PEDIATRICS Vol. 106 No. 6 December 2000, pp. 1397-1405

Longitudinal Neurologic Follow-Up in Neonatal Intensive Care Unit Survivors With Various Neonatal Morbidities

Margaret M. McGrath, DNSc^*, Mary C. Sullivan, PhD^*, Barry M. Lester, PhD, and William Oh, MD

From the ^* University of Rhode Island, College of Nursing, Kingston, Rhode Island; and  Department of Pediatrics, Brown University School of Medicine, Providence, Rhode Island.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  The purpose of this prospective longitudinal study was to examine neurocognitive and school performance outcomes of low birth weight infants with reference to neonatal morbidity and socioeconomic status. We further evaluated the cognition and school performance based on their neurologic status at the time of assessment.

Methods.  One hundred eighty-eight children (39 healthy full-term and 149 preterm infants) were classified into 4 subgroups based on their neonatal medical status: healthy, sick (without neurologic complications), small for gestational age, and neurologically compromised infants. Neurologic status was classified as normal, suspect, or abnormal at hospital discharge, 18 months, 30 months, 4 years, and 8 years of age. Socioeconomic status, cognitive, and school performances were assessed.

Results.  Neurologically, both full-term and healthy preterm groups did well during the 8-year period. There were significant fluctuations between suspect and abnormal neurologic classifications among the 3 preterm groups with neonatal complications. Preterms with neurologic abnormality during the neonatal period did the poorest with 45% of the group remaining abnormal at 8 years of age. Children who were neurologically normal had higher cognitive scores at ages 4 and 8 than those categorized as suspect or abnormal. Preterm infants with neurologic abnormality required significantly more academic resources in the school. Reading and math achievement scores were the lowest for the preterm groups classified as neurologically suspect or abnormal.

Conclusions.  Neonatal morbidities exert a significant impact in neurologic outcomes among preterm children during the 8 years of assessment. Compromised neurologic status adversely affects cognitive and school performances. Neonatal medical status is an important variable indicating neurocognitive and school performance outcomes in low birth weight infants.  Key words:  low birth weight, NICU survivors, neurologic classification.

The most consistent predictor of poor outcome in preterm infants is the degree of maturity at birth as indicated by birth weight.^1,2 Historically, in low birth weight (LBW) infant follow-up research, the variable of birth weight provided a global index of risk. More recently, researchers have found that medical complications of prematurity have added to our understanding of developmental outcome. Taylor et al,² in a regional sample of very low birth weight (VLBW) children, reported that birth weight alone is inadequate in accounting for neurodevelopmental impairment at early school age. In the study, a composite Neonatal Risk Index including both medical and neurologic complications was the best predictor of outcomes. The present investigation expands the previous work by prospectively grouping infants and sorting out medical complications from severe neurologic compromise. Considering the range of major neonatal illness associated with poor school and academic performance, our objective was to relate neonatal morbidities to neurodevelopmental outcomes at ages 18 months, 30 months, 4 years, and 8 years.

Standard neurologic assessment is considered an indicator of the range of compromise related to neonatal complications of prematurity.³ There has been a difference of opinion, however, about the predictability of early neurologic examinations for preterms with mild to moderate medical complications and later neurologic status.^4,5 Sostek⁶ argued that the outcomes connected with degrees of intraventricular hemorrhage (IVH) supported the theory of sleeper effects because she found delays in cognitive and motor skills at age 5 years which were not evident earlier, at 2 years of age. Sleeper effects are thought to reveal insult, recovery, and potential plasticity of the brain not identifiable until later ages. In a heterogeneous sample of premature children, sleeper effects may be even more important beyond 5 years of age because a recent report estimated that half of these children demonstrate learning disturbances to be the primary result of prematurity.² The opposite interpretation is that the majority of neonatal intensive care unit (NICU) survivors aged 2 to 51/2 years who are initially exposed to an abnormal environment and take longer to acquire necessary skills will catch up by age 51/2 when given time for consolidation.⁷

The purpose of this longitudinal study was to examine the neurologic outcome at various time points through 8 years of age of LBW survivors who were prospectively grouped by neonatal medical status during the neonatal period. A secondary objective of the study is to assess the effect on cognitive and academic achievement.

