JAMA Journal of the American Medical Association 
Copyright 1997 by the American Medical Association.  All Rights Reserved. 
Applicable FARS/DFARS Restrictions Apply to Government Use.  American Medical
Association, 515 N. State St, Chicago, IL 60610.

----------------------------------------------
Volume 277(2)             8 January 1997             pp 126-132
----------------------------------------------

Results at Age 8 Years of Early Intervention for Low-Birth-Weight Premature
Infants: The Infant Health and Development Program
[Original Contribution]
McCarton, Cecelia M. MD;  Brooks-Gunn, Jeanne PhD;  Wallace, Ina F. PhD;  Bauer,
Charles R. MD;  Bennett, Forrest C. MD;  Bernbaum, Judy C. MD;  Broyles, R. Sue
MD;  Casey, Patrick H. MD;  McCormick, Marie C. MD, ScD;  Scott, David T. PhD; 
Tyson, Jon MD;  Tonascia, James PhD;  Meinert, Curtis L. PhD, Infant Health
Development Program Research Grp
From the Albert Einstein College of Medicine, Bronx, NY (Dr McCarton); Center
for Children and Families, Teachers College, Columbia University, New York, NY
(Dr Brooks-Gunn); Research Triangle Institute, Research Triangle Park, NC (Dr
Wallace); Department of Pediatrics, University of Miami School of Medicine,
Miami, Fla (Dr Bauer); Department of Pediatrics, University of Washington,
Seattle (Drs Bennett and Scott); Children's Hospital of Philadelphia, University
of Pennsylvania (Dr Bernbaum); Department of Pediatrics, University of Arkansas
for Medical Sciences, Little Rock (Dr Casey); Department of Maternal and Child
Health, Children's Hospital, Boston, Mass (Dr McCormick); Department of
Pediatrics, University of Texas Southwestern Medical Center, Dallas (Drs Tyson
and Broyles); and Department of Biostatistics and Epidemiology, Johns Hopkins
University, Baltimore, Md (Drs Tonascia and Meinert).

----------------------------------------------

Outline

Abstract
METHODS
RESULTS

Cognitive Outcomes
School Performance Outcomes
Behavior Outcomes
Health Outcomes

COMMENT

Cognitive Outcomes
School Performance Outcomes
Behavioral Outcomes
Health Outcomes

CONCLUSIONS
REFERENCES

Graphics

Table 1
Table 2
Table 3
Table 4
Table 5

----------------------------------------------

Abstract

Objective: To reevaluate at age 8 years children who had participated during the
first 3 years of life in a randomized clinical trial of special services for
low-birth-weight (LBW) premature infants.

Design: Follow-up of a randomized controlled trial of premature infants
(<or= to 37 weeks' gestation), stratified by 2 LBW groups (lighter [<or=
to 2000 g] and heavier [2001-2500 g]) and divided into intervention (n=377) and
follow-up only (n=608) groups.

Setting: Eight sites serving diverse populations.

Participants: At age 8 years, 874 children were assessed: 336 in the intervention
group and 538 in the follow-up only group.

Intervention: The 3-year intervention consisted of home visits (birth to 3
years), child development center services (ages 1 to 3 years), and parent group
meetings (ages 1 to 3 years).

Primary Outcome Measures.Cognitive functioning (Weschler Intelligence Scale for
Children-III; Peabody Picture Vocabulary Test-Revised); academic achievement
(Woodcock-Johnson Tests of Achievement-Revised); and parental reports of school
performance, behavior (Child Behavior Checklist), and health (Child General
Health Survey).

: At age 8 years, in the entire cohort and in the lighter LBW stratum, the
intervention and follow-up only groups were similar on all primary outcome
measures. Differences favoring the intervention group were found within the
heavier LBW group: full-scale IQ score (4.4 points higher, P=.007), verbal IQ
score (4.2 points higher, P=.01), performance IQ score (3.9 points higher, P
=.02), mathematics achievement score (4.8 points higher, P=.04), and receptive
vocabulary score (6.7 points higher, P=.001). On a physical functioning
subscale, the whole intervention group received less favorable ratings, while
the lighter LBW intervention group had lower maternal ratings assessing social
limitations caused by behavior.

: Although at age 8 years there were modest intervention-related differences in
the cognitive and academic skills of heavier LBW premature children, attenuation
of the large favorable effects seen at 3 years was observed in both the heavier
and lighter LBW groups. This indicates a need to develop additional intervention
strategies for LBW premature children that can provide sustained benefits.

