Published online August 1, 2006
PEDIATRICS Vol. 118 No. 2 August 2006, pp. e449-e459 (doi:10.1542/peds.2005-3024)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dahl, L. B.
Right arrow Articles by Rønning, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dahl, L. B.
Right arrow Articles by Rønning, J. A.
Related Collections
Right arrow Premature & Newborn
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?

ARTICLE

Emotional, Behavioral, Social, and Academic Outcomes in Adolescents Born With Very Low Birth Weight

Lauritz Bredrup Dahl, PhDa,b, Per Ivar Kaaresen, MDa,b, Jorunn Tunby, MPHa, Bjørn Helge Handegård, CScb, Siv Kvernmo, PhDa,b and John A. Rønning, PhDa,b

a Child and Adolescent Clinic, University Hospital of North-Norway, Tromsø, Norway
b Institute for Clinical Medicine, University of Tromsø, Tromsø, Norway


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
BACKGROUND. Very low birth weight survivors are at increased risk of developing emotional and behavioral problems and low social and academic competencies. Information on such problems in very low birth weight adolescents is still sparse.

OBJECTIVES. Our purpose for this work was to study gender-specific emotional and behavioral problems and social and academic competencies in a cohort of very low birth weight adolescents in north Norway.

METHODS. Families with very low birth weight adolescents aged 13 to 18 years, born between 1978 and 1989 (n = 162) were addressed by mail and asked to complete the Child Behavior Check List and the Youth Self-Report. Data were compared with 2 normative adolescent populations (Child Behavior Check List, n = 540; Youth Self-Report, n = 2522). Scores given by very low birth weight adolescents and their parents on identical items in Child Behavior Check List and Youth Self-Report (cross-informant syndrome constructs) were compared in pairs. To explore predictive effects, demographic and early medical characteristics were entered into a hierarchical multiple regression analysis.

RESULTS. There were 156 eligible families, and 99 (63.5%) responded. All completed the Child Behavior Check List, and 82 (52.6%) completed the Youth Self-Report. Very low birth weight boys reported less externalizing and internalizing behaviors and thought and attention problems and higher activity score, whereas very low birth weight girls reported less externalizing behavior and less social, thought, and attention problems and higher activity score compared with normative adolescents. Very low birth weight parents, however, reported more social and attention problems and less social and school competence in boys and more internalizing behavior and social and attention problems and less school competence in girls compared with normative parents. They scored high proportions of both genders within the borderline/clinical range on all of the scales, except for externalizing behavior and social problems in girls. Female very low birth weight adolescents, in contrast to males, reported more problems than parents when compared in pairs, and externalizing problems in particular were not recognized by parents.

CONCLUSIONS. From parents' point of view, significant proportions of very low birth weight adolescents experience more emotional and behavioral problems and less competence than normative adolescents. In contrast, very low birth weight adolescents state less problems and similar or higher competence than normative adolescents. Very low birth weight adolescent girls report more emotional and behavioral problems compared with their parents than very low birth weight adolescent boys do. Externalizing problems in very low birth weight adolescent girls are often not recognized by parents. To better understand these seemingly paradoxical findings and to develop adequate intervention programs, there is a need for prospective longitudinal studies.

Key Words: very low birth weight • adolescent • behavioral problems • emotional problems • competence • gender-specific • Child Behavior Check List • Youth Self-Report

Abbreviations: VLBW—very low birth weight • HRQoL—health-related quality of life • YSR—Youth Self-Report • CBCL—Child Behavior Check List • SES—socioeconomic status • SGA—small for gestational age • ELBW—extremely low birth weight • FGR—fetal growth restriction

Introduction of neonatal intensive care in the 1960s led to progressive improvements in survival and morbidity rates of very low birth weight (VLBW; <1500 g) infants.1 Early outcome studies reported decreasing trends in the occurrence of neurosensory and neuromotor impairments.2,3 Later long-term outcome studies also reported on neuropsychiatric,4,5 cognitive,5–10 and emotional/behavioral problems and deficits in social competencies5–19 in VLBW survivors. Studies on health-related quality of life (HRQoL) and self-esteem were also reported.10,20–24 Common features in these reports were increasing trends in survival of small preterms, with a substantial number having subnormal mental and psychomotor development; cognitive, academic, and emotional and behavioral problems; and reduced competencies and self-esteem, which may persist into young adulthood.18,25 Along with a stable incidence of preterm birth, this inevitably results in an increased number of significantly affected survivors.26

Emotional and behavioral problems vary with birth weight6 and age at assessment and gender,27 and they follow different developmental trajectories.27 Such problems do not, however, change with culture.28 It is claimed that parent-reported problems and competencies should be supplemented with self-reports in older children and adolescents,27,29 but so far only a few studies have included both parent and youth self-reports. In the present study, we report on emotional and behavioral problems and social and academic competencies in VLBW infants of both genders, having reached adolescence, using parent-reported and self-reported information.30,31


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
VLBW Group
During 1978–1989, 265 VLBW (≤1500 g) infants were live born in the 2 northernmost counties of Norway (7.1 per 1000 live born). Of those, 175 infants were alive at 4 years of age (survival rate: 66.0%).32,33 During early spring 1998 and 2004, those adolescents aged 13 to 18 years (n = 162) and their parents were addressed and asked to complete the Youth Self-Report (YSR) and the Child Behavior Check List (CBCL), respectively. Mailed requests were sent without regard to degree of disability. We consistently use the term parent for any adult caregiver in this report. Maternal and neonatal characteristics of the cohort and follow-up data at 4 and 13 to 18 years of age are briefly described in Table 1.


