• Open Access

Fish oil supplementation, learning and behaviour in Indigenous Australian children from a remote community school: a pilot feasibility study

Authors


Correspondence to: Dr Natalie Sinn, Sansom Institute for Health Research, University of South Australia, GPO Box 2471, Adelaide, SA 5001; e-mail: natalie.sinn@unisa.edu.au

Indigenous Australian children have lower education outcomes than non-Indigenous children.1 They also have health problems associated with malnutrition.2 Nutrients are required for healthy brain development and function,3 including omega-3 fatty acids (n-3 PUFA). The long-chain n-3 PUFA docosahexaenoic acid (DHA) is highly concentrated in brain tissue.4 Traditional diets contained significantly higher levels of n-3 PUFA than current Western-style diets,5 and suboptimal levels have been implicated in mental health problems.6 Studies have reported improvements with supplementation on learning and behaviour in children6–9 with strongest effects in subgroups with learning difficulties.6–8

Children who are at risk for undernourishment and underachievement in general school populations may benefit from n-3 PUFA supplementation. We conducted an open label pilot study to investigate the feasibility of providing fish oil supplements containing n-3 PUFA to children in a remote, primarily Indigenous, Northern Territory (NT) school in preparation for a larger, placebo-controlled intervention.

All children from preschool to grade 7 were invited to take part. Teachers explained the study to the children and started taking fish oil themselves; information was sent to parents via the school newsletter. Consent forms were received for 47 children aged 3–14 – the majority of the school. Children had a total 12 weeks’ supplementation on school days with a one-week mid-semester break after 5 weeks. The capsules were ‘eye qTM’, each containing 40 mg fish oil and 100 mg evening primrose oil with active ingredients EPA (93 mg), DHA (29 mg), GLA (10 mg), and vitamin E (1.8 mg). Children were given 6 capsules (providing 750 mg EPA+DHA) per day. Teaching assistants placed stickers daily on weekly posters for each child to monitor compliance. The study was approved by Human Research Ethics Committees at the University of South Australia and the NT Department of Health and Menzies School of Health Research, and the NT Department of Education and Training; clinical trials registration ACTRN12609000753257.

Assessments included reading and spelling (Wide Range Achievement Test; WRAT), teacher questionnaires (Conners Behaviour Rating Scales; CBRS) and non-verbal cognitive assessments (Ravens Coloured Matrices; Draw-A-Person) as English is a second language for many of these children. Most children had no problems taking the capsules. Non-compliance was due to difficulty swallowing the capsules (mostly kindergarten), non-attendance or leaving the school. After excluding eight children who took fewer than 50 capsules, creating a skewed distribution, average compliance was 202.15 capsules (SD= 72.19), or 3.38/day. Two more children were excluded from analyses because they were under 5 years and their scores could not be age scaled; therefore the final data set contained n=37 (mean age 8.49, SD 2.29; 65% boys).

Average scores for baseline CBRS anxiety and mood disorder indicators were >70, in the clinical range. Thirty-four per cent had scores in the clinical range for disruptive behaviour, 30% for learning and language disorder, 73% for mood disorder, 51% for anxiety disorder, and 27% for ADHD indicators. During the study, the two main teachers who had completed the CBRS left the school and were replaced. Therefore these forms were not completed at 12 weeks. Linear mixed model analyses, considering all cases (n=37), showed significant improvements from baseline to 12 weeks in both raw and age-scaled scores on the Ravens (t=3.46, p<0.01; t=3.57, p<0.01, respectively), reading (t=2.38, p=0.03; t=2.98, p=0.01, respectively) and spelling (t=2.78, p=0.01; t=4.04, p<0.01, respectively). Intra-class correlations indicated very acceptable goodness of fit of the models to the data, particularly reading and spelling with more than 90% variance explained.

The first few days were fairly disruptive; however this was minimal once the process was streamlined. It would not have been possible without the full support of the principal and teaching staff. They all reported that it had been worthwhile, with anecdotal reports of calmer behaviour and better attention during class, e.g. “they are more co-operative and willing to contribute to discussions; numeracy and literacy have improved noticeably as they are able to concentrate on a task.”

While children improved their standing in relation to the standardisation samples, the study was not placebo-controlled and it is difficult to attribute the observed changes to supplementation. However, it is encouraging that we were able to detect significant improvements in achievement relative to age in a cohort of children generally described as underachieving over a relatively short period of fish oil supplementation, warranting follow-up in larger, controlled trials in this population.

Acknowledgements

We acknowledge the NT Department of Education and Training; Prof Sven Silburn, Menzies School of Health Research; the school principal, teachers, assistants and parents; Novasel Australia; and all children for making this study possible.

Ancillary