Journal of Pediatric Psychology, Vol. 27, No. 2, 2002, pp. 177-189
© 2002 Society of Pediatric Psychology
Observed and Perceived Dyadic and Systemic Functioning in Families of Preadolescents With Spina Bifida
Loyola University of Chicago
All correspondence should be sent to Grayson N. Holmbeck, Loyola University of Chicago, Department of Psychology, 6525 N. Sheridan Road, Chicago, Illinois 60626. E-mail: gholmbe{at}luc.edu .
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective: To examine dyadic and systemic family functioning across several domains (conflict, cohesion, and stress) in families of preadolescents with spina bifida in comparison to families of able-bodied preadolescents (8- and 9-year olds; n = 68 in each sample).
Methods: Mother-, father-, and child-reported questionnaire data and observational ratings of family behavior were employed.
Results: Findings revealed significant group and socioeconomic status (SES) differences, particularly for the observational family data. Compared to families of able-bodied children, families in the spina bifida sample were less cohesive and children from this sample were more passive during family interaction tasks. Additional analyses suggested that some of these significant associations between group status and family functioning were mediated by verbal IQ, indicating that a significant portion (42%-55%) of the overall group effect was due to variations in child cognitive functioning. Lower SES families demonstrated higher levels of observed mother-child conflict, less observed and perceived family cohesion, and more life events. Lower SES families from the spina bifida sample appear to be particularly at risk for lower levels of family cohesion.
Conclusions: Findings for the spina bifida sample support a
resilience-disruption view (Costigan,
Floyd, Harter, & McClintock, 1997
) of systemic functioning in
families of children with pediatric conditions.
Key words: spina bifida; physical disability; family; family system; SES; adolescence; conflict; cohesion; stress.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Several scholars have maintained that the impact of pediatric chronic illnesses on children's psychosocial functioning is best examined within the family context (Chaney et al., 1997
; Drotar,
1997; Kazak, Segal-Andrews,
& Johnson, 1995; Wallander
& Varni, 1998). Because most families with chronically ill
children appear to be healthy families who happen to be faced with difficult
circumstances (Kazak, 1997;
Kazak et al., 1995;
Quittner & DiGirolamo,
1998; Spaulding & Morgan,
1986), some have sought to examine how the presence of a child
with a pediatric illness may alter an otherwise healthy family system
(Kazak, 1997). When viewed
from a family systems perspective, a variable that is found to be maladaptive
in families with typically developing children (e.g., a highly structured
family environment, parental overprotectiveness) may be adaptive, at least
over the short term, in families of chronically ill children
(Anderson & Coyne, 1993;
Kazak et al., 1995;
Seiffge-Krenke, 1998).
Whereas the importance of investigating family functioning in pediatric
populations is well recognized (Kazak et
al., 1995; Wallander &
Varni, 1998), advances in this research area have been hampered
because most studies (Drotar,
1997; Kazak, 1986)
(1) have not examined family-level (e.g., dyadic, triadic, or systemic)
constructs (Wallander & Varni,
1998); (2) have not included matched comparison samples; (3) are
based only on information collected from a single source (usually mothers;
Holmbeck, Li, Schurman, Friedman, & Coakley, in press); and (4) have used
only a single method (usually questionnaires; Holmbeck et al., 2000).
Given past work, it appears that at least two types of family research
would be particularly informative. First, we need more process-level
information regarding dyadic- and system-level family variables that differ
between families of children with and without pediatric conditions. Such
information would increase our understanding of how families adapt to and are
transformed by the presence of children with these conditions
(Costigan et al., 1997).
Second, and based on results of the first type of research, we can examine
links between the most salient family-level process variables and medical and
nonmedical child adjustment outcomes. In this study, we focus on the first
type of research. Specifically, the purpose of this investigation was to
examine dyadic and systemic functioning in families with preadolescents who
have a physical disability (i.e., spina bifida) in comparison to families with
able-bodied preadolescents across a variety of family-level dimensions with
multisource and multimethod data.
Caring for a child with spina bifida frequently places long-term physical,
psychological, and financial strains on the family system
(Carr, 1991). Parents of
children with disabilities typically experience more stress than parents of
healthy children (Holmbeck et al.,
1997; Kazak et al.,
1995). Although several studies of families in the pediatric
literature have found no differences between pediatric families and comparison
families on familylevel variables (e.g.,
Ammerman, Van Hasselt, & Hersen,
1991; see Kazak et al.,
1995, for a review), one study found that roughly 15% of families
of children with spina bifida fell within the clinically problematic range on
a general measure of family functioning, and 25% demonstrated difficulties
with "the allocation and maintenance of defined roles and
responsibilities in the family"
(Ammerman et al., 1998, p. 457;
also see Loomis, Javornisky, Monahan,
Burke, & Lindsay, 1997). Differences across studies may be a
function of the type of data used in making comparisons across samples (e.g.,
normative data vs. data from matched controls). In this study, the family
functioning of a demographically matched sample of able-bodied adolescents was
compared to the functioning of families of children with spina bifida.
The selection of dyadic and systemic family functioning constructs for this
study was based on recent work that has identified conflict and
cohesion as central family process dimensions
(Cox & Brooks-Gunn, 1999;
Holmbeck, 1996). In addition,
stress within the family system has received considerable attention
in the pediatric literature (Quittner
& DiGirolamo, 1998;
Wallander & Varni, 1995).
All variables included in this study were dyadic or systemic (i.e., variables
that indexed the behavior of an individual in isolation were not
selected).
Families of children with spina bifida were expected to exhibit higher
levels of stress, given theoretical work that identifies childhood illness as
a factor that may increase the overall stress within the family system
(McCubbin & Patterson,
1982) and past findings that suggest that parents of children with
chronic illnesses exhibit more distress
(Holmbeck et al., 1997;
Thompson & Gustafson,
1996). With respect to conflict and cohesion, three perspectives
on families of children with disabilities yield different predictions; we
sought to examine the differential validity of each of these perspectives. The
first perspective is based on findings that children with physical
disabilities rely more on parents for help with tasks of daily living than
ablebodied children do (Murch & Cohen,
1989). It follows, then, that members of families that include
children with physical disabilities may be drawn closer together (in a
"centripetal" manner; Rait et
al., 1992; Stierlin,
1981), as the needs of the disabled child command the attention of
the rest of the family (Rolland,
1987). From this perspective, we expected levels of family
conflict (observed and perceived) to be lower and cohesion higher in families
of children with spina bifida than in families with able-bodied children. We
also predicted that children with spina bifida would be less influential in
family decision making and that parents from these families would be less
likely to solicit input from their children.