View this table:
[in this window]
[in a new window]
  		TABLE 1 Maternal Characteristics, Infant Data, and Data at Ages 4 and 13 to 18 Years

 
Control Groups
Because there are no Norwegian normative data available, Swedish data were used.34,35 Competencies and emotional/behavioral problems as reported by parents on the CBCL were examined in 1308 school-aged children/adolescents (627 boys and 681 girls) aged 6 to 16 years during 1992–1993 to create Swedish normative scores.34 The levels of emotional/behavioral problems were found to be comparable to those reported in not-normative Norwegian studies.36,37 We compared adolescents aged 13 to 16 years (n = 540) in this normative study with our cohort of VLBW adolescents aged 13 to 18 years.

Competencies and emotional/behavioral problems as reported by Swedish adolescents using the YSR were studied in 2522 subjects (1222 boys, 1274 girls, and 26 unknown) aged 13 to 18 years between 1995 and 1997 to create normative values.35 Problem scores were found to be similar to those from comparable, not-normative Norwegian studies.36,38 We compared our cohort of VLBW adolescents with the adolescents in this normative study. Because social class does not vary much in the Scandinavian countries, and because the population is rather homogeneous, we did not match the control groups for this or other demographic characteristics.

Measures
CBCL
The CBCL is a 120-item questionnaire that is completed by parents to reflect their view of the child's behavior at the time of administration and for the preceding 6 months.30 It provides a total problem score, 2 broadband scales (internalizing and externalizing) and the following dimensions (syndromes, subscales): withdrawn behavior, somatic complaints, anxiety/depressed behavior, delinquent behavior, aggressive behavior, social problems, thought problems, and attention problems. Parents rate each item from 0 (never true) to 2 (often or always true) for the child. The CBCL also contains competence scales, that is, total competence score with 3 subscales (activities, social, and school). High problem score thus means more emotional and behavioral problems, and high competence score means more competence. For the total problems score, internalizing, and externalizing, the borderline and clinical range starts at the 82nd and above the 90th percentiles, respectively, and for the subscales at the 95th and above the 98th percentiles, respectively. For the competence score, analogous cutoffs have been established for the low end of the scales, that is, at the 17th and below the 10th percentile for total competence, and at the fifth and below the second percentile for the subscales.30

YSR
The YSR is a self-report questionnaire for ages 11 to 18 years.31 It consists of 2 parts; the first contains 7 competence items, which include an activity scale, a social competence scale, and a total competence scale, which includes academic performance. The second part contains 103 problem items and 16 socially desirable items. The adolescents rate the items for how well they describe him or her at present or during the past 6 months on a 3-point scale from 0 (never true) to 2 (often or always true). The 103 problem items are combined to form total problems consisting of 8 subscales and 2 broadband dimensions (internalizing and externalizing). The subscales are withdrawn behavior, somatic complaints, anxious/depressed behavior, social problems, thought problems, attention problems, delinquent behavior, and aggressive behavior. Borderline and clinical ranges are as reported for the CBCL.30 As for the CBCL, a high problem score means more emotional and behavioral problems, and a high competence score means more competence.

Cross-informant Syndrome Constructs39,40
To analyze degree of agreement between informants, cross-informant syndrome constructs were established. They consist of 8 identical syndrome scales (subscales or dimensions) in YSR and CBCL. These have 82 identical items to enable correct comparisons among total problems, externalizing, internalizing, and syndrome scale sum score means in the 2 instruments. The 8 syndrome scales are attention problems, aggressive behavior, anxious/depressed, delinquent behavior, social problems, somatic complaints, thought problems, and withdrawn behavior. This kind of assessment has been developed by Achenbach39 to improve the coordination of data from the instruments collecting parent, self-, or teacher's report form. To this end, core syndromes were derived from those items most commonly contributing to specific syndromes on CBCL, YSR, and teacher's report form. Cross-informant syndrome constructs were then established from those core syndromes present on ≥2 of the instruments. Cross-informant syndrome constructs have been used to assess agreement on emotional/behavioral problems and competencies in normative populations.40,41

Socioeconomic Status
The socioeconomic status (SES) of the parents was assessed using the parent occupation classification of the Hollingshead index (A.B. Hollingshead, PhD, unpublished data, 1975) and in accordance with the guidelines of Statistics Sweden.42 Categories were merged into the following classes: 1 indicates upper status (enterprisers, lawyers, physicians, etc), 2 indicates middle status (civil servants, teachers, etc), and 3 indicates lower status (unskilled and skilled workers).

Statistics
When comparing responders and nonresponders, differences between groups of categorical data were analyzed by using {chi}2 tests, whereas t tests were used in continuous data, and a P < .05 was considered statistically significant. When comparing scores in the VLBW group with the normative CBCL and YSR samples, t tests were used. Because of multiple comparisons a P <.01 was considered significant. Cohen's d is given as an effect size estimate.43 An effect size of 0.4 to 0.6 was considered moderate, and >0.6 was considered high. Two-way analyses of variance, including an interaction term (age group by VLBW/normative control), were used to test whether differences in mean scores in CBCL and YSR between VLBW and normative adolescents of both genders were consistent at both ends of the chosen spectrum of adolescence. The correlations between CBCL and YSR cross-informant syndrome constructs in the VLBW cohort were examined using intraclass correlation coefficient. Paired-samples t tests were used to compare sum score means in the cross-informant syndrome constructs,39,40 and a P < .01 was considered significant. Hierarchical regression analysis was used to analyze the independent effect on behavioral outcome of variables known to influence outcome in VLBW infants. Three blocks were created. The first consisted of SES and year of birth, the latter chosen because of recruitment of subjects born over a long period of time, during which intensive neonatal care might have changed. The second block consisted of gestational age, gender, outborn delivery, and small for gestational age (SGA), and the third included sepsis, bronchopulmonary dysplasia, Apgar score at 5 minutes, and days on ventilator. Parent-reported and self-reported total problems, externalizing, internalizing, and total competence were used as dependent variables. For each of the outcome variables, the 3 blocks were evaluated sequentially.