	METHODS

Top Abstract Methods Results Discussion References

Sample

Infants who met an a priori medical criterion were prospectively enrolled into 5 cohorts. The sample included 188 children, 39 of whom were normal, healthy full-term (FT) infants with uncomplicated pregnancy, labor, and delivery and recruited from our normal nursery. One hundred forty-nine preterm infants were recruited from our NICU and classified into 4 subgroups based on their neonatal clinical diagnosis. One group of infants (n = 32) did not have significant clinical problems and was considered to be a healthy preterm group (HPT). Another group (n = 54) consisted of infants who were clinically ill (sick preterm group [SPT]) but did not have significant neurologic abnormality. The diagnosis of the SPT group included bronchopulmonary dysplasia, respiratory distress syndrome, sepsis, anemia, mild IVH (grades 1 and 2), jaundice of prematurity with indirect bilirubin concentration >10 mg/dL, chronic lung disease defined as dependence on oxygen until 36 weeks' corrected age, and necrotizing enterocolitis. Infants with mild IVH were included in this group because of previous findings which show that mild IVH was not significantly associated with adverse neurologic outcomes.⁸ A third group of infants (n = 34) had severe neurologic compromise in the neonatal period (neurologic preterm group, [NPT]). Infants in this group have experienced seizures, meningitis, or hydrocephalus and/or grade 3 or 4 IVH. The NPT group reflects the most severe degree of neurologic insult and were grouped together because they were thought to be the children who would experience the worst outcome.⁸ The fourth group of preterm infants (n = 29) was classified as small for gestational age (SGA), with or without medical problems. The diagnosis of SGA was based on birth weight of less than the 10th percentile of expected weight for gestation.⁹ The sample represents 8% of the hospital's NICU population during the years of recruitment and reflects the demographic profile and range of medical morbidity of the time. The infants in this sample were smaller than the mean birth weight of the hospital's NICU population.

At the time of enrollment, demographic and clinical data, including gender, race, gestation, birth weight, and Apgar scores, were obtained from medical charts. Hobel neonatal risk scores¹⁰ were gathered to measure the severity of perinatal risk. Differences in risk severity were found between study groups. In 1985 through 1989, the Hobel was the most frequently used risk index during the neonatal period. At discharge the duration of hospital stay was recorded (Table 1).

Neurologic Assessments

At hospital discharge, the neurologic status of the subjects was classified by the attending physician as normal (no neurologic abnormality), suspect (deviation of tone, posture, movement patterns, reflexes, or head growth), or abnormal (history of meningitis, seizures, grade 3 and 4 IVH, and shunted hydrocephalus).

At 18 and 30 months of age (corrected for prematurity) a neurologic assessment was conducted by 2 authors (M.M., M.S.) and/or the 2 physicians of the follow-up clinic using a standardized set of the neurologic examination based on the combined methods of Prechtl¹¹ and the Collaborative Perinatal Study.¹² There was 95% agreement on neurologic classification among all examiners at all time points. In addition to the earlier criteria for abnormal classification at 18 and 30 months, the criteria for abnormal neurologic status were expanded to include cerebral palsy, blindness, deafness, shunted hydrocephalus, or uncontrolled seizures.

At 4 and 8 years of age, neurologic status of the children was examined again by the same examiners and classified as normal, suspect, or abnormal. The criteria for abnormal classification were the same. Criteria for the suspect classification were expanded to include fine motor weakness, unilateral sensorineural hearing loss, and atypical neurologic findings in tone, reflexes, gait, or movement for which no specific diagnosis was available. At 8 years of age, the diagnosis of attention deficit hyperactivity disorder (ADHD) determined by a clinical psychologist was added to the abnormal category using the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.¹³ Although it is different from other signs of abnormal neurologic status, it was added to the 8-year criteria. Recently, the literature¹⁴ on the neurobiology of ADHD has classified it as a brain disease. In addition, ADHD has been found to affect the outcomes in this study.