JAMA.1997;277:126-132

----------------------------------------------

ON GLOBAL MEASURES of intelligence and other neurodevelopmental assessments,
low-birth-weight (LBW) premature children perform more poorly than normal-birth-weight
children. [1-3] In addition, by school age, LBW children are at increased risk
of learning disabilities, [4,5] academic difficulties, [6-8] and behavior
problems, [4,6,7,9] even in the absence of global intellectual deficits. Few
research efforts have been devoted to determining whether early intervention
could attenuate the long-term problems seen in LBW preterm infants. [10,11]

The Infant Health and Development Program (IHDP) was designed as a multisite
randomized clinical trial evaluating in the first 3 years of life the efficacy
of center-based educational intervention, home-based family support services,
and pediatric follow-up in reducing cognitive, behavioral, and health problems
among LBW premature infants. The cognitive and behavioral outcomes of the IHDP
trial at ages 3 and 5 years have been reported, [12-15] as have the health
outcomes at age 3 years. [12,16] At 3 years, the children in the intervention
group had significantly higher intelligence test scores and receptive vocabulary
test scores and lower scores on a parental measure of reported behavior problems
than the children in the follow-up only group. The rate of maternally reported
health conditions over the first 3 years was greater for the children in the
intervention group (difference of 0.27), although they were not hospitalized to
a greater extent than the children in the follow-up only group. The advantages
at age 3 years conferred by the intervention were more pronounced in the heavier
LBW stratum (2001-2500 g) than in the lighter LBW stratum (<or= to 2000 g)
in terms of IQ score (14 vs 7 points), receptive vocabulary score (9 vs 5
points), and behavior problem score (7 vs 2 points). In addition, the rate of
maternally reported health problems was greater in the lighter LBW intervention
group than in the lighter LBW follow-up only group (difference of 0.30); no
differences in the rate of maternally reported health problems between the
intervention and follow-up only groups were found in the heavier LBW stratum.

At age 5 years, 2 years after the intervention ended, we saw an attenuation of
IQ effects. Specifically, there were no significant overall differences in IQ
score, receptive vocabulary, reported behavior problems, or health measures
between the intervention and follow-up only children. However, within the
heavier LBW stratum, the intervention group had higher full-scale IQ scores (4
points) and verbal IQ scores (4 points) as well as higher receptive vocabulary
scores (6 points) than the follow-up only group.

The present report extends the follow-up to age 8 years. Age 8 years was chosen
because it is one age at which increased academic demands may be placed on a
child. [17] As such, it is a common time for grade failure and academic problems
to emerge. Our hypotheses were that enhancements of global measures of cognitive
function, such as IQ score, that were found at 3 years would be attenuated by
age 8 years, but that significant differences favoring the intervention group
would be found in school performance measures of reading and mathematics
achievement and in reduced rates of grade failure.

METHODS

Detailed information on the selection of participants, study design, and
components of the intervention program has been published previously [12-18] and
will be summarized here briefly. Infants were eligible for the study if they had
a birth weight of 2500 g or less, had a gestational age of 37 weeks or less,
resided in the catchment area, and did not have a severe medical illness or
neurological impairment. Patients were enrolled from October 1984 through August
1985. A total of 985 infants constituted the primary analysis group for the
trial. These infants were randomly assigned to the intervention group (n=377) or
the follow-up only group (n=608) using a design with 2 birth-weight strata:
lighter LBW (<or= to 2000 g [n=623]) and heavier LBW (2001-2500 g [n=362]).
[12,15,19] Although the baseline characteristics of the study sample varied
greatly across the 8 sites, the randomization procedure resulted in comparable
intervention and follow-up only groups at study entry. [12]

Infants in both the intervention and follow-up only groups received the same
periodic (through age 8 years) medical, developmental, and social assessments,
with referrals made for other services as needed. The early-intervention program
began at discharge from the neonatal nursery and continued until October
1988-until each child was at least 36 months of age, corrected for prematurity.
The intervention had 3 components [13] : home visits through age 3 years,
attendance at specially designed child development centers beginning at age 1
year through age 3 years, and a series of parent group meetings every other
month during the second and third years. Home visits were designed to occur
weekly in the first year and biweekly in the second and third years. Over the 3
years an average of 66.7 home visits were made. Attendance at the child
development centers was to be for full days 50 weeks per year. The children
attended the centers an average of 267 days per year (over both years). Parent
group meetings were to be held 4 times per year; on average (over 2 years),
there were 3.7 parent group meetings per year.