Step 1 evaluated block 1 only, step 2 evaluated block 2 controlled for block 1, and step 3 evaluated block 3 controlled for block 1 and 2. The proportion of explained variance (R2) is given as an effect size estimate in the regression analyses. When analyzing predictive value of single independent early predictors for outcome, P < .01 was considered significant. The statistical package SPSS 12.0 (SPSS Inc, Chicago, IL), was used for the analyses.

Ethics
The study was approved by the regional committee for medical research ethics.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Response Rates
Of the 175 VLBW adolescents who were alive at 4 years, 1 died later. Of the remaining, 162 were 13 to 18 years of age (Table 1). Six families could not be traced, giving 156 families eligible to complete the YSR and CBCL questionnaires. Eighty-two adolescents ([52.6%] 35 boys and 47 girls) completed the YSR, and 99 parents (63.5%) completed the CBCL for 44 boys and 55 girls. If the 6 untraced families are included in the calculations, the response rate for CBCL and YSR was 61.1% and 50.6%, respectively.

Responders Versus Nonresponders
Data at birth, during neonatal and postneonatal care, and at 4 and 13 to 18 years of age are given in Table 1. The infants of responders had significant shorter gestational age (29.3 vs 30.7 weeks; P = .003), and a higher proportion was female (55.6% versus 35.0%; P = .01). Other variables showed no significant differences. At follow-up, 20 adolescents (12.8%) were disabled, 12 (12.1%) among responders and 8 (12.7%) among nonresponders. Of the 12 responding families 7 (7.1%) completed the YSR. Disabilities were blindness because of retinopathy of prematurity (n = 2), hearing impairment (n = 2), cerebral palsy (n = 15), severe mental retardation (n = 4), and autism (n = 2).

Comparisons of Behavior and Competence on the YSR and CBCL
VLBW Versus Normative Male Adolescents
Data are given in Table 2. VLBW boys reported significantly lower than normative boys on total problems and both broadband scales and on all syndrome scales, except for withdrawn and social problems, where no differences were reported. On the competence scales, they reported no differences except for activities, where they reported higher than normative boys. The proportions of VLBW boys scoring within borderline/clinical range were similar to those of normative controls. When adjusting for SES there were no essential changes, except for attention problems, for which the adjusted score difference was no longer statistically significant (d = 0.40; P = .02).


View this table:
[in this window]
[in a new window]
  		TABLE 2 Self- and Parent-Reported Behavior Problems and Competencies in Male and Female VLBW Adolescents Compared to Normative Values

 
Parents of male VLBW adolescents scored their offspring significantly higher on the total problems and on the social and attention problems syndrome scales. On the competence scales, the differences were significant in favor of the normative group for social and school competencies. The proportions of male VLBW offspring scored by their parents within the borderline/clinical ranges were higher than those of normative controls for all of the scales (Table 2). Results were the same regardless of whether VLBW adolescents aged 13 to 18 years or 13 to 16 years were compared with the normative population aged 13 to 16 years. When adjusting for SES, there were no essential changes, except for delinquent behavior, for which the adjusted score difference became borderline statistically significant (d = 0.43; P = .01), and for social competence for which the adjusted score difference was no longer statistically significant (d = 0.37; P = .03). There were no significant interactions between age groups (13–15 vs 16–18 years) and mean score differences between VLBW and normal male adolescents in any CBCL or YSR scores, indicating valid results throughout the age range studied.

VLBW Versus Normative Female Adolescents
Data are provided in Table 2. VLBW girls reported significantly lower than normative girls on total problems and on externalizing, whereas internalizing was reported similar. They reported significantly lower on all of the syndrome scales except for withdrawn, somatic complaints, and anxious/depressed. On the competence scales they reported no significant differences except for higher scores on activities. The proportions of VLBW girls scoring within normative borderline/clinical ranges were low compared with those of normative controls, except for anxious/depressed behavior (15%; Table 2). When adjusting for SES, there were no essential changes in problem and competence score differences.

Parents of VLBW female adolescents reported significantly higher scores than normative parents on internalizing and on the subscales for anxious/depressed, social, and attention problems. There were no significant differences on externalizing and on the corresponding subscales. On the competence scales, parents reported significantly lower for total and school competence. The proportions of female VLBW offspring scored by their parents within the borderline/clinical ranges were higher than those of normative female controls for all of the scales except for somatic complaints and externalizing, including both delinquent and aggressive behavior (Table 2). When adjusting for SES, there were no changes, except for less differences on total problems (d = 0.23; P = .13) and total competence (d = 0.28; P = .07). There were no significant interactions between age groups (13–15 vs 16–18 years) and mean score differences between VLBW and normal female adolescents in any CBCL or YSR scores, indicating valid results throughout the age range studied.

Accordance Between Parents and Their VLBW Offspring
Total and gender-specific intraclass correlation coefficients of the CBCL and YSR cross-informant syndrome constructs are given in Table 3. The coefficients ranged from 0.21 (aggressive behavior in girls) to 0.81 (thought problems in boys). The correlation coefficient was 0.60 between total problem scores and 0.61 between total competence scores, with a higher correlation coefficient for boys than for girls. All of the correlations, except for externalizing and aggressive behavior in girls and somatic complaints in boys, were significant at the ≤.01 level. Female adolescents reported significantly higher mean scores than their parents did on all of the scales, except for social and thought problems. There were no significant differences between male adolescents and their parents.