Data on cognition, medical diagnosis, and socioeconomic status (SES) were collected at each assessment point. Cognitive assessments were conducted at 18 months with the Bayley Scales,¹⁵ at 4 years with the McCarthy Scales of Children's Abilities¹⁶ at 8 years, and the Weschler Intelligence Scales for Children, Third Edition¹⁷ at 8 years. Academic achievement was measured with the Wide Range Achievement Test,¹⁸ and school performance, including retention in grade, placement in a special education class, academic resources, and diagnosis of a learning disability, was determined at 8 years of age from the participants' school records. Deficient cognitive status (ie, borderline intelligence, intellectually deficient) was not used in the neurologic classification criteria of normal/suspect/abnormal.

	RESULTS

Top Abstract Methods Results Discussion References

As expected, and as shown in Table 1, there were statistically significant differences among the study groups on birth weight, gestational age, Apgar score, number of days hospitalized, and Hobel perinatal risk scores. The FT infants had the lowest Hobel risk score and the NPT group had the highest, thus reflecting the severity of neonatal illness. At birth, the SES of the groups did not differ and ranged from lower-middle class to upper-middle class as determined by the Hollingshead Four-Factor Index.¹⁹ The sample was 50% male and 50% female, and 89% white. At the time of the child's birth, mothers were older than 16 with no known psychiatric diagnosis or history of drug or alcohol abuse.

As shown in Table 2, the sample retention rate for the 5 study groups was consistently high throughout the 5 time points (mean: 94%; range: 82%-100%). The lowest retention rate was 82% at 18 months of age in the NPT group, whereas the highest retention rate was 100% at age 4 in 4 of the 5 study groups.

Figures 1A through E show the percentage of change in neurologic status by age for the study groups at discharge, 18 months, 30 months, 4 years, and 8 years of age. In the healthy FT group (Fig 1A), the majority remained normal over the 8 years. No infant in the FT group was classified as suspect or abnormal at the time of hospital discharge. There was a slight increase in this classification over the time span, with 11% classified as suspect and 5% as abnormal at age 8. The 2 children who were classified as abnormal at age 8 were diagnosed as ADHD.

View larger version (66K): [in this window] [in a new window]   Fig. 1.   A, neurologic status of healthy full-term group at various ages; B, neurologic status of healthy preterm group at various ages; C, neurologic status of sick preterm group at various ages; D, neurologic status of neurologic preterm group at various ages; E, neurologic status of SGA preterm group at various ages.

The distribution of neurologic classifications for the HPT group (Fig 1B) was similar to that of the healthy FT group, although more HPT than FT children were classified as suspect. The percentage of children categorized as normal declined steadily from 100% at birth to 61% at age 8, whereas those categorized as suspect increased to 25% at age 8. As in the FT group, 4 children were first classified as abnormal at age 8 (14%) because of ADHD diagnoses.

The distributions for the remaining 3 preterm groups were distinctly different from those of the FT and HPT groups as shown in Fig 1C. Fifty-seven percent (n = 31) of the SPT group were classified as normal at hospital discharge, and this number increased to a peak of 73% at 30 months and dropped to 54% classified as normal at age 8. Thirty-nine percent of the SPT group was classified as suspect at hospital discharge, decreasing to a low of 19% at 30 months and increasing to 23% at age 8. The lowest percentage for abnormal classification in the SPT group was 3% at hospital discharge. This increased at each age point, with a total of 23% classified as abnormal at age 8. Seven of these 12 children were diagnosed with ADHD. With ADHD removed from the abnormal criteria, 4 of the 8 remained abnormal (Fig 2).

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Neurologic status by study groups at 8 years with ADHD diagnosis removed from abnormal criteria.

Because of predetermined criteria, all infants in the NPT group (Fig 1D) were classified as abnormal at discharge. From age 18 months to the assessment at age 4, the percentage of infants classified as neurologically suspect or abnormal fluctuates significantly. At 8 years of age, 41% of the infants were normal, 14% were suspect, and 45% were classified as abnormal. With ADHD removed from the abnormal criteria, 9 of these children remained abnormal (Fig 2).

