Child psychiatry consultations to pediatric inpatients: A literature review

Critical Review and Update From time to time, the Editors of the Journal solicit articles providing a critical review and update of the work in a particular field over the last decade. Such submissions are circulated in the customary manner for prepublication review.

Child Psychiatry Consultations to Pediatric Inpatients: A Literature Review Margaret A. Shugart, M.D.

Abstract: Wlzile the prevalezzce of psychiatric disorders in clirzical pediatric populatiozzs is high, only a few studies have systematically examined child psychiatry cozzsultations to pediatric inpatients. This paper reviews this body of literature and indicates important areas for fzlrther study. Some of the studies did not report basic data such as the number of consults, age, sex, or race of referrals, arzd referralrate. Without such basic data, if is inzpossible to make adequate conzparisons between studies. Several studies did not report economic status or admission diagnosis of the child, psychiatric interz~entionltreatmerzt on the ward, or psyclziatric referral after discharge. Mostly the samples were small, including data on less tlzarz 100 inpatient consults. There is a need for greater consistency in the classification of reasozzs for referral, so that accurate azzd mearzingful conzparisozzs can be made across sites. The examinatiorz of psychiatric diagvzoses was complicated by the wide razzge of diagnostic classifications that have been used. Sfandardizatiorz on the DSM-III-R/IV diagnostic system would make for greater comparability between studies. It would be useful in planning for services if future studies irzcluded cost as well as efficacy studies witlz a randonzized control group.

Recent studies have reported community prevalence rates of childhood psychopathology ranging from 17.6% to 22.0% [l-5]. Most of these children have not received treatment for their psychiatric disorders [1,4-71. A number of studies have found that the prevalence of psychological disorders is even higher among ill than among healthy children (see [8] and [9] for reviews). For example, three independent community surveys (the Isle of Wight [7], the English National [lO,ll], and the Rochester From the Division of Child Psychiatry, Department of Psychiatry, Duke University Medical Center, Durham, North Carolina. Address requests for reprints to: Margaret A Shugart, MD, John Umstead Hospital, Butner, NC 27509. Genmzl

Hospkd

Psychiatry

13,

325-336,

1991

1Y91 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010

0

Child Health Surveys [12]) all found more psychiatric problems in chronically ill children than controls. The Ontario Child Health Study of children aged 4-16 found that the chronically ill children had rates of psychological disorders that were twice as high as the rates for healthy children [13], while in the Isle of Wight study children with epilepsy and neurological disorders had rates of psychiatric disorder three or four times the rate in the general population [7]. It is, therefore, quite clear that chronic physical disorders are associated with psychiatric disturbances in children and adolescents. The prevalence of depressive disorders in hospitalized pediatric patients and in several specialty pediatric clinical settings has also been the subject of a number of studies. Kashani et al. [14] found that 7% of hospitalized pediatric patients aged 712 were diagnosed as having Major Depression using DSM-III criteria and that these children had a significantly higher frequency of somatic complaints, headache, and abdominal pain than the nondepressed hospitalized patients. Kashani et al. [15] also studied depression in 100 children, aged 7 to 12, admitted to the hospital for orthopedic procedures, using semistructured interviews of the patient and parent as well as reviewing other data, and found 23% of these children showed evidence of depression, using DSM-III criteria. Kashani and Hakami [16] studied 35 outpatient children and adolescents with cancer, using a semistructured interview with the child and a parent as well as reviewing other pertinent data, and found 17% had a DSM-III diagnosis of Major Depression. Kashani et al. [17] studied 100 children and adolescents in an outpatient pediatric cardiology department, utilizing a semistructured interview of the child and a parent. Thirteen percent of the pa325 ISSN 0163.8343/411$3.50

M. A. Shugart

tients met DSM-III criteria for Major Depression. All four patients with isolated chest pain were found to be depressed, and they concluded that chest pain was an accompanying symptom of depression in some children. In these three studies, the prevalence of depression in the children and adolescents was higher than the 1.7%-5.9% prevalence found in the general population [l-4,6,18]. Ling et al. [19] retrospectively evaluated 25 children presenting with headache to a neurology service and found 40% had a depressive illness. None of these studies indicated whether psychiatric consults had been obtained as part of their evaluation, but it seems clear that a variety of medical presentations are associated with depression in children. A further link between pediatrics and child psychiatry is provided by the psychosomatic diseases, those physical disorders in which psychological factors play a role in the causation or severity of the somatic abnormality, such as eczema and bronchial asthma [7]. In addition, physical disorders may present with psychiatric symptoms. For instance, disorders of the brain, such as tumors and epilepsy, may present as psychiatric disorders [20,21]. In the latter cases, a psychiatrist may be involved in the treatment of the patient prior to the pediatrician. While high rates of psychopathology in medically ill children have been reported repeatedly, the majority of children with psychopathology visiting their pediatrician do not receive any psychiatric treatment. For instance, Costello [22] found that while 22% of children visiting their primary care provider for a range of pediatric services had at least one DSM-III psychiatric disorder, only 5.7% were identified by their pediatrician as being disturbed and still fewer (3.8%) were referred for any mental health services by their pediatricians. Children with psychopathology who are hospitalized on inpatient pediatric services do not necessarily receive treatment for their psychiatric disorders either. As long ago as 1972, Stocking et al. [23] prospectively examined the rate of psychopathology in a random sample of pediatric hospital admissions over a lo-month period. They found that 64% of the 80 children studied had significant psychopathology, with 10% of those psychopathologic disorders being primarily related to adapting to their medical illness and 54% apparently unrelated to their medical illness. However, the pediatric staff requested psychiatric consults on only 11% of the overall sample, or 18% of those with signif326