A second perspective is that children with spina bifida have the same desires for independent functioning as their able-bodied counterparts, even prior to adolescence. Such autonomy strivings may cause tension in the parent-child relationship, particularly in families of children with pediatric conditions where parents may have exercised considerable control over their children's activities and medical regimens during the middle childhood years. From this perspective, we expected more conflict, less parent-child agreement and cohesion, more disengagement, and a tendency for family members to be less receptive to each other's opinions in families of children with spina bifida. Given that we are studying 8- and 9-year-old children, a major question is whether conflicts over independent functioning begin in late childhood, prior to the onset of adolescence.
Finally, a third perspective is based on the resilience-disruption
hypothesis of family adaptation advanced by Costigan et al.
(1997) in their study of
families of children with mental retardation. Costigan and colleagues argue
that "families are both disrupted by and resilient to the stress
associated with raising a child who has a disability" (p. 56). Based on
this perspective, we expected family process to be disrupted as children with
spina bifida would be less involved and influential in family interactions;
such families would, therefore, appear less cohesive than families in the
able-bodied sample. On the other hand, these families were likely to evidence
resilience by exhibiting low levels of family conflict and family stress.
Given recent findings that outcomes in pediatric populations vary as a
function of socioeconomic status (SES; e.g.,
Frank, Blount, & Brown,
1997), we also examined whether SES main effects and group (spina
bifida vs. control) by SES interactions were associated with the family
variables. As low SES can be viewed as a stressor that often disrupts
parenting and family functioning, we anticipated that lower SES families would
exhibit lower levels of cohesion and higher levels of family stress and
conflict.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Participants
Participants were 68 families with 8- and 9-year-old preadolescents with spina bifida (37 boys, 31 girls; M [age] = 8.34) and a matched comparison group of 68 families with 8- and 9-year-old able-bodied preadolescents (37 boys, 31 girls; M [age] = 8.49), who were part of a larger study on the transition to adolescence in families with children who have spina bifida (Holmbeck et al., 1997
,
1998;
Holmbeck, Shapera, & Hommeyer,
2002). Complete demographic information for both groups is
provided in Table I. A wide
range of family incomes is represented in both samples. The majority of
participants were Caucasian (91% in the able-bodied group; 82% in the spina
bifida group). Although biological mothers from all families from both groups
participated in the study, only 55 (81%) fathers/step-fathers from the spina
bifida group and 52 (76%) fathers/step-fathers from the able-bodied group
participated. The groups were successfully matched on all 10 demographic
variables (see Table I).
|
Information on a number of physical status variables for the spina bifida group was obtained based on maternal report or from information gleaned from the child's medical chart: (1) spinal lesion level: 32% sacral, 54% lumbosacral or lumbar, 13% thoracic; (2) spina bifida type: 82% myelomeningocele, 12% lipomeningocele, 6% other; (3) shunt status: 71% shunt, 29% no shunt; and (4) ambulation: 19% no assistance, 63% assistance with braces, 18% assistance with a wheelchair. The average number of shunt surgeries among those with shunts was 2.50 (SD = 2.91).
As expected, a significant difference was found between the samples on a
measure of receptive language (Peabody Picture Vocabulary Test, Revised
[PPVT]; Dunn & Dunn, 1981;
M = 92.49, SD = 18.49 for the spina bifida sample and
M = 108.97, SD = 15.06 for the able-bodied sample). This
finding parallels results based on verbal IQ test scores; children with spina
bifida typically score in the low average range (e.g.,
Wills, Holmbeck, Dillon, & McLone,
1990). Because lower receptive vocabulary scores were viewed as
part of the symptom presentation in children with spina bifida and because
children with spina bifida are typically mainstreamed into classrooms with
able-bodied children, we made no attempt to match the samples on this
variable. On the other hand, given that group status and PPVT scores were
confounded, we were interested in whether significant group differences would
continue to be significant after accounting for PPVT scores. Thus, we re-ran
all analyses controlling for PPVT scores.
Participant Recruitment
Participating families in the spina bifida group were recruited from lists
provided by four sources: (1) a children's hospital, (2) a children's hospital
that cares exclusively for youngsters with physical disabilities, (3) a
university-based medical center, and (4) a statewide spina bifida association.
A recruitment letter was sent to all parents of children within the 8- to
9-year-old age range (and those who would reach this age within the following
year). Letters were followed up with phone calls. Out of 310 nonoverlapping
child names from the four sources, 72 families lived too far away (greater
than 120 miles from the laboratory), 64 declined to participate, 56 could not
be reached (due to invalid addresses and phone numbers), 16 did not speak
English, 14 children had turned 10 years old before a visit could be
scheduled, 11 children did not have spina bifida, and 8 were excluded for
miscellaneous reasons. Sixty-nine families remained. One additional family was
dropped following the family visit, as the child was 13 years old (due to an
error on the original participant list). A comparison of participating
children with children from families that declined to participate (n
= 64) revealed no differences with respect to lesion level (
2
[2] =.62, p >.05) or type of spina bifida (myelomeningocele vs.
lipomeningocele) (2 [1] = 1.63, p >.05).
Participating families from the able-bodied comparison group were recruited by contacting schools where the children with spina bifida were enrolled. To obtain a comparison group the same size as the spina bifida group, we did not need to contact all possible schools. Instead, the initial list of schools was based on school enrollment information for the first 42 children with spina bifida who agreed to participate in our study. This list provided us with the necessary number of potential able-bodied participants to yield a satisfactory matching of groups. Of these 42 schools, 24 were ruled out for various reasons (e.g., the community was too far away to run multiple families in that community, the average family income or racial distribution in the school would have produced matching difficulties). Of the remaining 18 schools, 12 agreed to participate. At the participating schools, recruitment letters (as well as self-addressed, stamped envelopes) were sent home with "control" children in our age range; parents could then return a slip indicating their consent to participate. To obtain the sample used in this study, we sent roughly 1,700 letters. The low recruitment rate is attributable, at least in part, to the longitudinal nature of the study that was described in detail in the recruitment letter.
