Journal of Pediatric Psychology, Vol. 28, No. 6, 2003, pp. 423-432
© 2003 Society of Pediatric Psychology
Inpatient Pediatric Consultation-Liaison: A Case-Controlled Study
1 University of Louisville School of Medicine, 2 Indiana University School of Medicine
All correspondence should be sent to Bryan D. Carter, Division of Child and Adolescent Psychiatry, University of Louisville School of Medicine, 200 East Chestnut Street, Louisville, Kentucky 40202. E-mail: bdcart01{at}louisville.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective To conduct a prospective case-controlled study of pediatric inpatients referred for consultation in a tertiary care children's medical center. Method Referrals (n = 104) were matched with nonreferrals (n = 104) for age (4 to 18 years), gender, and illness type/severity and completed parent- and self-report (dependent on age) behavioral rating scales to assess for adjustment/functioning. Nurses completed in-hospital ratings of behavioral/adjustment difficulties. Goal attainment and satisfaction ratings were obtained from the referring physicians, parents/guardians, and the consultant. Results Referrals exhibited more behavior/adjustment/coping difficulties than nonreferrals by parent, nurse, and self report. Frequently employed interventions included coping-strategies intervention, cognitive and behavioral therapies, and case management. Referring physician and consultant ratings of goal attainment were high, as were physician ratings of satisfaction and parent/guardian ratings of overall helpfulness. Conclusions Pediatric inpatients referred by their physicians had significantly more internalizing and externalizing disturbances than their nonreferred hospitalized peers. Many of the behavioral and adjustment problems that lead to in-hospital consultation referral were evident in global behavior difficulties prior to hospitalization. Referring pediatricians, parents/guardians, and consultants rate the outcome as benefiting the patients via assisting in the overall management of their health concerns, coping, and adjustment.
Key words: pediatric consultation-liaison; children; coping; hospitalization.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Pediatric psychology and consultation-liaison child psychiatry have emerged as subspecialty fields of clinical child psychology and child psychiatry, respectively, in response to the psychosocial needs of children with medical problems and their families. One of the most common vehicles for the delivery of pediatric psychological and psychiatric services is inpatient pediatric consultation-liaison (C/L) services. Consultations for hospitalized children typically follow a medical model whereby the consultant conducts an assessment of the referred pediatric patient and advises the referring physician and medical team about the findings and management of psychosocial aspects of the patient's care (Drotar, Spirito, & Stancin, in press
). This is often a complex process involving
multiple interviews of child, family, and staff, repeated behavioral
observations, perhaps formal psychological assessment, communication of
findings to the hospital staff via written, telephone, and face-to-face
contacts, and implementation of intervention procedures often under tight time
constraints of competing medical procedures and insurance limitations
(Drotar, 1995). Additionally,
C/L services are often confronted with a wide array of referral questions,
including differential diagnoses of organic versus psychogenic contributors to
symptom presentation, adherence problems to medications and treatments, coping
and adjustment to chronic illness and trauma, pain management, decision making
for organ transplantation, and arranging for posthospitalization
follow-up.
Despite the centrality of C/L services to hospital-based pediatric
psychology and child psychiatry, there is a relative dearth of studies
characterizing the array of services provided by the C/L team. In one of the
few large C/L studies, Olson and her colleagues
(1988) conducted a
retrospective review of the records of 749 inpatient referrals seen by the
pediatric psychology service at Oklahoma Children's Hospital over a 5-year
period. Referrals seen, in order of greatest frequency, were depression or
suicide attempt, adjustment problems to chronic illness, and behavior
problems. General pediatrics requested consultations most frequently, followed
by surgery and adolescent medicine. Almost a third of the children seen for
in-hospital consultation were subsequently seen for outpatient follow-up.
Health care professionals making referrals were generally very satisfied with
the services of the pediatric C/L team and expressed a high likelihood of
making future referrals for consultation.
Rodrigue and colleagues
(1995) conducted an archival
review of 1,467 records of in-hospital (n = 448) and outpatient
(n = 1,019) referrals to a health center-based pediatric psychology
service at the University of Florida Health Sciences Center from 1990 to 1993.
Accounting for most of the inpatient referrals were general pediatrics (40%),
pediatric hematology/oncology (31%), adolescent psychiatry (15%), pediatric
intensive care (5%), and the burn unit (4%). The most common reason for
referral was assessment of cognitive or neuropsychological functioning
(reflecting the strong psychological assessment orientation of this particular
pediatric C/L service), followed by externalizing behavior problems,
comprehensive psychological evaluation, presurgery/pretransplant evaluation,
and adjustment problems to chronic illness. A retrospective survey of 143
referring health professionals indicated generally high overall satisfaction
with service quality.
