AIM: Early intervention services (EIS) for psychosis are being implemented, internationally. It is important to learn from established examples and define the components and intensity of services that provide good value for money. This study aims to assess the cost-effectiveness of EIS according to how closely they adhered to the recommendations of the English Department of Health 2001 Policy Implementation Guide (PIG). METHODS: EIS from the National Eden Study were assessed using a measure of fidelity to the PIG that rated the presence or absence of 64 recommended items relating to team structure and practice. EIS were then classified into three groups: those with fidelity of 75-80%, 81-90% and 91-95%. Patient-level resource use and outcomes were measured 1 year following inception into the service; costs were calculated and combined with quality-adjusted life years (QALYs) gained. RESULTS: At a threshold of £20 000 per QALY, the 81-90% fidelity group had a 56.3% likelihood of being the most cost-effective option followed by 75-80% fidelity at 35.8% and 91-95% fidelity group (7.9%). CONCLUSIONS: The results from England suggest that striving to maximize fidelity may not be warranted, but that dropping below a certain level of fidelity may result in inefficient use of resources.