View this table:
[in this window]
[in a new window]
  		TABLE 3 Intraclass Correlations and Mean Paired Differences in Cross-informant Syndrome Constructs of CPCL and YSR

 
Effects of Maternal, Perinatal, and Neonatal Data on Problems and Competencies in Adolescence
Data are shown in Table 4. On the CBCL problem and competence scales, only small proportions of the outcome variance (5% and 3%, respectively) are explained by the SES and year of birth. The proportion of the explained variances increased significantly when introducing block 2 (16% for both scales); introduction of block 3, however, contributed insignificantly (4% for both scales). When analyzing single explanatory variables, SGA was of significant importance in the CBCL scales, except for internalizing behavior. Also on the YSR problem scales, insignificant, small proportions of the outcome variance could be explained by SES and year of birth; on the total competence scale, however, SES and year of birth explained 15%. The introduction of blocks 2 and 3 did not increase the proportion of the explained variance significantly for total problems (10% and 11%, respectively). When evaluating the association of the variables in block 2 with self-reported scores on internalizing behavior, the proportion of the explained variance increased significantly (20%). When analyzing single explanatory variables, gender was significantly associated with self-reported scores on total problems and internalizing. For total competence, year of birth was the most important predictor. Birth years 1979–1982 predicted low score on total competence.


View this table:
[in this window]
[in a new window]
  		TABLE 4 The Association Between Early Variables Known to Affect Outcome in VLBW Infants and Parental- and Self-Reported Behavior Problems and Competencies in Adolescence Is Evaluated Using Hierarchical Multiple Regression Analysis

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The main results in this study are: (1) the high prevalence of parent-reported mental health problems in VLBW adolescents (Table 2); (2) the discrepancies between self-reported (Table 2) and parent-reported (Table 2) problems and competencies in VLBW adolescents compared with normative adolescent populations; (3) the disagreement between female VLBW adolescents and their parents in contrast to male VLBW adolescents (Table 3); and (4) that behavioral problems and competencies in VLBW adolescents were predicted by early characteristics, such as SGA (CBCL) and gender and birth year (YSR; Table 4).

Comparison of VLBW and Normative Adolescents on YSR and CBCL
One main result of this follow-up study of VLBW adolescents was high behavioral problem scores and low or equivalent competence scores as reported by parents and concurrently recorded, low self-reported behavioral problem scores and high or equivalent self-reported competence scores compared with normative adolescent populations. This was found for both genders. Others have reported similar results.17,19,44 Rickards et al17 found no differences in self-reported emotional and behavioral problems in VLBW adolescents compared with randomly selected children of normal birth weight at 14 years, whereas their parents reported more such problems than did parents in the control group. Saigal et al19 reported similar results in adolescents aged 12 to 16 years who were of extremely low birth weight (ELBW). By adolescent self-report, there were no significant differences between ELBW and control adolescents. However, parents of ELBW adolescents reported significantly higher scores than parents of controls for depression and attention problems. In the study by Indredavik et al,44 VLBW adolescents reported similar to controls on YSR, in contrast to parents of VLBW adolescents who reported higher scores than parents of controls on several subscales on CBCL. Such parental overreporting may partly be because of a greater awareness and concern in parents and particularly in mothers of preterm offspring compared with mothers of normative adolescents. It is claimed that mothers of VLBW children adopt a more active and controlling strategy as a consequence of an emerging developmental delay in their offspring, which may influence their completion of the CBCL.13 Why do VLBW adolescents, as one would expect, not report more emotional and behavioral problems than controls on YSR? Some19,44 suggest that there might be a tendency for underreporting of problems by VLBW adolescents, which may also be suspected from reports on HRQoL and self-esteem in VLBW adolescents.10,20–24 When investigating psychiatric symptoms in VLBW offspring by questionnaires, as well as psychiatric interview, more psychiatric symptoms were found in VLBW offspring compared with controls by interview.44,45 Questionnaires completed by parents, however, were more in line with results obtained by interview of their offspring,44 indicating that VLBW adolescents may underreport on this type of questionnaire. What factors that may contribute to a positive self-perception and, thus, lead to underreporting by VLBW adolescents in studies on HRQoL, self-esteem, and behavioral problems and competencies are unknown. However, both denial and a recalibration of expectations have been suggested.19

Self-reports on emotional and behavioral problems in VLBW adolescents are sparse and to some extent conflicting.5,17,25 In their follow-up of VLBW offspring until young adulthood, Hack et al25 found no differences in self-reported total problems and internalizing and externalizing behaviors in boys and no differences in self-reported attention deficit problems for any gender, compared with normal birth weight controls. Young women in the study by Hack et al,25 however, reported more withdrawn behavior, and a higher proportion had internalizing behavior above the borderline/clinical cutoff compared with control female subjects. These findings are in line with our findings of abnormally high proportions of VLBW girls reporting within borderline/clinical range for internalizing behaviors (anxious/depressed and withdrawn; Table 2). The lower occurrence of self-reported delinquent behavior in VLBW young adults of both genders is also in accordance with our findings. In contrast with our findings, Elgen et al5 describe more self-reported attention problems, problems with spare time activities, and more aggressive behavior in low birth weight school children (<2000 g) at age 11 compared with normal birth weight controls. At age 11, however, the need to present oneself as not different from peers is probably not as developed as it may be among older adolescents and young adults. In line with this speculation, Rickards et al17 did not find any differences in self-reported behavioral problems at 14 years between VLBW offspring and controls, irrespective of gender. Apparently inconsistent results in studies on VLBW offspring at different stages of life may alternatively be explained by similar developmental trajectories in VLBW children and adolescents to those found in normative populations27 and, thus, emphasize the importance of selecting strict corresponding age distributions when comparing study results.