In the Knapp and Harris 10-year review of clinical reports
(1998a) and treatment outcome
studies (1998b) on pediatric
C/L child psychiatry, the authors surveyed both the categorical
(illness-specific) and noncategorical investigations into the psychiatric care
of medically ill children. They concluded that pediatric C/L services are
increasingly playing a role in meeting the emotional and behavioral needs of
pediatric inpatients via facilitation of individual and family adaptation to
the stressors associated with chronic illness. Thompson and Gustafson
(1996) have incorporated these
illness-related stressors into a model of adaptation to chronic childhood
illness. Their topology includes multiple biomedical factors, including
disease prevalence, age of onset, mobility-activity, course of illness, impact
of illness on cognitive and sensory functioning, and visibility of the
illness, all of which differentially impact on the adjustment and functioning
of the child and family.
Roberts (2001) has urged
that increased efforts be made toward the development of evidence-based
interventions in clinical practice to improve service delivery within medical
systems. In order to progress toward this goal, the next stage of research
into pediatric C/L services needs to verify the disturbances in adaptation
experienced by hospitalized children and their families by applying
standardized instruments that measure adjustment and psychological functioning
in prospective samples of inpatient pediatric referrals, via multiple
informants, to avoid the limitations found in measurement by other
investigators of pediatric C/L services
(Harris, Canning, & Kelleher,
1996). In particular, there is a need for information as to the
accuracy of the referral process in appropriately identifying those pediatric
inpatients who can benefit from psychological/psychiatric services,
identifying the types of psychiatric diagnoses represented among referred
pediatric inpatients, and further identifying the clinical issues and needs of
these children. Furthermore, more specific information is needed as to the
efficacy of pediatric C/L services in attaining the goals set in the
consultation contract between the referring physician and the consultant, as
well as general satisfaction with the services from the perspective of the
referring physician and the pediatric inpatient's parent/guardian. To answer
these questions, multisource assessment with both referred and nonreferred
inpatients is necessary.
With this in mind, the goals of the present prospective study were to (a) provide a descriptive analysis of referral patterns, consultant roles/functions/services, and the medical and psychological/psychiatric conditions of children referred to a general inpatient pediatric C/L service at a tertiary care children's medical center; (b) investigate the psychological characteristics, via multi-informant standardized psychological instruments, of referred pediatric inpatients and compare them with matched nonreferred medically hospitalized children in order to assess accuracy in making referrals; and (c) assess goal attainment and general satisfaction with the consultation from the perspective of the referring health care professional, consultant, and the parent/guardian of the hospitalized child.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Program Description
The pediatric consultation-liaison service of the Division of Child and Adolescent Psychiatry of the University of Louisville School of Medicine is a clinical and training/teaching unit that provides consultations to pediatric inpatients at Kosair Children's Hospital, as well as to several other hospitals within the University of Louisville Health Sciences Center. This service has been in existence since 1974. The pediatric C/L team consists of the service director, who is a doctoral pediatric psychologist, an attending child psychiatrist, two pediatric psychology postdoctoral fellows, a child psychiatry fellow, and a predoctoral clinical child/pediatric psychology intern. Referrals are assigned to team members on a rotating basis, with the service director supervising on all cases. Kosair Children's Hospital is a 250-bed teaching hospital that provides a full range of specialized pediatric services, including bone marrow and organ transplantation.
Participants
Referrals consisted of 104 pediatric inpatients hospitalized at a regional
tertiary care children's medical center (Kosair Children's Hospital in
Louisville) and referred to the pediatric C/L service. Nonreferrals consisted
of 104 inpatients selected from the daily hospital admission sheet who were
matched to referrals for age (±2 years), gender, and illness
type/severity (see Table I). In
order to reduce sampling bias, nonreferred patients were selected from the
first available admissions roster following the enrollment of a referral case.
Inclusion criteria for all participants were ages 4 through 18 years and an
available parent/guardian for informed consent and completion of study
instruments. Referrals were not eligible for the study if they were to be
immediately discharged from the hospital, had insufficient time or ability to
complete the study instruments, and/or had severe developmental delay.
Pediatric inpatients were not approached for participation if there were
clinically sensitive issues (e.g., end-stage disease, life support, extreme
life-threatening trauma, or ventilator dependence with paralytic medications).
Of the referred cases approached for participation in the study, 31 refused
(compared with 2 nonreferred families), while 104 agreed to participate. The
most common reason given was "not interested." Self-report
measures were not administered to subjects under 8 years of age due to the age
limitations of these measures. Overall, the distribution of participants and
nonparticipants within the referred group did not differ by referring
service—that is, hematology/oncology, general pediatrics, surgery, etc.
