjustment of variables. Principal components analysis was used to create a socioeconomic score. The slope index and relative index of inequality were calculated. Results Measles immunization coverage ranged from 54.1% to 98.5% in the cantons of Ecuador. Measles immunization coverage was inversely associated with unsatisfied basic needs ( P  = 0.0007) in urban areas and proportion of indigenous and African-Ecuadorian residents in the canton ( P  = 0.015), and directly associated with unemployment rate in the canton ( P  = 0.037). The distribution of immunization coverage across the cantons was heterogeneous, indicating spatial dependence. In cantons in the lower socioeconomic stratum, the rate of non-immunization was 71% higher than in cantons in the upper stratum, with a prevalence ratio of 1.71 (95% CI: 1.69–1.72), and an absolute difference of 6.44 percentage points. The slope index of inequality revealed a difference of 10.6 percentage points in immunization coverage between the canton estimated by linear regression as the one with the best socioeconomic level as compared to the canton with the worst socioeconomic level. In turn, the relative index of inequality showed that immunization coverage was 1.12 times higher in the canton estimated to have the best socioeconomic level compared to the canton with the worst socioeconomic level. Discussion We observed a spatial distribution pattern for immunization coverage and socioeconomic indicators that suggest socially and economically vulnerable populations are also more susceptible to epidemic outbreaks. In cantons with lower immunization coverage, a higher proportion of UBN, lower employment rate, and higher proportion of indigenous and African-Ecuadorian residents were detected. Neighboring cantons were equally vulnerable to measles. Previous studies have shown an association between low immunization coverage and poverty indicators, lower maternal schooling, higher proportion of racial or ethnic minorities, and limited access to health services. The results of the present study must be interpreted with care. It is not possible to make inferences at the individual-level. However, these results seem consistent with those of studies having individuals as the unit of analysis. Conclusions The spatial dependence between measles vaccination coverage and health inequalities suggests clusters of vulnerable populations for outbreaks. Health social inequalities must be considered to achieve and maintain measles elimination.