| Kryger [32] (1996) | Canada | Case–control | 97 obese patients with OSA | Hospital length of stay, physician visits and total health care expenditure 2 years prior to OSA diagnosis. | OSA group had 251 nights in hospital, compared with 90 nights in the control group Total expenditures from physician claims were $82,238 in the OSA patients versus $41,018 in the control group | Small study size Limited to patients with severe OSA Limited adjustment for potential confounders |
| Bahammam [29] (1999) | Canada | Prospective cohort study | 344 OSA patients | Physician claims and hospitalisations 2 years before and after OSA diagnosis | The difference in physician claims between the patients and their matched controls 2 years after diagnosis and treatment was mean±se $174±32.4 per year This was significantly less than the difference in the year before diagnosis $260±35.7 | Sample exclusively male Limited adjustment for potential confounders |
| Ronald [34] (1999) | Canada | Case–control | 181 OSA patients | Physician claims and hospital length of stay 10 years prior to OSA diagnosis | OSA patients used twice the resources in the 10 years prior to their diagnosis Physician claims for cases totalled $686,365 ($3,972 per patient) compared with $356,376 ($1,969 per patient) for the controls during the length of the study | Small study size Limited adjustment for potential confounders |
| Smith [35] (2002) | Canada | Case–control | 773 OSA patients | Physician fees, physician visits and hospital nights 5 years prior to OSA diagnosis | OSA patients used 23–50% more resources in the 5 years prior to diagnosis than did control subjects | Limited to patients with severe OSA Limited adjustment for potential confounders |
| Kapur [37] (2002) | USA | Cross-sectional study | 6440 Sleep Heart Health participants | Determined by a modified chronic disease score based on medication data | Patients with OSA had an 18% increase in healthcare use compared with patients with no OSA | Outcome definition was an indirect measure of healthcare utilisation |
| Tarasiuk [36] (2005) | Israel | Prospective case–control | 218 OSA patients | Hospitalisation days, consultations and cost of drugs 2 years prior to OSA diagnosis | Healthcare utilisation was 1.7-fold higher in the OSAS patients compared with controls | Limited adjustment for potential confounders |
| Greenberg-Dotan [31] (2007) | Israel | Case–control | 289 males and 289 females with OSA | Hospitalisations, emergency department visits, visits to specialists and prescriptions supplied 5 years prior to diagnosis | Compared with controls, total 5-year healthcare costs were 1.8 times higher for both females and males with OSA Compared with males with OSA, expenditures for women with OSA are 1.3 times higher | Limited adjustment for potential confounders |
| Reuveni [33] (2008) | Israel | Case–control | 117 young and 117 middle-aged male OSA patients | Hospitalisations, emergency department visits, visits to specialists and prescriptions supplied 5 years prior to diagnosis | Healthcare utilisation for the 5-year period was ≥1.9 times higher in young and middle-aged male OSA patients compared with controls | Sample exclusively male Limited adjustment for potential confounders |
| Banno [30] (2009) | Canada | Retrospective observational study | 223 obese females with OSA | Physician fees and visits in the years prior to OSA diagnosis | Physician visits 1 year before diagnosis in the OSA cases were more frequent than in the obese controls: 13.2±0.73 versus 7.26±0.49 visits | Sample exclusively female Limited adjustment for potential confounders |
| Ronksley [38] (2011) | Canada | Cross-sectional study | 2149 patients referred for sleep testing | Physician claims, emergency department visits and hospitalisations | Patients with severe OSA and daytime sleepiness had a 22% increase in healthcare use compared to those with less severe OSA | Referred population resulting in limited generalisability of results |