Our finding that parents of VLBW adolescents report more behavioral problems than parents of controls was as expected, and is in accordance with other reports on ELBW and VLBW offspring at various ages.8,15,45–48 Adjusting for SES did not change the results substantially. This was as expected, because differences in SES in Nordic countries are small.

Accordance Between Parents and Their Adolescent Offspring
In the general population, adolescents normally report many more problems about themselves than their parents do.29,36,40,41,49 In the study by Saigal et al,19 comparisons within adolescent/parent dyads revealed that both the ELBW and the control cohorts of adolescents reported significantly higher scores than their parents did. When comparing sum score mean values of cross-informant syndrome constructs in CBCL and YSR in the present study (Table 3), we also found that this was the case in VLBW female adolescents but not in males. Despite the fact that VLBW female adolescents themselves report significantly more problems than their parents do, still a fair degree of agreement between parents and offspring is supported by significant correlations between the parent reports and self-reports. Recent studies have confirmed a low to moderate parent-adolescent agreement in normative populations.41,49,50 Cantwell et al29 found {kappa} values for parent-adolescent agreement on psychiatric disorders ranging from 0.19 (poor) for alcohol abuse and alcohol dependence to 0.79 (excellent) in conduct disorders. They found good agreement for attention deficit disorders and poor agreement for depression and anxiety disorders, and agreement was better for externalizing than for internalizing disorders. This is not strange, because internalizing disorders are of a more subjective character, whereas externalizing disorders are more socially recognizable. Our findings in VLBW adolescents are not in line with this. Parent-daughter correlations for internalizing behaviors were high, whereas those for externalizing behaviors were low (Table 3), indicating that such problems and, in particular, aggression, may be more difficult to recognize by parents of VLBW female adolescents compared with parents of normative female adolescents. For VLBW boys, however, correlations were high for internalizing, as well as for externalizing behaviors. Verhulst and van der Ende40 compared parents' reports with adolescents' self-reports of problem behaviors in 883 children ages 11 to 19 years from the general Dutch population using cross-informant syndrome constructs. Discrepancies were found larger for girls than for boys. This is in line with our findings of larger mean differences between parents' and VLBW daughters' syndrome means compared with those between parents and VLBW sons. The correlations on externalizing problems (0.25; Table 3) was far lower in VLBW girls in the present study than in girls from the general adolescent Dutch population (0.53),40 indicating that VLBW girls have externalizing behavioral problems not recognized by parents. The study by Sourander et al41 compares well with that by Verhulst and van der Ende,40 strengthening our assumption that VLBW female adolescents differ from normative female adolescents in this respect. Parent-daughter correlations on internalizing problems, however, were high in all 3 of the studies (VLBW girls, 0.67; Dutch adolescent population, 0.58; Finnish adolescent population, 0.46).

Early Predictors of Emotional/Behavioral Problems in VLBW Adolescents
We used a hierarchical multiple regression analysis (Table 4) introducing risk variables reflecting predetermined conditions (block 1) followed by situations occurring more or less by chance during pregnancy (block 2) and variables describing potential risk occurring in the perinatal and neonatal period (block 3). As reported by parents, the most important independent early variable predicting behavioral problems and competencies in adolescence was SGA. Gender was the most important single variable predicting self-reported total problems and internalizing behavior, whereas year of birth was a significant predictor for self-reported total competence (Table 4). To separate effects of prematurity from those of fetal growth restriction (FGR) may be difficult.51 In the present study we used the term SGA, which was defined as birth weight <10th percentile. Obviously our group of VLBW infants being SGA is a more heterogeneous group than preterms suffering from FGR; however, because effects from common perinatal and neonatal confounders were controlled for, our finding should grant for an independent effect of SGA on parent-reported competence and behavioral problems in the VLBW adolescents. This result also corresponds with a recent study by Tolsa et al,52 showing that placental insufficiency with FGR has specific structural and functional consequences on cerebral cortical brain development. The finding of a significant independent effect of gender on self-reported total problems and internalizing behavior are in accordance with Hack et al,25 and also in line with results displayed in Table 2 in the present study showing that self-reported total problems and internalizing symptoms are generally scored much higher in girls compared with boys.

In our hospital, prenatal maternal steroid treatment for fetal lung maturation was taken into use in the late 1970s, resulting in reduced mortality.53 During the 1980s, the use of prenatal steroids declined because of reports on potential negative adverse effects, and improvements in survival and initial morbidity ceased.53,54 However, neonatal care for those surviving initial lung disease gradually improved, leading to a decline in individuals surviving with severe disability,33 which might explain why year of birth predicts self-reported total competence (Table 4). This interpretation, however, is somewhat speculative, because there was no corresponding effect of year of birth on parent-reported competence.

Control Population
Because of a lack of solid normative studies from Norway, we used Swedish normative values in the present study.34,35 In the study by Heyerdahl et al,38 information on competence was lacking. In other Norwegian studies, there were low response rates,36,37 and some were designed for special purposes, like ethnic influences on occurrence of emotional/behavioral problems in multiethnic areas.36,38 The results of these Norwegian epidemiologic studies can, thus, not be considered very suitable as national normative values, and because the Nordic populations are very similar,34–38,55–57 we considered the Swedish normative values more appropriate.

