Poster group
Details of individual items:
poster
Purpose: The recent improvement in survival of very low birth weightinfants (< 1000 grams) has raised questions about long term prognosis specificto this population. The present study reports scores at 6 mos., 18 mos.and 30 mos. on the Mental (MDI) and Motor (PDI) scales of the BayleyScale of Infant Development II for a population of 85 very low birthweight infants (< 1000 grams) participating in an Early InterventionProgram (EIP) at the University of Miami School of Medicine. Infantsidentified in the NICU came for repeated follow up visits which includedpsychometric, medical, auditory, social, and speech evaluations. Infantsrequiring intervention were referred to appropriate resources whichincluded developmental intervention. The EIP clinic serves a primarilyMedicaid population and includes a large number of ethnic minorities.Compliance in the follow up program is 92%. The study addressed thestability of Bayley II scores over time for this population as a drop inscores at 18-24 months has been previously reported.Procedure: Medical data included sex, birthweight, gestational age,presence of prenatal care, apgar score <5, use of drugs inresuscitation, the presence of IVH, grade of IVH, diagnosis ofleukomalacia, necrotising enterocolitis, days on the ventilator, days onoxygen and number of days to discharge. Repeated measures analysis ofvariance was conducted on MDI and PDI scores for the total populationand interaction effects were investigated for birth weight category (<750 grams compared to 750-1000 grams) and the sociodemographicvariables of maternal education, family income, marital status andethnic background (expressed as white versus nonwhite).Results: MDI scores dropped over time with the significant dropoccurring between the first(6 mos) and second (18 mos) visit in the total population ( p. <.000)and then appeared to stabilize between the18 and 30 mos.visits. PDIscores did not show this pattern. There were significant differencesbetween the MDI scores of the extremely low birth weight (< 750 grams)and very low birth weight (750-1000 grams) infants on scores at allages.(p. < .04) but no interaction effect was evident as the scores ofboth groups dropped similarly over time( p. <.000). There were no significant relationships between any of themedical variables, prenatal care, or gender in the drop in scoresobserved over time. However, there were significant interaction effectsfor maternal education and ethnic background. Marital status and incomecategory followed a similar pattern but differences were notstatistically significant.Discussion: These data suggest that sociodemographic variables arerelated to MDI scores in VLBW infants from a Medicaid population, and donot appear to have the same relationship to motor skills as there was noparallel drop for PDI scores. Medical variables did not show arelationship between the pattern of scores over time. Howeverbirthweight data suggest that infants < 750 grams are more likely tohave lower MDI scores on their initial visit than those between 750 and1000 grams. Discussion of these results includes the possibility thatthese data are affected by the fact that infants are participating in afollow up program which provides opportunities for intervention. Inaddition, the appropriateness of the Bayley II for this population willbe discussed as children from impoverished backgrounds appear to scoreconsistently below expectations at 18 months of age in contrast to theBayley norms for this age group.
poster
This report presents the results of a meta-analysis comparingthe developmental outcomes of extremely low birthweight(<1000g; ELBW) and moderately low birthweight (1501-2500g;ModLBW) samples assessed at preschool age (3 years) andabove. Information about selected outcomes (prevalence ofcerebral palsy (CP), cognitive functioning, visual-motorperformance) of 37 ELBW cohorts (2,406 subjects) and 14ModLBW cohorts (12,123) are presented from the 33 studiesincluded in this review. Meta-analytic techniques (i.e.,pooled estimates of the overall prevalence rate of CP and ofthe overall mean IQ and visual-motor scores) were conductedto aggregate outcomes among birthweight groups. Secondaryanalyses were also conducted to examine the relativecontribution of associated demographic factors (e.g.,geographic origin, increasingly ELBW) as well as social riskto low birthweight outcome. Results revealed an overallprevalence rate (OPR) of cerebral palsy for ELBWs of 9.1%(95% CI, 7.9-10.3) compared to 1.3% (95% CI, 0.8-1.7) amongModLBWs (p<.001). There was a slight, but non-significantdifference in the prevalence of CP among ELBWs of NorthAmerican vs. non-North American origin (10.3% vs. 8.5%,respectively). There was also no difference in theprevalence of CP between ELBW cohorts of high and low socialrisk (9.1% vs. 10.5%, respectively). In the area ofcognitive functioning, the mean IQ for ELBW cohorts (at 3years and older) was 93.6 (95% CI, 86.0-101.1), significantlylower than that for those of ModLBW (100.3; 95% CI, 96.5-104.1). There were no significant differences by geographicorigin or between cohorts of high and low social risk (93.9vs. 94.0, respectively). The pooled estimate of visual-motorfunctioning for ELBW cohorts was 6.8 (95% CI, 4.0-9.6),significantly lower than that for ModLBW cohorts (12.1; 95%CI, 10.3-13.8). Results reveal significant differences inthe developmental outcomes of ELBW and higher LBW samples,particularly in the area of neurosensory impairment (i.e.,rate of CP) and more subtle neurological deficits.
