Prevalence of confirmed child abuse and the use of resources in child psychiatric out-patients

This paper explores the extent of known child sexual and physical abuse in child psychiatric out-patient services in Leeds. We surveyed a one-month period, looking at all out-patient contacts, and found that at least 17.9% of patients had been physically or sexually abused and these patients took up 25.8% of out-patient time. When the 70 abused children were compared to a group matched for age group, clinical team and diagnosis it was found that they were more likely to be followed-up for longer and they received more therapeutic time over a given period than non-abused children. The implications for resources are discussed.

Children who have been physically or sexually abused may suffer a variety of adverse psycholo gical and developmental outcomes (Lynch & Roberts, 1982;Taitz & King, 1988;Claussen & Crittenden, 1991). Not surprisingly therefore child psychiatric out-patient services may see significant numbers of abused children (Lanktree et al 1991). Hobbs & Wynne (1987) identified sexual abuse as a major child mental health problem requiring new resources in training and personnel. In Leeds, professionals working within child mental health services began to develop a treatment service for abused children in 1984. This was limited by manpower and resulted in many children and families being offered little or no therapeutic help (Frothingham et ai, 1993). Anecdotal reports have frequently suggested that many of the families referred to child psychiatric services have a high degree of complexity often requiring intense, frequent or prolonged work. As a result, in Leeds, the purchasers wished to have more information about service provision for this group of children. They requested an urgent audit of child psychiatric services for abused children. We therefore set out to assess how much of our work involved children who had definitely been abused and whether these chil dren used a similar amount of child psychiatric resources to those not abused.

The study
We conducted a survey of all child psychiatric out-patient services in Leeds. At the end of November 1993 all workers involved in multidisciplinary teams from three bases were asked to fill in returns relating to out-patient working during that month. Only face-to-face contacts with patients or their families were included and were recorded as therapist minutes. Information collected included demographic, psychiatric and consultation details. Time allocated to patients who did not attend was noted but not included as therapeutic time. In addition, a working diag nosis was established based on ICD-10 criteria (World Health Organization, 1992). Diagnoses were not given in difficult cases or if the patient had not been fully assessed at that point. Specific questions were asked about the presence of sexual and/or physical abuse. We recognise the importance of emotional abuse, but found, like other researchers (Claussen & Crittenden, 1991), that it was difficult to establish clear criteria which could be used to define and assess emotional abuse in a consistent way. This category was therefore not included in this study. A child was recorded as having suffered abuse if there had been either: (i) clear disclosure; (ii) conviction of the abuser; (iii) protective legal orders, or (iv) definite evidence of non-accidental injury (Taitz & King, 1988;Lanktree et ai, 1991). We used Baker & Duncan's (1985) definition of child sexual abuse and included only those children where abuse had involved physical contact.
Each child with confirmed abuse was matched by age band, diagnostic category and consultant team with a child who had no suspected abuse. Where there was a possibility of two or more children for the match, matching was carried out using random number tables. Where it was not possible to match exactly by diagnosis, matching by broad diagnostic group was carried out (e.g. elimination disorders, conduct disorders). These two groups were compared at eight months' follow-up.
Psychiatric Bulletin (1996), 20, 207-209 Findings A 100% response rate was achieved by personal contact. Thirty-two therapists organised in five clinical teams arranged to see 390 patients and succeeded in seeing 325 (83.3%). Seven hundred and ninety-seven contacts were arranged for these patients and 641 (80.4%) took place. The total time spent with all patients who were seen was 690 hours. The length of a single session ranged from 30 minutes to three hours, with a mean of 65 minutes. Fifty-eight per cent of sessions were 60 minutes in length. One hundred and fourteen patients (29.2%) cancelled or did not attend without prior notice at least once during the month. Forty-nine of these attended subse quent appointments within the month. The maximum number of sessions missed by one person in the month was five. Of the 390 patients offered appointments, 70 (17.9%) had a known, substantiated history of physical or sexual abuse, 48 with sexual abuse, 40 with physical abuse and 18 with both.
The mean time spent per patient over the month was 126 minutes. This varied depending on whether or not they had been abused. The mean time spent in the month per abused child was 178 minutes. If no abuse had taken place the mean time was 116 minutes. If both physical and sexual abuse had occurred it was 196 minutes. Twenty-six per cent of the total time was spent with the 17.9% of abused children.
At the time of follow-up it was not possible to match 11 children because of unusual diagnosis, no diagnosis or missing notes, and so these were left out of the analysis. Fifty-nine abused children were matched with non-abused children. Three pairs were aged between 0-5, 29 were between 6-11 and 27 were 12 and over. The median age for both groups was 11. At eight months' follow-up 40 of the 59 patients in the non-abused group had been discharged compared to 22 of the abused group (Pearson f, P=0.00005). Of those who had been discharged at this point, 36.4% of the abused group had been under the service for more than one year compared to 12.5% of the non-abused group (P<0.05). If those not yet discharged were included the figures were 62.7% and 25% respectively (P< 0.00005). The mean length of time in therapist hours over the month for attending patients was 183 minutes in the abused group and 109 minutes in the nonabused group. This was not due to non-attendance without warning which was not significantly different between the abused and non-abused group, but the abused group were less likely to cancel appointments by telephone than the nonabused group (P< 0.05), although all of this group received a subsequent appointment. Baker & Duncan (1985) reported that approxi mately 5% of the population under the age of 16 might be expected to have suffered from sexual abuse involving physical contact. In our group of attenders at child psychiatry out-patient clinics this figure was 12% with nearly a fifth of out patient children having suffered definite physical or sexual abuse. These figures represent con servative estimates since they include only abuse which has been clearly established and involving contact. In many more children abuse was suspected but not confirmed. Child mental health services do not work with all children who have suffered child abuse. This may be because they are not referred or referral is discouraged. Alternatively abuse may not be identified in the clinic. Lanktree et al (1991) showed that when clinicians in child psychiatric out-patients speci fically asked about abuse in a systematic way they were 4.5 times more likely to identify it.

Comments
In this study, abused children used more time in the representative month and were followed up for longer than non-abused children. This gives an indication of the increased resource implica tions involved in treating abused children. It is likely that this increased use of the service is because these children have complex problems requiring greater input, but it may be that clinicians choose to devote more time to these children for other reasons, such as pressure from another agency.
Previous studies have highlighted the cost involved with the medical examination and investigation of suspected child abuse (Summers & Molyneux, 1992). It should be remembered that for many of these children identification of abuse is the first stage in a process which subsequently involves various professional groups with their own (additional) resource implications. This paper gives some early indications of the in creased resource needs of such children referred to child psychiatry services. This is particularly worthy of attention at a time when social workers have been removed from multidisciplinary child mental health teams (Kurtz et al, 1994). In Leeds, as a result of this survey, a greater focus on consultation to social services has been devel oped, as well as more sophisticated input to therapeutic children's homes and regular train ing seminars, as a means of providing a better service. If we are to reach more children and their families with increased effectiveness, the resource implications need to be recognised when planning services. If more resources are needed for particular groups of children, this may need to be reflected in contracting arrangements. We are now using a bespoke patient information system which has been designed to include a range of information. One of its functions, prompted by this study, will be to identify more clearly those children who will have significant impact upon resource allocation. The group of children studied here falls into that category. This information can then be used for planning, contracting and monitoring of services. KURTZ,S.. THORNE. A. & WOLKIND. S. (1994)