icant psychopathology. They noted that pediatricians were most likely to overlook emotional disorders in early infancy or in the presence of an acute, treatable medical disorder. Sack and Blocker [24] hypothesized that children whose emotional or family problems did not interfere with, or complicate, their medical care remained “hidden” from the pediatrician. Costello et al. [25] also found that 83% of psychiatric morbidity in an outpatient pediatric primary care setting was “hidden.” So the problem of under-referral seems to be equally severe across the spectrum of clinical settings. Pediatric patients referred for treatment have been reported to have more psychiatric and physical pathology than those not referred for psychiatric treatment. Awad and Poznanski [26] noted more psychopathology in 30 hospitalized children referred for psychiatric consultation than in the control group of 60 hospitalized children who were not referred. Tiller [27] noted that all of the children who were referred for consultation in his study had multi-problem (physical and psychiatric/family) conditions. Costello et al. [28] found that children identified by pediatricians as disturbed had more than twice as many physical illness episodes as nonidentified children. It appears that children with multiple problems, psychiatric and/or physical, are more likely to be identified for psychiatric consultation. However, Awad and Poznanski [26] also found that 20% of their control group, who were not referred for consultation, had relatively serious psychiatric disorders. Factors associated with referral were a longer stay in the hospital, many previous hospitalizations, older age, and more severe and ambiguous medical diagnoses. Froese [29] also found that pediatric inpatients aged 2-18 referred for psychiatric consultation were older, more likely to be female, and had longer hospital stays than those who were not referred. A similar pattern was observed in nonreferred patients with psychiatric discharge diagnoses. Despite the fact that children with medical illness have significant levels of psychopathology, little systematic research has been done in the area of child psychiatry consultations to pediatric inpatients. Several reviews noted a paucity of such work in the 1970s [20,27,29-321. Since then, relatively little has been published in this area. To my knowledge, only 14 studies have examined child psychiatry consultation patterns to pediatrics, as shown in Table 1[20,24,27,30-401. In the remainder of the paper, I will review the findings from this

Child

Table 1. Published

studies of child psychiatry

Study Bolian [ 331 Schowalter

[34]

Monnelly et al. [30]

Site

N

I/P

100

I/P

?

I/P

79

[36]

I/P

95

Sack et al. [35]

I/P

98

Sack and Blocker [24]

I/P

?

Seligman and Raub [31]

I/P + O/P

Jankowski

consultations

to inpatient

Majority

12-18 yr

lo-18

yr

2-15 yr Majority

13-16 yr

1 mo-15 Majority

yr

6-18 yr

Consultations

pediatrics Referral

Age

Psychiatry

rate

Time period

10%

14 mo

?

?

1%

11 mo

14%

11 mo

7%

12 mo

1.7%3.4%

60 mo

12-19 yr

?

24 mo

3-15 yr

3.4%

?36 mo

?

3 mo

(Iip5419) I/P

112

I/P + O/P

128 (I/P 42)

Tiller [27]

I/P

191

2%

6 mo

Wrate and Kolvin [ 321

I/P

50

?

1%’

12 mo

Jeliinek et al. [38]

I/P

72

?

?

9 mo

Fine and Tonkin [ 391

I/P

?

10%

?

Rait et al.

I/P + O/P cancer

?

12 mo

Simonds Lewis

[20] [37]

1401 UP, inpatient;

Neonatesadolescents 14 days-17

Adolescents 2-25 yr

yr

(I,?@)

O/P, outpatient

body of literature further study.

and indicate important

areas for

Referral Rates Referral rates for child psychiatry consults to inpatient pediatrics have ranged from 1% to 14% of pediatric admissions, though not all studies have reported this statistic. This range of referral rates is similar to that found in adult studies of psychiatry consultations in general hospital [41]. Three studies reported changes in referral rate over time. Tiller [27] found a rate of 2% of pediatric admissions 30% higher than the previous year. He felt that this increase in referrals was secondary to changing social attitudes, increased psychiatric consultant availability, and improved education. Bolian [33] found a relatively high referral rate (lo%), which represented a IO-fold increase in the frequency of referrals during a period in which a change occurred so that consults were done by a full-time, hospital-based child psychiatrist rather than

through the volunteer services of community child psychiatrists. He stated that the rate of referrals could be titrated by visibility on the ward, interest shown, time spent on the ward, and the relationship with the pediatric staff. On the other hand, Sack and Blocker [24] found that the referral rate decreased from 3.4% to 1.7% of pediatric hospital admissions with the development of new liaison services, which included ward management, family intervention, and a Child Life Therapy Program providing supportive contact with selected patients. These new services took care of certain needs that had previously been requested through psychiatric consultation. Sack et al. [35] felt that their relatively high (7%) referral rate in a retrospective study resulted from their outreach efforts in child abuse and failure to thrive, as well as their maintaining high visibility on the pediatric ward. Lewis [37] also noted visibility as a positive factor in increasing referral rates. Fine and Tonkin [39] reported that consultation to pediatric staff was an important aspect of their con-

327

M. A. Shugart

sultation work. It appears that when the psychiatric consultant is visible, available, and interested in evaluating referrals and working with the pediatric staff, referral rates increase. Adult studies [42,43] also have reported an increase in psychiatric consultations with more liaison contact. A number of factors have been linked with nonreferral, including opposition of families to psychiatry [27], dissatisfaction with psychiatric services [27], attitudes of physicians toward psychiatry [27,33], concern about cost [33], the power structure of the hospital allowing only physicians to order consults [33], and the fact that some psychiatric problems could be managed by pediatricians [27]. However, the Costello [22] study suggested that nonidentification is the biggest problem in nonreferral. Failure to detect psychiatric morbidity seems to be particularly apparent when psychiatric problems do not interfere with medical treatment or do not result in physical or behavioral symptoms that would attract the attention of pediatricians. Often the time necessary to conduct a psychological evaluation of the child is not available to the pediatrician, whose primary focus of treatment is acute physical problems. Furthermore, by the time it is clear that there is a psychiatric problem, the child may be about to be discharged from the hospital.

