Vaccine advertising: preach to the converted or to the unaware?

Masha Krupenkin, Elad Yom-Tov, David RothschildNpj Digit. Medicine 2021 · DOI · OpenReview

Abstract

Digital advertising for public health often focuses on "bottom-of-the-funnel" users—those already searching for health information. This study investigates the impact of "priming" unaware users who search for unrelated terms. Through a large-scale randomized controlled trial on Bing, researchers found that priming can increase future vaccine-related searches by up to 150x, suggesting a significant opportunity to expand immunization coverage by reaching beyond the already-convinced.

Introduction

Despite the availability of effective vaccines, global immunization rates remain suboptimal. In the United States, for example, only 35% of eligible adults receive the shingles vaccine. Public health outreach increasingly relies on digital advertising, yet these campaigns often suffer from a strategic bias: they target "congruent" keywords (e.g., "flu shot"). While this captures individuals already considering vaccination, it neglects the "unaware" population. This research explores whether digital ads can effectively "prime" these users to seek vaccines and how to optimize advertising budgets between interested and uninterested groups.

Methodology

The researchers conducted a pre-registered randomized controlled trial (RCT) using Microsoft Advertising (Bing) targeting influenza, human papillomavirus (HPV), and herpes zoster (shingles). The study involved a treatment group of approximately 69,000 users and a control group of 8.6 million users.

The experimental design utilized two keyword strategies:

  1. Congruent Keywords: Terms directly related to the disease or vaccine.
  2. Priming (Tangential) Keywords: Terms unrelated to health but common among the target demographic (e.g., "SAT" for HPV-eligible youth, "pumpkin" for flu).
Experimental flow showing the relationship between campaign keywords, impressions, and user interactions.

The team employed Cox proportional hazard models to measure the time to future vaccine searches and logistic regression to analyze ad click probabilities.

Key Results

The study revealed a massive priming effect. While congruent ads increased future search likelihood by up to 116%, the impact on "unaware" users was even more dramatic. For HPV, the primed group was over 150 times more likely to perform a future vaccine search than the control group.

Cumulative hazard plots illustrating the significant increase in future vaccine searches for treated users across different conditions.

Specific Hazard Ratios (HR) for future vaccine searches among the unaware group were:

  • Influenza: HR = 6.9
  • Herpes Zoster: HR = 30.5
  • HPV: HR = 157.3

Discussion and Insights

The findings suggest a "Conversion Funnel Trap." Commercial advertisers often prioritize high Click-Through Rates (CTR), which are naturally higher for congruent keywords. However, this approach "preaches to the converted." Because vaccines often lack strong brand recognition among consumers, they behave like indirect sales. Broad, categorical, and tangential keywords are more effective at moving people into the awareness funnel.

To maximize total vaccinations, the authors propose a mathematical framework that shifts budget away from high-CTR congruent keywords toward lower-CTR priming keywords.

Limitations

The study uses search intent and clicks as proxies for actual vaccination ("needles in arms"). While search behavior is a strong predictor, it is not a perfect substitute for clinical data. Additionally, the anonymized nature of search data limits demographic verification, and the results are specific to search engine behavior rather than social media or display ads. Future research should explore different ad creatives and messaging hooks for priming audiences.

Summary updated

Abstract

Encouraging people to vaccinate is a challenging endeavor, but one which has tremendous public health benefits. Doing so requires overcoming barriers of awareness, availability, and (sometimes) vaccine hesitancy. Here we focus on nudging people to vaccinate through online advertising. We conducted a pre-registered online ads campaign encouraging people to vaccinate against three diseases: influenza, human papillomavirus, and herpes zoster. Ads were shown to ~69,000 people and were compared to similar ads shown to 8.6 million people. Outcome measures were clicks on ads and future searches for relevant terms. We find that ads have two main effects: First, a congruence effect whereby ads increase the likelihood of clicks and future searches by up to 116% in people who express an interest in the disease or the vaccine. Most commercial vaccine advertising is aimed entirely at this population. Second, we observed a priming effect, where ads shown to people who were searching for terms unrelated to the vaccine could be encouraged to click on them (odds ratios of 7.5–33.0) and, more often, search for the vaccine later (hazard ratios of 6.9–157.3). We provide analysis for optimizing vaccine advertising campaign budgets to balance the two populations. These findings demonstrate that digital advertising campaigns should consider not just advertising to direct keywords or to individuals that look exactly like existing customers, but consider tangential keywords that draw a wider target population who are likely earlier in their conversion funnel, thus increasing the number of people who vaccinate and maximizing vaccines uptake.
BibTeX
@article{krupenkin2021vaccine,
  author = {Masha Krupenkin and Elad Yom-Tov and David Rothschild},
  title = {{Vaccine advertising: preach to the converted or to the unaware?}},
  journal = {Npj Digit. Medicine},
  year = {2021},
  doi = {10.1038/s41746-021-00395-7},
}
Figure