poster
There is much interest both in the literature and clinically as to whetherthe poorer outcomes reported for this population mean that mostvery-low-birthweight (VLBW) children are at some sort of mild disadvantagecompared to peers, or whether a few of these children have multipleimpairments. Data from this study has already been published separatelyregarding the cognitive/educational and the psychiatric outcomes of a groupof 136 VLBW children aged 12 years compared to peers (Botting, Powls, Cookeand Marlow 1997; Botting, Powls, Cooke and Marlow, 1998). However, groupdata of this sort makes it difficult to judge the distribution of impairmentand related factors. Thus, the present investigation also attempted toexamine this using within-group analysis.Using a cluster analysis technique, it was possible to determine fourdistinct groups of children experiencing a different profile of impairment.Full scale IQ, parent-rated depression questionnaire score, parent-ratedanxiety questionnaire score, parent-rated behaviour and activities checklistscore and parent-rated Rutter behavioural questionnaire scores were used asdependent variables for the VLBW group. The four groups emerging showed thecombinations which are summarised below:Cluster 1: (n13) 'Multiple impairments'Low IQ scores and high (less favourable) scores on psychologicalquestionnaires.Cluster 2: (n 38) 'Normally developing'Normal IQ score and low (normal) questionnaire scores.Cluster 3: (n54) 'Cognitively impairments'Low/normal IQ scores and low (normal) psychological questionnaire scores.Cluster 4: (n23) 'Psychologically impairments'Normal IQ scores and high psychological questionnaires.Each group was significantly different from each other on all measures atp<0.0001. Importantly for the validity of these clusters very similar groupsemerged using child questionnaire data. In addition, a number of childrenfall into each cluster group.It appears therefore that impairment is fairly widely distributed, witharound one quarter of the children developing normally in both psychologicaland cognitive areas. Although the remaining three quarters experience someunfavourable outcome from being VLBW, only a quarter of the total group haveimpairments in both areas of outcome. Those with multiple impairments onquestionnaire based data were also more likely to meet clinical criteria forattention deficit hyperactivity disorders, have poorer reading comprehensionscores and maths scores (all p<0.001) and to have more motor impairment(p<0.05).Interesting demographic and family factors emerged showing the most severelyaffected children to have smaller houses (p<0.05) which were more likely tobe council accommodation (p<0.05) and to have experienced poorer parentingskills (as defined by a clinical interview index; p<0.01). This group didnot, however, show lower family incomes or poorer parental education levels.The distribution of impairment has implications for identifying those mostin need of preventative strategies and remediation within this clinicalpopulation.
