BACKGROUND. Using prediction rule is by far the easiest and least costly approach in screening patients with Obstructive sleep Apnea (OSA). A devised prediction rule summarized as likelihood ratios for Filipinos was reported in 2000.
OBJECTIVE: To validate the prediction rule, St. Lukes Medical Center Obstructive Sleep Apnea Clinical Score (SLMCOSACS)
STUDY DESIGN: Prospective validation study
MATERIALS AND METHOD: 50 patients with OSA and 50 normal volunteers matched for age and sex were included in the study. SLMS-OSACS was accomplished by a single investigator in all patients prior to a full night polysomnography. Data was analyzed using SPSS software
RESULTS: The clinical score of /- 8 as best cut-off value determined from the previous SLMC-OSACS study produced an overall accuracy of 89 percent, sensitivity of 84 percent, specificity of 92 percent, false positive rate of 8 percent and likelihood ratio of 14. This validation study revealed a lower cut off value of /- 5 with a sensitivity of 100 percent, specificity of 92 percent, false positive rate of 8 percent with a likelihood ratio of 12.5.
CONCLUSION: Body mass index, snoring affecting others and bothersome daytime sleepiness are significant predictors of OSA in Filipino patients. The SLMC-OSACS gives a valid reference for likelihood ratios for different scores which helps a clinician in his decision to refer a patient for full night polysomnography.