Responders Versus Nonresponders
The response rate in the present study was rather low, which may have several reasons. Low response rates are quite common in studies relying on mailed requests as in our study, in contrast to school-based sampling of information.34,35 Also, low response rates with increasing child age are reported.34,57 A higher proportion of parents of very preterm infants tend to have lower SES than average. We did not have sufficient information at the time of preterm delivery to explore whether parents' SES may have been different between responders and nonresponders. Low SES in families is associated with low response rates, and high scoring families on CBCL problem scales tend to have low SES and to be underrepresented in mail surveys.34,58 Thus, we can not exclude low SES as one reason for the high attrition rate. The statistically significant difference between responders and nonresponders in mean gestational age (1.4 weeks) is probably of no great clinical importance in our context. A disability rate within the expected range and identical proportions among responders and nonresponders exclude disability as one reason for the high attrition rate.

Limitations, Strengths, and Conclusions
The limitations of this study are shared with other studies that use questionnaires. Certainly, research of this sort would benefit from the use of independent, professional, impartial assessment of children's behavior. A questionnaire approach lacks the specificity and additional depth that more formally structured diagnostic interviews might provide. We are concerned about the high attrition rate, which may have caused differences between responders and nonresponders, which may be unmeasured. Strength of this study is that it is population based, including all of the VLBW infants of mothers residing in the 2 northernmost counties in Norway. To better understand why VLBW children and adolescents develop problems according to parents (Table 2) and why adolescents deny this (Table 2), there is a need for prospective longitudinal studies starting in infancy. More attention should be given to VLBW adolescent girls, who report more emotional and behavioral problems than VLBW adolescent boys (Table 2) and suffer from behavior problems and, in particular, externalizing problems, unrecognized by their adult caregivers (Table 3). In addition, prevention procedures should be tested out within randomized, controlled trials addressing the need to ameliorate the trajectories of problem development.


   ACKNOWLEDGMENTS
 
We thank Bo Larsson and Anders Broberg for providing us with normative data and the participating parents and adolescents for completing questionnaires.


   FOOTNOTES
 
Accepted Mar 31, 2006.

Address correspondence to Lauritz Bredrup Dahl, Child and Adolescent Clinic, University Hospital of North-Norway, N-9038 Tromsø, Norway. E-mail: lauritz.dahl{at}unn.no