poster
Many studies have compared the interactions of mothers and their VLBW infant with those of mothers and their normal birthweight infants, and found that the interactions of the VLBW dyads are qualitatively different than those of NBW dyads. Very little is known about what variables are predictive of the quality of mother's interactions with their VLBW infants. Empirical evidence provides increasing evidence that the quality of parent-child relationships is multiply determined. However, those few studies that have examined predictors of mother-VLBW infant interaction have assessed only one or two predictors. Studies are needed that examine multiple predictors of mother-VLBW infant interaction.The purpose of this prospective study was to examine the combined influence of a set of infant, parent, and contextual variables on the quality of mother's interactions with their VLBW infant when the infant was 9 months of age (corrected). Belsky's (1984) model of the determinants of parenting was used to guide the selection of the predictors considered in this study. The model proposes that parenting is determined by a myriad of factors emanating from within the child, parent and the social context in which the parent and child are situated. The infant variables examined were birthweight and severity of illness (Neurobiological Risk Score; Brazy et al., 1991), while the parent variables considered were state anxiety (State/Trait Anxiety Inventory; Spielberger et al., 1983), and sense of competence in the parenting role (Parenting Sense of Competence Scale; Gibaud-Wallston and Wandersman, 1979). The contextual variables examined were maternal received support (Support in Parenting Questionnaire) and maternal dyadic adjustment (the Dyadic Adjustment Scale; Spanier, 1976). The participants were 74 VLBW infants and their mothers. Dyads were included if the infant weighed < 1500 grams at birth and was born at < 32 weeks gestation. Infant, mother and contextual predictors were assessed when the infants were 3 months of age. Mother-infant interaction was assessed 6 months later when the infant was 9 months old. At this time, mother-infant teaching interactions were videotaped in the family's home, and later coded using the Nursing Child Assessment Teaching Scale (Barnard & Eyers, 1979).Data collection and coding have been completed for 58 of the 74 dyads, and analyses are now underway. Hierarchical multiple regression analysis will be used to examine the relationship between the dependent variable (quality of mother-infant interaction) and the set of infant, parent and contextual variables. Additional regression analyses will also be conducted to test selected propositions of Belsky's model of parenting. For example, do the mother variables (e.g., anxiety and sense of competence) contribute more to the prediction of the quality of mother-infant interaction than the contextual variables (e.g., received support and marital adjustment)? The results of these analyses will be presented at the ICIS meeting.
poster
Objective: Parents with very small preterm infants (<1500g) were offered a psychotherapeutic intervention programme. After participation in the programme, parents were asked to evaluate the intervention, the results of which are presented here.Method: 88 mothers and their preterm infants were recruited into the study. After randomisation fourty-four received a programme of psychotherapeutic intervention and fourty-four did not. The intervention consisted of four components: individual psychotherapy, a sensitivity training by video feedback, group psychotherapy and a home visit. The first two components focused on attachment, the latter two on coping. The Intervention Component Rating Scale (ICRS) was developed to measure the parental evaluation of the intervention. The ICRS is comprised of 16 statements. In this rating scale parents are able to assess any positive or negative impact they felt the four components of the intervention may have had. Results: Mothers rate the individual attachment focused psychotherapy as most helpful, whereas fathers assess the coping oriented parent group to be the most supportive. Overall the intervention programme is rated as more helpful by fathers as compared to mothers, however this difference is not significant. Both mothers and fathers find the intervention to be most helpful in the area of 'emotional support at the time of crisis' and 'confidence in the future development of the child'. The intervention seemed to have the least impact on 'relationship to the partner' as assessed by both parents.The children's biological risk also plays a part in the parental assessment of the intervention. Children's gestational age is associated with the mother's assessment of the intervention. Mothers whose children are born later rate the individual psychotherapy and the home visits as more helpful than parents whose children are born earlier. There is no significant association between father's assessment of the intervention and the children's gestational age.Discussion: Reasons for mothers' and fathers' preference of individual psychotherapy and the parent-group in favour of the sensitivity training and a home visit will be discussed. An evaluation of the results will take the timing of the intervention, the number of contacts and the biological risk of the child into account.
poster
Objective: There are contradictory results regarding the distribution of the attachment quality of premature infants in comparison to samples of full-term infants. Some studies report more insecure children in the preterm sample, whereas others do not find this difference. It has been suggested that children with a high neurobiological risk are especially likely to show insecure attachment. Many studies do not take the children's birthweight or their neurobiological risks into account. The aim of the ULM STUDY was to investigate the development of the attachment quality in a sample of high risk very low birthweight preterms (<1500g) with respect to maternal attachment representation, anxiety and coping as well as the children's neurobiological risk factors.Method: 80 mothers and their high risk preterm infants were recruited into the study. The quality of attachment of the infants (Strange Situation, Ainsworth et al., 1969) was analysed at 14 mths (corrected gestational age). The classification for disorganisation was included. The Adult Attachment Interview (George et al., 1985) was conducted to assess the maternal representations of attachment including the classification for unresolved trauma/loss.Results: The distribution of the preterm children's quality of attachment and maternal representation of attachment will be reported. The influence of the mothers' anxiety and coping at this time of crisis as well as the children's neurobiological risk factors will be taken into account when investigating the associations.Discussion: It will be discussed how children from a high risk sample develop emotionally, taking their own biological risk factors as well as the psycho-emotional risks of their mothers into account.References:Ainsworth, M. et al. (1969) Attachment and the exploratory behavior of one-year-olds in a strange situation. In: Foss, B.M. (ed.) Determinants of Infant Behavior, Basic Books, New York, 113-136.George, C. et al. (1985) The Adult Attachment Interview. Unpublished Manuscript. Dept. of Psychology, University of California, Berkeley.