The SGA group was the most variable in neurologic classification over age points (Fig 1E). Thirteen of the children (44%) were classified as normal at hospital discharge. This increased to 76% at age 4 and returned to 46% at 8 years of age. In the SGA group, the highest frequency of children classified as suspect was at the hospital discharge time point (38%). This decreased until age 4 (14%) and increased again to 32% at age 8. In the SGA group, originally, 17% were abnormal at hospital discharge. This decreased to 10% at both 30 months and 4 years, and increased to 21% at age 8. With ADHD removed from the abnormal criteria, 4 of the children remained abnormal (Fig 2).

To determine whether ADHD as an abnormal criterion for neurologic classification significantly influenced the findings, those children who were diagnosed with ADHD were reclassified as normal, suspect, or abnormal using the remaining criteria for classification. Figure 2 shows that at 8 years of age, the NPT group had the worst neurologic status. The other preterm groups were better but still had higher rates of abnormal neurologic status than the FT group. The ² was significant for the distribution of 8-year neurologic status by group with ADHD in the criteria (² = 12.31; P = .001) and without ADHD in the criteria (² = 20.77; P = .001). This analysis has the benefit of showing that these results for the abnormal category cannot be attributed to ADHD, and, moreover, this detailed information informs us of other indicators of abnormal status for ADHD children.

Statistical Analysis of Neurologic Findings

Hierarchical log linear analysis was used to examine changes over age in neurologic status by different study groups. The 3 categories of the change index were: 1) no change of neurologic status over time; 2) positive change, improvement in neurologic status; and 3) negative change, decline in neurologic status. We found that preterm heterogeneity influences neurologic status change over time in support of the hypothesis. Three groups, HPT, SPT, and SGA, showed significant neurologic fluctuation. No significant change was found in the FT and NPT groups (² = 14.32; P = .07).

Cognitive Outcome With Reference to Neurologic Status and Neonatal Morbidity

Tables 3 and 4 present cognitive scores gathered at 4 and 8 years of age, divided by study group and neurologic status. Given the instability of infant testing performance and lack of predictability to later cognitive outcome, the Bayley Mental Development Index scores gathered at age 18 months were omitted from this analysis.²⁰

At 4 years of age, children who were classified as abnormal were from the SPT, NPT, and SGA groups only. The mean scores were the lowest for these groups. Children classified as suspect had scores ranging from 88 to 99, with those in the FT and HPT group scoring higher than the SPT, NPT, and SGA groups. Those children categorized as normal had the highest scores despite their study group assignment. This was also found at age 8. Mean scores for the NPT and SGA groups were the lowest, >1.5 standard deviation (SD) below the standardized mean, which is within the diagnostic range for borderline intelligence.

Significant differences were found between neurologic classification for the full sample. Those children classified as normal had significantly higher cognitive scores than those classified as suspect or abnormal at both ages 4 (F = 12.68; P = .0001) and 8 (F = 4.63; P = .01).

Significant differences in cognition between the study groups at 4 years were found for those children classified as normal. The FT group had significantly higher cognitive scores than did all the preterm groups by direct contrast (F = 4.81; P = .03). No significant differences in cognition were found between study groups within the suspect and abnormal categories. No significant differences were found in cognition at 8 years within the normal, suspect, or abnormal categories.

School Performance

School achievement in reading (Table 5) and math (Table 6) at 8 years of age is presented by neurologic status group and study group. No significant differences were found between study groups. Significant differences were found between neurologic status in reading (F = 14.06; P = .0001) and math (F = 5.58; P = .004). Children classified as normal had higher achievement scores than did those classified as suspect and abnormal. Scores for the suspect groups were lower in reading only. Scores for those classified as abnormal were lower in reading and math.

Across study groups, FT children classified as neurologically normal had significantly higher math achievement scores than did all of the preterm groups (F = 9.56; P = .002). Two children in the FT group classified as abnormal had scores in the normal range but >1.5 SD below their cognitive mean scores, indicating a score discrepancy large enough to suggest a learning disability. Three of the 6 scores for the HPT, SPT, and NPT groups on reading and math were >1.5 SD below the standardized mean, indicating significant underachievement.

Table 7 shows school performance by study groups including grade retention rate, special education placement, additional academic resource help, and learning diagnosis. Academic resources and learning diagnosis were significantly related to study group classification, with the NPT group having highest need for academic resources and highest number of children with learning diagnoses.

