We-PAP, A novel, couples’ - based sleep health intervention for older adults with obstructive sleep apnea: Rationale and study protocol for a web-based randomized clinical pilot/feasibility trial of the We-PAP intervention compared to an educational control


 BackgroundObstructive sleep apnea (OSA) is a serious health condition that affects approximately 30–50% of older adults and contributes to risk for cardiometabolic disorders and dementia. Despite the well-documented role of partners in treatment seeking and adherence to positive airway pressure (PAP), treatments for OSA have nearly exclusively focused on the patient and current treatments for OSA do not address co-existing sleep problems such as insomnia that are prevalent in both patients with OSA and their partners. Therefore, the goal of this study is to develop and test a novel couples-based sleep health intervention to promote adherence to PAP and improve sleep health of the couple.MethodsWe are conducting a two-arm, parallel group, single blind, randomized controlled pilot/feasibility trial to compare our novel couples-based sleep health intervention (We-PAP) to an information control group (IC). We-PAP is based on a transdiagnostic model and uses a dyadic approach including increasing effective partner support, communication skills, and couple-level goal-setting. We-PAP involves 3 sessions and is delivered via telehealth in weekly sessions. The IC includes standardized patient educational materials. Both groups receive the usual follow-up with their medical team. The study involves assessments at pre-treatment, post-intervention (approximately 1 month after starting PAP and completing We-PAP sessions or IC) and 3 months after starting PAP. Our main outcomes are feasibility and acceptability ratings. Secondary outcomes include comparing We-PAP to IC for PAP adherence, sleep quality (self-report and objective) and cognitive measures.DiscussionWe-PAP is the first couples-based transdiagnostic sleep health intervention for patients with OSA and their partners. Results of this study will be used to inform the design of a subsequent fully adequately-powered clinical trial. If successful, this intervention could significantly advance current clinical practice in the treatment of OSA and sleep health more comprehensively in older adults. Moreover, this intervention may be useful for improving sleep in other aging populations with multiple sleep and other health problems, including patients with chronic illnesses or those at risk for Alzheimer’s disease and their caregivers.Trial registrationNCT04759157