poster
Objective: In the ULM STUDY we started a preventive psychotherapeutic intervention programme for parents with very small preterm infants (<1500g). The aim of the intervention program is to promote adequate coping strategies in the parents, support a positive parent-child relationship and be manifest in the cognitive and somatic development of the child.Method: 88 mothers and their high risk preterm infants (M3D979,6g, range:450-1490g) were recruited into a two year follow-up study. After randomisation fourty-four received a programme of psychotherapeutic intervention and fourty-four did not. The intervention consisted of focal attachment oriented individual psychotherapy, supportive group psychotherapy, a home visit and a sensitivity training by video feed back. The main focus of the individual psychotherapeutic intervention was on maternal attachment experiences of former loss, separation and on supportive bonding. To assess maternal representations of attachment the adult attachment interview was conducted half a year after the preterm birth. At fourteen months (corrected age) the children were assessed in the strange situation test to determine the quality of attachment to their mother.Results: The distribution of the children's quality of attachment whether secure, insecure avoidant or insecure ambivalent in the intervention group as well as the control group will be presented. How maternal representation of attachment classified as secure, insecure enmeshed, or insecure dismissing is related to the children's quality of attachment will be explored, both in the intervention and the control group. Ratings for disorganisation in the children(D) and unresolved maternal loss (U) will also be presented. The influence of the children's neurobiological risk factors as well as parents' sociodemographic data will be taken into account in the analyses to be conducted.Discussion:The role the psychotherapeutic intervention had on any differences found between the intervention and the control group will be explored. Special emphasis will be placed on the neourobiological deficits of the premature children and the influence this may have on a developing relationship between an infant and his parents.ReferenceBrisch, K.H., Buchheim, A., K88hntop, B., Kunzke, D., SchmB8cker, G., K89chele, H. & Pohlandt, F. (1996): Early preventive psychotherapeutic intervention program for parents after the delivery of a very small premature infant: The Ulm Study. Infant Behav Dev 19: (special ICIS issue) 356.
poster
no abstract
poster
The aim of the study is to describe the outcome of very low birth weight infants who were born in our Institute in the year 1997.Forty one infants were followed. Their mean birth weight was 1243.9 +/-215.1 g, the mean gestational age 31.3 +/-2.4 weeks, the mean length of hospitalization 56.3 +/-17.4 days, and the mean length of mechanical ventilation 5.7 +/-8.2 days.The infants were examined repeatedly at 3, 6, 9 and 12 months of life. For the assessments of psychological development, Bayley Scales of Infant Development (Czech Edition 1983) were administered. Neurological examinations were carried out according to Vlach (1978).On the basis of these assessments, developmental status was categorized as: normal (category I), mild developmental retardation (category II), moderate developmental retardation mostly also with abnormal neurological status (category III), serious psychological and neurological impairment (category IV).At the age of 12 months, developmental status of 33 infants (80.5%) could be ranked into I or II categories. Five infants (12.2%) belonged to the category III, and three infants (7.3%) were seriously psychologically and neurologically impaired (category IV).The developmental status at 3 months did not predict the outcome at 12 months. Especially BayleyB4s Mental Developmental Indexes (MDI) at this age had very poor predictive value. This may indicate the importance of follow up all very low birth weight infants in later months.In our very low birth weight infants, neither birth weight, gestational age, length of hospitalization nor length of mechanical ventilation influenced significantly the developmental status at 12 months.More problems in the neuromotor than in the mental development were observed in our infants: the mean value of BayleyB4s Psychomotor Developmental Index (PDI) at 12 months was lower than the mean value of MDI [PDI 99 (range 111-86) : MDI 112 (range 134-96)] for category I, PDI 79 (range 93-71) : MDI 102 (range 125-86) for category II, and PDI 55 (range 63-50) : MDI 80 (range 112-58) for category III. Performances of 3 infants in category IV were very poor and could not be classified according to the BayleyB4s norms.Our results suggest that in majority of our very low birth weight infants (in 80.5%), normal developmental status or only mild developmental retardation was ascertained. The assessments of mental development at 3 months had very low predictive validity. We suppose that neuromotor and behavioral assessments are very important in developmental follow up.
