Archives of Disease in Childhood 1991; 66: 862-865

862

Referrals to child psychiatry-a

staff attitudes

survey

of

S Oke, R Mayer

Abstract
cians' questionnaire asked paediatricians how
A questionnaire study was conducted in a often they would refer to child psychiatry a
health district to evaluate the attitudes of range of clinical problems seen as inpatients and
paediatricians and child psychiatry staff as to outpatients. Part 1 of the child psychiatry queswhich categories of problems should be refer- tionnaires asked child psychiatry staff how often
red to child psychiatry. In the majority of they think paediatricians should refer these
categories the two groups disagreed as to the problems to the child psychiatry department.
frequency with which the problem should be
The clinical problems were selected for inclureferred. In the categories relating to child sion after a review of the literature"8 and after
sexual abuse responses were often not in discussions with our colleagues. Our list was not
accord with Department of Health and Social exhaustive but represented most of the common
Security guidelines. Reasons for not referring reasons for referral to child psychiatry.
were also looked at and again it was found that
Respondents were asked to indicate how frethere were a number of significant differences quently they felt a category of problem should
in opinion as to what are reasons for not refer- be referred by answering 1-4 where l=rarely or
ring to child psychiatry. Both groups agree never, 2=sometimes, 3=frequently, and 4=
that lack of communication is a reason for always or nearly always. The categories of probnon-referral. Some suggestions are made as lems are shown in table 1.
to how this problem could be addressed.
The association between child psychiatry and
paediatrics has been widely commented on during the past 30 years.1-5 We wished to explore
whether locally there was agreement between
disciplines as to the appropriateness of referral
of various clinical problems and to look at factors that might lead to not referring.
Method
The study was a survey, using specially
designed questionnaires, of paediatricians (all
grades) and child psychiatry staff (psychiatrists,
clinical psychologists, child psychotherapists,
and social workers) in an inner London health
district. The decision to survey only medical
staff in paediatrics but all professional child
psychiatry staff reflects the fact that whereas
referral, although influenced or initiated by
other professions, is primarily a medical decision, the child psychiatry departments assess
and treat on multidisciplinary lines.

St Mary's Hospital and

Paddington Green
Department of
Child Psychiatry,

THE QUESTIONNAIRES
Two questionnaires were used, one for paediatricians and one for child psychiatry staff. They
varied slightly in format, but were identical in

content.

Both questionnaires consisted of

two

London
S Oke
R Mayer

parts.

Correspondence to:
Dr S Oke,
Wonford House Hospital,
Dryden Road,
Wonford,
Exeter EX2 5AF.
Accepted 28 February 1991

Part I
The aim of part 1 was to determine whether
paediatricians and child psychiatrists agree as to
which type of clinical problem should be referred to child psychiatry. Part of the paediatri-

Part 2
This section looked at factors that might negatively influence referral. The list of possible
reasons we used was compiled after a review of
the literature and after discussions with our
colleagues. These possible reasons were given in
the form of a list of statements and the respondents were asked to indicate 'yes' or 'no' to each
of these depending on whether or not they
thought the reason relevant in the decision not
to refer to child psychiatry. These statements
are shown in table 2.
In this part of the questionnaire we were
asking paediatricians direct attitudinal questions, but the child psychiatry staff were being
asked to speculate whether they felt certain factors affected the paediatricians' decision to
refezr. Both parts of the questionnaires included
a section inviting further comments from the
respondents. Questionnaires were sent to all the
staff indicated above with an explanatory letter
and prepaid reply envelopes. A reminder and a
second copy of the questionnaire were sent if
there was no response within three weeks. We
indicated that respondents could reply anonymously if preferred.
Results
RESPONSE RATES

The response rate to the questionnaires was 23/

37 (67-6%) for paediatricians and 25/37 (67-6%)
for child psychiatry staff.
We originally recorded the results by subgrouping the respondents according to department (paediatrics or child psychiatry), hospital
or centre and either status, if paediatric, or discipline, if child psychiatric (for example, senior

863

Referrals to child psychiatry-a survey ofstaff attitudes

Table I Results to part I of the questionnaire: which type of clinical problem should be referred to child psychiatry
Total
Department* Respondents (%) who refer:
Signifcance
No of