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Escobar GO, Littenberg B, Petiti DB. Outcome among surviving very low birthweight infants: a meta-analysis. Arch Dis Child. 1991;66 :204 –211[Abstract/Free Full Text]
  2. Lee K, Kim B, Khoshnood B, Hsieh H, Chen T, Herschel M, Mittendorf R. Outcome of very low birth weight infants in industrialized countries: 1947–1987. Am J Epidemiol. 1995;141 :1188 –1193[Abstract/Free Full Text]
  3. Stevenson DK, Wright LL, Lemons JA, et al. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, January 1993 through December 1994. Am J Obstet Gynecol. 1998;179 :1632 –1639[CrossRef][Web of Science][Medline]
  4. Roth S, Wyatt J, Baudini J, et al. Neurodevelopmental status at 1 year predicts neuropsychiatric outcome at 14–15 years of age in very preterm infants. Early Hum Dev. 2001;65 :81 –89[CrossRef][Web of Science][Medline]
  5. Elgen I, Sommerfelt K, Markestad T. Population based controlled study of behavioral problems and psychiatric disorders in low birthweight children at 11 years of age. Arch Dis Child Fetal Neonatal Ed. 2002;87 :F128 –F132[Abstract/Free Full Text]
  6. Horwood LJ, Mogridge N, Darlow BA. Cognitive, educational, and behavioral outcomes at 7 to 8 years in a national very low birthweight cohort. Arch Dis Child Fetal Neonatal Ed. 1998;79 :F12 –F20[Abstract/Free Full Text]
  7. Taylor HG, Klein N, Minich NM, Hack M. Middle-school-age outcomes in children with very low birthweight. Child Dev. 2000;71 :1495 –1511[CrossRef][Web of Science][Medline]
  8. Saigal S. Follow-up of very low birthweight babies to adolescence. Semin Neonatol. 2000;5 :107 –118[CrossRef][Medline]
  9. Bhutta AT, Cleves MA, Casey PH, Cradock MM, Anand KJS. Cognitive and behavioral outcomes of school-aged children who were born preterm: a meta-analysis. JAMA. 2002;288 :728 –737[Abstract/Free Full Text]
  10. Finnström O, Gäddlin PO, Leijon I, Samuelsson S, Wadsby M. Very-low-birth-weight children at school age: academic achievement, behavior and self-esteem and relation to risk factors. J Matern Fetal Neonatal Med. 2003;14 :75 –84[Medline]
  11. Levy-Shiff R, Einat G, Mogilner MB, Lerman M, Krikler R. Biological and environmental correlates of developmental outcome of prematurity born infants in early adolescence. J Pediatr Psychol. 1994;19 :63 –78[Abstract/Free Full Text]
  12. Schothorst PF, Van Engeland H. Long-term behavioral sequelae of prematurity. J Am Acad Child Adolesc Psychiatry. 1996;35 :175 –183[CrossRef][Web of Science][Medline]
  13. Wolke D. Psychological development of prematurely born children. Arch Dis Child. 1998;78 :567 –570[Free Full Text]
  14. Stewart AL, Rifkin L, Amess PN, et al. Brain structure and neurocognitive and behavioral function in adolescents who were born very preterm. Lancet. 1999;353 :1653 –1657[CrossRef][Web of Science][Medline]
  15. Stevenson CJ, Blackburn P, Pharoah POD. Longitudinal study of behavior disorders in low birthweight infants. Arch Dis Child Fetal Neonatal Ed. 1999;81 :F5 –F9[Abstract/Free Full Text]
  16. Hille ETM, den Ouden AL, Saigal S, et al. Behavioral problems in children who weigh 1000 g or less at birth in four countries. Lancet. 2001;357 :1641 –1643[CrossRef][Web of Science][Medline]
  17. Rickards AL, Kelly EA, Doyle LW, Callanan C. Cognition, academic progress, behavior and self-concept at 14 years of very low birth weight children. J Dev Behav Pediatr. 2001;22 :11 –18[Web of Science][Medline]
  18. Hack M, Flannery DJ, Schluchter M, Cartar L, Borawski E, Klein N. Outcomes in young adulthood for very-low-birth-weight infants. N Engl J Med. 2002;346 :149 –157[Abstract/Free Full Text]
  19. Saigal S, Pinelli J, Hoult L, Marie Kim M, Boyle M. Psychopathology and social competencies of adolescents who were extremely low birth weight. Pediatrics. 2003;111 :969 –975[Abstract/Free Full Text]
  20. Saigal S, Feeny D, Rosenbaum P, Furlong W, Burrows E, Stoskopf B. Self-perceived health status and health-related quality of life of extremely low-birth-weight infants at adolescence. JAMA. 1996;276 :453 –459[Abstract/Free Full Text]
  21. Tideman E, Ley D, Bjerre I, Forslund M. Longitudinal follow-up of children born preterm: somatic and mental health, self-esteem and quality of life at age 19. Early Hum Dev. 2000;61 :97 –110[CrossRef][Web of Science]
  22. Donohue PK. Health-related quality of life of preterm children and their caregivers. Ment Retard Dev Disabil Res Rev. 2002;8 :293 –297[CrossRef][Web of Science][Medline]
  23. Feingold E, Sheir-Neiss G, Melnychuk J, Bachrach S, Paul D. HRQL and severity of brain ultrasound findings in a cohort of adolescents who were born preterm. J Adolesc Health. 2002;31 :234 –239[CrossRef][Web of Science][Medline]
  24. Cooke RWI. Health, lifestyle, and quality of life for young adults born very preterm. Arch Dis Child. 2004;89 :201 –206[Abstract/Free Full Text]
  25. Hack M, Youngstrom EA, Cartar L, et al. Behavioral outcomes and evidence of psychopathology among very low birth weight infants at age 20 years. Pediatrics. 2004;114 :932 –940[Abstract/Free Full Text]
  26. Jobe AH. Predictors of outcomes in preterm infants: which ones and when? J Pediatr. 2001;38 :153 –156
  27. Bongers IL, Koot HM, van der Ende J, Verhulst FC. The normative development of child and adolescent problem behavior. J Abnorm Psychol. 2003;112 :179 –192[CrossRef][Web of Science][Medline]
  28. Crijnen A, Achenbach T, Verhulst FC. Comparisons of problems reported by parents of children in 12 cultures: total problems, externalizing and internalizing. J Am Acad Child Adolesc. 1997;36 :1269 –1277
  29. Cantwell D, Lewinsohn P, Rohde P, Seely JR. Correspondence between adolescent report and parent report of psychiatric diagnostic data. J Am Acad Child Adolesc Psychiatry. 1997;36 :610 –623[CrossRef][Web of Science][Medline]
  30. Achenbach TM. Manual for the Child Behavior Checklist/4-18 and 1991 Profile. Burlington, VT: Department of Psychiatry, University of Vermont; 1991
  31. Achenbach TM. Manual for the Youth Self-Report and 1991 Profile. Burlington, VT: Department of Psychiatry, University of Vermont; 1991
  32. Holt J, Weidle B, Kaaresen PI, Fundingsrud HP, Dahl LB. Very low birthweight infants: outcome in a sub-Arctic population. Acta Pædiatr. 1998;87 :446 –451[CrossRef]
