"I know a little, but I can't do more": a COM-B-based qualitative study of CPAP use and sleep behaviour change among hospitalised patients with COMISA in China.

BACKGROUND: Co-morbid insomnia and sleep apnoea (COMISA) is a highly prevalent condition associated with worse health outcomes and greater treatment complexity than either disorder alone. Patients with COMISA demonstrate lower acceptance and use of continuous positive airway pressure (CPAP) therapy compared with those who have obstructive sleep apnoea (OSA) alone, yet the subjective mechanisms underlying this adherence penalty remain poorly understood. Qualitative research on CPAP use and sleep behaviour change in patients with COMISA is scarce, the COM-B (Capability, Opportunity, Motivation-Behaviour) model has rarely been applied to this population, and the influence of non-Western cultural norms on COMISA treatment engagement has received little attention. METHODS: A descriptive qualitative study guided by the COM-B model and Theoretical Domains Framework (TDF) was conducted at a tertiary sleep medicine centre in Hunan Province, China. Sixteen hospitalised patients with polysomnography-confirmed moderate-to-severe OSA (AHI ≥ 15) and clinically significant insomnia (ISI ≥ 15) were purposively sampled using a 2 × 2 maximum variation matrix based on inpatient CPAP adherence (good vs. poor) and receipt of a brief nurse-led inpatient CBT-I programme (with vs. without structured CBT-I). Individual semi-structured interviews were analysed using directed content analysis, with reporting following the COREQ checklist. RESULTS: The analysis generated 1,160 reference points across 20 coding nodes, synthesised into three themes and nine sub-themes aligned with the COM-B dimensions. Motivation accounted for the largest share of coded data (589 reference points) and revealed a pattern of "insomnia dominance" in which participants systematically prioritised insomnia over OSA, four patterns of fixed causal attribution that led patients to rule out particular treatments, and an emotion-identity complex linking stigma, age-related identity, and treatment rejection. Within the Opportunity dimension, an "adherence illusion" was identified: some inpatients with good CPAP adherence had not internalised motivation for post-discharge self-management. Culturally specific barriers extended beyond rural CPAP-related death stigma to include urban obligatory banquet drinking norms. CONCLUSIONS: Hospitalised COMISA patients face qualitatively different behavioural barriers from those with OSA alone, requiring COM-B-guided individualised assessment rather than uniform adherence interventions. Nursing practice should incorporate pre-discharge environmental screening, proactive CPAP experience management during the first two to three nights, and culturally sensitive approaches spanning both rural and urban social norms. The Behaviour Change Wheel provides a systematic pathway for translating these behavioural diagnoses into targeted, nurse-led intervention design.