*!F The age ranges reported were quite varied (see Table 1) and, not surprisingly, depended upon whether pediatric and/or adolescent wards were examined. Several studies found a greater prevalence of depression in older compared with younger children with physical illness. This parallels findings for depression in the general population and in samples referred to psychiatric programs where there is an increase in the prevalence of depression in adolescence, particularly in females [l&44]. Kashani and Hakami [16] and Rait et al. [40], studying pediatric cancer patients, found that patients presenting with primarily depressive features were older than those with predominant anxiety. Likewise, Kashani et al. [17] found that among children with cardiovascular symptomatology the mean age of depressed children was higher than that of nondepressed children. In Tiller’s study [27], affective disorders predominated in the group of children over 14 years old, where they were present in 13 out of the 14 patients.

328

Sex of Referrals Several studies [20,27,30,31,33] have found an unexpected predominance of females given the base rates of admission to inpatient pediatric wards. Tiller [27] and Bolian [33] both found the male to female referral ratio was approximately l:l, compared to the hospital admission ratio of 3:2. Sack et al. [35] found that as many females as males were referred, and Monnelly et al. [30], Seligman and Raub [31], and Wrate and Kolvin [32] all found a predominance of female referrals with M:F ratios of l:l.l, 1:1.6, and 1:1.4 respectively; however, these studies did not report on pediatric admission ratios. This is in marked contrast to the predominance of males in child psychiatry clinics and hospitals [5,32,33,45]. Suggested explanations of the relative excess of female referrals have included a greater social acceptance of somatization [33] and emotional problems [27] in females, a reluctance to diagnose emotional problems in males [27], a preponderance of female staff who might be more sensitive to female psychological issues [33], and tolerance of psychiatric symptoms in boys [20]. Another possibility that deserves attention is that depressed girls might be more likely to present to the pediatrician with somatic symptoms because of a greater prevalence of somatization in females. The relative excess of female referrals is also found in adult studies [41].

Race Few studies have reported the race of referrals, but in those that did [20,31,33,40], all found the majority of referrals to be Caucasian (65%-94%) regardless of the percentage of admissions by race. Bolian [33] and Simonds [20] found that the racial mix was consistent with that of admissions, while Seligman and Raub [31] found that racial mix of referrals was inconsistent with that of admissions; 70% of the hospital population was black yet only 35% of the psychiatrically referred population was black.

Economic

Status

Bolian [33] and Simonds [20] reported the centage of total consults and admissions by nomic status. In Bolian’s study [33], “nonprivate” category included patients with care as well as those who were billed only

perecothe free the

Child Psychiatry Consultations

Table 2. Admission percentage

diagnoses by organ systemof patients with diagnosis

Organ system CNS GI Endocrine Other

Monnelly et al. [30]

Sack et al. [35]

33 22 15 30

36 28 5 31

amount of insurance coverage available, since insurance coverage seldom allowed for psychiatric “Private” patients were billed full consultation. fees. He found the percentage of “nonprivate” pediatric admissions (42%) was comparable with the percentage of “nonprivate” consultations (50%), as did Simonds [20] with 54% of consults in the categories of “part pay” or “no pay.“ Bolian reported that these data reflected a situation in which billing was kept to a minimum since he was a salaried employee as a consultant; he considered that cost was a major factor in the decision to request psychiatric consultation.

Admission

Diagnoses

Adult studies of psychiatric consultation to inpatient wards have found CNS disease, GI disease [42,46], or both [47-491 as frequent admission diagnoses of the referred patients. Only two child studies [30,35] reported the percentage of admission diagnoses by organ system and both found CNS and GI diagnoses to be the most common (Table 2). Simonds [20] found seizure disorders (18%) and diabetes (10%) to be the most frequent physical disorders. Bolian [33], Tiller [27], and Sack and Blocker [24] all reported an excess of medical compared with surgical referrals, as have a number of adult studies [41,47,48,50]. Sack and Blocker [24] found that over 90% of the referrals were from the medical service, while Bolian [33] found medical referrals more than five times as frequent as surgical referrals in a hospital in which the number of surgical and medical admissions was about equal. Tiller [27] found a predominance of medical compared with surgical referrals of more than 2:l. Bolian [33] suggested some reasons for the excess of medical referrals: surgical admissions involve shorter lengths of stay (LOSS) and possibly the assumption that the problem will go away with surgery, while medical ad-

missions involve often chronic disorders, treatment of the whole child, and difficult differential diagnoses.

Reason for Referral The examination of reasons for referral is complicated by the heterogeneity of the classification schemes used by different authors. Frequently cited reasons for referral reported from several studies in which they could be compared are given in Table 3 as percentage of referrals. Several studies [20,24,30,31,35,37,38,40] found that the most frequent reasons for referral included symptoms of unknown etiology (diagnostic problems), behavior problems (including ward management of behavior problems), and suspected depression. Sack and Blocker [24] found that these reasons and the presence of a parent/family problem were the most common reasons for referral in 1973-1974; however, in 1976-1977, when other new services were available for ward management and family intervention, consultations were requested mainly for diagnostic clarification. Tiller’s study [27] noted that the stated reason for referral was often a clinical medical diagnosis, but that over half of the referred children actually had behavior disorders or family social problems. In the absence of behavior or family problems, he found that it was sometimes emotions that the patient aroused in staff members that led to the referral. Sack et al. [35] found family disorder in over half of

the children referred for consultation; this was twice as common as in the control group of pediatric inpatients who were not referred for psychiatric consultation but were matched for age, sex, socioeconomic status, and need for surgical or medical services. Seligman and Raub [31] found that frequent reasons for referral from an adolescent unit also included a plea for help (37%) and hysteria (16%). Both Rait et al. [40] and Jellinek [38] found suicidal risk among the frequent reasons for referral, 12% and 14% respectively. Jellinek [38] found the most frequent reason for referral was concern about the child’s reaction to illness and hospitalization (38%). Wrate and Kolvin [32] found that the majority of children were referred on the basis of positive evidence of psychological disturbance (86%) while 14% were referred because there was no evidence of an organic basis for symptoms. Hengeveld et al.[41] found that the reasons for

