Abstract
BACKGROUND: Digital vaccination campaigns are increasingly used to address declining vaccine confidence, yet evidence from large-scale, real-world interventions in middle-income countries is limited. Meta's Brand Lift Studies (BLS), which use randomized test-control exposure, provide Bayesian estimates of attitudinal shifts resulting from digital content. Mexico, with over 88.6 million active internet users, provides a setting to evaluate the impact of targeted campaigns on vaccine attitudes. OBJECTIVE: This study evaluated the impact of 5 digital vaccination campaigns implemented by the Asociación Mexicana de Vacunología (@Vacunologia) on Facebook (Meta Platforms Inc) and Instagram (Meta Platforms Inc) in Mexico between 2021 and 2022 on key attitudinal constructs related to COVID-19 vaccine confidence. METHODS: This study used a retrospective ecological design. We analyzed aggregated BLS results for 5 campaigns targeting different audiences and vaccination themes. Measured outcomes included standard ad recall, perceived importance, perceived safety, perceived efficacy, and concerns about side effects. Statistical significance within the BLS framework was defined as an incremental lift of ≥2 percentage points with ≥90% posterior probability of replication-a threshold consistent with Meta's operational Bayesian approach. Exploratory comparisons across campaigns were conducted using 1-way ANOVA, unpaired 2-tailed t tests, and Fisher exact tests. RESULTS: Campaigns reached 84.9 million accounts and generated 179.4 million impressions with a total investment of US $215,600. All campaigns produced statistically significant improvements in at least one attitudinal outcome (Bayesian threshold ≥90%). Standard ad recall increased in 4 campaigns (ANOVA, P<.001), and concerns about side effects decreased in 2 campaigns (t test, P=.049; P=.006). Perceived safety, importance, and efficacy improved in selected audiences, with stronger effects observed among younger users and women (ANOVA, P=.005). No direct behavioral outcomes (eg, vaccination uptake) were measured; therefore, the findings reflect attitudinal rather than behavioral change. However, these constructs are recognized as proximal predictors of vaccine decision-making and constitute health-related outcomes. CONCLUSIONS: Large-scale digital vaccination campaigns can meaningfully strengthen attitudinal determinants of vaccine confidence in a middle-income context. These social media advertising campaigns effectively increased standard ad recall and improved perceptions of vaccine importance and safety, particularly among younger audiences and women in urban areas. However, changes in efficacy perceptions and concerns about side effects were limited. The innovation and implications of this study lie in evaluating large-scale, real-world digital vaccine campaigns in Latin America using experimental BLS data. Findings highlight that audience segmentation yields stronger perceptual shifts, suggesting that tailored digital strategies can complement traditional public health communication. While the BLS does not measure behavioral end points, the observed attitudinal improvements represent foundational steps toward influencing vaccine-related behaviors. Future work should link digital attitudinal metrics with vaccination and epidemiological data to assess real-world health impact.