Mange patienter bliver fulgt i de danske ambulatorier efter hospitalsbehandling. Da denne løsning er for ressourcekrævende i fremtiden, skal flere overdrages til opfølgning uden for sygehusene. Der er udviklet en model for, hvordan en sådan opfølgning for prostatakræft bedst tilrettelægges for patienter, almen praksis og hospitalsafdelingerne. Men den valgte implementeringsmetode kan være afgørende for, hvor succesful modellen bliver i den virkelige verden. Denne artikel beskriver implementeringsprocessen via observationer og interviews på to forskellige hospitalsafdelinger, der anvendte henholdsvis en ”byg ovenpå-strategi” og en ”indbygningsstrategi”. I ”byg ovenpå”-afdelingen afhang implementeringen af ekstra tilførte ressourcer og etablering af specifikke nye aktiviteter (iværksat ovenfra). Modellens nye procedurer og rutiner for, hvordan patienter blev overdraget til almen praksis, blev kun fulgt i projektperioden. I "indbygnings”-afdelingen blev modellen tilpasset eksisterende arbejdsgange og tilgængelige ressourcer. Implementeringen afhang dermed af det kliniske personales perspektiv, og her fortsatte en lokal tilpasset model for tidligere overdragelse til almen praksis også efter projektperioden. Et udgangspunkt i den eksisterende kliniske praksis og aktiv inddragelse af de fagprofessionelle viste sig at være afgørende for implementeringen af blivende forandringer. ; BACKGROUND: Timely discharge is a key component of contemporary hospital governance and raises questions about how to move to more explicit discharge arrangements. Although associated organisational changes closely intersect with professional interests, there are relatively few studies in the literature on hospital discharge that explicitly examine the role of professional groups. Recent contributions to the literature on organisational studies of the professions help to specify how professional groups in hospitals contribute to the introduction and routinisation of discharge arrangements. This study builds on a view of organisational and professional projects as closely intertwined, where professionals take up organising roles and where organisations shape professionalism. METHODS: The analysis is based on a case study of the introduction and routinisation of explicit discharge arrangements for patients with prostate cancer in two hospitals in Denmark. This represents a typical case that involves changes in professional practice without being first and foremost a professional project. The multiple case design also makes the findings more robust. The analysis draws from 12 focus groups with doctors, nurses and secretaries conducted at two different stages in the process of the making of the local discharge arrangements. RESULTS: From the analysis, two distinct local models of discharge arrangements that connect more or less directly to existing professional practice emerge: an 'add-on' model, which relies on extra resources, special activities and enforced change; and an 'embedded model', which builds on existing ways of working, current resources, and perspectives of professional groups. The two models reveal differences in the roles of professional groups in terms of their stakes and involvement in the process of organisational change: whereas in the 'add on' model the professional groups remain at a distance, in the 'embedded model' they are closely engaged. CONCLUSIONS: In terms of understanding the making of hospital discharge arrangements, the study contributes two sets of insights into the specific roles of professional groups. First, professional interests are an important driver for health professionals to engage in adapting discharge arrangements; and second, professional practice offers a powerful lever for turning new discharge arrangements into organisational routines.