Optimising Donor Human Milk Programs Using a Psychology-Informed Grey Wolf Optimiser: A Mixed-Methods Study from a Tertiary Care Hospital in India

Background This study aims to identify the reasons behind the low rate of donor human milk donation and proposes an advanced optimisation model based on behaviour to improve donor identification in a tertiary care centre in India. Of 800 lactating mothers identified to be eligible to donate breast milk after 48 hours of delivery, only 233 mothers (29.1%) donated breast milk, while 567 mothers (70.9%) did not donate breast milk in spite of counselling. The rate of donation was higher among mothers who had very low birth weight babies below 1500 g compared to mothers who had normal birth weight babies, i.e., 63.9% versus 31.4%. The rate of donation was higher among mothers who had sick babies compared to mothers who had normal babies, i.e., 34.4% versus 25.1%. The differences were statistically significant, meaning they were very unlikely to have been due to chance. The reasons cited by mothers who did not donate breast milk were lack of enough milk to donate, lack of family support, religious reasons, and social stigma, accounting for 44.7%, 15.6%, 14.5%, and 11.6%, respectively. Method The psychology-based Grey Wolf Optimiser model was developed to identify potential donors to breast milk banks and to guide counselling. The model performed well and improved donor prioritisation. Results This model provides a tool to improve human milk bank services to vulnerable babies. Out of the 800 eligible lactating mothers, 233 mothers (29.1%) donated their human milk, while 567 (70.9%) did not donate, despite being eligible and having received counselling. This provided the baseline level of inefficiency of the traditional donor recruitment process, which served as a basis to justify the use of the optimization framework. Conclusion This study proves that human milk donation is fundamentally a behaviour-driven process, rather than a process based on eligibility or awareness.