Procedure
Assessments of the participating families were conducted by graduate and
undergraduate research assistants during 3-hour home visits. After the family
members signed informed consent blanks, parents and child were asked to
complete a set of questionnaires as well as one hour of audiotaped and
videotaped family interaction tasks. Questionnaires were read aloud to
children and all Likert-scale formats were presented on large laminated cards.
Upon completion of the questionnaires and interaction tasks, families were
paid $50.
Measures
Questionnaire Measures of Family Functioning
Intensity of Parent-Child Conflict. The 15-item Parent-Adolescent
Conflict Scale (PAC) is a brief version of the Issues Checklist
(Robin & Foster, 1989).
The PAC is composed of a list of potential conflicts often discussed in
families with preadolescents (e.g., whether or not he or she does chores
around the house). Each item requires three responses. The family member first
responds "yes" or "no" according to whether or not the
issue was discussed during the last 2 weeks. If an issue was discussed, the
family member indicates the number of times. Finally, if an issue was
discussed, respondents rate on a 5-point Likert scale (ranging from
"calm" to "angry") how intense these discussions were
on average. Mothers, fathers, and children completed this questionnaire. Only
the intensity ratings were used in this study. Alphas for child, mother, and
father report for the spina bifida group were.81,.76., and.77, respectively.
For the able-bodied group, the corresponding alphas were.68,.65, and.55,
respectively.
Family-Level Conflict and Cohesion. Mothers and fathers completed
a shortened version of the FES, a 90-item self-report measure that assesses
social-environmental characteristics of the family system
(Moos & Moos, 1986). The
FES, composed of 10 subscales, is administered in a true/false format. For
this study, items from the conflict and cohesion subscales were used. Based on
data from this study, Cronbach alphas for the spina bifida group for the two
subscales were as follows: mother report of conflict (.60), mother report of
cohesion (.71), father report of conflict (.67), and father report of cohesion
(.74). The same alphas for the able-bodied comparison sample were.74,.71,.78,
and.57. The somewhat modest alphas for some subscales were due to the
restricted true-false response format
(Roosa & Beals, 1990).
Dyadic Decision-Making Agreement. Decision-making agreement was
assessed using the individually completed questionnaires included as part of
the Structured Family Interaction Task (SFIT;
Ferreira, 1963; see
description later). All family members completed a 5-item form that required
respondents to indicate their first and second choices for several family
activities. To compute dyadic decision-making agreement, we summed all
possible agreements among the first two choices for members of each dyad. For
example, if for a given family, the child's second choice for item 1
corresponded to the mother's first choice for item 1, this was counted as an
agreement for this item. Scores for each dyad could range from 0-10 (two
possible agreements across each of the five items). For families where no
father was present, scores for dyads that included fathers were not
computed.
Family Life Events. The Family Inventory of Life Events (FILE) was
employed as a parent-report measure that assesses the frequency of life events
and the degree to which events have an impact on the family system
(Olson et al., 1985). In its
original form, the FILE includes 71 items that assess life events in the
general areas of family conflicts, marital relations, births/pregnancies,
money, jobs, moves, deaths, and other. The version employed in this study
includes an additional 19 items (for a total of 90 items) added by another
investigator to tap additional relevant stressors (Judy Garber, personal
communication, October 1993). Of these 19 added items, 16 were new (e.g.,
"child experienced increased conflict with peers," "your
home was robbed"); three items were added because three of the original
items were divided into two items each (e.g., "spouse/parent was
separated or divorced" was split into two items: one that assessed
"separation" and one that assessed "divorce"). Each
item has two parts: first, the family member responds "yes" or
"no" as to whether the event has occurred in the past 12 months;
second, if the item has occurred in the past 12 months, family members rate on
a 5-point Likert scale (ranging from "no effect at all" to a
"very big effect") how much of an impact the event has had on
them. Only the total event score was used in this study. The scale reliability
for the mother's total event score was.86 for the spina bifida sample and.87
for the comparison sample, while the corresponding alphas for fathers were.70
and.83, respectively.
Observational Measures
Videotaped Family Interaction Tasks. Three tasks from the
videotaped family session were coded (the order of which were counterbalanced
across families): an unfamiliar board game task (developed for this study), a
conflict task (Smetana, Yau, Restrepo,
& Braeges, 1991), and the Structured Family Interaction Task
(Ferreira, 1963). Prior to
these tasks, families participated in a warm-up task that consisted of two
parts. The first part included a series of anagrams, which children were asked
to work on for 5 minutes. The second part involved assembling a series of five
puzzles.
For the unfamiliar board game task, families were asked to play an educational game purchased through a mail order catalog (not available for retail purchase). No families reported that they had ever seen the game. Families spent 10 minutes engaged in this task, during which time they were asked to establish their own rules and then play the game.
The conflict task was based on a procedure employed by Smetana et al.
(1991). During the
questionnaire portion of the home visit, parents and child completed a short
form of the Issues Checklist (Robin &
Foster, 1989). Prior to the beginning of the interaction tasks,
research assistants tabulated weighted conflict scores (i.e., intensity
x frequency) for each issue endorsed by each family member. The five
issues that received the highest total weighted conflict score across family
members were presented to the family for discussion during the conflict task.
Family members were asked to select three of the five issues and discuss them
for a total of 10 minutes.
Families also completed the Structured Family Interaction Task (SFIT;
Ferreira, 1963). As noted
earlier, during the questionnaire portion of the home visit, each parent and
child completed a five-item questionnaire with each item containing five
response options. Respondents recorded their first and second choices for
commonly discussed family activities (e.g., what TV show they would watch).
During the videotaped family interaction portion of the home visit, the family
was again handed a copy of this questionnaire and was asked to come to a group
consensus and select a first and second choice for the same items.
