Fig. 3From: Clusters of medical specialties around patients with multimorbidity – employing fuzzy c-means clustering to explore multidisciplinary collaboration22 identified subgroups of the identified clusters with the mean membership degree in percentages. Each circle represents one subgroup. The subgroup’s membership(s) to one or more clusters is/are depicted by the circle’s color(s), each color corresponds with one cluster. In addition, every subgroup is labelled with the subgroup’s name, the mean membership for the cluster(s) and the number of patients in the subgroup. The six larger circles depict the subgroups with a total of 17,728 patients (80.1%) who had a membership of ≥80% (full membership) for one of the six identified clusters. There are five subgroups (2-a to 6-a) with patients with a dominant membership of 60-80% for one of the clusters, but with no clinically relevant membership for another cluster. The other eleven subgroups (1-3, 1-5-a, 1-5-b, 1-5-c, 1-6-a, 1-6-b, 1-6-c, 3-5-a, 3-5-b, 5-6-a and 5-6-b) are combinations of two clustersBack to article page