Public health · 2026
Content analysis of output from generative artificial intelligence chatbots when prompted about breast cancer and alcohol consumption.
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.
Anbari A B, Maturino K, Goth O, Jesudasan G H, Bautista J R
Аннотация
OBJECTIVES: Alcohol is a group one carcinogen and contributes to breast cancer risk for women. Awareness of this relationship remains low. Generative artificial intelligence (GenAI) chatbots are an increasingly used source of health information. It is imperative to know what information about alcohol and breast cancer is provided by GenAI chatbots to the public. The objective was to characterize GenAI chatbot output when queried with pre-existing prompts about the relationship between alcohol consumption and breast cancer risk.
STUDY DESIGN: Content analysis to characterize GenAI output based on (1) clinical/public health content, and (2) information presentation and formatting.
METHODS: We extracted 66 GenAI chatbot outputs about alcohol and breast cancer risk from 22 configurations of 11 GenAI chatbots in June 2025. We used three prompts from the WHO, CDC, and National Breast Cancer Foundation websites regarding alcohol and breast cancer risk. Content analysis was conducted in Dedoose.
RESULTS: Similarities and variations in content and presentation and formatting were identified. Nearly all output included information about causal mechanisms (n = 65, 98.48%), and the majority included information about dose dependency (n = 58, 87.88%). About half (n = 35, 53.0%) presented content using a numbered list. Thirty-two (48.48%) did not provide citations/references, while 31 (4.97%) included statistics or data about breast cancer and alcohol consumption. Four (6.06%) included phrasing about alcohol's group 1 carcinogen status.
CONCLUSIONS: The variability present in GenAI output regarding alcohol consumption and breast cancer risk is a public health concern. Greater standardization of GenAI output and GenAI prompt and output literacy are needed.
STUDY DESIGN: Content analysis to characterize GenAI output based on (1) clinical/public health content, and (2) information presentation and formatting.
METHODS: We extracted 66 GenAI chatbot outputs about alcohol and breast cancer risk from 22 configurations of 11 GenAI chatbots in June 2025. We used three prompts from the WHO, CDC, and National Breast Cancer Foundation websites regarding alcohol and breast cancer risk. Content analysis was conducted in Dedoose.
RESULTS: Similarities and variations in content and presentation and formatting were identified. Nearly all output included information about causal mechanisms (n = 65, 98.48%), and the majority included information about dose dependency (n = 58, 87.88%). About half (n = 35, 53.0%) presented content using a numbered list. Thirty-two (48.48%) did not provide citations/references, while 31 (4.97%) included statistics or data about breast cancer and alcohol consumption. Four (6.06%) included phrasing about alcohol's group 1 carcinogen status.
CONCLUSIONS: The variability present in GenAI output regarding alcohol consumption and breast cancer risk is a public health concern. Greater standardization of GenAI output and GenAI prompt and output literacy are needed.