AINAR trial: A randomized controlled trial to assess the usefulness of artificial intelligence-based game to conquer needle fear in blood donors.

Abstract

Background: Needle fear and anxiety are important contributors to vasovagal reactions (VVR) and poor retention among blood donors. This study evaluated the usefulness of Artificial Intelligence for Needle Anxiety Reduction (AINAR), an artificial intelligence-based game intervention, in reducing needle fear and VVR during blood donation.

Methodology: Single-centre, parallel-group randomized controlled trial conducted at AIIMS, New Delhi (2025). Total of 3000 eligible whole blood donors were randomized equally into control and AINAR groups. Donors in the intervention arm played the AINAR game during needle insertion and removal. The primary outcome was immediate VVR incidence. Secondary outcomes included needle fear, anxiety scores, and AINAR colour feedback responses. Fear and anxiety were assessed using a 5-point Likert scale before and after donation.

Results: The overall incidence of VVR was 1.47%. VVR occurred in 1.8% of donors in the control group and 1.13% in the AINAR group, corresponding to a relative risk of 0.63, representing a reduction that did not reach statistically significance. Needle fear significantly decreased after donation in both groups, with a greater reduction in the AINAR group. Moderate-to-severe pre-donation needle fear was strongly associated with VVR (5.6 vs. 1.3%, P<0.001). Persistent post-donation fear was strongly associated with VVR (aOR 9.19, 95% CI 3.19-26.48), while prior VVR also remained independently associated with VVR (aOR 4.65, 95% CI 1.15-18.82). Exploratory subgroup analyses suggested lower VVR incidence with AINAR among some higher-risk donor groups. AINAR colour changes correlated with donor-reported fear and anxiety.

Conclusion: AINAR was associated with a greater reduction in donor-reported needle fear, whereas the reduction in immediate VVR did not reach statistical significance. Exploratory findings among high-risk donors warrant further investigation. Larger multicentre studies are required to determine whether AI-assisted behavioural interventions can reduce VVR and improve donation experience.