Nutrition as a modifiable causal factor in antisocial behaviours (360G-Wellcome-078667_Z_05_Z)

We propose to build on our previous double blind randomised study which demonstrated that prisoners committed 26.3% (8.3 - 44.33%, 95% CI, p=0.03, two tailed) fewer offences compared to placebo, when for nine months their diet was supplemented with recommended daily intakes of vitamins, minerals and essential fatty acids. Such behavioural effects are not currently considered in published standards of dietary adequacy. In view of the important public health implications, we have permission to extend this study from one to three institutions housing 1200 finally sentenced male prisoners aged 16 to 21 years to explore the psychophysiogical mechanisms. Projected statistical power to detect a significant reduction in offending is 98.4%. Blood samples will be collected to investigate dose response relationships. The high power allows testing for more specific effects of the nutritional intervention on: types of offence e.g. violence, drugs, property, incidents of self-harm. Mediating mechanisms will also be investigated by assessments of interpersonal relating, frontal lobe mediated tasks and heart rate variability. The studies will take three years to complete with six-month baseline and experimental periods followed by a three-month partial crossover with ongoing monitoring of disciplinary offences. Analysis will be by negative binomial (mixed Poisson) regression.