329

M. A. Shugart

Table 3. Frequent

Study

reasons

for referral-percentage

Symptoms of unknown etiology/diagnostic problem

of patients

Suspected depression

with reason

Behavior problems/ward management

for referral

Family problems

Other

Monnelly et al. [30]

30

19

13

9

29

Sack et al. [35]

17

14

29

-

40

22 64

14 3

29 18

23 3

12 12

Seligman and Raub [311

5

16

11

-

68

Simonds [20]

31

11

19

-

39

Rait et al. [40]

5

24

26

9

36

Sack and Blocker 1241 1973-1974 1976-1977

referral also varied widely in adult studies. The median percentage of referrals was 32% for assessment following suicide attempt, and was 18% for probable psychogenesis of physical symptoms. Diagnosis and management were the most frequent reasons for referral in some adult studies [4648,501. Karasu et al. [51] found depression most frequent, and Sobel et al. [49] found suicidal behavior most frequent. These reasons for referral (diagnosis, management, depression, and suicidal risk) parallel the findings in child studies.

Psychiatric Diagnoses Reasons for referral are often closely related to the psychiatric diagnosis of the referred patient. A wide range of diagnostic classifications have been used, reflecting the dates of these studies, during which at least three official diagnostic classification systems have been used (DSM-II, GAP, DSM-III). Comparisons of psychiatric diagnoses as percentages of referrals, where such comparisons are possible, are shown in Table 4. Hysterical reactions, depression, hyperactive child syndrome, and organic brain syndrome were frequent diagnoses in three studies [30,35,37]. Affective disorders predominated in the 14 patients over 14 years of age in Tiller’s study [27]. Rait et al. [40] found adjustment disorder with depressed mood alone accounted for 31% of the diagnoses of the cancer patients in their study, and depression for another 15%. Reactive or adjustment disorders were frequent diagnoses in several studies [20,27,36,37,40], as were personality disorders [20,30,35,36]. Diagnoses found by Lewis [37] in the general pediatric 330

ward included depression, unsocialized aggressive reaction, immaturity, stress reaction, and learning disorder. Some studies gave descriptive classifications, such as Wrate and Kolvin [32] who found somatic manifestations of psychological disturbance (34%) and suicide attempts (14%) were most frequent. Psychological reactions to physical disorders (32%), psychological reactions with associated changes in bodily functions (20%), and psychological reactions leading to self-inflicted injury or illness (19%) were the most common diagnostic groups in Bolian’s study [33]. He commented on the high incidence of self-inflicted injury and emotional sequelae of chronic disease. These diagnostic categories are similar to those seen by Schowalter [34] in patients referred for consultation who described the following major categories: physical disorders resulting in great psychological stress; illness triggered or worsened by emotional upset; and physical symptoms due to malingering, depressive neuroses, or anxiety neuroses. Depression and organic brain syndrome were also frequent psychiatric diagnoses found in psychiatric consultation referrals in several adult studies [41,42,46-511. The rate of depression reported in comparable adult studies was 52%-57% of referrals [48,52-541, a good deal higher than the rate reported in child studies. Studies of the prevalence of depression in adult medical inpatients have noted that 22%-26% were significantly depressed [50,55-571. Two adult studies [50,58] report consistency of psychiatric diagnoses over time in consultation referrals. Depressive disorders and organic brain syndromes predominated consis-

Child Psychiatry

Table 4. Psychiatric

diagnoses-percentage Monnelly et al. [30]

Diagnosis Hysterical reaction Hyperactive syndrome

child

Impulse disorder Depression Organic brain syndrome Mental retardation Other personality disorders Anorexia nervosa

Consultations

of patients with diagnosis Jankowski

Sack et al.

Simonds

]361

]351

WI

19

5

12

-

Lewis [37] (adolescents)

Tiller [27]

Rait et al. [40]

10 -

13

-

9 9 5

6 4 -

10 18 7 9

3 1

8 4

29 -

11 2

16 -

2 1

6

2 2

-

-

10 15 5

-

-

5

14 13 3 -

15 14

-

Anxiety reactions, neurosis, disorders Enuresislencopresis Adjustment reactions Reactive disorder Psychoneurotic disorders Developmental deviations Undiagnosed or other

-

psychiatric disorder No psychiatric illness

11 19

tently [50]. To my knowledge, have reported this information.