successful, this intervention could signi cantly advance current clinical practice in the treatment of OSA and sleep health more comprehensively in older adults. Moreover, this intervention may be useful for improving sleep in other aging populations with multiple sleep and other health problems, including patients with chronic illnesses or those at risk for Alzheimer's disease and their caregivers.
Trial registration NCT04759157 Background Obstructive sleep apnea (OSA) is a major public health problem that affects 30-50% of older adults 1,2 and is associated with signi cant morbidity and mortality. Older adults with OSA have a signi cantly increased risk for cardiovascular disease 3 and stroke 4,5 and have ve times the risk for Alzheimer's disease (AD) and related dementias 6,7 . The physiological consequences of OSA, including intermittent hypoxia and oxidative stress are key mechanisms that drive neurodegeneration and AD pathophysiology 8,9 . Given that 61% of adults share a bed with a partner 10 , the consequences of OSA, including fragmented sleep, reduced quality of life, and increased marital con ict, affect both the patient and partner 11,12 . Further, given that sleep fragmentation is also mechanistically linked with increased risk for cognitive decline and is a symptom experienced by both the patient with OSA and their bedpartner 13 , the public health consequences of OSA are far greater than just that experienced by the patient alone.
The rst-line treatment for OSA, positive airway pressure (PAP), is highly effective at treating OSA symptoms in the patient and improving both patient and bedpartner sleep. There is a dose-response relationship between PAP adherence and improvements in both patient and bedpartner sleep 14 , daytime sleepiness, and QOL, as well as reductions in OSA patients' cardiometabolic and AD risk factors, including hypertension 15,16 . PAP treatment has been shown to slow cognitive decline in patients with dementia 17 . Unfortunately, however, up to 80% of patients are nonadherent 16,18,19 . Patterns of PAP use in the rst 30 days strongly predict later adherence 20,21 and reimbursement of this treatment hinges on documenting adherence at 30-90 days. Therefore, interventions focusing on early use among new PAP users are vital to treatment success and to promote patient and bedpartner health.
There is a strong scienti c premise for targeting PAP adherence at the couple-level. First, a consistent body of evidence demonstrates that couples' sleep is highly interdependent, meaning that sleep in one partner affects and is affected by the other partner's sleep 12 . In fact, a PAP clinical trial showed that treating OSA was associated with a 50% reduction in bedpartners' nocturnal arousals 22 . Second, bedpartner's sleep disruption is a primary motivator for patients to seek OSA diagnosis. In contrast, 50% of OSA patients reported they would not use PAP if it disrupted their partners' sleep 23 . Third, evidence from other chronic illness populations (e.g., cancer, HIV, diabetes) shows that couples-based interventions 24 effective at improving adherence, symptom management, and patient and partner health outcomes. Finally, partner support is critical to promote PAP adherence, whereas relationship con ict may reduce adherence 25,26 . Our previous studies have shown that collaborative support (e.g., helping with the PAP machine) strongly predicted greater adherence, whereas pressure (e.g., nagging) to use CPAP 26 and relationship con ict predicted lower adherence 25 .
Only one previous pilot investigation 27 examined use of a couples' education and support intervention compared to patient-oriented education and usual care groups. Results demonstrated improvements in patient PAP adherence as well as moderate to large improvements in sleep quality and reductions in daytime sleepiness in patients and partners in the couples' group; however, only 6 of 10 couples randomized to this condition completed the intervention, highlighting the challenge of having both couple members attend in-person sessions. Collectively, these ndings provide a strong theoretical rationale and preliminary empirical support for integrating the partner into PAP adherence interventions; however, new treatment approaches are needed to reduce couples' burden and enhance treatment engagement.
Our proposed intervention is the rst treatment for OSA that uses a couples-based treatment model to target PAP adherence as well as the broader sleep health issues affecting both patient and partner. Beyond OSA, older adults are at increased risk for poor sleep health, including insomnia/poor sleep quality, circadian disruptions (typically phase advances), and irregular sleep-wake schedules, which can signi cantly increase risk for cardiometabolic disorders, cognitive decline, poor relationship functioning, and reduced QOL 28-32 . Further, OSA and insomnia commonly cooccur, with 22-55% having both conditions 33 . Moreover, partners of individuals who snore are three times more likely to have insomnia as compared to those living with non-snorers 49 . Finally, patients with insomnia have poorer PAP adherence 34 , and partners' sleep disturbances are associated with reduced PAP adherence51. Recent studies have tested insomnia treatments in patients with OSA and demonstrate promising results 33,35,36 . Importantly, no existing treatments focus on improving sleep health in partners of OSA patients, who are also a high-risk group for sleep disturbances 23 . Therefore, our intervention lls a critical gap by addressing sleep health in older adults with OSA and their partners, with the goal of improving PAP adherence, QOL, and the sleep and health of both partners.

Study design overview
This is a two-arm, parallel-group, single-blind, randomized controlled pilot trial to evaluate the feasibility and preliminary effectiveness of "We-PAP", a novel, couples-based sleep health intervention for patients with OSA and their partners compared to an information control (IC). Participants include patients who are newly diagnosed with OSA and starting PAP therapy and their partners (n = 40). After completing a pre-treatment baseline assessment, couples are randomly assigned to either a 3-session online sleep health intervention (We-PAP) or to a patient-focused information control group (IC) arm. Patients and partners complete follow-up assessments at post-treatment (approximately 1 month after starting PAP) and 3 months after starting PAP treatment. Approval for this on-going study was provided by the University of Utah Institutional Review Board (IRB_000135927). The study is registered with ClinicalTrials.gov (NCT04759157, 2/13/2021).

Theoretical framework
We-PAP is based on the transdiagnostic sleep and circadian model (i.e., a treatment that addresses common components of multiple disorders rather than a single disorder) 37 . Consistent with key, evidence-based principles from couples-based interventions for other chronic health conditions, We-PAP conceptualizes PAP adherence and both partners' sleep health as a "shared" experience and delivers education and intervention through the lens of the couples' shared sleep goals and challenges. The treatment provides education on sleep regulation and the importance of sleep for health, then uses speci c techniques for both OSA and insomnia symptoms based on best practices in cognitive behavioral (CBT) strategies to promote PAP adherence and Brief Behavioral Treatment for Insomnia (BBTI, e.g. consistent schedules, limiting napping) 38-40 . The conceptual framework of the intervention emphasizes education, as well as two couples-based themes that are integrated throughout the intervention. First, dyadic coping refers to the recognition that the illness (OSA) affects both members of the couple, which evokes a coping response in both members. Therefore, We-PAP helps OSA patients and partners conceptualize PAP adherence and both of their sleep health issues as a couple-level problem (i.e., a "we problem"). The second theme focuses on enhancing communication to promote collaborative support within the couple and reduce con ict. Substantial research in couples shows that while a supportive partner is crucial for supporting an array of health behaviors, including diet, physical activity, sleep, and treatment adherence, a high con ict relationship is a primary source of stress, which can undermine health behaviors 12,41−43 .