Global Coding of Family Interaction. Observational data for the
three tasks described were coded using a global-coding method developed by
Holmbeck, Belvedere, Gorey-Ferguson, and Schneider
(1995), based on a system
developed by Smetana et al.
(1991). As is typically done
with global coding systems, coders viewed a single family interaction task and
then provided 5-point Likert scale ratings on a variety of dimensions for that
task. The manual that accompanies this coding system includes behavioral
descriptions for each of the points along the Likert scale. The coding system
assesses several areas of parenting behavior, child behavior, and parent-child
relationships across six dimensions: (1) interaction style, (2) conflict, (3)
affect, (4) control, (5) child-centered and collaborative problem solving, and
(6) several family-level codes. Only items that assessed dyadic-level
(mother-child, father-child, mother-father) or family-level interactions were
selected for use in this study. Following is a list of the dyadic codes (with
the number of codes and domain assignment [i.e., cohesion, conflict, stress]
included in parentheses): (1) family member requests input from another family
member (six dyadic codes; cohesion), (2) how receptive a family member is to
another family member (six dyadic codes; cohesion), (3) the level of conflict
between two family members (three dyadic codes; conflict), and (4) the degree
to which parents present a united front (one code; cohesion). Family-level
variables included four codes that assessed the degree to which a family was
(1) impaired (cohesion; reverse-scored; assesses how well the family is able
to respond to the task and how well they can communicate and discuss
differences); (2) disengaged (cohesion; reverse-scored); (3) open or warm
(cohesion); and (4) able to reach a resolution or agreement (cohesion).
Undergraduate and graduate student coders were trained for approximately 10
hours until they obtained at least 90% agreement with an expert graduate
coder. All coders were "blind" to the specific hypotheses of this
study, but not necessarily to the group status of the child. For each of the
three tasks, two coders rated dyadic and family behaviors. Item-level means of
the two raters for each task were summed across the three tasks to yield a
single score for each coding item for each family.
Satisfactory interrater reliability
(Fleiss, 1981) was found
between coders across the 20 codes for both samples. The only exceptions
occurred for "mother requests input from father" and "father
requests input from mother," which yielded low intraclass correlations
for the spina bifida sample (rs <.30). Thus, these two items were
dropped from further analysis. For the spina bifida group, the ranges and
means for the item-level intraclass correlations were as follows:
mother-child.60 to.76 (M =.67); father-child.51 to.77 (M
=.69); mother-father.46 to.70 (M =.64; and family.71 to.78
(M =.73). The corresponding item-level intraclass correlations for
the comparison sample were mother-child.52 to.79 (M =.67);
father-child.61 to.79 (M =.69); mother-father.39 to.70 (M
=.55); and family.65 to.85 (M =.75).
Given high intercorrelations among the four family-level items (mean r =.70 in the spina bifida sample and.71 in the comparison sample), these items were combined into an "observed family-level cohesion" composite (scored in the direction of higher cohesion).
Home Observation Ratings. One research assistant completed the 44-item Home Visit Report measure at the end of each home visit. Most items were developed for this study, with 17 based on the work of another investigator (Judy Garber, personal communication, October 1993). This measure assesses eight global scales of the home environment (level of familial organization, parent friendliness, child friendliness, degree to which the home environment is child-centered, noise level, home messiness, quality of parenting behavior, and validity of child self-report data). All items were measured on 5- point Likert scales (ranging from "not at all" to "very much") with an option of "not applicable." To assess "family stress," subscales that tapped family-level constructs and those subscales that were expected to vary as a function of stress levels in the home were selected. As such, level of family organization, noise level, and home messiness were all considered measures of family stress (or lack of stress). Unfortunately, home messiness could not be used due to low frequencies and low interrater agreement. The family organization scale includes four items and had scale alphas of.85 and.80 for the spina bifida and comparison groups, respectively. The noise level scale also includes four items and had scale alphas of.85 and.70 for the spina bifida and comparison groups, respectively. Rater reliability data for the family organization and noise level scales were based on 15 randomly selected families and were.80 and.70, respectively, for the spina bifida sample and.85 and.85, respectively, for the comparison sample.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Analyses of group (spina bifida vs. comparison), SES (high [
45] vs.
low [< 45] based on a median split of Hollingshead scores;
Hollingshead, 1975), and
gender (of child) differences with respect to the family variables were
conducted with MANOVAs and ANOVAs. Child-, mother-, and father-reported
dependent variables were run in separate analyses because listwise deletion
defaults for MANOVA reduce the N to those participants with complete
data on all variables (e.g., a MANOVA that included mother- and father-report
variables in the same analysis would have eliminated all single-parent
families as well as all families where only one of two parents responded to
the research protocol). Although this strategy of running analyses separately
by family member allowed us to make use of all available data, this approach
also necessitated the use of ANOVAs (rather than MANOVAs) for some analyses.
Because no gender differences were found, all analyses included only group and
SES (and their interaction) as independent variables. Significant effects were
followed up with appropriate univariate post-hoc tests. All findings are
reported separately within each of the dependent variable domains: conflict,
cohesion, and stress (see Table
II).
|
Conflict
As can be seen in Table II,
there were no significant group or SES main effects or interactions for child
report of conflict intensity or for mother report of conflict. Although a
significant MANOVA emerged for father report of conflict, F(2, 98) =
3.55, p <.05, neither of the univariate F tests was
significant. The only other significant effect for the conflict domain was for
observed mother-child conflict, F(1, 128) = 5.31, p <.05.
For this dyadic variable, there was a significant difference between SES
groups, with higher SES participants (M = 1.46) scoring lower than
low SES participants (M = 1.64).
Cohesion
For cohesion, there was a significant SES main effect for mother report of
family cohesion, F(1, 126) = 8.36, p <.01, with lower SES
families (M =.78) scoring lower than higher SES families (M
=.88). No significant effects were found for father report of family cohesion.