10 25

-

-

59 2 1

3 24

1 18

9 6

no child studies

Length of Hospital Stay The average length of pediatric hospitalization for psychiatrically referred children was 14 days in the Monnelly et al. [30] and Sack et al. [35] studies. Both studies found that this was 5 days longer than the mean LOS for other pediatric patients. The average LOS was 22 days in the Tiller [27] study, more than three times the average total LOS for all pediatric patients. Tiller also found the mean LOS prior to referral to be 11 days, again more than the average total hospital LOS for all pediatric inpatients. He interpreted these data as reflecting an increased probability of referral with increased length of hospitalization. Perhaps difficulties with treatment and patient management resulted in longer LOS and subsequent referral. Both Awad and Poznanski [26] and Froese [29], in comparing referred and nonreferred pediatric inpatients, found that referred patients had a longer hospital LOS than nonreferred patients, with mean LOS of 11 and 8 days respectively in the Froese [29] study. Froese [29] also found that nonreferred patients discharged with psychiatric

-

5

10 -

10 30

12 39 12 12 6 2

52 -

5 2

diagnoses had hospital LOSS similar to the overall referred group, with a mean LOS of 11 days. However, the group of referred patients discharged with psychiatric diagnoses had a shorter mean hospital LOS, 9 days. Therefore, he felt that psychiatric referral in this group of patients hastened discharge, and that the longer LOSS for the overall referred group was related to the nature of their illnesses rather than being referred solely after no organic etiology had been found. Likewise, Froese [59] found that the mean hospital LOS for those patients referred with chronic illness was significantly longer than the mean LOS for those referred with acute illness. In addition, he found chronicity of illness correlated with referral delay. Awad and Poznanski [26] also stated that the longer hospital LOSS for the referred group in their study may have resulted from the chronicity of their illness.

Intervention,

Treatment,

and Referral

Tiller [27] found that hospital treatment resources were used by 88% of the child consultation patients who were referred after the initial assessment. Wrate and Kolvin [32] found that a treatment recommendation was made in 72% of the inpatient child consultations. Psychotropic medication was 331

M. A. Shugart

recommended in 4%-l 1% of referrals [20,27,30,36]. Monnelly et al. [30] found that no patients were transferred to inpatient psychiatric treatment. Jankowski [36] and Simonds [20] recommended inpatient psychiatric treatment in 7% and 12% of consultations respectively. Outpatient psychiatric treatment was recommended in 26%-40% of consultation referrals [20,30,36]. Simonds [ZO]also recommended counseling by a social worker in 45% of cases. Hengeveld et al. [41] found 9%-38% (median = 27%) of adult referrals received psychiatric treatment on the inpatient ward. Other adult studies have reported from 11% to 53% of the referrals received continued psychiatric care after the initial consult [41,48,50,60]. Adult studies have also noted much higher rates of psychotropic medication (14%-75%; median = 38%) than in child studies [41,46]. Hengeveld et al. [41] found 5%-31% (median = 13%) of adult referrals were subsequently admitted to a psychiatric inpatient ward, lo%-48% (median = 29%) of adult referrals to an outpatient psychiatric clinic, and 2%-15% (median = 8%) of referrals to a social worker.

Outcome and Follow-up Studies Few studies have addressed follow-up after discharge from the hospital and even fewer have reported on efficacy of treatment. Stocking et al. [23] conducted home visits with mother and child 6 months after discharge, to assess whether initial screening had resulted in unwarranted recording of the presence of pathology, and they determined that in no case did this occur. Sack et al. [35] sent a follow-up questionnaire at l-2 years after discharge, with a 54% return rate. They found that almost half of the “disordered” families in their study who responded had a change in family living patterns after hospitalization. The disordered group was less likely to follow through with counseling but more likely to have contact with social agencies. Simonds [20] reported that 55% of those referred for further psychiatric treatment received the recommended services. Wrate and Kolvin [32] found that parental cooperation in attending most or all subsequent appointments was poor, with only 53% attending regularly for treatment. Burstein [61] reported overall noncompliance rates for outpatient visits (psychotherapy and/or medication) following psychiatric consultation in a general hospital for adult patients was 49% initially, 62%

332

and 66% within three within two sessions, sessions. Seligman and Raub’s [31] O-24-month outcome follow-up of both inpatients and outpatients revealed improvement in symptoms and functional adjustment in the majority (52%) of adolescents. The remainder had no change (26%), deterioration (4%), or no follow-up (18%). Jankowski [36] obtained follow-up data 6-16 months after discharge by telephone interviews of the patient and a familiar adult, and found improvement of function in 55% of patients, no change in 35%, and worsened function in 7%, while 3% of patients were lost to follow-up. Popkin et al. [62] developed the ConsultationLiaison Outcome Evaluation System (CLOES) to address the lack of outcome data and evaluate the effectiveness of consultations on an adult consultation service. They examined three areas; psychotropic medication recommendations, diagnostic action recommendations, and verbatim or near verbatim representations of the consultants’ psychiatric diagnoses in the discharge summary. Consultees’ concordance with consultants’ recommendations ranged from 43%, in the diagnostic representation study, to 63%, in the psychotropic medication study. To my knowledge, there have been no randomized controlled trials to assess efficacy of child psychiatry consultations to pediatric inpatients.

Discussion While the prevalence of psychiatric disorders in clinical pediatric populations is high, only a few studies have systematically examined child psychiatry consultations to inpatient pediatrics, and some of these studies did not report basic data such as the number of consults, age, sex, or race of referrals, and referral rate. Without such basic data, it is impossible to make adequate comparisons between studies. Several studies did not report economic status, admission diagnosis, psychiatric intervention/treatment on the ward, or psychiatric referral after discharge. Mostly the samples were small, including data on less than 100 inpatient consults. It is essential to document age and sex of referrals in order to determine whether there are any age or sex trends in patterns of referral or in psychiatric diagnoses. This is important because age and sex significantly affect the rates of psychiatric

