Remote Diagnosis and Management of Obstructive Sleep Apnea

Obstructive sleep apnea (OSA) is estimated to be the third most common chronic disease in Veterans and is associated with an increased risk of hypertension, heart attacks, strokes, depression, and driving accidents. Current in-person management at a sleep center limits Veterans' access to care, prolongs patient wait times, and requires Veterans living in remote areas to travel long distances. This proposal will evaluate an innovative web-based clinical pathway to diagnose and manage Veterans with OSA. In Fiscal Year 13, the investigative team received a Veterans Health Administration (VHA) Innovation Award to create the Remote Veteran Apnea Management Portal (REVAMP), a personalized, interactive website that, when combined with other emerging telehealth technologies, is designed to improve access to care, reduce patient wait times, and allow Veterans to receive care without travelling to a sleep center. Veterans complete intake and follow-up questionnaires on the REVAMP website and perform an unattended home sleep test (HST) without in-person instructions. Sleep specialists review the findings with the patient during an initial phone clinic. REVAMP auto-populates the Veteran's questionnaire responses into templated progress notes that are exported to CPRS, the electronic medical record. Veterans diagnosed with OSA are treated with automatically adjusting positive airway pressure (APAP) units. These devices transmit data wirelessly to the website where treatment use and its effectiveness can be monitored by both Veterans and practitioners, thereby promoting patient self-management and productive patient-practitioner interactions.

The investigators' proposed prospective, randomized intervention will compare the clinical and cost-effectiveness of REVAMP management of Veterans with OSA to in-person care. Aim 1 will determine if management with REVAMP is clinically non-inferior to in-person care in terms of improvement in functional outcomes and APAP adherence. Non-inferiority of clinical effectiveness following 3 months of APAP treatment will be expressed in terms of improvement in the score of the Functional Outcomes of Sleep Questionnaire (FOSQ-10), the investigators' primary outcome measure. Adherence to APAP will be objectively monitored by wireless transmission of data from the participant's home unit. In Aim 2, patient preference, medical service use and cost will be collected every 3 months for the entire observation period to compare cost effectiveness of the two managements. Preference will be assessed by the Short Form-6D (SF 6D), and the EuroQol-5D (EQ-5D). Differences in the ratio of cost and quality-adjusted life years saved by REVAMP compared to in-person management will test the hypothesis that REVAMP management will have lower cost and equivalent outcomes. The results of Aims 1 and 2 will provide evidence to support widespread dissemination of REVAMP. Formative evaluation in Aim 3 will use qualitative (targeted phone interviews) and quantitative measures (attrition, work alliance, and patient satisfaction) to inform clinicians, administrators and other stakeholders how to implement this innovative chronic disease pathway.

Aim 1 (primary). To compare functional outcomes following 3 months of APAP treatment in Veterans with OSA randomized to REVAMP versus in-person management. The primary outcome measure in this modified intent-to-treat analysis (i.e., subjects initiated on APAP with at least one FOSQ follow-up score) will be the change from baseline in the FOSQ-10 score. Analysis will also compare the mean daily hours of APAP use in participants in the two groups initiated on APAP treatment.

Hypothesis 1a: Mean change in FOSQ-10 score among participants randomized to REVAMP management will be no more than one point less than that in participants receiving in-person management.

Hypothesis 1b: Mean daily hours of APAP use among participants receiving REVAMP management will be no more than 0.75 hour less than that in participants randomized to in-person management.

Aim 2 (secondary): To compare the differences in cost and quality-adjusted life years (QALY) between REVAMP management and in-person management. The perspective of the analysis will be that of the VA and the intention to treat analysis set will include all randomized participants.

Hypothesis 2a: Average total health-care delivery cost will be lower for participants receiving REVAMP compared to in-person management.

Hypothesis 2b: The 90% lower limit of cost per QALY ratio comparing in-person versus REVAMP manage-ment will be > $100,000 (i.e., the investigators will have 90% confidence that REVAMP is good value for the cost).

Aim 3 (exploratory): To conduct a mixed methods formative evaluation that will guide REVAMP's widespread implementation. Quantitative component: The investigators will track quantitative outcome measures across both groups including attrition, participant- and practitioner-rated therapeutic alliance (Working Alliance Inventory-Short Revised [WAI-SR]),9 and participant treatment satisfaction (Client Satisfaction Questionnaire [CSQ-8]).10 The investigators will compare the scores of the WAI-SR and CSQ-8 and attrition rates between treatment arms. Qualitative component: The investigators will explore participant- and practitioner-level perspectives, attitudes, and preferences regarding REVAMP versus in-person management, as well as barriers and facilitators to participation in either clinical pathway through phone interviews with participants from the two intervention groups, participants who withdraw from either intervention, and staff who provide care through REVAMP.

Our primary aim was to demonstrate that the efficacy of REVAMP management is not clinically inferior to in-person care. Efficacy was measured as the change in FOSQ-10 score from pretreatment to 3 months following initiation of APAP treatment. An increase of more than 1 point in mean FOSQ score is felt to indicate an important clinical difference. Based on previous studies, we assumed a common SD=2.80. Similar calculations were performed for the APAP adherence outcome based on a non-inferiority delta of -0.75 hr/day and assuming a SD=2.2. 114 participants/group were estimated to be required to achieve at least 85% power to reject the null hypothesis, i.e., find that the change in FOSQ score among patients receiving the telehealth-based management is > 1 point lower than that in participants assessed in-person and that APAP adherence among participants receiving the telehealth-based management is > 0.75 hours lower that that in participants assessed in-person.