Obstructive sleep apnea (OSA) is associated with diabetes, hypertension, stroke, coronary artery disease, and premature death. Positive airway pressure (PAP) is the mainstay of therapy. Despite its effective treatment with PAP therapy, noncompliance remains high. Many factors determine compliance. The role of severity of OSA measured by the apnea-hypopnea index (AHI) remains controversial. Meta-analysis of studies examining this role of AHI was performed. A systematic review of the medical literature was conducted using PubMed and Cochrane library by utilizing different combinations of key words: sleep apnea, AHI, compliance, and nonadherence. Inclusion criteria were English articles; Studies with adult population; with 2 groups of patients (compliant and noncompliant); Studies utilizing objective definition of compliance (PAP usage of >4 hours per night for 70% of days or usage >5 d/wk and for >4 hours per night). Studies were analyzed by standard methods of meta-analysis. The studies were heterogeneous for AHI; therefore, the random effect model was used. Six hundred forty-one manuscripts were found. Of these, 230 were found to be appropriate for full text evaluation. Thirty-one met inclusion criteria. Twelve of these studies used objective criteria for PAP compliance and were hence included in meta-analysis. All the subjects had OSA determined by polysomnography, for whom PAP was employed. Compliance to PAP therapy was evaluated after a period of time ranging from 4 weeks to 8 years. There were 1438 subjects included in the meta-analysis; 886 subjects were PAP compliant, whereas 552 subjects were noncompliant. A greater AHI was found in PAP compliant patients. The mean difference between compliant and noncompliant groups was 5.9 (95% confidence interval: 0.19-11.67, P < 0.05). Patients with mild OSA are less likely to be compliant with PAP therapy. These patients should receive aggressive management particularly at the start of therapy with close follow-up to increase compliance.