Several definitions were proposed to diagnose Positional Obstructive Sleep Apnea (pOSA). However, the comparison between these definitions according to their diagnostic value is scarce in the literature. Thus, we decided to conduct this study to compare between the four criteria according to their diagnostic value. Between 2016 and 2022, 1092 sleep studies were performed at the sleep lab at the Jordan University Hospital. Patients who had an AHI <5 were excluded. pOSA was described according to the four definitions; Amsterdam Positional OSA Classification (APOC), supine AHI twice the non-supine AHI (Cartwright), Cartwright plus the non-supine AHI <5 (Mador), and overall AHI severity at least 1.4 times the non-supine severity (Overall/NS-AHI). Furthermore, 1033 polysomnographic sleep studies were retrospectively analyzed. The prevalence of pOSA according to the reference rule was 49.9% among our sample. The Overall/Non-Supine definition had the highest sensitivity, specificity, positive predictive value, and negative predictive value, which were 83.5%, 99.81%, 99.77%, and 85.88% respectively. The Overall/Non-Supine definition also had the highest accuracy among the four definitions (91.68%). Our study showed that all the criteria had a diagnostic accuracy above 50%, indicating that they were accurate in forming the diagnosis of pOSA. The Overall/Non-Supine criteria had the highest sensitivity, specificity, diagnostic odds ratio, and positive likelihood ratio as well as the lowest negative likelihood ratio, indicating the superiority of this criterion over the other definitions. Choosing the right criteria for diagnosing pOSA would result in fewer patients being assigned to CPAP and more being assigned to positional therapy methods.