The Delphi exercies carried out in Poland constituted part of DEMoB.inc project founded by EC and conducted within 10 European countries. Polish part of Delphi exercise fulfils major method's assumptions and aims to define items of long-term psychiatric care in recovery model of treatment.

Delphi method has not been widely applied in research across Poland. Method is characterized by defined panel of experts (EP), information flow, regular feedback and participants’ anonymity. Delphi has proven prognostic potential in health care, social sciences, new technologies and policy forecasting. However informality of EP selection and limited complex forecasts with multiple factors and downward responses-rate tendencies in following rounds constitute its crucial shortcomings. Criticized basic statistics has been lately overcome by implementation of sophisticated fuzzy logic theory. Three stages Delphi exercise has been carried out in four groups of respondents: service users, carers, advocates and professionals. Wide range of EP representatives has promoted indication of diverse elements of care within partner sites.

Polish mental health care network varies from developed European countries. Out of 11 initially appointed domains only 2 (SPHRandA; SMA) have been underrepresented. Advocates achieved significant consensus in the largest number of items: 23 out of 104 which covered 9 domains whereas professionals in the lowest number: 11 out of 130 which covered only 4 domains. Moreover domains supported with advocates’ opinions fully overlapped with domains pointed by service users. The surprising outcome might be due to diversity of professions in “professional group”, differences in occupational responsibilities and work experience.