Effects of AI-generated misinformation about the anticancer effectiveness of dietary supplements on health beliefs and behavioral intentions

Effects of AI-generated misinformation about the anticancer effectiveness of dietary supplements on health beliefs and behavioral intentions among the Japanese general public: a randomized controlled trial - AI Supplement Misinformation RCT (Cancer)

Status

Active, not recruiting

Phases

Unknown

Study type

Interventional

Source

JPRN

Registry ID

JPRN-UMIN000061487

Enrollment

800

Registered

2026-05-08

Start date

2026-03-10

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Exposure to dietary supplement misinformation related to cancer

Interventions

Participants in the intervention group view AI generated misinformation claiming that one of four dietary supplements, placenta, euglena, Asian ginseng, or chlorella is effective for cancer treatment,

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: Individuals aged 18 to 64 years Registered panel members of a Japanese survey company who are able to complete an online survey Individuals who have experience using generative AI Individuals who provide informed consent online

Exclusion criteria

Exclusion criteria: Individuals who are not aware of dietary supplements Individuals who have no experience using generative AI Health care professionals, including physicians, nurses, pharmacists, registered dietitians, and dietitians Individuals working in the AI field, including AI researchers or developers Individuals with a personal or household history of cancer Individuals currently receiving dietary therapy under the guidance of health professionals Individuals who fail the attention-check question

Design outcomes

Primary

Measure	Time frame
Beliefs about the anticancer effectiveness of dietary supplements. This outcome is assessed using 12 items covering four dietary supplements: placenta, euglena, Asian ginseng, and chlorella. For each supplement, three items assess perceived effectiveness for cancer treatment, inhibition of cancer metastasis, and cancer prevention. Each item is rated on a 6-point Likert scale ranging from 1 = strongly disagree to 6 = strongly agree. Outcomes are measured before and immediately after viewing the intervention or control message. The primary analysis uses the post-message score as the primary endpoint.	—

Measure

Time frame

Beliefs about the anticancer effectiveness of dietary supplements. This outcome is assessed using 12 items covering four dietary supplements: placenta, euglena, Asian ginseng, and chlorella. For each supplement, three items assess perceived effectiveness for cancer treatment, inhibition of cancer metastasis, and cancer prevention. Each item is rated on a 6-point Likert scale ranging from 1 = strongly disagree to 6 = strongly agree. Outcomes are measured before and immediately after viewing the intervention or control message. The primary analysis uses the post-message score as the primary endpoint.

—

Secondary

Measure	Time frame
Secondary outcomes include: 1) general beliefs about the health-promoting, disease-preventive, and disease-treatment effects of dietary supplements; 2) intentions to purchase or use dietary supplements for cancer treatment, inhibition of cancer metastasis, or cancer prevention; 3) intentions to share AI-generated health information via social media or word of mouth; and 4) intentions to consult health professionals about dietary supplement use. Each item is rated on a 6-point Likert scale ranging from 1 = strongly disagree to 6 = strongly agree. Outcomes are measured before and immediately after viewing the intervention or control message.	—

Measure

Time frame

Secondary outcomes include: 1) general beliefs about the health-promoting, disease-preventive, and disease-treatment effects of dietary supplements; 2) intentions to purchase or use dietary supplements for cancer treatment, inhibition of cancer metastasis, or cancer prevention; 3) intentions to share AI-generated health information via social media or word of mouth; and 4) intentions to consult health professionals about dietary supplement use. Each item is rated on a 6-point Likert scale ranging from 1 = strongly disagree to 6 = strongly agree. Outcomes are measured before and immediately after viewing the intervention or control message.

—

Countries

Japan

Contacts

Public ContactMingxin Liu

Graduate School of Medicine, The University of Tokyo Department of Health Communication

liumingxin98@g.ecc.u-tokyo.ac.jp09062421998

Outcome results

None listed

Source: JPRN (via WHO ICTRP) · Data processed: Jul 3, 2026