• Ken M Kunisaki, Nancy Greer, Wajahat Khalil, Erin Koffel, Eva Koeller, Roderick MacDonald, and Timothy J Wilt.
• Minneapolis Veterans Affairs Health Care System and University of Minnesota, Minneapolis, Minnesota (K.M.K., W.K., E.K.).
• Ann. Intern. Med. 2018 Feb 6; 168 (3): 195-202.

BackgroundObstructive sleep apnea (OSA) diagnosis and care models rely on sleep specialist physicians (SSPs) and can be expensive and inefficient.PurposeTo assess OSA case-finding accuracy and comparative effectiveness of care by non-sleep specialists (NSSs) and SSPs.Data SourcesMEDLINE and CINAHL from January 2000 through July 2017.Study SelectionEnglish-language trials or observational studies comparing case finding or care by SSPs versus providers not specifically trained as SSPs (NSSs) for adults with suspected or diagnosed OSA.Data ExtractionOne investigator extracted data and assessed risk of bias and strength of evidence, with confirmation by a second investigator. Primary outcomes were patient-centered (mortality, access to care, quality of life, patient satisfaction, adherence, symptom scores, and adverse events). Intermediate outcomes included resource use, costs, time to initiation of treatment, and case finding.Data SynthesisFour observational studies (n = 580; mean age, 52 years; 77% male) reported good agreement between NSSs and SSPs on appropriate diagnostic testing and classification of OSA severity (low-strength evidence). Five randomized trials and 3 observational studies (n = 1515; mean age, 52 years; 68% male) found that care provided by NSSs and SSPs resulted in similar quality of life, adherence, and symptom scores (low-strength evidence). Evidence was insufficient for access to care and adverse events.LimitationsMany outcomes were reported infrequently or not at all. Many NSSs had extensive training or experience in sleep medicine, which limits generalizability of findings to providers with less experience.ConclusionCare by NSSs and SSPs resulted in similar outcomes in adults with known or suspected OSA. Studies are needed to determine care model implementation and reproducibility of results in nonacademic settings and among less experienced NSSs.Primary Funding SourceDepartment of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Quality Enhancement Research Initiative. (PROSPERO: CRD42016036810 [full Veterans Affairs Evidence-based Synthesis Program report]).

10 keywords...

hide…