Child

disorder in the general population. For instance, depression is more common in adolescents than in children, and more common in adolescent females than adolescent males [18]. It seems that a similar trend is discernable from the consultation literature also. However, the reasons for the overall excess of girls in consultation samples contrast sharply with what is found in general population studies or other child psychiatric treatment settings. In order to compare studies across settings or changes over time in the same setting, it is essential to know not only the number of referrals but also the referral rate as a percentage of pediatric admissions. Changes in referral rates must be documented, but we also need more formal evaluation of the possible causes of these changes. To date, our understanding of the reasons for increased referral rates is based on a few impressionistic reports. Examining changes in referral rates and the reasons for such changes would be helpful in designing improved inpatient psychiatric consultation services for children. While some studies report more psychopathology in patients who are referred than in those not referred, this is not always the case. For example, Costello and Janiszewski’s findings [63] suggested that many nontreated children are no less impaired than those who receive psychiatric treatment. Clearly the reasons for referral are complex and require further study in order to ensure that more children in need of psychiatric treatment enter the mental health system and receive such treatment. Even simple procedures like having the referring physician complete a brief problem checklist would help in quantifying the real reasons for referral. There is also a need for greater consistency in the classification of reasons for referral, so that accurate and meaningful comparisons can be made across sites [41,64]. In general, assistance with diagnosis, management, and behavior problems are frequently cited reasons for referral and occur when psychiatric problems interfere with medical diagnosis or treatment. Consultations are also more commonly requested if parent or family problems interfere with the development or medical treatment of the child. Clearly educating pediatricians about child psychopathology and available treatment methods is essential if we are to improve the detection of psychiatric disorders. Standardization on the DSM-III/III-R/IV diagnostic system will make for greater comparability between studies. Also, psychiatric classification

Psychiatry

Consultations

would be more comparable across settings if wellrecognized standardized interviews were employed in assessing patients. While a few studies have used semistructured interviews of the child and/or parent, which particular semistructured interview was used was not reported [14-17,231. With repeated interviews, treatment efficacy could be more accurately studied. Krener and Simmons [65] developed the Child Consultation Rating Scale for assessing psychopathology in hospitalized pediatric patients and their families in order to improve detection, monitoring, and teaching of child and family psychological functioning. They reported high interrater reliability and that repeated measurements, based on clinical evaluations of the child and family, would be helpful in evaluating treatment efficacy. It would be useful in planning for services if future studies included cost as well as efficacy studies examining hospital LOS, total cost of inpatient treatment, type of payment for treatment, and cost of follow-up treatment. Both hospital LOS and economic status are important factors involved in costeffectiveness and cost containment. Hales [66] reviewed studies of the benefits of adult psychiatric consultation-liaison services in a general hospital and reported decreased hospital LOSS of medical and surgical patients, decreased utilization of medical services, and decreased mortality resulting from consultation-liaison services. Although only a few studies addressed the pediatric hospital LOSS, this is a critical factor in the cost of the total treatment of the child. It might be that earlier psychiatric consultation would result in a decreased hospital LOS and/or in decreased total treatment cost. This might be particularly true with regard to extensive, expensive diagnostic testing for patients with somatization or conversion disorders, or depression. In addition, if the need for psychiatric consultation were recognized early, rather than after the child had already been in the hospital for the usual pediatric hospital LOS, the total length of hospital stay might be decreased. If indicated, patients could be referred to day hospital or outpatient treatment and discharged from the hospital more quickly. Stocking et al. [23] examined the necessity for pediatric hospitalization and found the frequency of unnecessary hospitalization in a random sample of pediatric admissions was rated as 30% by a child psychiatrist and 12.5% by a pediatrician. In the group of children with hospitalization rated as unnecessary, there was a high

333

M. A. Shugart

incidence (89%) of psychopathology. Houts et al. [67] reviewed 572 pediatric inpatient charts and estimated that 10.5% of the cases had one or more hospital admissions that could have been partly or fully preventable by psychiatric treatment. Sack [45] found that the pediatric hospital served as a mental health facility for many children and their families who did not use other portals of entry into the health care system. While only Bolian [33] and Simonds [20] addressed economic status, Bolian [33] felt that cost was a significant factor in requesting consultations. Certainly, the cost of consultations and the availability of insurance coverage for this service are important aspects of planning and providing necessary services for pediatric patients. Once again, formal evaluation of the barriers to adequate psychiatric care being requested and provided is urgently needed, in a field where only opinions and impressions are available as yet. Likewise, treatment interventions and services provided during hospitalization as well as referral and recommended treatment after hospitalization must be documented in order to plan and make available personnel and funding for these treatment services, to plan for the education of pediatricians and mental health clinicians, and to assess the efficacy of treatment. Follow-up studies of treatment outcome would also be helpful in documenting the cost effectiveness of psychiatric consultation. There is helpful literature (Von Korff et al. [68], Mitchell and Thompson [69], Pincus [70], and Lyons et al. [71]), on the design of studies of consultation-liaison services, but few of its recommendations have been incorporated into studies of pediatric consultation-liaison. In short, we need systematic studies of consultation to pediatric inpatients in which at least the following information is collected: basic demographic information; referral rate as a percentage of pediatric admissions; standard categorization of the reasons for referral; standardized categorization of the admission diagnosis; standardized psychiatric diagnosis using the current DSM and/or ICD classification system and structured diagnostic techniques; standard categorization of treatment interventions and further referral for psychiatric care; and standardized outcome measures. Information from systematized studies of psychiatric consultation to inpatient pediatrics will ultimately improve the quality of care for children and adolescents with psychiatric disorders. Information obtained from such studies is vital in plan334

ning for necessary services in the most costefficient manner. In addition, this information is pertinent to the education of pediatric and mental health clinicians and to our overall understanding of children’s psychiatric problems and needs. The author thanks Dr. Adrian Angold for his comments and critical reading of the manuscript, Dr. James Lee for his initial contribution to this work, and Ms. Martha Patrick and Ms. Mildred Crabtree for their editorial assistance.

References 1.

2.

3.