(chi-square = 1.79; p > .10). All general and specialty admitting
services were represented in the range of referrals, with the exception of the
neonatal intensive care unit, due to the age limitation constraints of the
study.
|
In the present study, we found that some patient populations were frequently referred for pediatric C/L services (e.g., relapsed oncology patients) and were therefore difficult to match with a patient of the same age and gender with an identical medical diagnosis. Table I represents our attempt to arrive at a consensus of categories of illness severity for purposes of matching nonreferrals to referrals. When a nonreferral could not be found with the same medical diagnosis as a referral, then at least two consultants on the pediatric C/L team had to agree, after reviewing the patient's medical record, on the illness category of the nonreferred patient for purposes of matching. The categories of acute non-life threatening, acute life threatening, chronic non-life threatening, and chronic life threatening were selected for these purposes. In only two cases did a third consultant become involved to resolve a disagreement in illness categorization.
Procedures
Permission was obtained from the attending staff physicians, via mail, to
approach their hospitalized patients and their parents/guardians about
participation in the study. No physician refused. Hospitalized children and
their parents/guardians were asked by a research assistant to participate in a
study of children's experience of being an inpatient in the hospital. All
subjects were assured confidentiality; that is, that none of the research
information would be shared with their physicians or treating health care
professionals. In addition, referred patients, their parents/guardians, and
their attending physicians/residents were assured that none of the
questionnaire or rating information they provided would be shared with the
consultant, to increase the likelihood of their making an honest and unbiased
rating. Referring physicians were contacted by the research assistant within a
few days after their patient was discharged from the hospital to obtain goal
attainment and satisfaction ratings. Patient participants were given $20 for
participation in the study upon completion of all instruments. Although not
all subjects completed all questionnaires, for those completing each measure,
there were no significant differences between the referred and nonreferred
groups on age, sex, race, or medical condition. The study had the approval of
the Human Studies Committee of the University of Louisville, and was funded by
Alliant/Norton HealthCare Community Trust Fund Grant No. 97-10.
To assess goal attainment and satisfaction information, the referring physicians and consultant established goals to be attained in the conduct of the consultation at the time of initiation, which were coded on a goal attainment/satisfaction sheet. All goals established between the referring physician and the consultant were idiographic and open-ended, that is, specific to the particular referral concerns of the physician and health care team. Up to three goals to be attained were listed by the consultant on the form. When the consultation had been completed, the research assistant contacted the referring health care professional and requested that he or she rate the attainment of each goal (up to three goals per subject) on a 5-point scale (1 = not attained to 5 = clearly attained), as well as the helpfulness of the consultation (1 = not helpful to 5 = very helpful). The consultant was given a separate sheet with the listed goals and asked to rate goal attainment on the same 5-point scale as the referring health care professional. Parents/guardians were also asked by the research assistant to rate the consultation on helpfulness to the child and to the family, but on a 4-point scale (1 = not helpful to 4 = extremely helpful), as well as answer yes or no to the more global question "Did the consultation aid in your child's recovery?"
Finally, the consultant was asked to select from the following list the types of interventions provided: talk and support; arrangement of follow-up care; coping-strategies intervention; helping the patient/family communicate with medical staff; prescribing medication; helping family members communicate with each other; in-hospital behavioral plan; cognitive behavioral intervention; crisis intervention; grief intervention; therapeutic play; and "other," to capture less frequently used interventions.
Axis I diagnoses from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) were assigned to referrals via the pediatric C/L team member's clinical assessment and nonstructured history and interview, with agreement being arrived at between the team member and his or her supervisor. Reliability of DSM-IV diagnoses was not established.
Study Instruments
The Kaufman Brief Intelligence Test (KBIT;
Kaufman & Kaufman, 1990) is
an intellectual screening instrument that takes about 20-30 minutes to
administer and consists of two subtests. The Vocabulary subtest contains
questions about single-word expressive vocabulary (picture naming) and
identification of words from verbal and letter-placement cues. The Matrices
subtest consists of pictorial analogies and matching problems. KBIT results
provide an estimate of verbal (Vocabulary) and nonverbal (Matrices) ability,
and also yield an overall IQ composite score, which was used in the present
study as a measure of global intellectual ability. KBIT composite and
Vocabulary IQ scores correlate strongly with scores on the Wechsler
Intelligence Scale for Children-Revised
(Kaufman & Kaufman, 1990).
The research assistant, who was a doctoral student in clinical psychology,
administered the KBIT to all subjects in their hospital rooms (72 referrals
and 73 nonreferrals).
The parent/guardian (100 referrals, 102 nonreferrals) completed the Child
Behavior Checklist (CBCL; Achenbach,
1991), a widely used, standardized checklist of children's problem
behaviors and adjustment. This parent-report measure was employed due to its
frequent use in both clinical settings and research, allowing more ready
comparison to data from other relevant studies. The CBCL is ideal for research
use across broad age ranges since its items and subscales do not change with
the age of the child. The 118 behavioral descriptions are rated on a 3-point
scale, with subscale T-scores derived based on the child's age and gender. In
addition to the internalizing and externalizing second-order scales, the CBCL
consists of the following subscales: withdrawn, somatic concerns,
anxious/depressed, social problems, thought problems, attention problems,
delinquent behavior, and aggressive behavior. All CBCL subscales were used in
this study.