Category
of problem

(A) Self poisoning

P
CP
P
(B) Emotional/
CP
behavioural problems
P
(C) Obesity
CP
P
(D) Helping families/children cope
CP
with terminal ilness
(E) Helping famillies/children cope with P
CP
physical handicap/chronic illness
(F) Repeated admissions (because mother P
cannot cope)
CP
P
(G) Physical illness exacerbated by
CP
psychological problems
P
(H) piagnostic difficulties
CP
P
(I) Child sexual abuse+behavioural/
CP
emotional problems
P
(J) Help with diagnosing child
CP
sexual abuse
P
(K) Encopresis
CP
P
(L) Enuresis
CP
P
(M) Admission problems/
CP
separation anxiety
P
(N) Parental neglect/
CP
non-accidental injury
P
(0) Drug/alcohol problems
in child
CP
P
(P) Anxiety over medical/
CP
nursing procedures
P
(Q) Behaviour disturbance
CP
on ward
P
(R) Mother has difficulties with
CP
baby feeding/sleeping
P
(S) Don't know what
CP
else to do

*P=paediatrics, CP=child

Always

Frequently

Sometimes

Never

responses

71
88
4-5
56
0
12
22
8
0
0
0
20
4
20
16
53
53
88
27
44
17
16
0
12
0
20
4
32
32
44
0
8
14
50
4
36
4
24

24
8
32
28
0
24
13
48
17
16
22
8
26
48
36
22
19
8
27
36
39
64
0
60
19
40
13
28
55
40
5
20
24
33
13
28
4
16

0
4
54 5
11
55
60
38
40
61
80
48
64
52
8
40
20
19
4
14
16
31
20
47
28
48
28
48
40
13
16
43
52
52
17
48
36
57
44

5
0
9
7
45
4
17
4
22
4
30
8
17
0
8
5
9
0
32
4
13
0
53
0
33
12
35
0
0
0
52
20
10
0
35
0
8
16

21
25
22
25
22
25
22
25
23
25
23
25
23
25
25
22
21
25
22
25
23
25
23
25
21
25
23
25
22
25
21
25
21
25
23
25
23
25

NS

p<0-01
NS
NS
NS

p<0-01

NS
p<0-01

P<0 01
NS

0.01

psychiatry.

Table 2 Results to part 2 of the questionnaire: factors that might negatively influence referral
(1) Referral not acceptable
(2) Stigmatises child or family

(3) Delay before first appointment too long

(4) Delay before treatment too long
(5) Treatment range narrow/inappropriate
(6) Psychiatric process damages child
(7) Lack of child psychiatry inpatient/
day patient facility
(8) Inadequate communication between
departments
(9) Case is 'lost' to child
psychiatry after referral
(10) Medical opinion soughtt
(11) Opinion of consultant psychiatrist sought
but difficult to obtain
(12) If all cases referred child psychiatry
system would be flooded
(13) Referral does not achieve much

P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP
P
CP

100
100
14
75

4525

52
41
50
50
68
5
41
48
65
77
24
40
62-5
41
52
27
56
32
36
48
72

0
0
86
25

5585

48
59
50
50
32
95
59
52
35
23
16
60
37-5
59
48
73
44
68
64
52
28

22
NS
25
22
P<0-01
24P<0
22
P<0-01
25
22
NS
24
22
P<0-01
25P<0
21
P<0-01
22P<0
21
NS
23
22
NS
25
22
NS
24
22
NS
25
22
NS
25
22
NS
25
21
NS
25

*P=paediatrics, CP=child psychiatry.

tA medical opinion sought but child frequendy seen by non-medical child psychiatry staff.

house officer or psychologist). However, there

superficial differences between these
subgroups within each department and the
numbers in some of the subgroups were very
small and therefore the responses were pooled
according to department for analysis.

given. These totals vary as a number of questionnaires were only partially completed. The
results were analysed using non-parametric correlation coefficient Kendall's T for tied ranks to
see if there was a significant difference between
the responses of the paediatricians and the child
psychiatric staff.9 The rationale for using this
method is discussed by Priest.'0 The result of
Part I
this analysis is given in table 1.
Results are given in table 1. The percentage of
There was a significant difference in the
paediatricians and child psychiatry staff responses to 13 out of 19 categories. Particularly
responding with either 1, 2, 3, or 4 to each striking disagreement was seen in the categories
category of problem A to S is given. The total of obesity, enuresis, and 'behavioural problems
number of responses to each category is also referred directly to the paediatrician'. When
were no