For child-mother agreement, there was a group main effect, F(1, 128)
= 10.80, p <.001, such that families in the spina bifida group
(M = 4.63) scored lower than families in the able-bodied group
(M = 5.44). The only significant group x SES interaction was
found for mother-father agreement, F(1, 101) = 11.58, p
<.001. Duncan post-hoc analyses revealed that high SES families from the
spina bifida group and low SES families from the able-bodied group tended to
score lower than other families on this agreement variable. Significant
MANOVAs emerged for the group main effect for observed mother-child cohesion,
F(4, 125) = 4.20, p <.01, and father-child cohesion,
F(4, 98) = 2.85, p <.05. For both sets of follow-up
univariate analyses, "child requests input from mother/father" was
lower in the spina bifida sample (Ms of 2.62 and 2.49 for
mother-child and father-child, respectively) than in the able-bodied sample
(Ms of 2.95 and 2.77 for mother-child and father-child,
respectively). The univariate statistics were as follows: (1) mother-child,
F(1, 128) = 10.01, p <.01, and (2) father-child,
F(1, 101) = 6.86, p <.01. No significant MANOVAs were
found for observed mother-father cohesion.
Finally, two main effects were found for the observational measure of family-level cohesion: (1) group, F(1, 128) = 4.28, p <.05, and (2) SES, F(1, 128) = 10.62, p <.001. Families in the spina bifida group (M = 4.12) and in the lower SES group (M = 4.07) scored lower than their counterparts in the able-bodied group (M = 4.32) and the higher SES group (M = 4.36). Inspection of the means reveals that the low SES, spina bifida sample scored significantly lower than the other three groups (a finding confirmed with a Duncan post-hoc analysis).
Stress
Although there were no significant effects for father report of life
events, there was a significant SES main effect for mother report of stress,
F(1, 126) = 4.16, p <.05, with higher SES families
(M = 12.72) scoring lower than lower SES families (M =
15.09). Finally, a significant MANOVA group main effect emerged for the
observational measures of stress, F(2, 126) = 3.46, p
<.05. Univariate follow-up analyses revealed a significant group main
effect for level of noise observed in the home, F(1, 127) = 6.28,
p <.01, with families from the able-bodied group (M =
1.61) scoring lower than families from the spina bifida group (M =
1.92).
PPVT Analyzed as a Covariate and Mediator
Given that the groups differed on the PPVT, all analyses were re-run with
PPVT scores as a covariate. All of the univariate effects remained
statistically significant, except for the following three effects for group
status (nonsignificant p values from these covariate analyses are
provided in parentheses): "child requests input from mother"
(p =.06), "child requests input from father" (p
=.12), and the "family cohesion composite" (p =.32). Such
findings raised the possibility that PPVT scores mediated associations between
group status and these family variables (i.e., we were interested in whether
group status was associated with PPVT scores, which were, in turn, associated
with the three family outcomes; group 
PPVT family functioning;
Greene & Ernhart, 1991).
To determine whether there was significant mediation, four criteria had to be
met (Holmbeck, 1997): (1) the
effect of group (the predictor) on PPVT scores (the mediator) should be
significant; (2) the effect of group on the family outcome should be
significant; (3) the effect of PPVT scores on the family outcome should be
significant, after controlling for the group effect; and (4) the group
family outcome total effect should drop significantly when PPVT scores are
included in the model. Based on multiple regression analyses, the first three
conditions were met for all three of these family outcomes. The fourth
condition was examined by testing the significance of the indirect effect
(Holmbeck, in press), by using Sobel's
(1988) equation for the
standard error of the indirect effect. Significant mediation was found in all
three cases (z > 1.96) and PPVT scores accounted for between 42%
and 55% (M = 46.9%) of the total effect of group status on these
three family outcomes.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
The purpose of this study was to examine observed and perceived family functioning in a sample of preadolescents with spina bifida in comparison to families of able-bodied children. Findings revealed significant group and SES differences, with most of the significant effects for the observational data. In the spina bifida sample, families were observed to be less cohesive, children were perceived as more passive during the family interaction tasks, there was less mother-child agreement on a measure of activity preferences, and noise levels were higher in the home environment than was the case in the able-bodied sample. On the other hand, levels of conflict and reports of life events did not differ across the two groups. Lower SES families demonstrated higher levels of observed mother-child conflict, less observed and perceived family cohesion, and more life events. Finally, PPVT scores were found to mediate some of the associations between group status and observed family functioning.
Findings of this study appear to be most consistent with the
resilience-disruption hypothesis proposed by Costigan et al.
(1997). With respect to
disruption, families of children with spina bifida were viewed as less
cohesive; indeed, four of the five significant group effects emerged in the
cohesion domain. Such results support the findings of previous work based on
self-reports of family functioning as compared to normative data
(Loomis et al., 1997).
Families in the spina bifida sample demonstrated resilience as well. That is,
levels of conflict and rates of life events did not vary across the groups. In
other words, having a preadolescent with spina bifida appears to be associated
with some disruption in family relationships, but such disruptions do not
translate into higher levels of family conflicts and stress
(Costigan et al., 1997). As
Kazak (1997) noted,
"families of children with serious illness and physically handicapping
conditions are different but not deviant" (p. 145). (It is important to
note that we found support for the "resilience" portion of
Costigan et al.'s resilience-disruption hypothesis via the absence of
significant effects for certain variables. Conclusions based on null findings
should be viewed cautiously.)
Although the lower levels of observed family cohesion in the spina bifida
sample support the "disruption" side of Costigan et al.'s
(1997) resilience-disruption
hypothesis, the link between group status and cohesion was significantly
mediated by PPVT scores. Moreover, the fact that PPVT scores remained
significantly associated with family cohesion even after accounting for group
status suggests that the association between PPVT and family functioning is
robust. It may be that communication in families of children with lower verbal
IQs is disrupted because of reduced involvement of these children in family
conversations. Alternatively, because the PPVT tended to be associated only
with the observational variables, it may be that the raters who coded the
videotapes for this study were biased in some way with respect to verbal IQ
level (a possibility that should be investigated in future work). These
interpretations aside, although PPVT scores served a significant mediational
role, roughly half of the group effect remained even after accounting for PPVT
scores. Thus, it appears that both group status and verbal IQ level have an
impact on observed family functioning. (In interpreting the findings for the
PPVT, we have implied that the PPVT is a valid proxy for verbal IQ scores,
given that the PPVT is highly correlated with verbal IQ scores, especially
vocabulary subtests. There is some controversy over this issue
[Gregory, 2000].)
