Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Cost-effectiveness of treatments for major depression in children

Study
Results
Comment

Few health economic studies are amenable to simple précis, mainly because they use information from so many different sources, but also because they use different perspectives. They are also often applicable mainly to one health care system, and translate poorly to others. Occasionally one swims into our ken that is really worth a look at. One looking at different treatments for major depression in children [1] is interesting.

Study

The setting for the study was the whole of Australia in 2000, where there were 48,500 new episodes of major depressive disorder in children and adolescents aged six to 17 years. Most (65%) do not consult, and of the 35% who do consult, a third already have an evidence-based treatment. It is those who consult but do not receive evidence-based treatment that formed the cohort for the economic assessment.

The interventions modelled as first-line treatments were 12 one-hour sessions of cognitive behavioural therapy plus two family sessions over 14 weeks, or nine months of treatment with SSRI, three months in the acute phase and six months in a continuation phase, with appropriate clinic visits.

Measurement of health gain was used to calculate disability-adjusted life years. Information on health gain was taken from systematic reviews of cognitive therapy and SSRI in children performed for the study. Costs were taken from Australian standard sources.

Results

The main results are in Table 1. Cognitive therapy was more effective than SSRI, based on limited information. The costs in Table 1 are total costs to the healthcare system if no costs were borne by patients. Incremental cost effectiveness ratios for cognitive therapy and SSRI are for their first line use compared with current practice.



Table 1: Major results of cost-effectiveness analysis of cognitive therapy for depression in children and adolescents in Australia, comparing cognitive therapy with SSRI. Costs are cost per disability adjusted life year



Parameter
CBT (psychologist)
CBT (psychiatrist)
SSRI
Efficacy
Remission with intervention (%)
62
62
46
Remission with control (%)
39
39
30
Imputed NNT
4.3
4.3
6.3
Cost
Total cost ($Aus, million)
3.7
12
5.4
Incremental CE ratio ($A)
9,000
32,000
23,000
ICER as £UK
3645
12960
9315
ICER as € euro
5220
18560
13340
ICER as $US
6984
24832
17848


Cognitive therapy using psychologists is cheaper than that using the more expensive psychiatrists, with SSRI intermediate between the two. Using current exchange rates, the values for the ICER are also given as £UK, euro, and $US for easy comparison.

In Australia there is an informal agreement that interventions with costs that fall below $Aus 50,000 per disability adjusted life year saved are probably worthwhile. Iterations used to develop the overall cost fell below this level 100% of the time for cognitive therapy using psychologists, 96% of the time for SSRIs, and 81% of the time for cognitive therapy using psychiatrists.

Comment

What is really interesting about this paper is that it uses the health economic analysis only as the first stage of a process to judge what is best, and what to do. It goes on to use what it calls ‘second stage filters' looking at quality of evidence, equity, feasibility and acceptability, and compares these aspects for cognitive therapy versus SSRI.

This takes health economics a fair bit down the road of operational research, and begins better to address issues of overall value, looking at aspects not usually seen in health economic studies. This paper does not just extend the concept that cognitive therapy can be cost effective in a healthcare setting, but it also makes one think outside our usual little box.

References:

  1. MM Haby et al. Cost-effectiveness of cognitive behavioural therapy and selective serotonin reuptake inhibitors for major depression in children and adolescents. Australian and New Zealand Journal of Psychiatry 2004 38: 579-591.

previous or next story