Children aged 8 to 18 (91 referrals, 92 nonreferrals) completed the
Children's Depression Inventory (CDI;
Kovacs, 1992), a 27-item
self-report measure of depressive symptoms that addresses cognitive,
affective, motivational, and vegetative symptoms of depression. The CDI has
good psychometric properties, with demonstrated high internal consistency and
test-retest reliability, and has been used extensively with children and
adolescents. Construct and criterion validity studies show that the CDI
relates to self-esteem, hopelessness, cognitive processing, and depression as
measured by other instruments (Craighead,
Smucker, Craighead, & Ilardi, 1998;
Kazdin, 1990;
Kovacs, 1992). The CDI total
score was used as a more global self-report measure of depressive symptoms in
this study. Due to age limitations, children in the study under age 8 did not
complete this measure.
Children aged 8 to 18 (84 referrals, 88 nonreferrals) completed the
Behavioral Assessment System for Children, Self-Report of Personality
(BASC-SRP; Reynolds & Kamphaus,
1992), a true-false inventory that assesses multiple emotional and
behavioral domains, including both positive (adaptive) and negative (clinical)
dimensions. Coefficient alphas and test-retest reliabilities for all
individual and composite scores are in the .80s and .90s. Convergent and
discriminant validity have also been demonstrated. The clinical maladjustment
subscales of the BASC-SRP (somatization, social stress, anxiety, locus of
control, and atypicality) were used for this study, along with the depression
and sense of inadequacy subscales, which also have strong loadings on this
factor. As with the other self-report measure, the CDI, age limitations on the
BASC-SRP precluded children in the study under age 8 from completing this
measure. By employing both the CDI and BASC-SRP depression subscale, as
somewhat psychometrically different self-report measures of depressive
symptoms, this study addresses the need for multimethod assessment strategies
that promote the generalizability and stability of results
(Holmbeck, Li, Schurman, Friedman, &
Coakley, 2002).
The child's primary nurse (89 referrals, 91 nonreferrals) completed the
Pediatric Inpatient Behavior Scale (PIBS;
Kronenberger, Carter, & Thomas,
1997). The PIBS is a 47-item nurse-completed measure of a child's
behavior during medical hospitalization. PIBS items yield 10
factor-analytically derived subscales covering a variety of internalizing and
externalizing behaviors that may directly impact the child's ability to
function in the hospital setting. The PIBS has acceptable interrater
reliability, high internal consistency, and strong validity
(Kronenberger, Carter, & Thomas,
1997; Kronenberger, Causey,
& Carter, 2001). Kronenberger, Carter, and Lombird
(1999) found that 7 of the 10
PIBS subscales (oppositional-noncompliant, positive-sociable, withdrawal,
conduct problems, distress, anxiety, overactive) had very strong internal
consistency reliability and discriminant validity, leading to the
recommendation that the 3 remaining PIBS subscales be used with caution.
Kronenberger, Causey, and Carter
(2001) also used these 7
subscales to validate the PIBS in an inpatient child psychiatry sample.
Therefore, the present study employed these same 7 subscales due to their
proven psychometric properties. PIBS subscale raw scores are derived by
summing all of the items on a subscale and then dividing them by the number of
items to get an average item score. Hence, a score of 1.0 would indicate that
the average item was rated a 1. Items are rated on a 0-1-2 scale
(never-sometimes-often) of frequency of observed occurrence.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
As indicated in Table II, consistent with the matching procedure, referrals and nonreferrals did not differ significantly on age [t(206) = 0.08, p < .94], ethnicity (Fisher exact test comparison of groups on white vs. nonwhite ethnicity, p < .65), IQ [t(143) = 1.09, p < .65], or gender (groups were matched on gender and were therefore identical; Fisher exact p < 1.0). Referrals were more likely to have problems with school refusal (Fischer exact p < .001) in response to the question on the background questionnaire "Has your child ever had a problem with refusing to go to school?"
|
Characterization of Referrals and Pediatric C/L Team Activities
The source of referrals was fairly diverse, with the 104 referrals being
distributed as follows: hematology/oncology (22), general pediatrics (21),
surgery/trauma (16), pulmonology (14), rehabilitation/physiatry (7),
cardiology (5), neurology/neurosurgery (4), gastroenterology (3),
endocrinology (2), nephrology (2), infectious diseases (1), ophthalmology (1),
and all others (6). These represented all the clinical attending specialty
services and units in the hospital, with the exception of the neonatal
intensive care unit, due to age limitations imposed in the study. Nonadmitting
services (e.g., radiology, pathology, etc.) were not represented. As indicated
in Table III, the medical
diagnoses of referrals and nonreferrals reflect the pattern of the referring
specialties.