poster
Infants born at very low birth rate (VLBW) are at risk for subtle to moderate handicaps that relate to the organization of arousal. General behavioral destabilization, including disorganized arousal, is known to occur in response to caregiving routines in the neonatal intensive care unit. An individualized approach to handling designed to reduce destabilization has been shown to moderate arousal during the caregiving episode as a whole. However, caregiving involves activities that vary in intensity of stimulation, thus comprise differing contexts for infant response. The purpose of this study was to determine whether arousal responses to the individualized handling intervention as reflected in the distribution of behavioral states varied by level of stimulus intensity and to explore the pattern of response across age. Behavioral state responses were examined in 38 infants (< 1700 gms, 53% male, 85% white) observed at 28, 32, and 36 weeks post conceptional age. Subjects served as their own controls. At each age interval, infants were assessed during 4 caregiving sessions, alternating the use of stress-reducing handling techniques (e.g., providing positional support and aides for self-regulation) (individualized condition) and no use of these techniques (traditional condition). The intervention was administered in an ABBA design, order assigned randomly, and administered according to detailed protocols by research nurses. Caregiver behaviors and infant behavioral state were coded from direct observation and videotapes. Behavioral states included alert, waking activity, drowse, sleep-wake transition, active sleep, quiet sleep, and fuss/cry. Level of caregiving stimulation was categorized as light (nasogastric or bottle feeding), moderate (e.g., vital signs, oral care), and intense (e.g., diapering, suctioning). Data were analyzed with a repeated measures crossover design with mixed modeling. Results show that at 28 weeks, there was less waking activity, drowse, and fuss/cry and more active sleep in the individualized than traditional condition during moderate and intense stimulation. The amount of alert was minimal; differences were not significant for transition. At 32 weeks, infants spent less time in alert, waking activity, and fuss/cry, and more in active sleep at each stimulus level in the individualized condition than in the traditional condition. There was less drowse during light stimulation, but more during moderate and intense stimulation, and more transitional state during intense stimulation. At 36 weeks, there was less alert and more active sleep during light and moderate stimulation, more drowse during moderate and intense stimulation, and less waking activity at all stimulus levels in the individualized than in the traditional condition; levels of fuss/cry did not differ. Quiet sleep overall occurred rarely. Across age, differential change by handling condition was most evident for moderate and intense stimulation.Results demonstrate that VLBW infants benefit from support for behavioral state organization at all stimulus levels. Effects were greatest for moderate and intense stimulation contexts, but arousal also was reduced during minimally intrusive caregiving interactions. The findings underscore the vulnerability of VLBW infants to the repetitive handling required for routine caregiving, and suggest further research on whether intervention to moderate arousal may reduce the incidence of later regulatory disorders.