4.

5.

6.

7. 8.

9. 10. 11. 12.

13.

14.

15.

Anderson JC, WilliamsS, McGee R, Silva PA: DSMIII disorders in preadolescent children. Arch Gen Psychiatry 44:69-80, 1987 Bird HR, Canino G, Rubio-Stipec M, Gould MS: Estimates of prevalence of child maladjustment in a community survey in Puerto Rico. Arch Gen Psychiatry 45:1120-1126, 1988 Velez CN, Johnson J, Cohen P: A longitudinal analysis of selected risk factors for childhood psychopathology. J Am Acad Child Adolesc Psychiatry 28:861-864, 1989 Costello EJ, Costello AJ, Edelbrock C, et al: Psychiatric disorders in pediatric primary care: Prevalence and risk factors. Arch Gen Psychiatry 45:1107-1116, 1988 Offord DR, Boyle MH, Szatmari I’, et al: Ontario Child Health Study: II. Six-month prevalence of disorder and rates of service utilization. Arch Gen Psychiatry 44:832-836, 1987 Costello EJ: Developments in child psychiatric epidemiology. J Am Acad Child Adolesc Psychiatry 28:836-841, 1989 Rutter M, Tizard J, Whitmore K: Education, Health, and Behavior. London, Longman, 1970 Pless IB, Roghmann KJ: Chronic illness and its consequences: Observations based on three epidemiologic surveys. J Pediatr 79:351-359, 1971 Barbarin OA: Adjustment to serious childhood illness. Adv Clin Child Psycho1 13:377-403, 1990 Douglas JWB, Blomfield JM: Children Under Five. London, Allen and Unwin, 1958 Douglas JWB: The Home and the School. London, Mackgibbon and Kee, 1964 Roghmann KJ, Haggerty RJ: Rochester Child Health Surveys: I. Objectives, organization and methods. Med Care 8~47, 1970 Cadman D, Boyle M, Szatmari I’, Offord DR: Chronic illness, disability, and mental and social well-being: Findings of the Ontario Child Health Study. Pediatrics 79:805-813, 1987 Kashani JH, Barber0 GJ, Bolander FD: Depression in hospitalized pediatric patients. J Am Acad Child Psychiatry 20:123-134, 1981 Kashani JH, Venzke R, Millar EA: Depression in children admitted to hospital for orthopaedic procedures. Br J Psychiatry 138:21-25, 1981

Child Psychiatry

16. Kashani J, Hakami N: Depression in children and adolescents with malignancy. Can J Psychiatry 27474-477, I982 17. Kashani JH, Lababidi Z, Jones RS: Depression in children and adolescents with cardiovascular symptomatology: The significance of chest pain. J Am Acad Child Psychiatry 21:187-189, 1982 18. Angold A: Childhood and adolescent depression: I. Epidemiological and aetiological aspects. Br J Psychiatry 152:601-617, 1988 19. Ling W, Oftedal G, Weinberg W: Depressive illness in childhood presenting as severe headache. Am J Dis Child 120:122-124, 1970 20. Simonds JF: Psychiatric consultations for 112 pediatric inpatients. South Med J 70:980-984, 1977 21. Graham I’J: Specific medical syndromes. In Rutter M (ed), Developmental Neuropsychiatry. New York, The Guilford Press, 1983, pp. 68-82 22. Costello EJ: Child psychiatric disorders and their correlates: A primary care pediatric sample. J Am Acad Child Adolesc Psychiatry 28:851-855, 1989 23. Stocking M, Rothney W, Grosser G, Goodwin R: Psychopathology in the pediatric hospital: Implications for community health. Am J Public Health 62:551-556, 1972 24. Sack WH, Blocker DL: Who gets referred? Child psychiatric consultation in a pediatric hospital. Int J Psychiatry Med 9:329-337, 1978 25. Costello EJ, Edelbrock C, Costello AJ, Dulcan MK, Burns BJ, Brent D: Psychopathology in pediatric primary care: The new hidden morbidity. Pediatrics 82:415-424, 1988 26. Awad GA, Poznanski ED: Psychiatric consultations in a pediatric hospital. Am J Psychiatry 132:915-918, 1975 27. Tiller JW: Psychiatric consultations in a children’s hospital. Med J Aust 1:532-535, 1978 28. Costello EJ, Burns BJ, Costello AJ, Edelbrock C, Dulcan MK, Brent D: Service utilization and psychiatric diagnosis in pediatric primary care: The role of the gatekeeper. Pediatrics 82:435-441, 1988 29. Froese AI’: Pediatric referrals to psychiatry: I. Comparison of referrals and non-referrals. Int J Psychiatry Med 7:241-247, 1977 30. Monnelly El’, Ianzito BM, Stewart MA: Psychiatric consultations in a children’s hospital. Am J Psychiatry 130:789-790, 1973 31. Seligman R, Raub JL: Psychiatric consultation in a medical setting for adolescents: Descriptions of one modus operandi. Clin Pediatr (Phila) 13:117-120, 1974 32. Wrate RM, Kolvin I: A child psychiatry consultation service to pediatricians. Dev Med Child Neurol 20:347-356, 1978 33. Bolian GC: Psychiatric consultation within a community of sick children: Lessons from a children’s hospital. J Am Acad Child Adolesc Psychiatry 10:293-307, 1971 34. Schowalter JE: The utilization of child psychiatry on a pediatric adolescent ward. J Am Acad Child Psychiatry 10:684699, 1971

