OBJECTIVES: To assess the extent to which the development steps of the evaluation of care pathways (CP) can be represented in the framework for the design and evaluation of complex interventions. METHODS: The framework is composed by 5 phases: theoretical (pre-clinical), identification of components of the intervention (phase I), definition of trial and intervention design (phase II), main trial execution (phase III), promoting effective implementation (phase IV). RESULTS: The framework was applied to the evaluation of CP for strokes. Pre-clinical phase was aimed in synthesizing the evidences: 3 reviews were selected and showed that CP are theoretically applicable in stroke care and that mortality should be the main outcome to be analyzed. Phase I was done through a descriptive pilot. A total of 253 consecutive patients admitted for strokes in 29 hospitals were analyzed. Overall in-hospital stroke mortality was 19.76%. Stroke teams (OR=0.25; p=0.025), antithrombotic therapy (OR=0.26; p=0.009) and complications (OR=6.40; p<0.001) were independent predictors of in-hospital mortality. Therefore these variables were selected as components of CP or treated as covariates. Because CP are active both on organizational (units) and individual (patients) level, a two-arm cluster Randomized Controlled Trial with hospitals and long-term rehabilitation facilities as randomization units was designed in phase II. 14 units were randomized either to arm 1 (CP) or to arm 2 (usual care) including 238 patients per group. The primary outcome measure was mortality, the CP were also analyzed with key quality indicators. The trial has been successfully performed (phase III) and in-hospital mortality has been reduced (OR=0.10; p=0.04). Because the adjusted results are not available yet, it was not possible to identify the active components of the CP and therefore phase IV has not been performed. CONCLUSIONS: Even if the results are still partial, it seems possible to apply this framework to the study of CP.