poster
Research has shown that the birth of a VLBW infant evokes considerableparental distress. Even after the the infant leaves the hospital parentsmay continue to experience distress as a result of the challenges thataccompany caring for a biologically fragile infant. Unfortunately, thecurrent literature yields meager conclusions about the psychologicaladjustment of mothers of VLBW infants as few studies have included controlgroups. This study sought to fill this research gap by comparing theparenting stress experienced by mothers of VLBW infants and full terminfants in the second half of the infants' first year of life. The primarygoal of this study was to examine the moderating effects of child-rearingattitudes on the relation between parenting stress and infant behavioralcharacteristics for mothers of VLBW and full term infants. Fifty-six 9-month-old infants (23 VLBW and 33 full term) and their motherswere the participants. Mothers completed measures of parenting stress,child-rearing attitudes, infant temperament, infant behavioral problems,and were interviewed at home about their child's health vulnerability. Analyses of covariance (ANCOVAs) with SES and marital status as thecovariates were completed to test for group differences in infant andmaternal characteristics. ANCOVAs showed that the VLBW group had asignificantly higher total behavior problem score, F (1, 52) 3.28, p <.05, and a significantly higher moodiness score, F (1, 52) 2.96, p < .05,than the full term infant group. No significant group differences werefound for infant temperament characteristics. The ANCOVAs also revealedthat while the infants did not differ significantly in their currentfrequency of illness, the mothers of the VLBW infants worried significantlymore about their infant's health, F (1, 52) 3.41, p < .05, and theyreported changing their plans more often due to infant health concerns, F(1, 52) 3.38, p < .05, than the mothers of the full term infants. Nosignificant differences were found in parenting stress or in parentingattitudes. Multiple regression procedures were used to examine the extent to whichparental strictness moderated the relations between both infanttemperamental fear and distress, and parenting stress for the VLBW and fullterm infants. The analysis regressing parenting stress on infant distresswas statistically significant (R .67, R2 .44, F (9,46) 4.07, p <.01), and the 3-way interaction term regression coefficient was alsostatistically significant (B -1.986, t -2.79, p < .01). These findingswere then investigated more closely by examining the relation betweenparenting stress and infant distress as moderated by parental strictnesswithin each infant group. The slopes of parenting stress on infantdistress at three levels of parental strictness for the VLBW and full termgroups were examined along with their corresponding t ratios. The datarevealed that for the VLBW infants, the slope of parenting stress on infantdistress was statistically significant when parental strictness was mediumand high. The mothers of VLBW infants who reported using a stricterparenting approach, whose infants' where higher in temperamental distress,experienced greater parenting stress than mothers whose infants were lowerin temperamental distress. The relation between parenting stress andinfant distress did not vary significantly at low levels of parentalstrictness. In contrast to the findings for the VLBW group, for the fullterm group, only a statistical trend was found for the slope of parentingstress on infant distress when parental strictness was low. The results of this study revealed that mothers of VLBW infants hadgreater health concerns about their infants than mothers of full terminfants, but few significant group differences in infants' behavioralcharacteristics were detected. Although the mothers did not differ in theabsolute amount of parenting stress they experienced, infantcharacteristics and parenting attitudes contributed to parenting stress.The degree to which the mothers of the VLBW infants experienced parentingstress was a function of the congruence between their infant's behavioralcharacteristics and their own child-rearing attitudes.
poster
One of the developmental tasks of children is to learn to interact withothers socially. For infants much of this learning first takes placethrough play interactions with their mothers. Research has demonstrateddifferences in the play characteristics of preterm infant-mother dyads andfull-term infant-mother dyads. With the increased survival of high-riskVLBW infants, studies now have greater opportunity examine playinteractions of VLBW infant and their mothers. The purposes of this studywere: (1) To compare the play behaviors of VLBW and full term infant-motherdyads at nine months of age, and (2) To examine the contributions ofparenting stress and child temperament to the quality of play behaviors. Study participants included 56 9-month old infants (23 VLBW <1500 gm and33 full term normal birth weight >2500 gm) and their mothers. During ahome visit infants completed a behavioral temperament assessment procedure,and mothers and their infants participated in a 10-minute free playsession. The temperament and play procedures were videotaped and latercoded. Mothers also completed questionnaire measures of infantstemperament and parenting stress. From the play interactions thefrequencies of specific mother and infant play behaviors were recorded.Independent coders also rated the quality of mothers' and infants' play on9-point ratings scales which were then summed to form Total Play Qualitysummary scores. Analyses