Recent studies have demonstrated the efficiency of collaborative dementia care, which aim to improve post-diagnostic support. However, tasks carried out of such models are currently unknown, hindering its implementation. To describe the tasks of an efficient collaborative dementia care management and analyze the association of these tasks with patients' characteristics. The analysis was based on 183 persons screened positive for dementia (PwD), who received dementia care management conducted by nurses with a dementia-specific qualification. A standardized computer-assisted assessment, patients' and caregivers' unmet needs were identified, based on these, pre-specified algorithms generated an individual list of intervention tasks for each PwD. Tasks performed to meet unmet needs were documented and categorized and descriptively analyzed. We used multivariate regression models to identify socio-demographic and clinical factors that are associated with a specific subgroup of tasks or a higher number of tasks. On average 22.7 tasks were carried out per dyad (PwD and caregiver). 38%, 19% and 13% of tasks could be categorized to cooperation with other healthcare providers, medical and nursing care, and social and legal support, respectively. A lower mental and physical health-related quality of life, age of PwD, lower education, higher deficits in daily living activities, higher depressive symptoms, and a higher number of drugs taken of the PwD, as well as female sex of the caregiver were associated with a significantly higher number of tasks carried out. Socio-demographic and clinical characteristics and the subjective perception of health determine the intensity of intervention needed to address unmet needs of PwDs. A variability of the intensity of collaborative care should be considered in the development of future intervention studies and in the implementation into routine care.