Impact of high-intensity polio eradication activities on children's routine immunization status in Northern India.

The objective of this article is to analyse and quantify the side effects of the Polio Eradication Initiative on routine immunization performance in India. Past studies have faced methodological challenges in assessing these side effects. This article offers a methodological alternative for health policy analysts. The research uses secondary household survey data from the Indian District-Level Household and Facility Survey (DLHS), focusing on children aged 10-30 months in the Northern Indian states of Uttar Pradesh (n = 34 327) and Bihar (n = 20 525). Covering the years 2002-08, this is the latest large-scale data from India that enables the matching technique used in this article. District-level programme intensity data of the Polio Eradication Initiative in India were reconstructed using publicly available resources. The methodological innovation compared with previous studies consists of matching each child in the DLHS data set with a child-specific value of programme exposure depending on its district of residence, its birth date, and the date of the survey interview. Average and age-specific associations between polio programme exposure and children's full immunization status were assessed using logistic regression, controlling for other determinants of immunization. The regression results show that the link is negative in Uttar Pradesh and positive in Bihar. Age-specific analysis shows that the positive association diminishes for older children in Bihar and that a negative association emerges and becomes increasingly pronounced for older children in Uttar Pradesh. This indicates that heterogeneous results emerge across two neighbouring states with similar programme intensity and suggests that the catch-up of unvaccinated older children may be a channel through which negative effects accrue. The method described in this article, based on an analytical focus on individual-level programme exposure, can therefore help health policy implementers and evaluators to illuminate positive or negative interactions between a health intervention and a health system.