Chronic medical comorbidities, more commonly found in patients with obstructive sleep apnea, have been associated with higher readmission and healthcare utilization rates.
Data from a nationwide, retrospective case-control cohort study showed that the presence of obstructive sleep apnea (OSA) was associated with significantly higher rates of future ICU admissions, hospital admissions, emergency room visits, and outpatient visits compared with individuals without OSA.
Senior author Amir Sharafkhaneh, MD, PhD, professor, Department of Medicine, Baylor College of Medicine, and colleagues wrote that "the presence of OSA may be a valuable marker of higher future healthcare utilization, and additional resources could be directed toward these patients during the index hospitalization. However, prospective cohort studies are required to replicate these findings."
Although OSA has been previously associated with cardiovascular comorbidities and other chronic conditions, little had been researched on the inpatient and outpatient healthcare utilization rates for these patients. To do so, investigators compared patients with and without OSA between 2009 and 2014 who were matched for demographic variables, geographic location, hospital environment, reason for admission, and severity of illness during hospitalization.
After matching, the study produced 171,824 total hospitalizations across 335 facilities and 142,151 unique patients comprised of 2 large, equally sized OSA and control cohorts (n = 85,912). Future healthcare utilization, the primary outcome, was transformed from “pure counts” to counts per 100 patient-years to adjust for the exposure time after hospitalization and to make interpretation intuitive. At admission, those in the OSA cohort had a higher proportion of cardiovascular comorbidities, such as coronary artery disease, stroke, hypertension, and heart failure.
After the index admission, overall healthcare utilization was low; however, those with OSA demonstrated significantly higher utilization parameters than controls. These patients had a higher number of admissions to the inpatient unit, admissions to the ICU, emergency room visits, and outpatient visits, even when no other vascular comorbidities were present (all, P <.001). Compared with controls, there was a greater percentage of those with OSA in the 80th and 90th percentile in all the future healthcare utilization outcomes observed, compared with controls.
Investigators also conducted an additional analysis that matched patients 1:1 for relevant cardiopulmonology diagnoses.Results remained somewhat similar, with significant increases in long-term inpatient and ICU admissions among those with OSA.
Between the 2 cohorts, short-term outcomes during the index admission were not statistically significant. The length of the hospital stay between the OSA cohort was 5.5 (±5.5) days, while the control cohort was 5.5 (±5.5) days as well (P = .02). Although numerically the same, due to the large study population, the OSA cohort had a small, but significantly higher length of hospital stay. Specifically, 395 patients with OSA and 486 controls stayed in the hospital during the index admission for 30 days or more (Proportion of patients who stayed more than 30 days, P = .002).
There were several noted limitations to the study noted by Sharafkhaneh et al, including the fact that healthcare utilization was only catered to the data captured by the database, thus limiting information on the number of future family medicine visits, medication usage, and other outcomes. Additionally, the investigators did not track the development of typical cardiovascular and cerebrovascular comorbidities or changes in OSA treatment status after the index hospitalization, as well as specifically assess their impact on future health care utilization. Lastly, the sensitivity and specificity of ICD-9 codes for OSA within the study were unclear.