  33. Dahl LB, Kaaresen PI, Fundingsrud HP. Very low birthweight infants in Troms and Finnmark counties 1978–1989 [in Norwegian]. Tidsskr Nor Laegeforen. 1999;119 :180 –185[Medline]
  34. Larsson B, Frisk M. Social competence and emotional/behavior problems in 6–16 year-old Swedish school children. Eur Child Adolesc Psychiatry. 1999;8 :24 –33[CrossRef][Web of Science][Medline]
  35. Broberg AG, Ekeroth K, Gustafsson PA, et al. Self-reported competencies and problems among Swedish adolescents: a normative study of the YSR. Eur Child Adolesc Psychiatry. 2001;10 :186 –193[CrossRef][Web of Science][Medline]
  36. Kvernmo S, Heyerdahl S. Influence of ethnic factors on behavior problems in indigenous Sami and majority Norwegian adolescents. J Am Acad Child Adolesc Psychiatry. 1998;37 :743 –751[CrossRef][Web of Science][Medline]
  37. Nøvik TS. Validity of the Child Behavior Checklist in a Norwegian sample. Eur Child Adolesc Psychiatry. 1999;8 :247 –254[CrossRef][Web of Science][Medline]
  38. Heyerdahl S, Kvernmo S, Wichstrøm L. Self-reported behavioral/emotional problems in Norwegian adolescents from multiethnic areas. Eur Child Adolesc Psychiatry. 2004;13 :64 –72[CrossRef][Web of Science][Medline]
  39. Achenbach TM. Integrative Guide for the 1991 CBCL/4–18, YSR, and TRF Profiles. Burlington, VT: University of Vermont, Department of Psychiatry; 1991
  40. Verhulst FC, van der Ende J. Agreement between parents' reports and adolescents' self-reports on problem behavior. J Child Psychol Psychiatry. 1992;33 :1011 –1023[Web of Science][Medline]
  41. Sourander A, Helstelä L, Helenius H, Parent-adolescent agreement on emotional and behavioral problems. Soc Psychiatry Psychiatr Epidemiol. 1999;34 :657 –663[CrossRef][Web of Science][Medline]
  42. Statistics Sweden. Swedish Socioeconomic Classification 1982. Stockholm, Sweden: SCB; 1982
  43. McCartney K, Rosenthal R. Effect size, practical importance, and social policy for children. Child Dev. 2000;71 :173 –180[CrossRef][Web of Science][Medline]
  44. Indredavik MS, Vik T, Heyerdahl S, Kulseng S, Brubakk A-M. Psychiatric symptoms in low birth weight adolescents, assessed by screening questionnaires. Eur Child Adolesc Psychiatry. 2005;14 :226 –236[CrossRef][Web of Science][Medline]
  45. Botting N, Powls A, Cooke RWL, Marlow N. Attention deficit hyperactivity disorders and other psychiatric outcomes in very low birthweight children at 12 years. J Child Psychol Psychiat. 1997;38 :931 –941[Web of Science][Medline]
  46. Whitfield MF, Grunau RVE, Holsti L. Extremely premature (≤800 g) schoolchildren: multiple areas of hidden disability. Arch Dis Child Fetal Neonatal Ed. 1997;77 :F85 –F90[Abstract/Free Full Text]
  47. Taylor HG, Hack M, Klein N et al. Outcomes of <750 g birth weight in middle childhood [abstract]. J Dev Behav Pediatr. 1998;19 :387
  48. Grunau RE, Whitfield MF, Fay TB. Psychological and academic characteristics of extremely low birth weight (≤800g) adolescents who are free of major impairment compared with term-borne control subjects. Pediatrics. 2004;114 (6). Available at: www.pediatrics.org/cgi/content/full/114/6/e725
  49. Seiffge-Krenke I, Kollmar F. Discrepancies between mothers' and fathers' perceptions of sons' and daughters' problem behavior: a longitudinal analysis of parent-adolescent agreement on internalizing and externalizing problem behavior. J Child Psychol Psychiatry. 1998;39 :687 –697[CrossRef][Web of Science][Medline]
  50. Rey JM, Schrader E, Morris-Yates A. Parent-child agreement on children's behaviors reported by Child Behavior Checklist (CBCL). J Adolesc. 1992;111 :8 –9
  51. Yanney M, Marlow N. Paediatric consequences of fetal growth restriction. Semin Fetal Neonatal Med. 2004;9 :411 –418[CrossRef][Medline]
  52. Tolsa CB, Zimine S, Warfield SK, et al. Early alteration of structural and functional brain development in premature infants born with intrauterine growth restriction. Pediatr Res. 2004;56 :132 –138[CrossRef][Web of Science][Medline]
  53. Dahl LB, Berge LN, Dramsdahl H, et al. Antenatal, neonatal and post neonatal deaths evaluated by medical audit. A population-based study in northern Norway - 1976 to 1997. Acta Obstet Gynecol Scand. 2000;79 :1075 –1082[CrossRef][Web of Science][Medline]
  54. Kaaresen PI, Dølen G, Fundingsrud HP, Dahl LB. The use of CRIB (clinical risk index for babies) score in auditing the performance of one neonatal intensive care unit. Acta Paediatr. 1998;87 :195 –200[CrossRef][Web of Science][Medline]
  55. Hannesdottir H, Einarsdottir S. The Icelandic child mental health study. An epidemiological study of Icelandic children 2–18 years of age using the child behavior checklist as a screening instrument. Eur Child Adolesc Psychiatry. 1995;4 :237 –248[CrossRef][Web of Science][Medline]
  56. Bilenberg N. The Child Behavior Checklist (CBCL) and related meterial: standardization and validation in Danish population based and clinically based samples. Acta Psychiatr Scand. 1999;100 :2 –52
  57. Helstelä L, Sourander A. Self-reported competency and emotional behavioral problems in a sample of Finnish adolescents. Nord J Psychiatry. 2001;55 :381 –385[CrossRef][Web of Science][Medline]
  58. Verhulst FC, Koot HM, Berden GF. Four-year follow-up of an epidemiological sample. J Am Acad Child Adolesc Psychiatry. 1990;29 :440 –448[Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Facebook Facebook   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?


This article has been cited by other articles:


Home page
 NeoReviewsHome page
E. E. Rogers and R. E. Piecuch
Neurodevelopmental Outcomes of Infants Who Experience Intrauterine Growth Restriction
NeoReviews, March 1, 2009; 10(3): e100 - e112.
[Abstract] [Full Text] [PDF]

Home page
 Arch Gen PsychiatryHome page
K. Raikkonen, A.-K. Pesonen, K. Heinonen, E. Kajantie, P. Hovi, A.-L. Jarvenpaa, J. G. Eriksson, and S. Andersson
Depression in Young Adults With Very Low Birth Weight: The Helsinki Study of Very Low-Birth-Weight Adults
Arch Gen Psychiatry, March 1, 2008; 65(3): 290 - 296.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dahl, L. B.
Right arrow Articles by Rønning, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dahl, L. B.
Right arrow Articles by Rønning, J. A.
Related Collections
Right arrow Premature & Newborn
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?