The finding that children with spina bifida were less likely to request
input from their parents during observed family interactions is in line with
past work on families of children with pediatric and cognitive conditions
(Costigan et al., 1997). In our
sample, it is not clear if this lack of involvement on the part of children
with spina bifida is a consequence of certain aspects of their
neuropsychological functioning (e.g., moderately lower IQ and receptive
vocabulary abilities, attentional problems;
Wills et al., 1990), a
manifestation of internalizing symptoms
(Ammerman et al., 1998;
Appleton et al., 1997), a form
of "learned passivity" that may have resulted from higher levels
of parental control in these families (see
Holmbeck et al.,
2002,Holmbeck et al.,
2002), or if the child's lack of interest in his or her parents'
perspectives is a passive strategy for gaining a certain degree of autonomy.
In support of the first possibility, the mediational analyses revealed that
the PPVT scores partially mediated the effect of group status on the
"child requests input from parent" variables. The meaning of such
passive behaviors among children with such conditions will need to be
addressed in future research.
Although, at a systemic level, families appear to adapt successfully to
having a child with a physical disability, past research from the same data
set suggests that stress is higher at the individual level
(Holmbeck et al., 1997). It
may be, then, that parents experience more stress as individuals, but that
this does not have an impact on level of satisfaction in the marriage or the
level of conflict in the family as a whole. Such parents may simply be less
engaged during interactions with their family. On the other hand, our findings
may be age-dependent; the lack of effects for certain family variables may
have been a consequence of our focus on preadolescents. If this is the case,
we may begin to see more group differences as the children in our sample move
into the adolescent transition.
Unlike the group differences findings, the analyses involving SES revealed
that this variable had an impact on all three domains of family functioning.
Lower SES families exhibited higher levels of observed mother-child conflict,
less cohesion, and more life events. It is also noteworthy that the findings
for SES emerged for both self-reports and observational data. Why
were the effects for SES more pervasive? The findings of this study suggest
that families may be less adept at adjusting to economic strain than to the
presence of a child with special needs. Unlike the effects for group status,
low SES appears to adversely affect the family's ability to stave off family
conflicts and life events. It also appears that parents are aware of the
impact of SES, given that significant SES effects were found for the
self-report data. These findings for SES parallel results of studies that have
focused on the impact of single parenting (e.g.,
Costigan et al., 1997) and
studies of SES and other outcomes (Bier,
Morales, Liebling, Geddes, & Kim, 1997).
The fact that the group and SES main effects were significant for
observed family cohesion suggests that these two variables have an additive
effect on this family variable. Specifically, the low-SES, spina bifida sample
appears to be particularly at-risk for low levels of cohesion. Additive
effects like this one have been found in other studies in the pediatric
literature (Carr, 1991).
Holmes, Yu, and Frentz (1999),
for example, found that disease status and negative life events predicted
child adjustment problems additively; diabetic children from environments
characterized by high levels of negative life events were more likely to have
adjustment difficulties than children experiencing only one of these
stressors.
Despite finding significant group and SES effects across several dyadic and
systemic variables, our study is not without limitations. First, the
generalizability of these findings to children and adolescents of other ages
is limited due to the narrow age range targeted in the present investigation
(8- and 9-year-olds). The age of the sample also may have limited the number
of significant results. As noted earlier, group differences in family dynamics
may emerge as the children move into adolescence. Second, 16 families were
excluded from participating because they did not speak English (most of these
families were Spanish-speaking). Thus, the results of this study are
generalizable primarily to Caucasian, English-speaking families. Future
studies should include a more representative sampling of Spanish-speaking
families to attend to this issue of external validity, particularly given the
high rate of spina bifida in Hispanic populations
(Lary & Edmonds, 1996).
Third, we have examined only one illness group; thus, we do not know the
degree to which the group differences apply more generally to the larger
population of chronically ill children. Finally, although the coders of the
observational data were blind to the specific hypotheses of this study, they
were not blind to the group status (spina bifida vs. comparison) of the
children. Thus, some coders may have been biased in their ratings. On the
other hand, all coders were instructed not to base their ratings on their
knowledge of the child's group status.
From a clinical perspective, the findings of this study identify low SES
families of children with spina bifida as a possible at-risk group (also see
Seefeldt et al., 1997).
Clinicians who work with such families may wish to bolster the stress
management skills of family members to reduce the risks of possible increases
in family conflicts and decreases in cohesion. With respect to families of
children with physical disabilities more generally, and given that there may
be a link between parenting stress
(Holmbeck et al., 1997) and
lower levels of family cohesion, the level of individual stress could also be
addressed with clinical interventions. Finally, although many families of
children with physical disabilities appear to be resilient as they adapt to
the stresses of managing a child with special needs
(Costigan et al., 1997), the
degree to which such resilience skills can be supported is likely to determine
whether members of these families exhibit adjustment difficulties in the
future.
|
Acknowledgments |
|---|
Completion of this manuscript was supported by Social and Behavioral Sciences Research Grants 12-FY93-0621, 12-FY95-0496, 12-FY97-0270, and 12-FY99-0280 from the March of Dimes Birth Defects Foundation and a Research Support Grant and Paid-Leave from Loyola University of Chicago. We thank Ann Walsh Johnson, Joy Ito, Pat McGovern, Pat Braun, Caroline Anderson, David McLone, John Lubicky, the Illinois Spina Bifida Association, and the staff of the spina bifida clinics at Children's Memorial Hospital, Shriner's Hospital-Chicago, and Loyola University of Chicago Medical Center. We also thank numerous undergraduates and graduate students for their help with data collection and data entry. Most important, we thank the parents, children, teachers, and health professionals who participated in our study.
Received July 17, 2000; revision received January 18, 2001; accepted April 17, 2001
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Ammerman, R. T., Kane, V. R., Slomka, G. T., Reigal, D. H., Franzen, M. D., & Gadow, K. D. (1998). Psychiatric symptomatology and family functioning in children and adolescents with spina bifida. Journal of Clinical Psychology in Medical Settings, 5, 449-465.
Ammerman, R. T., Van Hasselt, V. B., & Hersen, M.
