Original Article
Relationship between Test Scores Using the Second and Third Editions of the Bayley Scales in Extremely Preterm Children

https://doi.org/10.1016/j.jpeds.2011.09.047Get rights and content

Objective

To define the relationship between current Bayley Scales of Infant and Toddler Development, Third edition (Bayley-III) scores and the Bayley Scales of Infant Development, second edition Mental Development Index (MDI) to aid the comparison of population outcomes.

Study design

MDI and Bayley-III cognitive/language scales were administered concurrently in 185 extremely preterm children (≤26 weeks) at 29-41 months of age. Cognitive and language scores were combined (combined Bayley-III score [CB-III scores]) for comparison with MDI scores.

Results

Bayley-III cognitive and language scores were 10 and 3 points higher than MDI scores, respectively; CB-III scores were 7 points higher. The relationship between CB-III and MDI scores was not a simple offset: CB-III values were increasingly higher than MDI at lower scores. Bayley-III scores underidentified MDI scores <70 (sensitivity 58%; specificity 100%). An algorithm for converting Bayley-III scores into MDI scores improved predictive value (sensitivity 95%; specificity 97%). Bayley-III scores <80 were similarly predictive (sensitivity 89%; specificity 99%).

Conclusions

We recommend caution in the interpretation of Bayley-III scores in population studies as the correlation with the previous edition appears worse at lower test score values and the predictive value for IQ is as yet unclear.

Section snippets

Method

As part of a national study of babies born at ≤26 completed weeks of gestation in England during 2006 (EPICure-2), we identified children with English as a first language at home, without neurosensory disability (blindness, deafness, nonambulant cerebral palsy), who were assessed by 1 of 3 clinicians trained in the simultaneous use of both editions of the test.

Developmental assessment using our combined assessment was attempted in 225 subjects. Forty (18%) children were considered by the

Results

The participants had a mean average extrapolated CB-III score of 99.5 (SD 15.7); this was somewhat higher than in the remainder of the participants who attempted the Bayley-III tests (96.1) and was expected because of the requirement for the child to be able to concentrate for the extended period of time and exclusions.

Discussion

CB-III scores were on average 7 points higher than comparable MDI values from the second edition acquired concurrently in this sample of 185 children with adjusted ages between 29 and 41 months and mean 99.5 and SD 15.7. These results are consistent with the validity study reported by the publishers of the Bayley-III, in which 102 children aged 1-42 months were tested using both the second and third editions,12 and mean Bayley-III cognitive and language scores were observed to be 7 points

References (23)

  • S. Johnson et al.

    Developmental screen or developmental testing?

    Early Hum Dev

    (2006)
  • N. Bayley

    Bayley Scales of Infant Development

    (1969)
  • N. Bayley

    Manual for the Bayley scales of infant development

    (1993)
  • N.S. Wood et al.

    Neurologic and developmental disability after extremely preterm birth. EPICure Study Group

    N Engl J Med

    (2000)
  • N. Bayley

    Bayley scales of infant and toddler development

    (2006)
  • Vohr B, Stephens BE, Higgins RD, Hintz SR, Bann CM, editors. The impact of Bayley III assessment: are outcomes of ELBW...
  • P. Anderson et al.

    Underestimation of developmental delay by the new Bayley-III scale

    Arch Pediatr Adolesc Med

    (2010)
  • Classification of health status at 2 years as a perinatal outcome

    (2008)
  • Downward extended composite normative data from the Bayley Scales of Infant & Toddler Development (Bayley-III®)....
  • J.M. Bland et al.

    Statistical methods for assessing agreement between two methods of clinical measurement

    Lancet

    (1986)
  • P. Royston et al.

    Regression using fractional polynomials of continuous covariates: parsimonious parametric modelling

    J R Stat Soc Ser C Appl Stat

    (1994)
  • Cited by (192)

    • Risk Assessment and Neurodevelopmental Outcomes

      2023, Avery's Diseases of the Newborn
    View all citing articles on Scopus

    This research was funded by the Medical Research Council (MRC), London, UK. N.M. receives a proportion of funding from the Department of Health's NIHR Biomedical Research Centres funding scheme at UCLH/UCL. The authors declare no conflicts of interest.

    View full text