A cross-cultural comparison of the role of maternal mental health in the prediction of infant cognitive development and empirical investigation of the role of early caregiving in India.

Bluett-Duncan, Matthew
(2021) A cross-cultural comparison of the role of maternal mental health in the prediction of infant cognitive development and empirical investigation of the role of early caregiving in India. Doctor of Philosophy thesis, University of Liverpool.

[thumbnail of 200671807_Feb2021.pdf] Text
200671807_Feb2021.pdf - Unspecified

Download (12MB) | Preview


Background The association between postnatal depression and cognitive development has been extensively researched in high-income countries (HICs). The results of a systematic review conducted as part of this thesis indicated that more high-quality longitudinal studies are required to examine the association between postnatal depression and infant cognitive outcomes in LMICs. An empirical investigation was therefore conducted in India (Study 1) to evaluate this relationship and to examine the potential moderating role of early maternal sensitivity. Additionally, cross-cultural comparison of depression symptoms are increasingly common in the literature but differential item functioning (DIF) between groups may result in scale items taking on different characteristics across cultures and lead to skewed findings. This thesis describes the development of a set of anchoring vignettes (AVs) and their application to an Indian and UK dataset so that the impact of DIF on a popular self-report tool assessing perinatal depression could be determined and accounted for (Study 2). Methods Study 1 data was drawn from an existing longitudinal, community-based sample from India and the associations between exposure to early postnatal (n = 309) or chronic (n = 395) depression and infant cognitive development were examined. Mothers completed the Edinburgh Postnatal Depression Scale (EPDS) at 8 weeks postpartum and then at 6, 12 and 24 months of age, and infants were assessed using the cognitive and language subscales of the Bayley Scales of Infant Development (BSID-III) at 24 months. Maternal sensitivity was assessed and coded using the NICHD mother-infant interaction scales at 6 months. In the second part of the thesis, a set of anchoring vignettes (AVs) were developed to detect DIF on EPDS items between urban populations in India and the UK. In India, data was drawn from the same longitudinal cohort as study 1. 247 mothers completed the AVs, and 549 mothers completed the EPDS at 12 months. In the UK, 828 mothers from an existing longitudinal, community-based sample completed the EPDS at 12 months, and 252 mothers from a separate community-based sample completed the AVs. Results Study 1 revealed a significant small association between early postnatal depression and language subscale scores. Following adjustment for covariates neither postnatal nor chronic maternal depression significantly predicted scores on either the cognitive or language subscales of the BSID-III. There was also no evidence of a moderating effect by maternal sensitivity. However, there was a significant positive association between maternal sensitivity and language scores for boys only. In Study 2 the AV analyses indicated that UK mothers rated the vignettes and themselves more severely than Indian mothers. Prior to DIF adjustment, rates of depression were significantly higher in UK participants, but after DIF adjustment, rates were significantly higher in Indian participants. Conclusions Study 1 indicates that postnatal depression may not have the same impact on infant cognitive development in India as seen in HICs. However, in line with research in HICs, there was evidence that boys may be more reliant than girls on the early regulation provided through sensitive maternal caregiving. The findings also highlight the need to investigate the potentially protective or adverse roles of other nutritional or socio-cultural factors specific to this context. The results from Study 2 indicate that UK mothers have higher expectations regarding postnatal depression than Indian mothers and so give higher ratings to their own symptoms. The findings indicate that the EPDS does not function equivalently in these settings and extends the literature in support of AVs as a valid and helpful data harmonisation tool for multi-item scales of mental health. Further, the results suggest that cross-cultural comparisons without AV correction need to be interpreted with caution.

Item Type: Thesis (Doctor of Philosophy)
Uncontrolled Keywords: postnatal depression, maternal mental health, cognitive development, cross-cultural
Divisions: Faculty of Health and Life Sciences
Depositing User: Symplectic Admin
Date Deposited: 09 Sep 2021 14:24
Last Modified: 18 Jan 2023 21:36
DOI: 10.17638/03129982
  • Sharp, Helen
  • Pickles, Andrew
  • Kishore, Thomas
  • Satyanarayana, Veena
URI: https://livrepository.liverpool.ac.uk/id/eprint/3129982