The University of Edinburgh
Incomes and outcomes: preterm birth, socioeconomic status and neurodevelopment
Abstract
dc:description.abstractPreterm birth is a leading cause of atypical brain development, and is associated with long-term motor, cognitive, and behavioural outcomes, with effects across the life course. In the general population, low socioeconomic status (SES) is also associated with brain structure, cognition, education attainment, and behavioural and mental health diagnoses. However, the relative contributions of low SES and low birth gestational age (GA) to brain development from the neonatal period are unknown. Inflammation is associated with both preterm birth and SES independently and could be one mechanism through which they have effects on brain development. This thesis aimed to investigate (1) the associations between SES and preterm birth with neonatal brain morphology; (2) the associations between white matter microstructure, SES, and preterm birth in the neonatal period and at five years of age; and (3) differences in the proteome associated with preterm birth and SES inferred from epigenetic scores (EpiScores), DNA methylation (DNAm)-based surrogates that have been developed for multiple circulating proteins. These studies were undertaken with the Theirworld Edinburgh Birth Cohort (TEBC), a longitudinal cohort study of preterm and term infants. Several SES measures were gathered, and although these measures were correlated, the strength of correlation ranged from very weak to strong. This meant they were not sufficiently correlated to assume that one measure captures all dimensions of SES, so multiple measures of SES were examined in each study. In the first study, 85 regional volumes across the brain were investigated at term-equivalent age, alongside whole-brain measures of cortical morphology; gyrification index (GI), sulcal depth, curvature, thickness, and surface area (SA). Six regional brain volumes associated with SES measures (standardised β 0.06-0.17), and there were interactions between birth GA and SES measures (standardised β -0.05-0.02), although there were more widespread associations with birth GA (standardised β -0.20-0.24). Family-level SES measures, such as maternal education and occupation, were associated with more regional volumes than neighbourhood-level SES measures, namely the Scottish Index of Multiple Deprivation (SIMD). For the cortical measurements, birth GA was associated with SA and GI (standardised β 0.10-0.16); there were no associations between cortical measurements and SES. In the second study, tract-based spatial statistics (TBSS) were used to investigate differences in fractional anisotropy (FA) with birth GA and SES in the neonatal period and at five years of age. In the preterm subgroup in the neonatal period, FA positively associated with birth GA and maternal education, and there were associations with an interaction between birth GA and both maternal education and SIMD. In the preterm subgroup at five years of age, there were associations between FA and an interaction between birth GA and SIMD. The SES measure that best correlated with FA was maternal education in the neonatal period, but SIMD at five years of age. For the term cohort, in the neonatal period there were negative associations with maternal education and SIMD, but only a small region of positive association with SIMD at five years of age. In the final study, inflammation was investigated as a possible mechanism for the correlations between preterm birth and SES with brain development. This used EpiScores, developed in large association studies of DNAm patterns with protein abundance, as these may provide a more accurate measure of chronic inflammation than phasic and maturation-dependent acute phase proteins. 43 EpiScores associated with low birth GA (standardised β |0.14 to 0.88|). Only three EpiScores associated with SES (standardised β |0.06 to 0.13|), and these did not remain significant after adjustment for perinatal pro-inflammatory co-exposures (HCA, sepsis, BPD, and NEC). In conclusion, preterm birth and SES both associate with differences on brain MRI at term-equivalent age and at five years of age, with differences in volumetric and other structural measures, and in FA as a measure of white matter microstructure. From findings at different timepoints, it appears that family-level SES measures may be particularly important in the neonatal period, and neighbourhood-level SES measures at five years of age. A range of EpiScores associate with preterm birth, providing various proteins that warrant further study to understand their correlation with developmental outcomes. However, inflammation is unlikely to be the primary axis through which SES is embedded in the early development of preterm infants. There is time for interventions to reduce social disparities during the perinatal period, which could promote healthier brain development after preterm birth. The differences in SES measure associations at different ages suggests that the most beneficial socioeconomic interventions for brain development in this vulnerable population may change over time.
Degree
thesis:*- Grantor dc:publisher
- The University of Edinburgh
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Mckinnon, Katie May
- Advisors dc:contributor.advisor
-
- Boardman, James
- Richardson, Hilary
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Identifier URI
- http://dx.doi.org/10.7488/era/5780