After the intervention ended at age 3 years, the sites attempted to find
appropriate community education programs for children in both groups. Resources
available at each site differed greatly; however, within the sites there were no
intervention differences between the intervention and follow-up only groups in
maternal reports of enrollment in education programs at age 4 years.

Of the original 985 infants in the primary analysis group, 874 (89%) were
evaluated at age 8 years for a comprehensive assessment of cognitive functioning,
academic achievement, behavior, and health. Individual sites differed in the
percentage of children they were able to evaluate, ranging from 84% to 97%. The
percentages of children within the intervention and follow-up only groups were
similar, both overall and within the sites. The mean age of assessment was 8.1
years (range, 7.9-8.9 years).

The cognitive, academic achievement, and behavior evaluations at age 8 years
were performed by centrally trained assessors who were masked as to the child's
treatment group and history. Health status was assessed by clinic staff who had
access to the child's treatment group assignment and history. As with the other
evaluations, the assessment at age 8 years was to be conducted at the enrolling
IHDP site. Whenever possible, children who had moved away from their enrolling
site were assessed at alternative locations, either at a nearby IHDP site or in
their own community, by a masked assessor who was sent to evaluate them. Of the
874 children who participated in the masked portion of the assessment, all but
59 (7%) were seen at an IHDP site.

Primary outcome measures were in the domains of cognitive skills, school
performance, behavior, and health status. Several measures were administered in
each domain. The complete battery is listed in (Table 1), but only the primary
measures indicated in (Table 1) are reported here. The battery of tests given to
the child is listed in the order of administration. The battery generally took
no more than 3 hours to complete.

----------------------------------------------
Table 1.-Contents of the Age 8 Years Assessment for Children in the Infant
Health and Development Program [38-47] 
----------------------------------------------

Differences between the intervention and follow-up only groups with respect to
each baseline characteristic were assessed using t statistics for continuous
measures and chi squared statistics for categorical measures. Separate multiple
linear regression models [20] were developed for each outcome measure within
each of the 2 birth-weight strata. The same set of 8 explanatory variables
measured at enrollment was used in each model: intervention treatment group
assignment (intervention vs follow-up only), site (8 categories), birth weight
(grams), sex (male vs female), Neonatal Health Index (calculated based on length
of stay in the neonatal nursery, adjusted for birth weight, and standardized to
a mean of 100, with higher scores indicating better health), [21] maternal
education (did not complete high school, completed high school, postsecondary
schooling), maternal age (years), and maternal ethnicity (African American,
Hispanic, and white/other). Population marginal mean values and differences
(intervention vs follow-up only) for each outcome measure were computed from the
regression models to determine if the intervention group controlled for
potentially confounding effects of 1 or more of the 7 other baseline explanatory
variables. [22] The population marginal means estimate the expected outcome
measure in either the intervention or follow-up only group with the 7 other
explanatory variables equated to their average values within a particular
birth-weight stratum. Nominal P values derived from the regression models were
used for primary outcome comparisons; no adjustments were made for multiple
comparisons or multiple outcomes. [23]

RESULTS

(Table 2) compares the intervention and follow-up only groups at age 8 years for
baseline data obtained at enrollment. The baseline characteristics of the
intervention and follow-up only groups were comparable. Although within each
site the baseline characteristics were comparable for the intervention and
follow-up only groups, there were differences between the sites, particularly on
demographic characteristics, such as level of maternal education and ethnicity.

----------------------------------------------
Table 2.-Baseline Characteristics of Children in the Infant Health and
Development Program
----------------------------------------------

Cognitive Outcomes

Overall, we found no statistical differences between the intervention and
follow-up only groups (Table 3). However, the heavier LBW intervention group
displayed significantly higher scores than the heavier follow-up only group on
the Weschler Intelligence Scale for Children-III verbal IQ, performance IQ, and
full-scale IQ tests. The heavier LBW intervention group also had significantly
higher scores on the Peabody Picture Vocabulary Test-Revised. No differences
were found between the lighter LBW intervention and follow-up only groups.

----------------------------------------------
Table 3.-Cognitive Measures at Age 8 Years for Childen in the Infant Health and
Development Program* 
----------------------------------------------

We also examined whether there were sustained effects on IQ scores of the
intervention as a function of the mother's level of education. In the entire
group as well as the 2 LBW strata, there were significant IQ differences between
the children as a function of the mother's level of education. Children of
mothers with post-high school education had higher IQ scores than children of
mothers who had completed high school, while children of mothers with less than
12 years of education had the lowest IQ scores. The differences between the
intervention and follow-up only groups, however, were consistent across the 3
maternal education groups. Within the heavier LBW stratum, there was a 4-point
difference between the intervention and follow-up only groups for each of the
maternal education subgroups-the same intervention-related difference as within
the entire heavier LBW stratum. However, there were no IQ differences between
the intervention and follow-up only groups for any maternal education subgroup
in the group as a whole or in the lighter LBW stratum.