Oke, Mayer

864

there was a difference in responses it was always

that the child psychiatry staff indicated that the
case should be referred and the paediatricians
indicated that it should not.
There was agreement between the two
groups' responses to four of the categories. Firstly, both paediatricians and child psychiatry
staff agreed that category A (self poisoning) and
O (drug and alcohol problems) should be referred to child psychiatry. Both groups also agreed
that categories E (helping families and children
cope with physical handicap or chronic illness)
and F (children with repeated admissions
because the mother cannot cope) should not be
referred to child psychiatry.
Part 2
We recorded the percentage of 'yes' and 'no'
responses given by both groups of staff to each
of the statements 1 to 13. The results are given
in table 2. Again, some of the questionnaires
were incomplete and so the total number of
responses varied. These results were analysed
using the x2 test with Yates's correction for continuity. Significant differences were found in
the responses to two of the statements. To statement 2 (referral stigmatises the child or family)
paediatricians indicated that this was not a
reason for not referring whereas child psychiatry staff thought this was a reason for
not referring. Paediatricians responded to statement 6 (the psychiatric process damages the
child) by indicating that this was not a reason
for not referring whereas child psychiatry staff
were divided in their responses to this statement.
Both groups agreed that statements 1 (referral
not acceptable to the family) and 8 (inadequate
communication between the departments after
the child is seen) are reasons for not referring.
Both groups also agree that statement 12 (if all
appropriate cases were referred the child
psychiatry department would be flooded) was
not a reason for not referring.

* 'There is a cultural gap between the local

population and the staff of the child
psychiatry clinics-more medical approach
and perhaps more work with other professionls in the community might be more helpful'.
* 'Reluctance on the part of the child psychiatrist to see the patient'.
* 'Child psychiatry should be part of the range
of services provided by paediatrics and not a
separate department to whom referrals are
made'.
* 'A child psychiatrist is not present on site as
part of the paediatric team'.
* 'I feel that both paediatricians and child
psychiatric trainees should spend six months
in each other's professions in order to better
understand the job'.
* 'Heavy demand on parents with frequent
(two to three times a week) therapy sessions'. *
* 'General requirement that both parents
should be prepared to go to the initial

appointment'. *
* 'Low profile of psychiatric department
within paediatric department-but improving'.
Comments from child psychiatry staff
* 'Paediatricians sometimes think they can
offer the same help as child psychiatry would
give'.
* 'Resistance to conceptual basis of child
psychiatry/psychotherapy'.
* 'Some paediatricians (not all) may find it
hard to accept that they are not experts in all
aspects of illness in children. . . . in the good
old days paediatricians did have to cope with
all aspects of childhood illness, including
psychological aspects. Lack of child
psychiatric resources in some areas may
mean that paediatricians are still dealing with
what should ideally be dealt with by child

psychiatrists'.

Some of the comments made in response to the

question posed at the end of part 2 of the questionnaires: 'Do you think there are any other
reasons for not referring to the child psychiatric
department' are given below. The asterisked
comments were made by anonymous respondents.

* 'Paediatricians' own anxieties about (a) emotions being stirred up, (b) working with other
professionals, (c) losing control of the situation'.
* 'The department is seen as too narrow in its
range of work and intransigent'.
* 'Ignorance and lack of communication
between the two departments'.
There were no comments made in response to
part 1 of the questionnaire.

Comments from paediatnicians

* 'Psychiatrists seem aloof from everyday
problems that families face and from
problems on the ward'.
* 'The approach of the psychiatric staff is not
appropriate for the majority of our patients.
The families most often needing help are
emotionally damaged mothers in very deprived settings who need a behavioural
approach weighted with problem solvingJungian therapy is not accepted ... Middle
and upper middle class with support and
intelligence can make use of the service'.

