The relationship between prenatal nutrition and offspring autism related outcomes, and its socioeconomic context

Friel, Catherine Fiona (2024) The relationship between prenatal nutrition and offspring autism related outcomes, and its socioeconomic context. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2023FrielPhD.pdf] PDF
Download (5MB)


Background. Prenatal nutrition may be an aetiological factor in the development of offspring autism. However, findings are inconsistent, possibly because of methodological limitations, such as small sample size and retrospective study design. Additionally, understanding the socioeconomic context of prenatal diet-autism relationships may inform public health. The overall aim was to draw on causally informed approaches to investigate the association between prenatal nutrition and autism and measure the extent to which inequalities in autism may be explain by a ‘healthy’ prenatal dietary pattern (HPDP).

Methods. The research problem is introduced in Chapters 1-3, and the research aims were address in Chapters 4-7, which are in journal format. Chapter 4: in a systematic review and meta-analysis, I synthesised evidence on the association between prenatal multivitamin supplements and autism diagnosis and evidence of triangulation. Chapter 5: I measured the association between HPDP and offspring autism diagnosis and autism-associated traits in the Norwegian Mother, Father, and Child Cohort (MoBa). Results related to autismassociated traits were triangulated by replicating the analysis in the Avon Longitudinal Study of Parents and Children (ALSPAC). Chapter 6: Mendelian randomisation was applied to measure the relationship between the genetic instrument for HPDP and autismassociated traits in MoBa. Chapter 7: I measured the controlled direct effects of maternal socioeconomic deprivation on autism diagnosis and autism-associated traits when eliminating the proportion of the total effects which are attributed to HPDP.

Results. The probability of autism diagnoses reduced in relation to prenatal multivitamin supplement use compared to no/low use. In MoBa and ALSPAC, the probability of autism diagnosis and autism-associated traits reduced if mothers had a high adherence to HPDP compared to low adherence. Causally informed approaches considered in Chapters 4-7 strengthened the interpretation of results, particularly the cross-context comparison. However, no clear evidence of association emerged from the Mendelian randomisation analysis, though there was insufficient power to detect moderate to large associations. The controlled direct effects analysis demonstrated that a modest proportion of socioeconomic disparities in each outcome were explained by HPDP.

Conclusion. This thesis addressed some key limitations of the existing literature and applied causally informed approaches. Although these analyses provide stronger evidence, testing of causation was inconclusive. Future studies should continue to assess whether the associations between HPDP and autism diagnosis and autism-associated traits are causal. If causality were established, HPDP may be a target for intervention to reduce inequalities in the outcomes but when part of a larger strategy.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: R Medicine > RA Public aspects of medicine
R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry
R Medicine > RG Gynecology and obstetrics
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Health & Wellbeing > MRC/CSO Unit
Funder's Name: Medical Research Council (MRC)
Supervisor's Name: Dundas, Professor Ruth, Leyland, Professor Alastair and Anderson, Dr. Jana
Date of Award: 2024
Depositing User: Theses Team
Unique ID: glathesis:2024-84235
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 15 Apr 2024 11:25
Last Modified: 15 Apr 2024 11:25
Thesis DOI: 10.5525/gla.thesis.84235
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year