School Performance Outcomes

Although there were no overall differences between the intervention and
follow-up only groups on either composite of the Woodcock-Johnson Tests of
Achievement-Revised (Table 4), the heavier LBW intervention group had significantly
higher Woodcock-Johnson broad mathematics scores than the heavier LBW follow-up
only group. No intervention differences were found in either the reading or
mathematics scores within the lighter LBW stratum.

----------------------------------------------
Table 4.-School Performance at Age 8 Years for Children in the Infant Health and
Development Program
----------------------------------------------

Parents were asked whether children had repeated a grade or were classified as
needing special education. Results indicate that the percentages of grade
repetition and of children classified for special education were similar in the
overall intervention and follow-up only groups and within the 2 birth-weight
strata.

Behavior Outcomes

Results for the primary behavior measure are shown in (Table 5). When rated by
parents on the Child Behavior Checklist, the intervention group had scores
equivalent to those of the follow-up only group, and the 2 birth-weight strata
were also comparable. Moreover, the percentages of children with high scores
indicative of serious behavior problems were similar in the intervention and
follow-up only groups and within the 2 birth-weight strata, with between 8% and
12% rated as above the cutpoint.

----------------------------------------------
Table 5.-Behavior and Health Measures at Age 8 Years for Children in the Infant
Health and Development Program* 
----------------------------------------------

Health Outcomes

Responses to questions regarding illnesses and hospitalizations revealed that
overall the groups had similar numbers of hospitalizations, surgical procedures,
and school absences. As seen in (Table 5), the intervention and follow-up only
groups received similar ratings on most scales of the Child General Health
Survey. However, in the sample as a whole, the intervention group received lower
ratings on the Physical Functioning Scale, indicating that these children were
perceived as being more limited than those in the follow-up only group in the
performance of physical activities, such as playing sports, walking up stairs,
bending, lifting, or caring for themselves. While the heavier LBW intervention
group was not distinguished from the heavier LBW follow-up only group on any
measure of health, the lighter LBW intervention group received lower ratings
than the lighter LBW follow-up only group on the scale assessing role/social
limitations due to behavior. This indicates that their parents found them,
because of behavior problems, to be more limited in the kind, amount, and degree
of schoolwork and/or social activities they could perform.

COMMENT

The present report details the long-term follow-up of a multisite randomized
clinical trial of early intervention from birth to 3 years of age for LBW
children. While the intervention improved cognitive test scores and reduced
behavior problems at age 3 years, 5 years after the intervention ended, at age 8
years, overall IQ score, cognitive skills, school achievement, behavior, and
health indices were similar in the 2 groups of children. Both the intervention
and follow-up only groups had IQ and vocabulary scores substantially below
expectations based on standardized norms. Moreover, there was no reduction in
grade repetition or use of special educational services. On the other hand,
there were statistically significant, albeit clinically modest, differences
between the heavier LBW intervention and follow-up only groups on several
measures of cognition and school achievement. As at age 5 years, there were no
differences found on any primary outcome measure at age 8 years between the
lighter LBW intervention and follow-up only groups. The diminution of intervention
effects after termination of this program is consistent with the findings of
numerous studies of normal-birth-weight disadvantaged children. [24-28]

Cognitive Outcomes

The success of the intervention for the heavier LBW children can be gauged by
their higher test scores in several cognitive skill areas (ie, intelligence,
mathematics, and receptive vocabulary) as well as on some measures of nonreported
secondary outcomes (ie, nonverbal reasoning and aspects of organizing the
reproduction of a complex, visually presented design). These findings indicate
that the intervention had modest success across a range of intellectual
functions for this subgroup of infants.

The observed mean difference in full-scale, verbal, and performance IQ scores of
4 points favoring the heavier LBW intervention group is consistent with findings
from the Abecedarian project, a 5-year preschool educational intervention for
normal-birth-weight socially disadvantaged children that was the basis of the
IHDP educational intervention component. At age 8 years, a difference of 2.5
points was found for the Abecedarian intervention group. [29] In addition, the
mean difference in mathematics test scores of 5 points is similar to the 9-point
difference found in the Abecedarian project. This similarity in intervention
effects is notable given that the IHDP target population consisted of infants at
biological risk and that the intervention stopped at age 3 years.

