When two interventions are demonstrably equivalent in terms of clinical outcomes, it is unclear which should be provided. One obvious decision rule in this case is to implement the intervention that is least costly and therefore most cost-effective. A recent economic evaluation by Patel and colleagues estimates the cost-effectiveness of adherence therapy for people with schizophrenia, compared with simple health education.

Non-adherence rates are high for prescribed antipsychotics, and relapse rates for people with schizophrenia have been shown to be five times greater for non-adherent patients. Some evidence supports the presence of a weak positive effect of adherence therapy on adherence rates. However, a 2006 study by Gray et al showed no benefit of adherence therapy, compared with health education, in terms of quality of life. This new study by Patel et al reports on an economic evaluation of this trial.

Methods

409 patients were recruited to the multi-centre, multi-country trial. The intervention arm received adherence therapy consisting of five key interventions:

1. Medication problem solving
2. A medication timeline
3. Exploring ambivalence about medication
4. Discussing beliefs and concerns about medication
5. Using medication in the future

The control arm received didactic health education, without any adherence therapy techniques, in order to identify any benefits of adherence therapy beyond that of a therapeutic relationship.

The analysis adopted the net benefit approach of economic evaluation and costs were analysed from a health service perspective and a societal perspective. The cost of the interventions was recorded by using case notes and study proformas and attaching national unit costs to staff time and service use. Patient-reported service and resource use was collected every 3 months using a tailored version of the Client Sociodemographic & Service Receipt Inventory (CSSRI-EU). The primary outcome of the evaluation was quality-adjusted life years (QALYs). This was captured at baseline and at one year follow-up using both the SF-36 and EQ-5D instruments.

Results

Though the original study had already demonstrated the equivalence of clinical outcomes for adherence therapy and health eduction, this study adopted a decision-making perspective; identifying the intervention that maximises expected net benefit. However, it was not clear which intervention would maximise net benefit.

• On average, adherence therapy cost €192, while health education cost €138.
• The results indicate that, in addition to equivalence in outcomes, both interventions cost about the same from either a societal perspective or a health and social care perspective. As such, the probability of adherence therapy being more cost-effective than health education sat at around 50%, depending on the choice of outcome and cost perspective.
• However, the authors’ primary analysis, using QALYs estimated with the SF-36 and a societal cost perspective, showed that it is more likely that health education is the most cost-effective intervention.
• The authors suggest that it is possible that health education ‘dominates’ adherence therapy; that is to say it is more effective and less expensive.

Conclusions

The authors conclude that adherence therapy does not meet the need to provide cost-effective support to patients with schizophrenia. The two interventions are similar in cost and outcome, but adherence therapy appears to be less cost-effective than health education.

The authors highlight the possibility that, in some cases, one intervention may be demonstrably more cost-effective than the other. Different settings (e.g. different countries) may be able to provide therapy at a lower cost, for example. Furthermore, because a control intervention was used, it is not clear from this study whether or not adherence therapy is more cost-effective than treatment as usual.

The low cost of the interventions means that the results are largely driven by other costs; the intervention costs representing less than 1% of the total societal costs incurred in each group. Use of inpatient services alone averaged €7411 and €6976 in the intervention and control groups respectively.

The study is linked to a trial that was completed and published in 2006, but the original study did not report cost-effectiveness estimates. Though the results of this analysis are not definitive, the study demonstrates the potential value of carrying out post hoc economic evaluations of trialled interventions.

Links

Patel, A., McCrone, P., Leese, M., Amaddeo, F., Tansella, M., Kilian, R., Angermeyer, M., Kikkert, M., Schene, A. and Knapp, M, (2013) Cost-effectiveness of adherence therapy versus health education for people with schizophrenia: randomised controlled trial in four European countries. Cost Effectiveness and Resource Allocation 2013; 11(12). [PubMed]

Gray, R., Leese, M., Bindman, J., Becker, T., Burti, L., David, A., Gournay, K., Kikkert, M., Koeter, M., Puschner, B., Schene, A., Thornicroft, G. and Tansella, M. Adherence therapy for people with schizophrenia European multicentre randomised controlled trial. The British Journal of Psychiatry 2006; 189(6), 508-514. [PubMed]