The Bayley Scales of Infant and Toddler Development- 3rd Edition: A Quantitative Analysis for Application in an Australian Population

The Bayley Scales of Infant and Toddler Development- 3rd Edition (BSID-III) is widely used in Australia. Despite this, there are no Australian normative data and scant quantitative evidence regarding its clinical and predictive utility in an Australian context. This thesis used prospective, longitudinal data from a large cohort of general population 1- and 3-year-old Australian children and their mothers to investigate the clinical and predictive utility of the BSID-III in Australia. Specifically, this thesis consisted of four empirical studies that aimed to: (1) explore differences between 1-year-old Australian infant performance on the BSID-III and the US normative sample; (2) investigate the utility of the BSID-III to detect differences in infant performance based on indirect perinatal risk factors for neurodevelopmental delay; (3) explore differences in performance on the BSID-III between 3-year-old Australian children and the US normative sample; and, (4) examine the predictive utility of the BSID-III from 1-year to 3-years, as well as by sex. Information on maternal socio-demographics, birth and infant health was collected via structured maternal interview antenatally, and at 8-weeks, 1-year and 3-years (subsample) postnatally. The BSID-III was administered by trained assessors at 1- and 3-years. Study 1 found that at 1-year of age Australian children performed significantly higher than the US BSID-III normative sample on the cognitive domain and significantly lower on the gross motor domain. Study 2 showed that the BSID-III was able to detect some statistically significant, but not clinically relevant differences between infants at indirect risk of neurodevelopmental delay based on perinatal factors, and those at low risk/no risk. Study 3 demonstrated that Australian 3-year-old children performed significantly higher on the language (expressive and receptive) and motor domains (fine and gross) of the BSID-III compared to the US normative sample, but significantly lower on the cognitive domain. With regards to Study 4, results suggested that the BSID-III held greatest predictive utility from 1- to 3-years on the receptive language domain, and that predictive utility of the BSID-III improved once examined by sex. This thesis contributes to our knowledge of the clinical and predictive utility of the BSID-III for Australian infants and pre-schoolers. Taken together, results suggest that the current clinical practice in Australia of utilising US normative data to interpret infant performance is likely to result in sub-optimal identification of developmental delay. Clinical implications and directions for future research are discussed.