In contrast to expectations, no differences in cognitive or behavior measures
were detected at 5 years [15] or 8 years of age in the lighter LBW stratum. An
intervention lasting only through age 3 years may have been insufficient to
sustain long-term effects because the biomedical problems were greater in this
group of children. In both the intervention and follow-up only groups, the
lighter LBW children had higher rates of neurological abnormalities than the
heavier LBW children [30] and had lower IQ, receptive vocabulary, reading, and
mathematics scores.

Based on the results at 3 years of age, [31] we expected greater effects for
children of less socioeconomically advantaged backgrounds. Using maternal
education as a proxy, we found that there was no special benefit for more
socioeconomically vulnerable children at age 8 years. However, it should be
noted that in the earlier study maternal education was dichotomized, while in
the current study it was trichotomized.

School Performance Outcomes

Although it was hypothesized that the effects of early intervention would be
most evident in the prevention of school failure, no differences were found in
the percentage of children who repeated a grade or who were placed in special
education. The grade retention rates of 14% and 15% in the intervention and
follow-up only groups, respectively, were about the same as the rates (16% after
3 years of school) found in the Abecedarian program. [32] The kindergarten and
first-grade retention rates for LBW children (12.3%) reported in a large
epidemiologic study [33] are comparable to those found in the present study but
were lower (7.6%) for the overall distribution of children. However, the rates
in the present study are lower than that reported for 1 large LBW sample, in
which 24% of LBW children who were 9 years old had repeated a grade. [34] The
lower rates of grade retention in the present sample of LBW children may reflect
differences in school districts' criteria for grade retention and/or policy
changes limiting the number of children held back. It may also be that
differences in grade repetition may not emerge until the children are somewhat
older, when demands for academic performance are even greater, such as during
the later years of elementary school. Support for this hypothesis is garnered
from the follow-up of children in the Perry Preschool Program, [35] which was a
randomized controlled intervention program during the 1960s of half-day
preschool and weekly home visits for 3- and 4-year-olds from socioeconomically
disadvantaged backgrounds. Data from the follow-up indicated that differences in
rates of grade retention and special education placement were not detected
between the intervention and control groups until the children had completed
third grade.

Behavioral Outcomes

No intervention differences in parental ratings of behavior were found at age 8
years either in the group as a whole or within either birth-weight stratum. The
attenuation of differences in this domain is consistent with the findings at age
5 years. However, the percentage of children with high scores in the heavier
(7%) and lighter (12%) LBW strata is greater than the 2% expected based on the
standardization sample of the Child Behavior Checklist. A higher-than-average
rate of behavioral difficulties is consistent with the findings from other
follow-up studies of LBW premature children. [9,36]

Health Outcomes

No differences in any measure of illness or hospitalization were found at age 8
years. These results parallel the findings at age 5 years and indicate that the
modest elevation in minor illnesses found in the intervention group at age 3
years was limited to the first year in which children were in congregate child
care. However, a few differences were found in the Child General Health Survey,
which measures children's health-related quality of life. In the group as a
whole, the intervention group received lower ratings than the follow-up only
group on a measure of physical limitations in behavior. Although no intervention
differences in health behaviors were found in the heavier LBW stratum, within
the lighter LBW stratum the intervention group received a lower rating than the
follow-up only group on the scale assessing role/social limitations due to
behavior. The reasons for the few intervention-related differences in health
behaviors remain speculative. One possible explanation is that a great deal of
training and observation was done with the mothers in the intervention group,
making these mothers more accurate observers and reporters of their children's
health-related behaviors.

CONCLUSIONS

After an 8-year study of this nature one must ask what has been learned. In
particular, what does the average difference of 4 IQ points found in the heavier
LBW children mean? Individually, 4 IQ points would not produce a functionally
detectable difference between a child in the intervention group and one in the
follow-up only group. However, on a group basis, a 4-point IQ difference between
the intervention and follow-up only children might reduce the percentages of
children classified as intellectually deficient and of borderline intelligence.
For instance, approximately 11% of children in the heavier LBW follow-up only
group had IQ scores in the borderline range (ie, 70-80). If these children had
received the intervention and had gained the same 4-point increase in their IQ
score, only 8% of them would have scored between 70 and 80, a percentage closer
to that expected in the population at large (7% of 8-year-olds).

