Low immunization rates for children continue to be a significant global concern. A recent study in India illustrates that a cost-effective combination of strategies, such as text reminders and small financial incentives, can greatly improve vaccination rates.
Conducted by MIT economists, the research reveals that a combination of incentives and communication methods resulted in a remarkable 44% increase in child immunizations, all achieved at a low cost. Interestingly, using just text messages and community support—without financial incentives—yielded a 9% increase in immunizations at virtually no expense, marking it as the most cost-effective method identified in the study.
“The most effective overall strategy includes incentives, reminders, and enlisting local ambassadors to encourage vaccination,” explains MIT economist Esther Duflo, who contributed significantly to this research. “The costs are minimal, and a more budget-friendly option is relying solely on reminders through community networks. This enhances the existing immunization infrastructure and demonstrates that the small expenses of the program are outweighed by cost savings in administration.”
This experiment is particularly notable for its innovative methodology, developed by the research team to combine various approaches and analyze the impacts of different combinations.
“This study effectively links multiple pieces of evidence together,” adds MIT economist Abhijit Banerjee, another key figure in the project. “Our confidence in advocating this as a viable policy approach is crucial.”
A new paper, titled “Selecting the Most Effective Nudge: Evidence from a Large-Scale Experiment on Immunization,” is set to be published in the journal Econometrica. The paper is co-authored by Duflo, Banerjee, and several researchers from MIT’s Abdul Latif Jameel Anti-Poverty Lab (J-PAL).
Notably, both Duflo and Banerjee are founders of J-PAL, a prominent institution focused on field experiments aimed at combating poverty. In 2019, they received the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, along with Harvard’s Michael Kremer.
Evaluating 75 Strategies Simultaneously
Globally, approximately 2 million children die each year from diseases that vaccines could prevent. As of 2016, only 62% of children in India were fully immunized against diseases like tuberculosis, measles, and polio.
Previous studies by Duflo and Banerjee established the importance of innovative strategies to elevate immunization rates. For example, their earlier research found an increase in immunization rates for rural children in Rajasthan, India, from 5% to 39% when families were offered lentils as an incentive—a finding acknowledged in their Nobel citation. This has led many researchers to explore new methods for increasing vaccinations.
This latest study was conducted in collaboration with the Haryana state government over a period from 2016 to 2019, encompassing more than 900 villages.
The team centered the experiment on three primary methods to encourage vaccination: financial incentives, text reminders, and information from local ambassadors, or respected residents. They created a series of varied combinations for these elements, adjusting the levels of incentives, the frequency of text messages, and the amount of local information exposure.
Ultimately, researchers evaluated 75 different combinations, employing a novel methodology they termed treatment variant aggregation (TVA). This method uses an algorithmic, data-driven approach to consolidate variations that yield similar results while identifying ineffective combinations. Additionally, they adjusted for biases, known as the “winner’s curse,” that often favor the more successful options in empirical studies.
“This isn’t merely a comparative analysis; it evaluates multiple strategies simultaneously,” Banerjee explains. “Even when effects seem negligible, there’s valuable information to be gained. It’s essential to track all possible outcomes through the process, although complex mathematically.”
The researchers also identified that local demographic differences significantly affect the success of various strategies. Generally, communities with lower immunization rates exhibit greater responsiveness to financial incentives.
“We find ourselves revisiting earlier conclusions from our Rajasthan study, where low vaccination rates led to substantial positive responses to incentives,” says Duflo, who serves as the Abdul Latif Jameel Professor of Poverty Alleviation and Development Economics. “We have confirmed our previous findings in this context, yet the TVA methodology offers richer insights more swiftly.”
A Path Forward
The research team hopes that the TVA methodology will gain greater traction among researchers, paving the way for more comprehensive experiments that assess multiple potential solutions simultaneously. This approach can be applied across various fields, including antipoverty initiatives and medical trials.
Furthermore, the findings empower governments and organizations to predict the implications of different policy options in both healthcare and financial contexts.
“Our goal was to provide the Haryana government with a pragmatic strategy for the future,” states Duflo.
She adds, “Many believed that to draw definitive conclusions, one must evaluate only one intervention at a time, such as financial incentives or text messages. However, I’m pleased to reveal that it’s possible to analyze multiple strategies together, even though it requires considerable effort.”
Other contributors to this study include Arun G. Chandrasekhar, Suresh Dalpath, John Floretta, Matthew O. Jackson, Harini Kannan, Francine Loza, Anirudh Sankar, Anna Schrimpf, and Maheshwor Shrestha.
This research was made possible through the cooperation of the Haryana Department of Health and Family Welfare.
Photo credit & article inspired by: Massachusetts Institute of Technology
