In CPAP-naive people with severe sleep apnoea, low-quality evidence indicates that supportive interventions that encourage people to continue to use their CPAP machines increase usage compared with usual care. Moderate-quality evidence shows that a short-term educational intervention results in a modest increase in CPAP usage. Low-quality evidence indicates that behavioural therapy leads to a large increase in CPAP machine usage. The impact of improved CPAP usage on daytime sleepiness, quality of life and long-term cardiovascular risks remains unclear. For outcomes reflecting machine usage, we downgraded for risk of bias and inconsistency. An additional limitation for daytime sleepiness and quality of life measures was imprecision. Trials in people who have struggled to persist with treatment are needed, as currently little evidence is available for this population. Optimal timing and duration and long-term effectiveness of interventions remain uncertain. The relationship between improved machine usage and effect on symptoms and quality of life requires further assessment. Studies addressing the choice of interventions that best match individual patient needs and therefore result in the most successful and cost-effective therapy are needed.