(1991). Parent-child problem-solving interactions in families of
visually impaired youth. Journal of Pediatric Psychology,
16, 87-101.
Anderson, B. J., & Coyne, J. C. (1993). Family context and compliance behavior in chronically ill children. In N. A. Krasnegor, L. Epstein, S. B. Johnson, & S. J. Yaffe (Eds.), Developmental aspects of health compliance behavior (pp. 77-89). Hillsdale, NJ: Lawrence Erlbaum.
Appleton, P. L., Ellis, N. C., Minchom, P. E., Lawson, V., Boll,
V., & Jones, P. (1997). Depressive symptoms and self-concept
in young people with spina bifida. Journal of Pediatric
Psychology, 22,
707-722.
Bier, J. B., Morales, Y., Liebling, J., Geddes, L., & Kim, E. (1997). Medical and social factors associated with cognitive outcome in individuals with myelomeningocele. Developmental Medicine and Child Neurology, 39, 263-266.[ISI][Medline]
Carr, J. (1991). The effect of neural tube defects on the family and its social functioning. In C. M. Bannister & B. Tew (Eds.), Current concepts in spina bifida and hydrocephalus (pp. 180-192). New York: Cambridge University Press.
Chaney, J. M., Mullins, L. L., Frank, R. G., Peterson, L., Mace, L.
D., Kashani, J. H., & Goldstein, D. L. (1997). Transactional
patterns of child, mother, and father adjustment in insulin-dependent diabetes
mellitus: A prospective study. Journal of Pediatric
Psychology, 22,
229-244.
Costigan, C. L., Floyd, F. J., Harter, K. S. M., & McClintock, J. C. (1997). Family process and adaptation to children with mental retardation: Disruption and resilience in family problem-solving interactions. Journal of Family Psychology, 11, 515-529.
Cox, M. J., & Brooks-Gunn, J. (Eds.). (1999). Conflict and cohesion in families: Causes and consequences. Mahwah, NJ: Lawrence Erlbaum.
Drotar, D. (1997). Relating parent and family
functioning to the psychological adjustment of children with chronic health
conditions: What have we learned? What do we need to know? Journal
of Pediatric Psychology, 22,
149-165.
Dunn, L. M., & Dunn, L. M. (1981). Peabody Picture Vocabulary Test-Revised (PPVT-R). Circle Pines, MN: American Guidance Service.
Ferreira, A. J. (1963). Decision making in normal and pathological families. Archives of General Psychiatry, 8, 68-73.
Fleiss, J. L. (1981). The measurement of interrater agreement. Statistical methods for rates and properties (pp. 213-236). New York: John Wiley & Sons.
Frank, N. C., Blount, R. L., & Brown, R. T.
(1997). Attributional, coping, and adjustment in children with
cancer. Journal of Pediatric Psychology,
22, 563-576.
Greene, T., & Ernhart, C. B. (1991). Adjustment for cofactors in pediatric research. Developmental and Behavioral Pediatrics, 12, 378-385.
Gregory, R. J. (2000). Psychological testing: History, principles, and applications. Boston: Allyn and Bacon.
Hollingshead, A. A. (1975). Four factor index of social status. Unpublished manuscript, Yale University, New Haven, CT.
Holmbeck, G. N. (1996). A model of family relational transformations during the transition to adolescence: Parent-adolescent conflict and adaptation. In J. A. Graber, J. Brooks-Gunn, & A. C. Petersen (Eds.), Transitions through adolescence: Interpersonal domains and context (pp. 167-199). Mahwah, NJ: Lawrence Erlbaum.
Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599-610.[ISI][Medline]
Holmbeck, G. N. (2002). Post-hoc probing of
significant moderational and mediational effects in studies of pediatric
populations. Journal of Pediatric Psychology,
27, 87-96.
Holmbeck, G. N., Belvedere, M. C., Christiansen, M., Czerwinski, A. M., Johnson, S. Z., Kung, E., & Schneider, J. (1998). Assessment of adherence with multiple informants in pre-adolescents with spina bifida: Initial development of a multidimensional, multitask parent-report questionnaire. Journal of Personality Assessment, 70, 427-440.[ISI][Medline]
Holmbeck, G. N., Belvedere, M., Gorey-Ferguson, L., & Schneider, J. (1995). Family macro-coding manual: March of Dimes triadic version. Unpublished coding manual, Loyola University of Chicago.
Holmbeck, G. N, Gorey-Ferguson, L., Hudson, T., Seefeldt, T.,
Shapera, W., Turner, T., & Uhler, J. (1997). Maternal,
paternal, and marital functioning in families of preadolescents with spina
bifida. Journal of Pediatric Psychology,
22, 167-181.
Holmbeck, G. N., Li. S., Schurman, J. V., Friedman, D., &
Coakley, R. M. (2002). Collecting and managing multisource and
multimethod data in studies of pediatric populations. Journal of
Pediatric Psychology, 27,
5-18.
Holmbeck, G. N., Shapera, W., & Hommeyer, J. S. (2002). Observed and perceived parenting behaviors and psychosocial adjustment in pre-adolescents with spina bifida. In B. K. Barber (Ed.), Intrusive parenting: How psychological control affects children and adolescents (pp. 191-234). Washington, DC: American Psychological Association.
Holmes, C. S., Yu, Z., & Frentz, J. (1999). Chronic and discrete stress as predictors of children's adjustment. Journal of Consulting and Clinical Psychology, 67, 411-419.[ISI][Medline]
Kazak, A. E. (1986). Families with physically handicapped children: Social ecology and family systems. Family Process, 25, 265-281.[ISI][Medline]
Kazak, A. E. (1997). A contextual family/systems
approach to pediatric psychology: Introduction to the special issue.
Journal of Pediatric Psychology,
22, 141-148.
Kazak, A. E., Segal-Andrews, A. M., & Johnson, K. (1995). Pediatric psychology research and practice: A family/systems approach. In M. C. Roberts (Ed.), Handbook of pediatric psychology (pp. 84-104). New York: Guilford Press.
Lary, J. M., & Edmonds, L. D. (1996). Prevalence of spina bifida at birth-United States, 1983-1990: A comparison of two surveillance systems. Morbidity and Mortality Weekly Reports, 45, 15-26.