|
After talking with the referring physician by phone or face to face, the pediatric C/L team member seeing the case recorded the reasons for referral on the consult intake sheet using 17 precoded categories, including concern with parent adjustment/coping, differential diagnosis (r/o psychogenic cause), an "other" category for listing less frequent reasons for referral. Referring health care professionals often gave more than one reason for making a referral to the pediatric C/L service. The most common reasons for referral were: coping/adjustment (36), medication/treatment noncompliance (32), depression (31), pain management (16), anxiety (16), illness exacerbation (15), acute evaluation (14), new diagnosis (14), parent coping (15), ongoing treatment (13), decision making regarding treatment (10), differential diagnosis (10), acting out (8), general support (7), and family conflict (4). DSM-IV Axis I diagnoses assigned to referrals consisted primarily of adjustment disorders (55), followed by psychological factors affecting physical condition (10), major depression (9), depressive disorder NOS [not otherwise specified] (9), somatoform disorder (4), psychotic disorder due to substance abuse (3), acute stress disorder (3), eating disorder (3), mood disorder due to general medical condition (3), dysthymic disorder (2), and all other DSM diagnoses (3).
The most common interventions provided by the pediatric C/L service included supportive therapy (82 cases), coping-strategies intervention (68 cases), arranging posthospitalization follow-up (61 cases), therapeutic medical play intervention (28 cases), cognitive behavioral therapy (23 cases), improving communication within the family (17 cases), setting up a behavioral management plan (16 cases), arranging for psychotropic medication (16 cases), facilitating communication between patient/family and staff (13 cases), and grief intervention (11 cases). Many referrals received more than one form of intervention.
Parent-, Self-, and Nursing-Reported Behaviors
A multivariate analysis of variance (MANOVA) comparing referrals and
nonreferrals on parent-reported behavior problems outside of the hospital
(CBCL subscales) was statistically significant [F(8, 193) = 2.18,
p < .04]. Referrals scored higher than nonreferrals on all CBCL
subscales [T(200) > 2.04 for all subscales, p < .05
for all subscales; see Table
IV].
|
A MANOVA for self-reported behavioral problems on the BASC-SRP clinical maladjustment subscales (excluding the somatization subscale, as this subscale does not occur on all age versions of the BASC-SRP) was statistically significant [F(6, 165) = 2.41, p < .03), indicating that referrals and nonreferrals differed on the BASC-SRP clinical maladjustment subscales. Referrals scored higher than nonreferrals on BASC-SRP subscales for depression, sense of inadequacy, and social stress, but not on the anxiety, atypicality, and locus of control subscales (p > .05 for all; see Table V). On the CDI, referrals (mean T-score of 48.7) also differed from nonreferrals (mean T-score of 44.1) on the CDI total score [t(170) = 3.49, p < .001).
|
A MANOVA for nursing-reported in-hospital behavior problems (PIBS subscales) was statistically significant [F(10, 169) = 4.00, p < .001). Referrals scored higher than nonreferrals on all PIBS subscales (p < .05), with the exception of those for positive-sociable and conduct problems (p > .05 for all; see Table VI).
|
Clinical Maladjustment
To determine the relative extent to which referrals and nonreferrals
exhibited clinical levels of behavior/adjustment difficulties, clinical cutoff
scores were applied to the study measures. For the PIBS subscales, clinical
cutoff scores were defined as .5 standard deviations or higher above the mean
of the PIBS nonreferred development sample
(Kronenberger et al., 1997). A
clinical cutoff of .5 standard deviations above the mean was employed, as this
value corresponded closely to the average subscale elevation in a "high
intervention need" PIBS subsample
(Kronenberger et al.,1997). In
that sample, the high-intervention-need group scored .22 to .86 standard
deviations above the mean PIBS score reported for a general hospital sample
(.86 for oppositional-noncompliant, .64 for withdrawal, .57 for conduct
disorder, .48 for distress, .52 for anxiety, and .22 for overactive). Thus, a
value of .5 SD above the mean of the PIBS was felt to reflect a
clinical elevation in the form of a high intervention need. While this value
may seem lower than that typically used for clinical cutoffs for outpatient
behavior questionnaires, it is felt by the authors to reflect the high
baseline level of adjustment problems that is typical for hospitalized
children with serious illness or injury, as compared with what might be
expected in the lowered stress conditions of the home environment. The
positive-sociable subscale was not used to calculate the clinical cutoff,
because that subscale is a prosocial subscale that has not been shown to
reflect behavioral or adjustment problems
(Kronenberger et al., 1997).
