Robin sequence (RS) can be defined as the combination of micrognathia and upper airway obstruction/glossoptosis causing neonatal respiratory problems, with or without a cleft palate and either isolated or non-isolated. Pathogenesis varies widely. We hypothesize that optimal treatment depends on pathogenesis and therefore patients should be stratified according to diagnosis. Here, we evaluate diagnoses and (presumed) pathogeneses in an RS cohort. Medical records of all RS patients presenting between 1995-2013 in three academic hospitals were evaluated. Four clinical geneticists re-evaluated all information, including initial diagnosis. Diagnoses were either confirmed, considered uncertain, or rejected. If uncertain or rejected, patients were re-evaluated. Subsequent results were re-discussed and a final conclusion was drawn. We included 191 RS patients. After re-evaluation and changing initial diagnoses in 48 of the 191 patients (25.1%), 37.7% of the cohort had isolated RS, 8.9% a chromosome anomaly, 29.3% a Mendelian disorder, and 24.1% no detectable cause. Twenty-two different Mendelian disorders were diagnosed, of which Stickler syndrome was most frequent. Stratification of diagnoses according to (presumed) pathogenic mechanism in 73 non-isolated patients with reliable diagnoses showed 43.9% to have a connective tissue dysplasia, 5.5% a neuromuscular disorder, 47.9% a multisystem disorder, and 2.7% an unknown mechanism. We diagnosed more non-isolated RS patients compared to other studies. Re-evaluation changed initial diagnosis in a quarter of patients. We suggest standardized re-evaluation of all RS patients. Despite the relatively high diagnostic yield pathogenesis could be determined in only 59.7% (71/119), due to limited insight in pathogenesis in diagnosed entities. Further studies into pathogenesis of entities causing RS are indicated.