Uvulopalatopharyngoplasty (UPPP) is the most common surgical treatment for adult patients with obstructive sleep apnea (OSA). Its short-term efficacy, as measured through polysomnography, is well-established and has been demonstrated in two randomized controlled trials in recent years. However, less is known about subjective sleep quality, longterm efficacy, and side effects after surgery. In addition, the reliability of the most widespread clinical test for selecting patients for surgery-the Friedman staging system-is unclear. The Friedman staging system uses a combination of tonsil size and tongue position to predict the likelihood of successful surgery. The objective of studies I and II was to evaluate the staging system by determining its inter-examiner agreement. In study I, 15 doctors evaluated the system by using it on each other. In study II, 14 doctors evaluated the system by using it on 12 patients with OSA. Cohen's kappa analysis was used. Kappa values of 1 represents perfect agreement, and values of 0 represent no more agreement than would have been expected through random chance. In study I, the median kappa was 0.36 for tongue position. In study II, the median kappa was 0.32 for tongue position, 0.62 for tonsil size, and 0.38 for the Friedman staging system. These findings correspond to poor agreement, indicating that the system is an uncertain method for selecting patients for UPPP. Uvulopalatopharyngoplasty was first described in the eighties. Since then, there has been several modifications. The modified UPPP used in this thesis includes a tonsillectomy and is performed with cold instruments. It is less radical to the palate compared to the original procedure. The objective of study III was to investigate whether modified UPPP improves sleep quality by using the Functional Outcomes of Sleep Questionnaire and the Karolinska Sleep Questionnaire. The study consisted of two parts: Part 1 was a randomized controlled trial with two study groups (intervention and controls), and Part 2 was a post-operative follow-up of all patients (intervention and controls who received delayed surgery) with an analysis of outcomes at six and 24 months after UPPP. In eight out of nine subscales, there were significant improvements between the intervention and controls in favor of UPPP. In addition, at the six-and 24-month post-operative follow-ups of all patients, eight out of nine subscales were significantly improved compared to the baseline. These findings suggest a real and lasting beneficial effect of UPPP on subjective sleep quality, although a placebo effect cannot be excluded.