Discussion
The response rate was 67-6% for both paediatricians and child psychiatry staff. Some of the
junior paediatric staff gave the reason that they
felt too inexperienced to participate. There were
only two anonymous respondents and these
were both paediatricians.
One problem we came across that affected all
parts of the study concerned was what was
meant by referral. Some staff though that a discussion of the case with a member of the child
psychiatric department constituted a referral
whereas others defined referral as asking a

COMMENTS

Referrals to child psychiatry-a survey ofstaff attitudes

865

lem. Some interesting results were seen in the

responses to statement 5 ('range of treatment
too narrow or inappropriate') and statement 13
('referral to child psychiatry does not seem to
achieve much'). A large number of child
Part I of the questionnaire
There was agreement between paediatricians psychiatry staff responded 'yes' to both stateand child psychiatric staff on only a few categor- ments. This could mean that either this group is
ies of problems and significant disagreement dissatisfied with their work or that they feel that
was seen in the responses of the two groups to they are poorly perceived by their paediatric
most of the categories. In cases of child sexual colleagues. The comments received from the
abuse with emotional or behavioural disturb- paediatricians supports the second conclusion.
ance 72% of paediatricians and 96% of child
Both departments agreed that inadequate
psychiatry staff said they would always or fre- communication was a reason for non-referral.
quently refer. This difference is significant and The agreement on this issue is striking and is
contradicts Department of Health and Social reflected in some of the respondents' comSecurity (DHSS) guidelines that recommend ments. It is suggested that this may be underreferral to child psychiatry. " Interestingly, 80% lying some of the differences found in part one
of child psychiatry staff and 54% of paediatri- of the study.
cians favoured always or frequently referring
This could be readily addressed by more
cases requiring help with the diagnosis of child active attempts at communication between the
sexual abuse. This difference is significant and two groups. Joint ward rounds or clinical meetagain contradicts DHSS guidelines, which ings could be a suitable forum in which this
advise that such cases should not be referred to could take place. If all the cases that the child
child psychiatry. "
psychiatry staff suggest were referred it would
It was noted that when there was a significant undoubtedly result in a flooding of the service.
difference between child psychiatry staff and More consultative work and mutual education
paediatricians it was always that the child could take place in such meetings thereby
psychiatrists were in favour of referral and reducing the perceived need for many of these
paediatricians were against it. One possible referrals. Also, perhaps a jointly formulated
explanation of these findings is that paediatri- referral policy could be initiated.
cians and child psychiatrists may have different
ideas about these problems. It is likely that The authors wish to thank Dr IT Dresser, Professor TE Oppe,
paediatricians would see a broad range in sever- and Professor RG Priest for their advice and the paediatric and
ity of these cases but the child psychiatric staff child psychiatry staff for their help and participation in the study.
may only see the more severe cases. Another
possible cause of these differences may be the 1 Eisenberg L. The relationship between psychiatry and
apparent lack of communication between the
pediatrics: a disputatious view. Pediatrics 1967;39:645-7.
TF. Child psychiatry and pediatrics-the state of the
two departments. This idea is supported by 2 Anders
relationship. Pediatrics 1977;60:616-20.
the results of part 2 of the questionnaire and 3 Jellinek MS. The present status of child psychiatry in pediatrics. N Engl J Med 1982;306:1227-9.
by some of the comments made by the respon- 4 Taylor
DC. Child psychiatry/paediatric liaison: a discussion
dents.
paper. J R Soc Med 1986;79:726-8.
member of the child psychiatry department to
see the child and/or family.

Part 2 of the questionnaire

This part looked at reasons for not referring to
child psychiatry. Unfortunately, it is not clear
whether responding 'no' in this part of the questionnaire indicates that the respondent thinks
that the statement is incorrect or that the statement is correct but not relevant in influencing
non-referral. The questionnaire has compounded these two distinct questions and a
further study is needed to overcome this prob-

5 Fritz CK, Bergman AS. Child psychiatrists characteristics as

described by pediatricians in a national survey (1986-1987).
Int J Psychiatry Med 1986;16:91-100.
6 Mrazek D. Child psychiatric consultation and liaison to
paediatrics. In: Rutter M, Hersov L, eds. Child and adolescent psychiatry-modern approaches. 2nd Ed. Oxford:
Blackwell Scientific, 1976.
7 Lewis M. Child psychiatric consultations in pediatrics.
Pediatrics 1978;62:359-64.
8 Monnelly EP, Lanzito BM, Stewart MA. Psychiatric consultations in a children's hospital. Am J Psychiatry 1973;130:
789-90.
9 Kendall MG. Rank correlation methods. 3rd Ed. London:
Griffin, 1962.
10 Priest RG. The homeless person and the psychiatric services:
an Edinburgh survey. Br J Psychiatry 1976;128:128-36.
11 Department of Health and Social Security. Diagnosis ofsexual
abuse: guidance for doctors. London: HMSO, 1988.