Recruitment, screening and consent
A total of 40 couples are being recruited from the University of Utah Health Sleep Centers. Patients who are undergoing testing (in-lab or home sleep testing) are provided a letter introducing the study and notifying them they may be contacted about the study. Study staff contact potential patients via phone and email to provide study information and if interested, conduct screening for study eligibility. After completing screening, couples schedule an online consent visit over zoom to review and sign the online consent form and arrange to complete the baseline assessment.

Eligibility criteria
Inclusion criteria for the couple include: 1) Age 50-85 years; 2) Married or living with a romantic partner for at least 1 year; 3) Patient is PAP naïve or re-initiating PAP after 3 or more years; 4) Able to read and write in English; 5) Able to access online, video-conference capabilities (Wi-Fi or cellular plan) Exclusion criteria include: 1) Self-reported diagnosis of severe comorbid sleep disorders other than insomnia (e.g., moderate or severe RLS; REM Behavior Disorder, narcolepsy); 2) Presence of severe medical and psychiatric disorders that would interfere with participation in treatment (schizophrenia, bipolar disorder, hemodialysis); 3) Patient is using supplemental oxygen or adaptive servo-ventilation.

Randomization
Upon completion of the baseline assessment, couples are randomly assigned to either We-PAP or IC using the RedCAP randomization algorithm. The randomization table, developed by the statistician (BB), uses a 1:1 ratio of We-PAP to IC with equal strata for male and female patients using a random number generator. The study interventionist con rms with the study staff that the couple meet the eligibility criteria then contacts the couple via letter and email to notify them of their group assignment.

Assessment schedule
At baseline/pretreatment, couples complete questionnaires, 7 days of wrist actigraphy and daily sleep diaries (delivered either via text message or on paper), and a brief cognitive testing battery. Approximately 1 month after beginning PAP and upon completion of their intervention, couples complete a second set of questionnaires. Finally, at 3 months after starting PAP, couples complete questionnaires, 7 days of actigraphy and daily sleep diaries and a brief cognitive testing battery. PAP use data is collected via download using the corresponding cloud based system.

Intervention description
We-Pap was developed through an iterative process that included focus groups and a brief eld trial (n = 4 couples) before beginning the RCT. This intervention is a novel, couples-focused PAP adherence and sleep health treatment that combines a transdiagnostic sleep and circadian framework with a dyadic (i.e., couples) perspective. Together, patients and partners assigned to the We-PAP intervention complete three online Zoom-hosted, videoconferencebased educational sessions conducted weekly, which utilize structured PowerPoint slides. All sessions include the themes of sleep education, dyadic coping and communication. During each session, the interventionist reviews homework (if applicable), presents session content, engages the couple in discussion and planning and assigns homework for the next session (if applicable).

Information control (IC)
In this group, couples receive a packet of standardized patient educational materials about OSA and PAP published by the American Academy of Sleep Medicine on the website Sleepeducation.org. In addition to notifying participants in this group of their group assignment, the interventionist contacts couples to ensure they have received the study materials. If the couple has questions about PAP, the interventionist refers couples to direct questions to their sleep medicine provider or durable medical equipment (DME) company.
Interventionist training, supervision, and delity monitoring All We-Pap intervention sessions are recorded for training and delity monitoring. The interventionist delivering the We-PAP intervention content has a doctoral degree in clinical psychology and is currently participating in a Behavioral Sleep Medicine Fellowship program. Training included didactics, supervised telehealth patient sessions focused on CBTi at the University of Utah Sleep Wake Center, rehearsal of We-Pap sessions prior to the eld trial, role-playing, review of pre-recorded session format provided by the lead investigators, in-person review and coaching of digitally recorded eld trial sessions, and weekly supervision meetings. The interventionist is supervised by one of the lead investigators (WT). Fidelity is monitored via a structured checklist completed by the therapist at the end of each session and 10% of sessions will be rated for delity by one of the lead investigators (WT or KB). The checklist also includes an open-ended comments section, where the interventionist writes notes to review in weekly meetings with the supervisor.

Measures
Participants complete a brief demographic measure, including age, sex, race / ethnicity, marital status, education level, employment status, household income level, substance use, and bed-sharing.
OSA severity (AHI) will be extracted from medical records. awakening and sleep quality rating. Two additional items included in the diary assess whether partners shared a bed (yes/no) and the degree to which the couple worked together to use CPAP (rated on a scale 1-5).

