Published online August 1, 2005
PEDIATRICS
Vol. 116
No. 2
August 2005, pp.
491-492
(doi:10.1542/peds.2005-1061)
The Conundrum of Prediction
Glen P. Aylward, PhD, ABPP
Division of Developmental and Behavioral Pediatrics
Southern Illinois University School of Medicine
Springfield, IL 62794-9658
Abbreviations: BSID, Bayley Scales of Infant Development • K-ABC, Kaufman Assessment Battery for Children • MDI, Mental Developmental Index • ELBW, extremely low birth weight
The study by Hack et al1 in this issue of Pediatrics underscores the complexities inherent in prediction of developmental outcome. More specifically, in this study many children who scored >2 SDs below average on the Bayley Scales of Infant Development-II (BSID-II)2 at 20 months' corrected age subsequently did better on the Kaufman Assessment Battery for Children (K-ABC)3 given at 8 years. Greater stability in scores was found in those children with Mental Developmental Index (MDI) scores of <70 and who also had neurosensory impairment, with continued problems noted at both measurement points. Conversely, the vast majority of children whose initial MDI score was >70 also did well on the later assessment. As a result, high negative predictive values, but low positive predictive values, were found.
There are numerous advantages to the study: large sample size, consideration of environmental and biomedical variables, long-term serial evaluation, analyses with and without inclusion of children with neurosensory deficits, and emphasis given to whether the child had received early intervention services. The authors also identified correlates of stability or change in scores over time.
It is clear that if major handicap is considered to be a cognitive score of <70, then there is a marked reduction in the prevalence of such over time. However, children who shifted from the cognitive impairment group at 20 months to a K-ABC score of >70 at 8 years still were at risk for problems. The mean score in those children who originally scored <70 and whose status subsequently improved was 87.1. Scores in this range still place these children at distinct disadvantage when they have to compete in the classroom with peers whose scores are average or higher. It would be interesting to determine how many of these children had an 8-year score of 
100. Furthermore, based on the Flynn effect4 (with which mean scores are estimated to increase 0.5 points per year), the average K-ABC score is estimated to be 10 points higher, further accentuating this disadvantage. Moreover, of children whose later evaluation improved to a Mental Processing Composite score of >70, two thirds displayed poor adaptive functioning and more than half had an individual educational plan (IEP), suggesting significant school-performance problems (this occurring both in the neurosensory-intact group and total population).
These findings underscore the authors' point regarding the impreciseness of early developmental assessment.5 The data also point out that there may be change in the presentation of problems attributable to conditions such as extremely low birth weight (ELBW) and the fact that low cognitive scores in infancy may indicate increased risk of a problem, but they lack specificity as to what type of problem the child will manifest. Therefore, an early MDI of <70 might not precisely equate to a later IQ of <70, but it is suggestive of an increased likelihood of problems nonetheless (a point acknowledged by the authors). The current findings are in contrast to the often-cited trend of worsening outcome over time,6–8 underscoring the need for additional investigation.
Several other points are notable. The accuracy of the BSID-II may be affected by rules governing starting points and the fact that, because corrected age was used, children do not automatically receive credit for earlier items and therefore have a greater opportunity to fail items early on.9,10 This would also affect ceiling rules that mandate termination of testing although the child could possibly pass additional items had they been administered. Some of these issues may be resolved in the upcoming BSID-III. These data also underscore the ceiling or suppressor effect of significant biological risk, in that with children having MDI scores of <70 and accompanying central nervous system morbidities, self-righting and environmental influences were attenuated. Conversely, sociodemographic variables seem to be more influential in children scoring in the 70 to 84 range.
As the authors rightly emphasize, major care-related decisions regarding ELBW children should not be based solely on the results of early cognitive assessments: long-term data are critical. Their data indicate that below the age of 2 years, caution should be exercised when considering cognitive function of <70 as a major handicap without the presence of concomitant neurodevelopmental findings. Conversely, although these results are encouraging, they should not be overgeneralized beyond the authors' intent to suggest that ELBW children with early delays routinely have later normal cognitive function. A low score during infancy is a marker for subsequent risk, this risk often being in terms of high-prevalence, low-severity dysfunctions.11
 |
FOOTNOTES
|
|---|
Accepted May 5, 2005.
Reprint requests to (G.P.A.) Division of Developmental and Behavioral Pediatrics, Southern Illinois University School of Medicine, PO Box 19658, Springfield, IL 62794-9658. E-mail: gaylward{at}siumed.edu
No conflict of interest declared.
|
REFERENCES
|
|---|
- Hack M, Taylor HG, Drotar D, et al. Poor predictive validity of the Bayley Scales of Infant Development for cognitive function of extremely low birth weight children at school age.
Pediatrics. 2005;116
:333
–341[Abstract/Free Full Text]
- Bayley N.
Bayley Scales of Infant Development. 2nd Ed. San Antonio, TX: Psychological Corporation; 1993
- Kaufman AS, Kaufman NL.
Interpretive Manual for the Kaufman Assessment Battery for Children. Circle Pines, MN: American Guidance Service; 1983
- Flynn JR. Searching for justice. The discovery of IQ gains over time.
Am Psychol. 1999;54
:5
–20
- Aylward GP. Conceptual issues in developmental screening and assessment.
J Dev Behav Pediatr. 1997;18
:340
–349[CrossRef][ISI][Medline]
- Marlow N, Wolke D, Bracewell MA, Samara M. Neurologic and developmental disability at six years of age after extremely preterm birth.
N Engl J Med. 2005;352
:9
–19[Abstract/Free Full Text]
- O'Callaghan MJ, Burns YR, Gray PH, et al. School performance of ELBW children. A controlled study.
Dev Med Child Neurol. 1996;38
:917
–926[ISI][Medline]
- Dewey DG, Crawford SG, Creighton DE, Sauve RS. Long-term neuropsychological outcomes in very low birth weight children free of sensorineural impairments.
J Clin Exp Neuropsychol. 1999;21
:851
–865[ISI][Medline]
- Ross G, Lawson K. Using the Bayley-II: unresolved issues in assessing the development of prematurely born children.
J Dev Behav Pediatr. 1997;18
:109
–111[CrossRef][ISI][Medline]
- Gauthier SM, Bauer CR, Messinger DS, Closius JM. The Bayley Scales of Infant Development II: where to start [published correction appears in J Dev Behav Pediatr. 1999;20:197]? J Dev Behav Pediatr. 1999;20
:75
–79
- Aylward GP. Cognitive and neuropsychological outcomes: more than IQ scores.
Ment Retard Dev Disabil Res Rev. 2002;8
:234
–240[CrossRef][ISI][Medline]
PEDIATRICS (ISSN 1098-4275). ©2005 by the American Academy of Pediatrics
TOP
REFERENCES