Any participant scoring above the clinical cutoff on at least one subtest was
classified as "above clinical level." Any participant scoring
below the clinical cutoff on all subtests was classified as
"below clinical level." To determine clinical cutoffs for the
CBCL, any participant scoring at or above a T-score of 62 (90th percentile) on
any of the clinical scales was classified as "above clinical
level." Participants scoring below T = 62 on all
clinical subscales were classified as "below clinical level."
To determine clinical cutoffs on the CDI and BASCSRP, participants receiving a total T-score of 62 or above on the CDI total score or any BASC-SRP clinical maladjustment scale were classified as "above clinical level."
By nurse report on the PIBS, 47% of referrals versus 19% of nonreferrals met clinical cutoff criteria for in-hospital behavior/adjustment problems (Fisher exact p < .001). On parent report with the CBCL, 61% of referrals versus 39% of nonreferrals met clinical cutoff criteria (Fisher exact p < .01). On self-report with the CDI and BASC-SRP, 40% of referrals versus 27% of nonreferrals met clinical cutoff criteria (Fisher exact p < .10).
Goal Attainment, Helpfulness, and Satisfaction Ratings
Of the 88 referring physicians who were contacted, the mean goal attainment
rating on the 5-point scale was 4.34 (SD = .92) and the overall
helpfulness rating was 4.64 (SD = .57). The pediatric C/L team
consultants' (n = 94) mean rating on helpfulness was 3.97
(SD = .89). Parent/guardian (n = 88) mean rating for
helpful to child on the 4-point scale (1 = not helpful to 4 =
extremely helpful) was 3.45 (SD = .69), and for helpful to
family, 3.20 (SD = .77). When parents/guardians were asked
whether or not they felt the consultation aided in their child's recovery in
the hospital, 75 (85%) indicated yes and 13 (13%) indicated no.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
To our knowledge, this is the first prospective case-controlled investigation of inpatient pediatric patients referred for psychological consultation in a tertiary care pediatric hospital setting. By providing systematic, prospective psychological data on nonreferred as well as referred children from the perspective of multiple informants (nursing staff, parents/guardians, referring physicians, and patients), the results of this study build on previous work (Olson et al., 1988
; Rodrigue et al.,
1995) that has characterized the nature of cases referred for
consultation and gathered retrospective satisfaction data from referring
professionals. We further obtained goal-attainment and satisfaction ratings
from referring health care professionals, parents/guardians, and the
consultants at the time of completion of the consultation. One particular challenge of a matched control study with medically hospitalized children was to find nonreferred patients with similar medical conditions to the referred patients, in order to control for the impact of illness severity on patient functioning. Table I represents the results of these efforts, which may benefit from refinement in future studies. Consistent with the two previous studies of pediatric C/L services, the majority of inpatient referrals came from pediatric hematology/oncology, general pediatrics, and surgery (trauma). The other referring services (e.g., pediatric pulmonary medicine) reflect the specialty interests and involvement of our particular pediatric C/L service with the Louisville Cystic Fibrosis Center. Studies in other children's hospital settings will likely reflect the unique specializations of the different pediatric C/L teams.
One major limitation of the study was the primary reliance on DSM-IV psychiatric diagnostic codes. The majority of referrals were given DSM-IV diagnoses of adjustment disorders, suggesting that for most referred children, behavioral symptoms reflect difficulties in coping with the stressors and demands of various aspects of their medical condition. However, just as likely is that the assignment of these diagnoses reflects the economic need to provide a billable psychiatric diagnosis for insurance purposes. The relatively new consultation-friendly Current Procedural Terminology Health and Behavior Assessment and Intervention codes developed for psychological services to facilitate the prevention, treatment, or management of physical health problems (which are billed to the patient's medical, not mental health, benefits under their International Classification of Diseases, 9th Revision medical diagnosis) represent efforts to address the limitations and inappropriateness of assigning DSM diagnoses to these patients.
Most consultation requests were for coping/adjustment concerns, rather than
merely "acting out" or behavioral problems. For younger patients,
interventions often involved some form of medical play therapy intervention to
facilitate coping and mastery skill development. Interestingly, based on the
rating sheet data employed in this study, most consultants also saw themselves
as providing generally "supportive" interventions to a majority of
referrals. While behavioral management strategies were often an implicit part
of such interventions, more formal behavioral treatments were identified less
often on the rating form for interventions provided. The average case referred
to this particular pediatric C/L service is seen for an evaluation and 4.5
postevaluation contacts. Since Kosair Children's Hospital is a major regional
tertiary care medical center, almost half of the referred patients in this
study needed some kind of posthospital follow-up arrangement, often for
services closer to their home. Finally, a small proportion of referrals in our
study were administered psychotropic medications to assist with affective
symptoms, sleep difficulties, pain, etc. Future studies should focus on more
detailed analyses of actual consultant interventions, in line with Roberts'
(2001) call for validation
studies to build our armamentarium of evidence-based interventions in
pediatric psychology and consultation-liaison child psychiatry. Furthermore,
these studies should incorporate measures of adaptation and coping to focus
more on positive protective factors that facilitate patient and family
functioning in the face of hospitalization and disease-related stressors.