Loomis, J. W., Javornisky, J. G., Monahan, J. G., Burke, G., & Lindsay, A. (1997). Relations between family environment and adjustment outcomes in young adults with spina bifida. Developmental Medicine, 39, 620-627.
McCubbin, H., & Patterson, J. M. (1982). Family adaptation to crisis. In H. McCubbin, A. Cuble, & J. Patterson (Eds.), Family stress, coping, and social support (pp. 26-47). Springfield, IL: C. Thomas.
Moos, R. H., & Moos, B. S. (1986). Family environment scale manual. (2nd ed.). Palo Alto, CA: Consulting Psychological Press, Inc.
Murch, R. L., & Cohen, L. H. (1989). Relationships
among life stress, perceived family environment, and the psychological stress
of spina bifida adolescents. Journal of Pediatric
Psychology, 14,
193-214.
Olson, D. H., McCubbin, H. I., Barnes, H., Larsen, A., Muxen, M., & Wilson, M. (1985). Family inventories. St. Paul: University of Minnesota Press.
Quittner, A. L., & DiGirolamo, A. M. (1998). Family adaption to childhood disability and illness. In R. T. Ammerman & J. V. Campo (Eds.), Handbook of pediatric psychology and psychiatry (vol. 2, pp. 70-80). Boston: Allyn and Bacon.
Rait, D. S., Ostroff, J. S., Smith, K., Cella, D. F., Tan, C., & Lesko, L. M. (1992). Lives in the balance: Perceived family functioning and the psychosocial adjustment of adolescent cancer survivors. Family Process, 31, 383-397.[ISI][Medline]
Robin, A. L., & Foster, S. L. (1989). Negotiating parent-adolescent conflict: A behavioral-family systems approach. New York: Guilford.
Rolland, J. S. (1987). Chronic illness and the life cycle: A conceptual framework. Family Process, 26, 203-221.[ISI][Medline]
Roosa, M. W., & Beals, J. (1990). Measurement issues in family assessment: The case of the Family Environment Scale. Family Process, 29, 191-198.[ISI][Medline]
Seefeldt, T., Holmbeck, G. N., Belvedere, M., Gorey-Ferguson, L., Hommeyer, J. S., & Hudson, T. (1997). Socioeconomic status and democratic parenting in families of pre-adolescents with spina bifida. Psi Chi: The Undergraduate Research Journal, 2, 5-12.
Seiffge-Krenke, I. (1998). The highly structured
climate in families of adolescents with diabetes: Functional or dysfunctional
for metabolic control? Journal of Pediatric Psychology,
23, 313-322.
Smetana, J. G., Yau, J., Restrepo, A., & Braeges, J. L. (1991). Adolescent-parent conflict in married and divorced families. Developmental Psychology, 27, 1000-1010.
Sobel, M. E. (1988). Direct and indirect effects in linear structural equation models. In J. S. Long (Ed.), Common problems/proper solutions: Avoiding error in quantitative research (pp. 46-64). Beverly Hills, CA: Sage.
Spaulding, B. R., & Morgan, S. B. (1986). Spina
bifida children and their parents: A population prone to family dysfunction.
Journal of Pediatric Psychology,
11, 359-374.
Stierlin, H. (1981). Separating parents and adolescents: Individuation in the family. New York: Aronson.
Thompson, R. J., & Gustafson, K. E. (1996). Adaptation to chronic childhood illness. Washington, DC: American Psychological Association.
Wallander, J. L., & Varni, J. W. (1995). Appraisal, coping, and adjustment in adolescents with a physical disability. In J. L. Wallander & L. J. Siegel (Eds.), Adolescent health problems: Behavioral perspectives (pp. 209-231). New York: Guilford.
Wallander, J. L., & Varni, J. W. (1998). Effects of pediatric chronic physical disorders on child and family adjustment. Journal of Child Psychology and Psychiatry, 39, 29-46.[ISI][Medline]
Wills, K. E., Holmbeck, G. N., Dillon, K., & McLone, D. G.
(1990). Intelligence and achievement in children with
myelomeningocele. Journal of Pediatric Psychology,
15, 161-176.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. P. R. Vermaes, J. R. M. Gerris, and J. M. A. M. Janssens Parents' Social Adjustment in Families of Children with Spina Bifida: A Theory-driven Review J. Pediatr. Psychol., November 1, 2007; 32(10): 1214 - 1226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Coakley, G. N. Holmbeck, and F. B. Bryant Constructing a Prospective Model of Psychosocial Adaptation in Young Adolescents with Spina Bifida: An Application of Optimal Data Analysis J. Pediatr. Psychol., November 1, 2006; 31(10): 1084 - 1099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. DeLucia and S. C. Pitts Applications of Individual Growth Curve Modeling for Pediatric Psychology Research J. Pediatr. Psychol., November 1, 2006; 31(10): 1002 - 1023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. Greenley, G. N. Holmbeck, and B. M. Rose Predictors of Parenting Behavior Trajectories Among Families of Young Adolescents with and without Spina Bifida J. Pediatr. Psychol., November 1, 2006; 31(10): 1057 - 1071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. Greenley, R. M. Coakley, G. N. Holmbeck, B. Jandasek, and K. Wills Condition-Related Knowledge Among Children with Spina Bifida: Longitudinal Changes and Predictors J. Pediatr. Psychol., September 1, 2006; 31(8): 828 - 839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. C. Haber, J. K. Austin, G. R. Huster, K. A. Lane, and S. M. Perkins Relationships Between Differences in Mother-Father Perceptions and Self-Concept and Depression in Children With Epilepsy Journal of Family Nursing, February 1, 2003; 9(1): 59 - 78. [Abstract] [PDF]
|
|
|
|
|
|
R. M. Coakley, G. N. Holmbeck, D. Friedman, R. N. Greenley, and A. W. Thill A Longitudinal Study of Pubertal Timing, Parent-Child Conflict, and Cohesion in Families of Young Adolescents With Spina Bifida J. Pediatr. Psychol., July 1, 2002; 27(5): 461 - 473. [Abstract] [Full Text] [PDF]
|
|
| |||||||