revealed several group differences in play behaviors. The VLBWinfants were rated as having a poorer Total Play Quality than the full-terminfants (t(54) -2.55, p < .05), but the mothers did not differsignificantly in Total Play Quality. Mothers of full-term infants askedmore questions (t(51) -2.47, p <.05) and presented a greater number oftoys to their infants (t(53) -2.11, p <.05), compared to mothers of VLBWinfants.Correlational analyses were completed separately for each group to examinethe relations between both infant temperament and parenting stress, andTotal Play Quality of the mothers and infants. Within the full term group,the mothers of infants who were described as higher in IBQ activity levelhad a higher Total Play Quality (r .40, p <.05). For the VLBW infants,temperament was unrelated to mothers' play. Infant temperament was relatedto infant play quality in both groups. For the VLBW group, higher infantTotal Play Quality was associated with mothers' reports of greater infantduration of orienting (r .42, p <.05), and threshold attention (r.51,p<.05). For the full term infants, Total Play Quality was positivelycorrelated with the behavioral temperament measure of activity (r .40, p<.05). Additional analyses showed that for the VLBW infants biologicalfactors were significantly related to mothers' Total Play Quality. Themothers of infants with higher newborn risk scores (r -.57, p <.01) andlower gestational age (r .46, p<.05) showed poorer Total Play Quality. Distinct patterns of relations between parenting stress indicators andplay characteristics were also detected. Among mothers of full-terminfants, those who reported greater parent-child dysfunction (r-.47,p<.01), described their child as more difficult (r-.39, p<.05), and hadgreater total parenting stress scores (r-.40, p<.05), demonstrated poorerTotal Quality of Play. Surprisingly, for the mothers of VLBW infants,parenting stress indicators were not significantly related to their playquality. The findings of this study demonstrated several differences in the playcharacteristics of VLBW and full-term infants, and supported the idea thatVLBW infants are not as adept at social interaction as full term infants. Temperament and indicators of parenting stress were associated with playcharacteristics but distinct patterns of association were noted for eachinfant group. The results will be discussed with respect to the uniquecaregiving challenges that face parents of VLBW infants and theirimplications of social-emotional development.
poster
Advances in neonatal care have resulted in increasedsurvival rates for very low birth weight (VLBW) infants. Acommon sequalae associated with VLBW is bronchopulmonarydysplasia (BPD). The presence of BPD may increase the incidenceof feeding difficulties, irritability, and caretaker burdenparticularly in the absence of an adequate support network. Thepurpose of this study was to compare VLBW infants with andwithout BPD at 12 months of age adjusted for prematurity todetermine if BPD contributed to the mother's perception of infanttemperament and her feelings of attachment and competence. Therole of social support was also examined. The sample was drawn from a larger study of 354 infantsdischarged from a level III Neonatal Intensive Care Unit (NICU)in the southeastern United States who were being seen in adevelopmental follow-up clinic. All infants less than 30 weeksgestation were used for this analysis. The sample consisted of110 VLBW infants. Fifty-six of the infants had BPD and 54 didnot have BPD. The mean birth weight of all infants was 933 grams(range 550-1580), the mean gestational age was 27.5 weeks (range24-30 weeks), and the mean duration of hospitalization was 68.1days (range 10-255 days). There were significantly more males inthe BPD group. The BPD group also had significantly more infantswith birth weight < 1,000 grams, lower gestational age, longerperiods of mechanical ventilation, and longer hospitalization.There were no differences between the BPD group and the non-BPDgroup in race, maternal age, maternal educationl, marital status,or income of the mother. Infant temperament was measured using the Bates InfantCharacteristics Questionnaire (Bates, 1984) administered in thehospital and at the 12 month follow-up visit. The level ofsocial support was measured using the Family Support Scale(Dunst, Jenkins, & Trivette, 1994) administered in the hospitaland at 12 month follow-up. Maternal perception of attachmentand competence were measured using subscales of the ParentingStress Index (Abidin, 1990). There was no significant difference in perceived temperamentor social support between the mothers of the BPD and non-BPDinfants. Mothers in both groups rated their infants assignificantly more fussy at 12 months than before discharge(p3D.001). Both groups experienced a significant decline insocial support by 12 months when compared to the support levelthey experienced in the hospital (p3D.000). There was nosignificant difference in levels of maternal attachment orcompetence between the BPD and the non-BPD mothers. Maternalattachment was closely related to maternal competence. Stepwiseregression was used to examine the relative contribution ofattachment, perception of temperament, social support, length ofstay, birthweight and gestational age on feelings of maternalcompetence at 12 months of age. The best predictive modelincluded attachment and the mother's perception of infanttemperament at 12 months of age. This study suggests that thereare more similarities than difference between mothers of VLBWinfants with and without BPD.