The evaluation and management of OSA in medically hospitalized patients

The evaluation and management of OSA have traditionally taken place in the outpatient setting. Over the last couple of decades, however, an increasing number of studies suggest that OSA is highly prevalent in medically hospitalized patients, can feasibly be screened for or diagnosed during admission, and may present an opportunity for initiation of therapy, all with the intent of improving clinical outcomes.

Roughly 80% of patients with OSA are undiagnosed, suggesting an ongoing failure of the outpatient setting to identify and diagnose these individuals. Studies have reported that inpatient populations have prevalence rates of (mostly undiagnosed) moderate to severe OSA in the 25% to 84% range, depending upon the medical population studied. And while not consistent across all studies, a growing number of observational studies suggest worse inpatient outcomes for patients with undiagnosed OSA, particularly in those receiving opioids during admission.

In addition, higher one-to-three-month readmission rates have been reported in patients with cardiopulmonary disease and previously undiagnosed OSA compared with those without OSA. And finally, OSA is associated with a 17% increased length of stay per whole day increment and 67% higher costs, even after accounting for potential confounding factors. However, the impact of diagnosing OSA and intervening (ie, via sleep medicine consultation, enhanced monitoring, initiation of therapy, ensuring follow-up, etc) on inpatient outcomes has not been critically analyzed.

As a result of growing interest in this field and the concerns described, the American Academy of Sleep Medicine (AASM) convened a multidisciplinary task force of experts and methodologists to conduct a systematic review of the literature using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) methodology to develop, analyze, and assign strengths to recommendations. Following nearly five years of work, the AASM Task Force on the Evaluation and Management of OSA in Medically Hospitalized Patients published its systematic review and clinical practice guidelines (CPG) in December 2025.¹

A Good Practice Statement, based on expert consensus, preceded the recommendations. This statement specifies that, “For medically hospitalized adults with an established diagnosis of sleep-disordered breathing and on active treatment, existing treatment should be continued rather than withheld, unless contraindicated.” This was thought to be consistent with good patient care and should be the standard of care.

The subsequent recommendations are as follows:

1. For medically hospitalized adults at increased risk for OSA, the AASM suggests in-hospital screening for OSA as part of an evaluation and management pathway that incorporates diagnosis and treatment with PAP rather than no in-hospital screening.
   conditional recommendation, low certainty of evidence
   
   Remarks: Screening may include validated questionnaires and/or screening with overnight high-resolution pulse oximetry (HRPO). High risk for OSA is defined by signs and symptoms that suggest moderate to severe OSA (eg, excessive daytime somnolence of more than two of the following: diagnosed hypertension; habitual loud snoring; witnessed apnea, gasping, or choking and/or association of high-risk comorbidities).

2. For medically hospitalized adults with newly diagnosed OSA or with a prior established diagnosis of moderate to severe OSA but not currently on treatment, the AASM suggests the use of inpatient treatment with PAP rather than no PAP.
   conditional recommendation, low certainty of evidence

3. For medically hospitalized adults at increased risk for or with an established diagnosis of OSA, the AASM suggests that sleep medicine consultation be available as part of an evaluation and management pathway, rather than no sleep medicine consultation.
   conditional recommendation, very low certainty of evidence
   
   Remarks: It is recognized that there will be variability in the availability of hospital-based expertise and resources specific to sleep medicine consultation; therefore, suggestions are provided in terms of the construct of the model of care, which will need to be tailored according to resources and personnel available.

4. For medically hospitalized adults at increased risk for or with an established diagnosis of OSA, the AASM suggests a discharge management plan to ensure timely diagnosis and effective management of OSA, rather than no plan. 
   conditional recommendation, very low certainty of evidence

There was insufficient data to make a recommendation regarding enhanced inpatient physiological monitoring for medically hospitalized patients with known or suspected OSA.

The conditional nature of the recommendations was made based on the low degree of certainty about the evidence resulting from imprecision and potential for bias, as the randomized studies evaluated were small in nature, used different methodologies, and reported variable outcomes. Observational studies considered were generally consistent in terms of direction of the outcomes but lacked adequate controls, which contributed to the weak strength of the recommendations.

There are some additional caveats to be considered: 

1. The initial charter was to address sleep-disordered breathing (SDB), ie, inclusive of disorders such as central sleep apnea and sleep-related hypoventilation, in addition to OSA. However, most of the evidence was focused on OSA; limited and generally low-quality data addressed other forms of SDB in medically hospitalized patients. Thus, the task force did not believe it was appropriate to extrapolate evidence from OSA to non-OSA sleep-related breathing disorders.
2. The guideline is not intended to provide guidance on management of hospitalized patients with acute or chronic respiratory failure requiring noninvasive ventilatory support, nor are the recommendations crafted to address OSA considerations in the peri-operative surgical or procedural inpatient population.
3. It is recognized that the hospital environment can be disruptive to sleep in terms of interruptions of sleep, light exposures, and noise; however, these aspects and the consideration of sleep disorders other than SDB (eg, parasomnias, restless legs syndrome) are not addressed by this guideline.

The hope is that these recommendations will prompt institutions and inpatient/sleep providers to begin to consider and develop, if not already doing so, processes and protocols for screening, diagnosing, and potentially offering PAP therapy in medically hospitalized patient populations, particularly those considered at increased risk for OSA (ie, cardiac units, respiratory units, neurologic units). The default of not considering OSA in these inpatient populations may contribute to the status quo of persistently high rates of undiagnosed OSA and fail to improve patient outcomes.

Recognizing that a formal overarching protocol of screening for OSA, diagnosing OSA, and implementing therapy for previously unrecognized OSA may require significant resource investment and personnel that may not be available at many institutions, the task force discussed various models of how this process could be realized in real-world practice. A process as simple as a nursing-driven screening protocol utilizing standardized questionnaires with subsequent care coordination to ensure that those identified as high risk for OSA receive appropriate outpatient follow-up for testing and treatment could suffice based on the current recommendations.

Conversely, some institutions may be well-situated to offer a full inpatient sleep consult service with a team of consultants, standardized screening, an array of diagnostic options, and well-supported patient- centered inpatient initiation of PAP therapy with transition to outpatient care. It is expected that there will be variable institutional approaches to this, and there will be numerous local factors that will influence decision-making, but the AASM believes it is time to begin to address this issue in clinical practice.

There are many opportunities for future investigation to better inform the benefits and values of various approaches to diagnosis and management of SDB in the inpatient setting. The field needs large, rigorously conducted studies to examine optimal objective and subjective screening approaches, diagnostic testing approaches, and OSA interventions in the inpatient setting. Approaches to care for those with highly complex cardiopulmonary pathophysiology and SDB (eg, hypoventilation syndromes and central sleep apnea) are warranted.

The role of the inpatient sleep consultation service needs to be further explored. The utility of enhanced inpatient physiological monitoring to detect early warning signs of a deteriorating clinical state, cost benefit analyses, and models of workflow processes and consultative and peri-discharge care are also high priority areas for future studies. Last but not least, improved coverage of inpatient sleep testing and services will be key to addressing gaps in inpatient sleep medicine care.

This article was originally published in the Summer 2026 issue of CHEST Physician.

1. Mehra R, Auckley DH, Johnson KG, et al. Evaluation and management of obstructive sleep apnea in adults hospitalized for medical care: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2025;21(12):2193-2203. doi:10.5664/jcsm.11864