At the same time, it is reasonable to question the economic price involved in
sustaining a 4-point IQ difference. Unfortunately, no true cost-benefit test is
possible. A major limitation of the IHDP is that there were no prospectively
obtained cost data for implementing the program across 8 sites, although program
expenses were assessed at the Miami site. The cost of delivering the 3
programmatic components was estimated at $15 146 per year per child. [37] The
investigators suggested that this high cost could have been reduced to $8806 per
year per child if the centers were located in the community rather than at a
central location, cutting down transportation costs, and if the teacher-to-child
ratio at the child care center were increased from 2:6 to 2:8.

These costs would be offset if grade retention was decreased and the need for
special education was reduced. Although we did not detect such differences, it
may have been too soon to accurately assess school failure. The majority of
children were in first and second grades. The results of the Perry Preschool
Program [35] suggest that eventually more children may be retained or placed in
special classes. Only future evaluations of the IHDP cohort will be able to
assess whether economic benefits will justify the large costs per child.

These results have several implications for future early-intervention program
initiatives. First, our findings suggest that programmatic efforts need to
distinguish between heavier and lighter LBW infants. The program did help
children with birth weights of 2000 g or less at age 3 years, but, clearly, the
effects were smaller and not sustained. The lighter LBW group may contain a
higher proportion of neurologically impaired children who could not benefit from
the intervention. Second, assessments need to be refined enough to detect
qualitative differences between heavier and lighter LBW infants. For example,
maternal reports of attention deficit or distractibility may reflect environmental
conditions for the heavier LBW child but may be related to neurological
impairment in the lighter LBW child. The heavier LBW child may respond to
changes in environment that the mother can control, but the lighter LBW child
may need a more structured and professionally designed situation. Third, the
lack of sustained effect may reflect the need for different or continued support
of lighter LBW children. It is the task of future research to characterize the
populations needing ongoing support; to explore the type, intensity, and
duration of interventions needed to produce sustained effects; and to develop
policies to implement such interventions. Finally, the corresponding low mean IQ
scores and high rates of school failure of LBW children underscore the
importance of public health efforts to reduce premature births.

This study was supported by grant 91-01142-00 from the Pew Charitable Trusts,
Philadelphia, Pa; by grant 5R01 HD 27344 from the National Institute of Child
Health and Human Development, National Institutes of Health, Bethesda, Md; and
by grants from the Maternal and Child Health Bureau (Title V, Social Security
Act), Washington, DC; the Department of Health and Human Services, Washington,
DC; and the Robert Wood Johnson Foundation, Princeton, NJ.

For a complete list of participants in the Infant Health and Development
Program, see JAMA. 1990;263:3035-3070, and JAMA. 1994;272:1257-1262.

Reprints: Cecelia McCarton, MD, Albert Einstein College of Medicine, Rose F.
Kennedy Center, Room 820, 1410 Pelham Parkway South, Bronx, NY 10461.

REFERENCES

1. Aylward GP, Pfeiffer SI, Wright A, Verhulst SJ. Outcome studies of low birth
weight infants published in the last decade: a metaanalysis. J Pediatr.
1989;115:515-520. [Medline Link] 

2. Escobar GJ, Littenberg B, Petitti DB. Outcome among surviving very low
birthweight infants: a meta-analysis. Arch Dis Child. 1991;66:204-211. [Medline
Link] 

3. McCormick MC. Long-term follow-up of infants discharged from neonatal
intensive care units. JAMA. 1989;261:1767-1772. [Medline Link] [CINAHL Link] 

4. Hack M, Breslau N, Aram D, Weissman B, Klein N, Borawski-Clark E. The effect
of very low birth weight and social risk on neurocognitive abilities at school
age. Dev Behav Pediatr. 1992;13:412-420. [Medline Link] [PsycINFO Link] [BIOSIS
Previews Link] 

5. Sostek AM. Prematurity as well as intraventricular hemorrhage influence
developmental outcome at 5 years. In: Friedman SL, Sigman MD, eds. The
Psychological Development of Low-Birthweight Children: Annual Advances in
Applied Developmental Psychology. Norwood, NJ: Ablex Publishing Corp; 1992;6:259-274.


6. Klebanov PK, Brooks-Gunn J, McCormick MC. Classroom behavior of very low
birth weight elementary school children. Pediatrics. 1994;94:700-708. [Fulltext
Link] [Medline Link] [BIOSIS Previews Link] 

7. McCormick MC, Brooks-Gunn J, Workman-Daniels K, Turner J, Peckham GJ. The
health and developmental status of very low-birth-weight children at school age.
JAMA. 1992;267:2204-2208. [Medline Link] 

8. Klein NK, Hack M, Breslau N. Children who were very low birth weight:
development and academic achievement at nine years of age. Dev Behav Pediatr.
1989;10:32-37. 

