Economic evaluation of treatments for children with severe behavioural problems

Abstract

Background: Disruptive behaviour disorders, including conduct disorder, affect at least 10% of children and are the most common reasons for referral to children's mental health services. The long-term economic impact on society of unresolved conduct disorder can exceed pound sterling 1 million for one individual over their lifetime.

Aims of the study: The aim of this study was to estimate, from a multi-sectoral service perspective, the longer term cost-effectiveness of an intensive practice based parenting programme for children with severe behavioural problems as compared to a standard treatment, on a pilot basis.

Methods: A six-month pragmatic controlled trial was conducted involving forty-two families who had been referred to a Child and Adolescent Mental Health Service (CAMHS) because of severe child behavioural problems. The families were randomly allocated into either the standard or intensive, practice-based treatment arms of the trial. At baseline, children were aged 2 to 10 years. The externalising T-scale of the Child Behaviour Check List (CBCL) was used as the primary outcome measure. Follow-up studies were conducted at 6 months and four years post-intervention. At the four-year follow up point the two treatments were subjected to an incremental cost-effectiveness analysis. This analysis was enabled by the collection of cost data with respect to the provision of the intensive and standard treatments in terms of therapeutic contact time and also participants' use of health, special educational and social services usage by means of a Client Service Receipt Inventory.

Results: Both groups exhibited improved behaviour at six month follow-up, but only the practice based treatment group showed sustained improvement at the four-year follow-up. An independent t-test revealed a significant difference between group mean scores at four-year follow-up (p = 0.027). The research found a median bootstrap ICER estimate of pound sterling 224 From the cost-effectiveness acceptability curve (CEAC) it was found that 89.6% of the cost-effectiveness plane represented a cost saving over the control intervention while 99.9% represented an improvement in effect. Therefore the intensive intervention could not be said to differ significantly from the control intervention on the basis of costs or effects. However under certain circumstances requiring judgement the intensive intervention could be cost-effective.

Discussion: Training interventions for the parents of children with severe conduct disorders currently take the form of a number of behavioural strategies provided by a CAMHS team including the reinforcement of appropriate behaviours and parent record-keeping. An alternative treatment was then considered that included all the aspects provided by the standard treatment (with the exception of agreeing written goals) and also added five-hour sessions of unit-based treatment during which videotaped recording of parent-child interactions were used to give feedback to parents and to provide the opportunity for behavioural rehearsal. The findings of the current research indicate that this alternative treatment may be more cost-effective than the standard intervention under certain circumstances.

Conclusions: It can be concluded on a pilot basis that while the CEAC failed to show a significant difference between costs and effects for the intensive treatment, under circumstances of resource/effect trade offs the treatment could be said to be cost-effective. IMPLICATIONS FOR HEALTH CARE PROVISION, USE AND POLICIES: Health and social care policy and commissioning must be evidence based. Although the analysis in this paper should be considered a pilot due to the small sample size our results suggest that investment by health services and social services in practice-based parenting interventions may well be less costly and more effective in the longer-run than the standard practice involving giving management advice to parents.

Implications for further research: It would be of interest for further research to continue to follow up the work done in this study with a larger cohort of subjects to further establish the effective components of parenting programmes and their relative costs and benefits both at intervention and over time.