Vaccine · 2026
Estimating the impact of Shigella vaccines on growth outcomes and implications for clinical trial design.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Codi Allison, Rogawski McQuade Elizabeth, Benkeser David
Аннотация
BACKGROUND: The value proposition for Shigella vaccines is strengthened by the potential for vaccines to prevent linear growth faltering. However, because expected effect sizes in Phase 3 vaccine trials are small due to limited Shigella incidence, a simple comparison of growth by randomized vaccine arm is likely underpowered and may yield null or even inverse results.
METHODS: We consider a new approach that estimates vaccine effects in the subgroup that would be infected in absence of vaccination, termed the naturally infected. In simulations parameterized by multi-site studies of diarrhea, we compare power for detecting linear growth effects in the naturally infected versus the full study. We further quantified how power is impacted by trial design choices including immunization schedule, study site, and timing of growth measurements.
FINDINGS: Simple comparisons of height-for-age z-score (HAZ) by randomized vaccine arm have extremely limited power (<15%) at realistic trial sizes (n=2500 to 20,000) and carry risk of showing an inverse effect due to random chance. In contrast, naturally infected effects were often five to ten times larger and power was up to three times higher. Using a twelve month immunization schedule with a single growth endpoint in high-incidence settings maximized power to detect an effect.
INTERPRETATIONS: While realistically sized clinical trials may be underpowered to detect an effect of vaccination on growth, estimation using the naturally infected subpopulation and careful trial design improve chances of detecting an effect while mitigating risks of null or inverse results.
METHODS: We consider a new approach that estimates vaccine effects in the subgroup that would be infected in absence of vaccination, termed the naturally infected. In simulations parameterized by multi-site studies of diarrhea, we compare power for detecting linear growth effects in the naturally infected versus the full study. We further quantified how power is impacted by trial design choices including immunization schedule, study site, and timing of growth measurements.
FINDINGS: Simple comparisons of height-for-age z-score (HAZ) by randomized vaccine arm have extremely limited power (<15%) at realistic trial sizes (n=2500 to 20,000) and carry risk of showing an inverse effect due to random chance. In contrast, naturally infected effects were often five to ten times larger and power was up to three times higher. Using a twelve month immunization schedule with a single growth endpoint in high-incidence settings maximized power to detect an effect.
INTERPRETATIONS: While realistically sized clinical trials may be underpowered to detect an effect of vaccination on growth, estimation using the naturally infected subpopulation and careful trial design improve chances of detecting an effect while mitigating risks of null or inverse results.