9. Buka SL, Lipsitt LP, Tsuang MT. Emotional and behavioral development of
low-birthweight infants. In: Friedman SL, Sigman MD, eds. The Psychological
Development of Low-Birthweight Children: Annual Advances in Applied Developmental
Psychology. Norwood, NJ: Ablex Publishing Corp; 1992;6:187-214. 

10. Bennett FC. The effectiveness of early intervention for infants at increased
biologic risk. In: Guralnick MJ, Bennett FC, eds. The Effectiveness of Early
Intervention for At-Risk and Handicapped Children. Orlando, Fla: Academic Press
Inc; 1987:79-112. 

11. McCarton CM, Wallace IF, Bennett FC. Preventive interventions with low birth
weight premature infants: an evaluation of their success. Semin Perinatol.
1995;19:330-340. [Medline Link] [BIOSIS Previews Link] 

12. The Infant Health and Development Program. Enhancing the outcomes of
low-birth-weight, premature infants: a multisite, randomized trial. JAMA.
1990;263:3035-3042. [Medline Link] 

13. Ramey CT, Bryant DM, Wasik BH, Sparling JJ, Fendt KH, LaVange LM. Infant
Health and Development Program for low birth weight, premature infants: program
elements, family participation, and child intelligence. Pediatrics. 1992;89:454-465.
[Medline Link] 

14. McCormick MC, McCarton C, Tonascia J, Brooks-Gunn J. Early educational
intervention for very low birth weight infants: results from the Infant Health
and Development Program. J Pediatrics. 1993;123:527-533. [Medline Link] [BIOSIS
Previews Link] 

15. Brooks-Gunn J, McCarton CM, Casey PH, et al. Early intervention in
low-birth-weight premature infants: results through age 5 years from the Infant
Health and Development Program. JAMA. 1994;272:1257-1262. [Fulltext Link]
[Medline Link] [PsycINFO Link] [BIOSIS Previews Link] 

16. McCormick MC, Brooks-Gunn J, Shapiro S, Benasich AA, Black G, Gross RT.
Health care use among young children in day care: results in a randomized trial
of early intervention. JAMA. 1991;265:2212-2217. [Medline Link] 

17. Caldwell BM. The challenge of third-grade slump: special report: early
childhood education. Principal. May 1987;66:10-14. 

18. Constantine WL, Haynes CW, Spiker D, Kendall-Tackett K, Constantine NA.
Recruitment and retention in a clinical trial for low birth weight, premature
infants. Dev Behav Pediatr. 1993;14:1-7. [Medline Link] [PsycINFO Link] [BIOSIS
Previews Link] 

19. Kraemer HC, Fendt KH. Random assignment in clinical trials: issues in
planning (Infant Health and Development Program). J Clin Epidemiol. 1990;43:1157-1167.
[Medline Link] 

20. McCullagh P, Nelder JA. Generalized Linear Models. 2nd ed. London, England:
Chapman & Hall Ltd; 1989. 

21. Scott DT, Bauer CR, Kraemer HC, Tyson J. A neonatal health index for preterm
infants. Pediatr Res. 1989;25:263A. Abstract. 

22. Searle SR, Speed FM, Milliken GA. Population marginal means in the linear
model: an alternative to least squares means. Am Stat. 1980;34:216-221. 

23. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology.
1990;1:43-46. [Medline Link] 

24. The Consortium for Longitudinal Studies. As the Twig Is Bent: Lasting
Effects of Preschool Programs. Hillsdale, NJ: Lawrence Erlbaum Associates Inc
Publishers; 1983. 

25. McKey R, Conndelli L, Gansin H, Barrett B, McConkey C, Plantz M. The Impact
of Head Start on Children, Families, and Communities: Final Report of the Head
Start Evaluation, Synthesis and Utilization Project. Washington, DC: CSR Inc;
1985. 

26. Wasik BH, Ramey CT, Bryant DM, Sparling JJ. A longitudinal study of two
early intervention strategies: Project CARE. Child Dev. 1990;61:1682-1696.
[Medline Link] [PsycINFO Link] 

27. Lee VE, Brooks-Gunn J, Schnur E, Liaw F-R. Are Head Start effects sustained?
a longitudinal follow-up comparison of disadvantaged children attending Head
Start, no preschool, and other preschool programs.. Child Dev. 1990;61:495-507.
[Medline Link] [PsycINFO Link] 

28. Barnett WS. Long-term effects of early childhood programs on cognitive and
school outcomes. Future Child. Winter 1995;5:25-50. 