Consultations

35. Sack W, Cohen S, Grout C: One year’s survey of child psychiatry consultations in a pediatric hospital. J Am Acad Child Psychiatry 16:716-727, 1977 36. Jankowski JJ: Clinical child psychiatry services on a pediatric-adolescent unit. J Am Acad Child Psychiatry 13:95-109, 1974 37. Lewis MB: Child psychiatric consultation in pediatrics. Pediatrics 62:359-364, 1978 38. Jellinek MS, Herzog DB, Selter LF: A psychiatric consultation service for hospitalized children. Psychosomatics 22:29-33, 1981 39. Fine SH, Tonkin RS: Limited psychiatric consultation to an adolescent medicine unit. Child Psychiatry Hum Dev 13:48-54, 1982 40. Rait DS, Jacobsen PB, Lederberg MS, Holland JC: Characteristics of psychiatric consultations in a pediatric cancer center. Am J Psychiatry 145:363-364, 1988 41. Hengeveld MW, Rooymans HGM, Vecht-van den Bergh: Psychiatric consultations in a Dutch university hospital: A report on 1814 referrals, compared with a literature review. Gen Hosp Psychiatry 6:271279, 1984 42. Kramer BA, Spikes J, Strain JJ: The effects of a psychiatric liaison program on the utilization of psychiatric consultations. Gen Hosp Psychiatry 1:122128, 1979 43. Torem M, Saravay SM, Steinberg H: Changes in referrals for psychiatric consultation on two medical wards as a result of intensive liaison work. Unpublished material, 1978 44. Angold A: Childhood and adolescent depression: II. Research in clinical populations. Br J Psychiatry 153:476-492, 1988 45. Sack WH: Making the “second diagnosis”: A retrospective analysis of child psychiatry consultations in a pediatric hospital. Clin Pediatr (Phila) 17:548550, 1978 46. Malhotra S, Malhotra A: Liaison psychiatry in an Indian general hospital. Gen Hosp Psychiatry 6:266270, 1984 47. Tuason VB, Rhee YW: Psychiatric disorders in patients of a general hospital. South Med J 65:408-412, 1972 48. Shevitz SA, Silberfarb PM, Lipowski ZJ: Psychiatric consultations in a general hospital: A report on 1,000 referrals. Dis Nerv Sys 37:295-300, 1976 49. Sobel SN, Munitz H, Karp L: Psychiatric consultations in two Israeli general hospitals. Gen Hosp Psychiatry 10:298-304, 1988 50. Lipowski ZJ, Wolston EJ: Liaison psychiatry: Referral patterns and their stability over time. Am J Psychiatry 138:1608-1611, 1981 51. Karasu TB, Plutchik R, Steinmuller RI, Conte H, Siegel B: Patterns of psychiatric consultation in a general hospital. Hosp Community Psychiatry 28291-294, 1977 52. Kligerman MJ, McKegney FP: Patterns of psychiatric consultation in two general hospitals. Psychiatry Med 2:126-132, 1971 53. West ND, Bastani JB: The pattern of psychiatric referrals in a teaching hospital. Nebr Med J 15:438440, 1973

335

M. A. Shugart

54. Poe RO, Lowell FM, Fow HM: Depression: Study of 100 cases in a general hospital. JAMA 195345-350, 1966 55. Schwab JJ, Clemmons RS, Freemon FR, Scott ML: Differential characteristics of medical in-patients referred for psychiatric consultation: A controlled study. Psychosom Med 27:112-118, 1965 56. Moffic HS, Paykel ES: Depression patients. Br J Psychiatry 126346-353,

in medical 1975

in-

57. Lipowski ZJ: Consultation-liaison psychiatry: Past failures and new opportunities. Gen Hosp Psychiatry 1:3.-10, 1979 58. Wilson MS, Meyer E: Diagnostic consistency in a psychiatric liaison service. Am J Psychiatry 119:207209, 1962 59. Froese Al’: Pediatric referrals to psychiatry: II. Factors relating to delays in referral. Int J Psychiatry Med 7:249-254, 1977 60. Karasu TB: Utilization of a psychiatric consultation service. Psychosomatics 19:467-473, 1978 61. Burstein A: Psychiatric consultations in a general hospital: Compliance to follow-up. Gen Hosp Psychiatry 6:139-141, 1984 62. Popkin MK, Mackenzie TB, Callies AL: Consultliaison outcome evaluation system. Arch Gen Psychiatry 40:215-219, 1983

336

63. Costello EJ, Janiszewski S: Who gets treated? Factors associated with referral in children with psychiatric disorders. Acta Psychiatr Stand 81:523-529, 1990 64. Lipowski ZJ: Review of consultation psychiatry and psychosomatic medicine: II. Clinical aspects. Psychosom Med 29:201-224, 1967 65. Krener I’, Simmons MK: The child consultation rating scale. Int J Psychiatry Med 19:23-29, 1989 66. Hales RE: The benefits of a psychiatric consultationliaison service in a general hospital. Gen Hosp Psychiatry 7214-218, 1985 67. Houts CB, Turbett JA, Arnold LE, Kruse E: Costs of medical/surgical pediatric hospital days preventable by psychiatric treatment. J Am Acad Child Psychiatry 24~227-230, 1985 68. Von Korff M, Katon W, Lin EHB, Wagner EH: Evaluation of psychiatric consultation-liaison in primary care settings. Gen Hosp Psychiatry 9:102-110, 1987 69. Mitchell WD, Thompson TL: Some methodological issues in consultation-liaison psychiatry research. Gen Hosp Psychiatry 7:66-72, 1985 70. Pincus HA: Making the case for consultation-liaison psychiatry: Issues in cost-effectiveness analysis. Gen Hosp Psychiatry 6:173-179, 1984 71. Lyons JS, Hammer JS, Wise TN, Strain JJ: Consultation-liaison psychiatry and cost-effectiveness research: A review of methods. Gen Hosp Psychiatry 7:302-308, 1985