The results of this study strongly suggest that referrals for consultation from pediatricians and pediatric specialists are generally more distressed than their nonreferred hospitalized counterparts. Not surprisingly, pediatric inpatient referrals had significantly more behavioral and adjustment difficulties than nonreferred children by nurse, parent, and, to a lesser extent, self report. These included both internalizing and externalizing problems. Parent report of more global adjustment difficulties and behavior problems suggest that patient prehospital behaviors can play a significant role in predicting and identifying children that may need referral to pediatric C/L services early in their hospital stay. However, of note is the finding that a significant proportion (19%-39%, depending on rating source) of the nonreferred patients also had clinical levels of distress and maladjustment. This would suggest that frequent staff education and training (including possible use of cost- and time-effective screening instruments) are needed to sensitize health care personnel as to how to recognize when a patient may be in need of referral for consultation.
Our experience in the present study would suggest that both global behavioral ratings and hospital-specific measures play a role in assessment of pediatric inpatients. The PIBS was most useful for assessing in-hospital difficulties experienced by children, while the CBCL and BASC-SRP were sensitive to more global extrahospital behavioral difficulties. Both sources of information play an important role in helping to identify hospitalized children in need of psychological and psychiatric services.
Consistent with previous studies of pediatric C/L services, satisfaction
and goal attainment outcome findings suggest that overall, these services are
perceived as effective and valued by both referring health care professionals
and parents/guardians. While care was taken to ensure raters of the
confidentiality of their ratings, in order to encourage openness and honesty
in their assessments, the possibility remains that raters adopted a generally
positive set in responding to the request of the research assistant. Perhaps
more importantly, future studies need to focus on more long-term outcomes,
beyond the short-term goals of in-hospital consultations. This would include
assessing the need for extension of pediatric C/L services beyond the
in-hospital environment, as well as closer assessment of the long-range impact
of in-hospital interventions, patient/family follow-through on referrals to
outside professionals and agencies, etc. In light of the findings of Rodrigue
and colleagues (1995) of poor
follow-through of consultation cases in obtaining outpatient services, any
effort to systematically improve postconsultation adherence could be of great
clinical value. Health care utilization data would be valuable in this regard,
to determine if aggressive and more extensive posthospital involvement reduces
unnecessary medical utilization.
Clinical experience in a number of pediatric health care settings indicates
that family factors are a major part of both the referral concerns and the
provision of intervention in pediatric C/L cases (Kazak, 2002). The present
study did not directly investigate this dynamic, as it proved to be a
logistical challenge due to the time-consuming task of questionnaire
completion, and deferred it in order to avoid a higher rate of refusals to
participate in the study. Methods that do not rely solely upon the family
members themselves, or involve more innocuous problem-solving tasks for the
family, though more labor intensive for the investigators, might prove less
intrusive than pencil- and-paper inventories that inquire about family
dynamics and relationships. Observational and nursing-based rating scales
might also bypass family sensitivity objections, but would require careful
attention to issues of informed consent and research ethics. Future studies
need to focus more on the child in the family via the incorporation of family
measures (Knapp & Harris,
1998b).
Finally, future studies of pediatric C/L services need to be mindful of the
growing body of literature supporting the biopsychosocial model of
understanding the process of adaptation to chronic childhood illness, injury,
and the experience of hospitalization
(Thompson & Gustafson,
1996). Illness/injury and hospitalization often have a dramatic
and dynamic impact on children's interaction with their physical and social
environment, which is mediated by such factors as prevalence, age of onset,
mobility-activity, illness course, impact of illness/injury on cognitive and
sensory functioning, and visibility
(Thompson & Gustafson,
1996). These factors are just as important as symptoms and
premorbid adjustment in predicting the child's reaction to medical events and
interventions to facilitate coping and adaptation.
This study was limited by the age range of the children seen, as younger children below the age of 4 were excluded due to incompatibility and lack of measurement devices for such a broad age range. Also, this sample reflects the referral pattern of one particular pediatric C/L service and may not be applicable to services with a different focus. Additionally, because raters (parents, nurses, and patients) were not blind to the child's referral status, it is possible that some of their responses were biased by the knowledge that the child was referred to the pediatric C/L service. Finally, future studies should address the issue of early identification of hospitalized children in need of intervention. Targeting children and families "at risk" before or early in the child's hospitalization may help reduce the psychological morbidity and prevent behavioral difficulties that interfere with medical treatment and recovery. Use of instruments for routine nursing monitoring of behavior and distress among hospitalized children, much like the monitoring of physiological parameters (temperature, blood pressure, heart rate, etc.) and pain, may serve to increase hospital staff sensitivity to the child's psychological status and lead to more timely and appropriate consultation referrals.