Intervening Factors

Other intervening variables, such as SES, and medical sequelae other than neonatal morbidity may influence neurologic classification. We had the opportunity in this study to prospectively collect this information by study group. As seen in Table 8, the SES of the parents in the various study groups at different ages were relatively constant except at age 4. Parents in the FT, SPT, NPT, and SGA groups had higher SES than the parents of HPT children [F(4,183) = 3.79; P = .005] . 

As shown in Table 9, the number of children who were enrolled in early intervention was small. In Rhode Island, 60% of eligible 4-year-olds were not enrolled in Head Start because of transportation problems or waiting lists.²¹ Also, very few children had intervening medical problems through the 8-year period. They include 1 FT child with petit mal seizures, 1 HPT child with chronic history of febrile seizures, 1 SPT child with meningitis during infancy, 2 NPT children with meningitis and 2 with shunted hydrocephalus during their first year, and 1 SGA child diagnosed with neurofibromatosis at 18 months of age. Taking into account the small incidence of these possible intervening variables, we have reasonable confidence that the neurologic findings are supported.

	DISCUSSION

Top Abstract Methods Results Discussion References

The unique aspects of this study include its prospective classification of 5 study groups, a longitudinal study design, comprehensive assessments, long duration of follow-up, assessment of intervening factors, and a high compliance rate. These data utilize repeated neurologic categorization and developmental assessment including cognition, academic performance, SES, and medical status. This study meets the detailed methodologic requirements for preterm follow-up studies as prospective, not retrospective, longitudinal, not cross-sectional, and having a long duration of follow-up with minimal attrition (94% over 8 years). A 70% to 80% (90% is optimal) follow-up rate is recommended for valid interpretation of LBW outcome studies with adequate statistical power to have reasonable confidence in the results.²² This study had sufficient power (0.84) to discern statistical differences. We are not aware of another longitudinal study that prospectively stratified term and preterm infants based on neonatal morbidities.

It should be noted that, unlike many LBW studies, this study sample was recruited for research purposes, not in the context of health service delivery. Therefore, this study followed stringent research protocol (ie, researchers were blinded to the neonatal status of the children). Many follow-up studies are conducted as part of a clinical service follow-up program. Little is known about potential differences in the quality of data collected in a clinical versus research setting. For example, blindness to the child's history, and training and reliability of testers, including percentage agreement in reliability between study personnel, are frequently not reported in clinic-based follow-up studies.²³

Our findings confirm the hypothesis that change in neurologic classification over time varies as a function of neonatal morbidity, and the changes also affect the cognitive and school achievement outcomes. This investigation went beyond the broad classification of neonatal medical complications to specify the type and severity of neonatal morbidity most likely and least likely to have neurologic fluctuation. Children in the HPT, SPT, and SGA preterm groups showed fluctuation in neurologic classification over the 5 time points. The FT group showed the most stability. This finding verifies that different trajectories occur for NICU survivors based on factors other than gestational age and birth weight criteria.²

The study design allows for 2 control groups differentiated by gestational age. The FT group was healthy at 40 weeks, and the HPT group was healthy and <32 weeks' gestation. With the FT and HPT groups as controls, meaningful comparisons can be made with the other 3 preterm groups about preterm morbidity and neurologic classification. Comparing the HPT group with the 3 other preterm groups (SPT, NPT, SGA) illustrates the differences in long-term follow up of preterms. Bregman⁴ states that control groups are necessary in long-term follow-up if the effects of prematurity versus the impact of other multiple factors on outcomes are to be determined. The stratification into the 5 study groups, including the 2 control groups, was an attempt to more closely specify knowledge about the mechanisms of brain plasticity, normal development, and the modifying effects of intervening factors on later outcome.

This study's findings confirm previous research associating neonatal morbidity with school-age outcomes.²⁴ However, the study extends this body of research by identifying another marker for consideration of long-term outcomes in NICU survivors. This marker is repeated neurologic examination throughout childhood. Taylor et al² found that preterm children who experienced chronic lung disease had the same IQ equivalent score as VLBW children who experienced grade 3 or 4 IVH. We also found this to be true in the SPT and NPT groups at ages 4 and 8 for those children classified as neurologically normal only. The SPT and NPT group who were classified as neurologically suspect or abnormal had significantly lower cognitive scores compared with SPT and NPT neurologically normal groups.