Sleep disturbance and sleep-related impairment
Participants complete the PROMIS sleep disturbance and sleep-related impairment adaptive measures 45 Scores are presented as t-scores, with average of 50 and SD of 10. Scores > 60 are considered elevated. Self-reported sleep is measured using a standardized sleep diary and the PROMIS sleep disturbance questionnaire.

Other outcome measures
Relationship quality Participants complete the Couples Satisfaction Index (CSI-4) 46 to evaluate an individual's self-reported degree of satisfaction, happiness, warmth and comfort, and reward with the relationship. The 4-items on the scale are summed and range from 0 to 21. Higher scores indicate higher levels of relationship satisfaction. CSI-4 scores falling below 13.5 suggest notable relationship dissatisfaction.

Cognitive function
A study staff member administers the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) forms A and B via videoconference (Zoom) 47 . This 30 min validated clinical measure was designed to assess cognitive status in adults and older adults. The ve sub-scores include attention, language, visuospatial/construction, immediate and delayed memory as well as a total score. In order to complete this measure via teleconference, couples are mailed the coding subtest in a sealed envelope and told not to open it until the assessment. Then they are asked to place the completed test in a sealed envelope during the testing session.

Statistical analysis
Statistical analyses will be conducted using Stata v15 or higher. We will begin with conducting descriptive analyses, including evaluating whether there are any baseline group differences. Any variables that emerge as being signi cantly different between the two treatment groups will be included as covariates in models predicting outcome variables. If more than three variables emerge as being signi cantly different between the treatment arms, the full set of demographics and sleep related functioning variables are used to estimate a propensity score indicating likelihood of membership in We-PAP relative to information control (IC); this propensity score is included as a covariate in all tests of study outcomes. Given the pilot nature of the proposed study, signi cance tests of study hypotheses are performed using one-sided tests with α = 0.10. Feasibility analysis will include review of the adherence and treatment satisfaction measures. Models testing preliminary e cacy will be conducted using multilevel models with separate models will be run for each outcome. The primary outcome is PAP adherence. Secondary outcomes are objective and self-reported sleep quality. Tertiary outcome measures will be explored including relationship satisfaction and cognitive outcomes.

Sample Size
Power estimates were generated using descriptive statistics for behavioral and cognitive behavioral CPAP were generated using descriptive statistics for behavioral and cognitive behavioral CPAP interventions vs. IC as reported in Wozniak et al.,48 as well as in Buysse et al., 39 and Sweetman et al. 49 Maas and Hox's (2005) 50 canonical work on necessary sample sizes for group level effects in MLM determines that group sizes of 20 members and larger is su cient for generating unbiased estimates for our secondary measures.

Discussion
The We-PAP study is the rst intervention designed to improve PAP adherence and overall sleep health for older adult patients with OSA and their partners. This intervention was developed in collaboration with patients, partners, and sleep medicine providers and aims to create a collaborative approach to PAP adherence and sleep health for both the patient and partner. The brief 3-session format and the delivery of the intervention via telehealth is designed to maximize accessibility, by allowing couples to participate from their home at a convenient time.
The signi cance of this intervention is that We-PAP represents a new conceptual approach to treatment of patients of OSA, by taking a couples-based and transdiagnostic approach to improving sleep health for both the patient and partner. This pilot trial will provide important information regarding the feasibility and acceptability of this intervention. We will recruit equal numbers of male and female patients as well as participants from diverse backgrounds. Even in the rst few months, we have been able to enroll couples from rural areas located three or more hours from the clinic, which is important because these patients have greater barriers to attending follow-up visits. We will also be powered to evaluate preliminary e cacy for PAP adherence and improvement in sleep quality. If successful, our next steps will be to test the We-PAP intervention in a fully powered RCT to evaluate the We-PAP intervention to treatment as usual in a larger and more diverse patient sample. This intervention has the potential to improve aging-related health and cognitive outcomes for older adults. This intervention may also be useful for patients with cognitive di culties, including mild cognitive impairment (MCI) and Alzheimer's disease (AD), given that PAP adherence in these populations has the potential to improve cognitive function.

Trial status
The We-PAP study began recruiting in May 2021. To date, 20 participants (10 couples) have been randomized to We-PAP or Information Control.  CONSORTextensionforPilotandFeasibilityTrialsChecklist.doc