Received June 17, 2002; revision received October 1, 2002; revision received December 13, 2002; accepted December 20, 2002
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/4-18 and 1991 profile. Burlington: University of Vermont, Department of Psychiatry.
Craighead, W. E., Smucker, M. R., Craighead, L. W., & Ilardi, S. S. (1998). Factor analysis of the Children's Depression Inventory in a community sample. Psychological Assessment, 10, 156 -165.
Drotar, D. (1995). Consulting with pediatricians: Psychological perspectives for research and practice. New York: Plenum Press.
Drotar, D., Spirito, A., & Stancin, T. (in press). Professional roles and practice patterns. In M. Roberts (Ed.), Handbook of Pediatric Psychology (3rd ed.). New York: Guilford Press.
Harris, E. S., Canning, R. D. & Kelleher, K. J. (1996). A comparison of measures of adjustment, symptoms, and impairment among children with chronic medical conditions. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1025 -1032.[CrossRef][ISI][Medline]
Holmbeck, G. N., Li, S. T., 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.
Kaufman, A. S., & Kaufman, N. L. (1990). Kaufman Brief Intelligence Test. Circle Pines, MN: American Guidance Services.
Kazak, A. E., Simms, S., & Rourke, M. T. (2002).
Family systems practice in pediatric psychology. Journal of
Pediatric Psychology, 27,
133 -143.
Kazdin, A. E. (1990). Assessment of Childhood Depression. In A. M. La Greca (Ed.), Through the eyes of the child: Obtaining self-reports from children and adolescents (pp. 189 -233). Boston: Allyn & Bacon.
Knapp, P. K., & Harris, E. S. (1998a). Consultationliaison in child psychiatry: A review of the past 10 years. Part I: Clinical findings. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 17-25.[CrossRef][ISI][Medline]
Knapp, P. K., & Harris, E. S. (1998b). Consultationliaison in child psychiatry: A review of the past 10 years. Part II: Research on treatment approaches and outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 139 -146.[CrossRef][ISI][Medline]
Kovacs, M. (1992). Children's Depression Inventory (CDI) Manual. North Tonawanda, NY: Multi-Health Systems.
Kronenberger, W. G., Carter, B. D., & Thomas, D. (1997). Assessment of behavior problems in pediatric inpatient settings: Development of the Pediatric Inpatient Behavior Scale. Children's Health Care, 26, 211-232.[CrossRef]
Kronenberger, W. G., Carter, B. D., & Lombird, T. (1999). Correspondence of the Pediatric Inpatient Behavior Scale (PIBS) scores with DSM diagnosis and problem severity ratings in a referred pediatric sample. Poster presented at the Florida Conference on Child Health Psychology, Gainesville, Florida.
Kronenberger, W. G., Causey, D., & Carter, B. D. (2001). Validity of the Pediatric Inpatient Behavior Scale in an inpatient psychiatric setting. Journal of Clinical Psychology, 57, 1421-1434.[CrossRef][ISI][Medline]
Olson, R., Holden, E. W., Friedman, A., Faust, J., Kenning, M.,
& Mason, P. (1988). Psychological consultation in a
children's hospital: An evaluation of services. Journal of
Pediatric Psychology, 13,
479-492.
Reynolds, C., & Kamphaus, R. W. (1992). BASC: Behavioral Assessment System for Children manual. Circle Pines, MN: American Guidance Service.
Roberts, M. (2001). Service delivery in child health psychology. Paper presented at the Florida Conference on Child Health Psychology, Gainesville, Florida.
Rodrigue, J. R., Hoffman, R. G., Rayfield, A., Lescano, C., Kubar, W., Streisand, R., et al. (1995). Evaluating pediatric psychology consultation services in a medical setting: An example. Journal of Clinical Psychology in Medical Settings, 2, 89 -107.[CrossRef]
Thompson, R. J., & Gustafson, K. E. (1996). Epidemiology and classification. In R. J. Thompson and K. E. Gustafson (Eds.), Adaptation to chronic childhood illness (pp. 27-56). Washington, DC: American Psychological Association.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. C. Levy, W. G. Kronenberger, and B. D. Carter Brief Report: Illness Factors and Child Behavior Before and During Pediatric Hospitalization J. Pediatr. Psychol., April 29, 2008; (2008) jsn039v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Shaw, M. Wamboldt, B. Bursch, and M. Stuber Practice Patterns in Pediatric Consultation-Liaison Psychiatry: A National Survey Psychosomatics, February 1, 2006; 47(1): 43 - 49. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||