The differences between SGA and average for gestational age infants on cognitive ability has been investigated. Wallace and McCarton,³ in a neurodevelopmental assessment of SGA preterm through age 6, found SGA infants to have significantly lower cognitive scores at ages 1, 2, and 5/6 years compared with average for gestational age children. They hypothesized that the reason for these differences is related to retardation of brain growth, and this may have enhanced the adverse effect of perinatal complications on neurologic status. In the present study, children in the SGA group had lower cognitive scores if classified as neurologically suspect or abnormal at ages 4 and 8. SGA children classified as normal had cognitive scores no different from the FT group at ages 4 and 8. These findings indicate the need for continued surveillance of the mechanisms of intrauterine malnourishment on the developing brain.

Two points may be made about the comparison of neurologic status among the study groups. First, no child was classified as abnormal until age 8 in the FT and HPT groups, whereas there was a steady increase in abnormal classification in both the SPT and SGA groups over 8 years. Second, no child classified as abnormal at age 4 had improved at age 8 in any study group. The NPT groups showed the largest fluctuation in the suspect and abnormal categories, lowest cognitive scores, and poorest academic performance compared with the HPT and FT groups. This fluctuation has been reported in 2 longitudinal studies of infants who had neurologic insult and poor developmental outcome.^24,25 Our finding is consistent with those of Collin et al²⁴ that normal infant development is poorly predictive of continued normal development. Children born prematurely are at risk for the emergence of sleeper effects at later ages. The current study results serve to emphasize the need for continued neurodevelopmental follow-up.

Neurologic classification criteria for this study differ from other neurologic follow-up studies of LBW children by adding ADHD to the abnormal classification criteria at age 8. This addition is based on current neurologic studies of ADHD children; these studies recognize ADHD as a dysfunction in the orbital-limbic pathways of the frontal area as the probable impairment demonstrated in ADHD behaviors, particularly behavioral disinhibition and diminished sensitivity to behavioral consequences or incentive learning.²⁶ ADHD was a criteria for the abnormal classification based on current research findings that support a neurobiological foundation to this disorder. ^14,26,27 The 25 children diagnosed with ADHD in this study represented all study groups with the NPT group having the highest percentage (FT; 10.5%; HPT, 13%; SPT, 13%; NPT; 22%; SGA, 11%). Our rate of ADHD was 15% for the preterm children; 3 to 5 times the national prevalence rate.²⁸ Szatmari and colleagues²⁹ reported ADHD in 16% of their VLBW children. Although some studies have reported that preterm children are at higher risk for developing ADHD, others have found no difference between the VLBW children and a control group.³⁰ In this study at age 8, there was a significant relationship between neurologic status and the 5 study groups. The relationship was significant whether or not ADHD was included in the criteria.

Others have classified ADHD as a brain abnormality. Szatmari²⁹ labeled ADHD as a neurodevelopmental problem in a sample of preterm infants born between 1980 and 1982. We do acknowledge that ADHD is different from other signs of abnormal neurologic status such as meningitis, uncontrolled seizures, cerebral palsy, and shunted hydrocephalus. However, even if ADHD is an imperfect indicator of abnormal neurologic status, it is important to reiterate that the relationship between neurologic classification status and study group was significant whether or not ADHD was included in the criteria.