29. Ramey CT, Campbell FA. Poverty, early childhood education, and academic
competence: the Abecedarian experiment. In: Huston AC, ed. Children in Poverty:
Child Development and Public Policy. Cambridge, England: Cambridge University
Press; 1991:190-221. 

30. McCarton C, Bennett FC, Tonascia J, et al. Neurologic status at 36 months of
age. In: Gross RT, Spiker D, eds. Infant Health and Development Program. Palo
Alto, Calif: Stanford University Press. In press. 

31. Brooks-Gunn J, Gross RT, Kraemer HC, Spiker D, Shapiro S. Enhancing the
cognitive outcomes of low birth weight, premature infants: for whom is the
intervention most effective? Pediatrics. 1992;89:1209-1215. [Medline Link] 

32. Horacek HJ, Ramey CT, Campbell FA, Hoffman KP, Fletcher RH. Predicting
school failure and assessing early intervention with high-risk children. Am Acad
Child Adolesc Psychiatry. 1987;26:758-763. 

33. Byrd RS, Weitzman ML. Predictors of early grade retention among children in
the United States. Pediatrics. 1994;93:481-487. [Fulltext Link] [Medline Link]
[BIOSIS Previews Link] 

34. Klebanov PK, Brooks-Gunn J, McCormick MC. School achievement and failure in
very low birth weight children. J Dev Behav Pediatr. 1994;15:248-256. [Medline
Link] [PsycINFO Link] [BIOSIS Previews Link] 

35. Weikart DP, Bond DT, McNeil JT. The Ypsilanti Perry Preschool Project:
Preschool Years and Longitudinal Results Through Fourth Grade. Ypsilanti, Mich:
High/Scope Educational Foundation; 1978:3. High/Scope Educational Research
Foundation Monographs. 

36. McCormick MC, Workman-Daniels K, Brooks-Gunn J. The behavioral and emotional
well-being of school-age children with different birth weights. Pediatrics.
1996;97:18-25. [Fulltext Link] [Medline Link] [BIOSIS Previews Link] 

37. Fewell RR, Scott K. Cost analysis decisions for IHDP. In: Gross RT, Spiker
D, eds. Infant Health and Development Program. Palo Alto, Calif: Stanford
University Press. In press. 

38. Beery KE. Manual for the Developmental Test of Visual-Motor Integration. 3rd
revision. Cleveland, Ohio: Modern Curriculum Press; 1989. 

39. Wechsler D. Manual for the Wechsler Intelligence Scale for Children-Third
Edition. San Antonio, Tex: Psychological Corp; 1991. 

40. Waber DP, Holmes JM. Assessing children's copy productions of the Rey-Osterrieth
Complex Figure. J Clin Exp Neuropsychol. 1985;7:264-280. [Medline Link]
[PsycINFO Link] 

41. Woodcock RW, Johnson MB. Manual for the Woodcock-Johnson Tests of Achievement-Revised.
Allen, Tex: RCL Enterprises; 1990. 

42. Elliott CD. Manual for the Differential Ability Scales. San Antonio, Tex:
Psychological Corp; 1990. 

43. Dunn LM, Dunn LM. Manual for Peabody Picture Vocabulary Test-Revised. Circle
Pines, Minn: American Guidance Service; 1981. 

44. Sheslow D, Adams W. Manual for the Wide Range Assessment of Memory and
Learning. Wilmington, Del: Jastak; 1990. 

45. Brown L, Hammill DD. Manual for the Behavior Rating Profile. 2nd ed. Austin,
Tex: Pro-Ed; 1990. 

46. Achenbach TM, Edelbrock C. Manual for the Child Behavior Checklist and
Revised Child Behavior Profile. Burlington, Vt: Queen City Printers; 1983. 

47. Landgraf JM, Ware JE Jr, Schor E, Davies AR, R-Roh K. Comparison of health
status profiles in children with medical conditions: preliminary psychometric
and clinical results for the Children's Health and Quality of Life Project.
Presented at the 10th annual meeting of the Association for Health Services
Research; June 1993; Washington, DC. 

Child Behavior; Cognition; Health Status; Infant, Low Birth Weight; Infant,
Premature

----------------------------------------------