Our findings that cognitive scores at age 4 and academic achievement at age 8 are higher for full-term, neurologically normal children compared with other preterm groups are consistent with the literature. Academic achievement of NICU survivors in longitudinal studies at school age show that they may approach the average range; however, the mean scores still tend to be significantly lower than control groups.^31-33 When academic achievement scores are viewed with consideration to neurologic status, it is clear that children classified as normal had the highest scores and the preterm children classified as abnormal had the lowest scores. Neurologic status adds to our information about medical morbidity and academic achievement. The effect of neurologic status may be seen in deficits in specific processes such as attention and memory, and are demonstrated in academic difficulties. The consistently large number of SPT and NPT children who were in special education classes, needed additional learning resources, or who had a learning disability diagnosis underscores the increased cost to society of providing educational resources for these children. Almost 4.5 million students in the United States (birth to age 21) receive special education services, an increase of almost 20% in 20 years.³⁴ Twenty to 65% of children in VLBW and LBW samples without major handicap required special education or additional school services.²⁵ To our knowledge, only one other study² has documented the relationship of specific neonatal morbidity to school-age developmental outcomes in VLBW survivors. The findings from our study represent one regional perinatal center with the best ranking in the nation for survival and quality of outcomes for the lowest birth weight preterm infants.³⁵ Although this may be viewed as a limitation, it is not unusual as most LBW studies are single-site studies.

The finding that fluctuating neurologic status is a function of neonatal morbidity indicates that children with morbidities similar to those of the HPT, SPT, or SGA groups may be classified as normal at an early age and become suspect or abnormal later. A child with neurologic categorization of suspect or abnormal may require early intervention services at 18 and 30 months of age, speech or physical therapies at preschool age, and additional school resources at school age. There have been major technological and therapeutic advances in decreasing the mortality of NICU survivors; however, the rates of neonatal morbidity have remained stable.^1,2,31 Identification of additional markers, such as fluctuating neurologic status in determining later developmental outcomes, improves our understanding of the importance of neonatal sequelae.

Neonatal morbidities exert a significant impact in neurologic outcomes among preterm children during the 8 years of assessment. Compromised neurologic status adversely affects cognitive and school performances. Neonatal medical status is an important variable indicating neurocognitive and school performance outcomes in LBW infants.

	ACKNOWLEDGMENT

This work was supported by National Institutes of Health, National Institute of Child Health and Development Grant RO1 191985 and National Institute of Nursing Research Grants 02263 and 03695.

	FOOTNOTES

Received for publication Mar 30, 1999; accepted Feb 11, 2000.

Reprint requests to (M.M.M.) Infant Development Center, Women and Infants' Hospital, 111 Plain St, Providence, RI 02903.

	ABBREVIATIONS

LBW, low birth weight; VLBW, very low birth weight; IVH, intraventricular hemorrhage; NICU, neonatal intensive care unit; FT, full-term; HPT, healthy preterm group; SPT, sick preterm group; NPT, neurologic preterm group; SGA, small for gestational age; ADHD, attention deficit hyperactivity disorder; SES, socioeconomic status; SD, standard deviation.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Hack M, Klien N, Taylor HG. Long term developmental and health outcomes of low birth weight children. In: Shiono P, Behrman R, eds. The Future of Children. Palo Alto, CA: Packard Foundation; 1995
2. Taylor HG, Klein N, Schatschneider C, Hack M Predictors of early school age outcomes in very low birth weight children. Dev Behav Pediatr. 1998; 19:235-243 [Medline]
3. Wallace IF, McCarton CM Neurodevelopmental outcomes of the premature, small-for-gestational infant through age 6. Clin Obstet Gynecol. 1997; 40:843-852 [CrossRef][Medline]
4. Bregman J Developmental outcome in very low birthweight infants: current status and future trends. Pediatr Clin North Am. 1998; 45:673-690 [CrossRef][Medline]
5. Litt R, Joseph A, Gale, R Six year neurodevelopmental follow-up of very low birthweight children. Isr J Med Sci. 1995; 31:303-308 [Medline]
6. 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 Birth Weight Children: Annual Advances in Applied Developmental Psychology. Norwood, NJ: Ablex; 1992:259-274
7. Kitchen WH, Ford GW, Rickards AL, Lissenden JV, Ryan MM Children of birth weight <1000 g: changing outcome between ages 2 and 5 years. J Pediatr. 1987; 110:283-288 [CrossRef][Medline]
8. Vohr BR, Garcia-Coll C, Flanagan P, Oh W. Effects of intraventricular hemorrhage and socioeconomic status on perceptual, cognitive, and neurological status of low birth weight infants at 5 years of age. J Pediatr. 1992;280-285
9. Lubchenco LC, Hansman C, Boyd E Classification of newborns based on maturity and intrauterine growth. Pediatrics. 1966; 37:403 [Abstract/Free Full Text]
10. Hobel CJ, Hyvarinen MA, Okada DM, Oh W Prenatal and intrapartum high risk screening Am J Obstet Gynecol. 1973; 17:1-9
11. Prechtel H, Beintema D. Neurological examination of the full term newborn infant. London, United Kingdom: Heineman; 1967
12. Niswander KR, Gordon M. The Collaborative Perinatal Study: The Women and Their Pregnancies. Philadelphia, PA: WB Saunders Co; 1972
13. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994
14. Faraone SV, Bierderman J Neurobiology of attention deficit hyperactivity disorder. Biol Psychiatry. 1998; 44:951-958 [CrossRef][Medline]
15. Bayley N. Bayley Scales of Infant Development. New York, NY: The Psychological Corporation; 1969
16. McCarthy M. McCarthy Scales of Children's Abilities. New York, NY: American Psychological Association; 1972
17. Wechsler D. Weschler Intelligence Scales for Children. 3rd ed. San Antonio, TX: The Psychological Corporation; 1991
18. Jastak S, Wilkinson GS. Wide Range Achievement-Revised Administration Manual. San Antonio, TX: The Psychological Corporation; 1984
19. Hollingshead AB. Four-Factor Index of Social Status. New Haven, CT: Yale University Press; 1975
20. Bornstein, MH. Stability in early mental development: from attention and information processing in infancy to language and cognition. In: Bornstein MH, Krasnegor NA, eds. Stability and Continuity in Mental Development. Hillsdale, NJ: L Erlbaum; 1989:145-170
21. Msall ME, Bier J, LaGasse L, Tremont, M, Lester B The vulnerable preschool child: the impact of biomedical and social risks on neurodevelopmental function. Semin Perinatol. 1998; 5:52-61
22. Vohr BR, Msall ME Neuropsychological and functional outcomes of very low birthweight infants. Semin Perinatol. 1997; 21:202-220 [CrossRef][Medline]
23. Lester BM, Miller-Loncar C Biology versus environment. Clin Perinatol 2000; 27:461-481 [CrossRef][Medline]
24. Collin MF, Halsey CL, Anderson CL Emerging developmental sequelae in the "normal' extremely low birth weight infant. Pediatrics. 1991; 88:115-120 [Abstract/Free Full Text]
25. Vohr BR, Garcia-Coll CT Neurodevelopmental and school performance of very low birth weight infants: a seven-year longitudinal study. Pediatrics. 1985; 76:345-350 [Abstract/Free Full Text]
26. Barkley RA. Attention-Deficit Hyperactivity Disorder. New York, NY: Guilford Press; 1990
27. Milberger S, Biederman J, Faraone S V., Chen L, Jones J. Is maternal smoking during pregnancy a risk factor for attention deficit hyperactivity disorder in children? Am J Psychiatry. 1996; 153:1138-1142 [Abstract/Free Full Text]
28. Schnelle E More than inattention. Adv Nurs Pract. 1994; 2:9-12
29. Szatmari P, Saigel S, Rosenbaum P, Psychiatric disorders at five years among children with birthweights <1000 g: a regional perspective. Dev Med Child Neurol. 1990; 32:954-962 [Medline]
30. McCormick MC, Gortmaker SL, Sobol AM Very low birthweight children: behavior problems and school difficulty in a national sample. J Pediatr. 1990; 117:687-693 [Medline]
31. McCormick MC The outcomes of very low birth weight infants: are we asking the right questions? Pediatrics. 1997; 99:869-876 [Free Full Text]
32. Schraeder BD, Heverly MA, O'Brien C, McEvoy-Shields K Vulnerability and temperament in very low birth weight school-aged children. Nurs Res. 1992; 41:161-165 [Medline]
33. Hack M, Breslau N, Aram D, Weissman B, Klein N, Barowski-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]
34. Gerber MM, Levine-Donnerstein D. Educating all children: ten years later. Except Child. 1989:17-27
35. Women and Infants' Hospital of Rhode Island. Annual Report: The Value to Community. Providence, RI: Women and Infants' Hospital of Rhode Island; 1998