Adherence to Continuous Positive Airway Pressure Hugely Improved during COVID-19 Lockdown in France.

To the Editor:

The impact of the coronavirus disease (COVID-19) pandemic on adherence to chronic therapies is still debated and potential concerns have been raised regarding its effect on the short- and long-term outcomes of chronic diseases (1, 2). During the pandemic, access to routine care has been restricted as resources have been reallocated, prioritizing the response to COVID-19. In addition, nationwide home confinement has been extensively used to prevent the spread of COVID-19 infection, leading in some individuals to a deterioration in their mental health, greater anxiety, and unhealthy behaviors that might have impacted adherence to chronic therapies (3).

Obstructive sleep apnea (OSA) is one of the most frequent chronic diseases, affecting nearly one billion people worldwide, and is associated with cardiometabolic comorbidities and poor health outcomes (4, 5). The first line therapy for OSA is continuous positive airway pressure (CPAP), which reopens and stabilizes the upper airway during sleep. The CPAP devices can transmit individual measurements of nightly adherence to homecare providers and physicians through remote telemonitoring platforms. This digital care function provides access to robust and objective adherence data across well-defined periods of time. Preliminary data have suggested a slight increase in CPAP adherence during the first strict COVID-19 lockdown (6, 7).

In this study, we aimed to 1) provide detailed knowledge of shifts in CPAP adherence due to lockdown through subgroup analyses by age, sex, and CPAP adherence level before lockdown (reference adhesion) and 2) determine whether any changes in CPAP adherence were related to the level of excess mortality during the first wave of COVID-19 in the CPAP users’ county of residence.

Data were collected from the remote monitoring databases of homecare providers with restricted demographic data but including age and sex. All patients had consented to their CPAP data being used for research purposes. We analyzed the impact of the first COVID-19 national lockdown in France on CPAP adherence, measured objectively by remote telemonitoring in 14,543 patients (age: median [interquartile range (IQR)], 66 [57-73] yr; male, 68.3%). A mixed linear regression model adjusted to age, sex, and prelockdown CPAP adherence was used to compare data from a pre–COVID-19 period, springtime the previous year (March 15 to April 15, 2019), with data during the first COVID-19 wave national lockdown in France (March 15 to April 15, 2020; missing values in 2020: n = 6 without significant impact). The model was validated by checking the normality of the residuals. CPAP adherence significantly increased from 394 (IQR, 305–458) to 414 (IQR, 324–479) minutes (median percentage of variation from baseline: 3.7% [IQR, −4.9 to 14.7], P < 0.01). Our model indicated that men had a 48.7% greater increase in adherence between the pre–COVID-19 period and during lockdown compared with women (absolute difference, 7.45 min [95% confidence interval, 3.26–11.59 min], P < 0.01). The interaction between age and the assessment period was significant (P < 0.01). The mixed model showed that after the lockdown, the adherence of patients aged 65–75 years was increased 20.23 minutes more compared with CPAP-treated patients under 65 years. Patients over 75 years demonstrated a 10.26 minute increase in CPAP adherence compared with those under 65 years. There was an exemplary relationship between pre–COVID-19 adherence and increase in adherence during lockdown with, in particular, a huge increase among individuals with poor adherence in the pre–COVID-19 period (Figure 1). This increase of over 30 minutes in CPAP adherence exceeds the threshold of adherence improvement considered as clinically significant (8). The geographical spatial distributions of COVID-19 infection—incident cases, number of hospitalizations, number of ICU admissions, and the level of excess of mortality by county—are collected by the French national public health agency and are publicly available data on the French government website: https://www.data.gouv.fr/fr/organizations/sante-publique-france/. Our analysis showed that high levels of COVID-19 infection parameters were not significantly associated with changes in CPAP adherence between the pre–COVID-19 and lockdown periods per county (Figure 2).

Figure 1.

Waterfall plot of the change in CPAP adherence between a pre–coronavirus disease (COVID-19) period (March 15 to April 15, 2019) and during national COVID-19 lockdown (March 15 to April 15, 2020) in France, by adherence before COVID-19. CPAP = continuous positive airway pressure; IQR = interquartile range.

Figure 2.

The geographical distribution of the level of excess of mortality during the first coronavirus disease (COVID-19) wave by county was not significantly associated with changes in CPAP adherence between the pre–COVID-19 period and during the first strict lockdown period. Data on CPAP adherence and level of excess of mortality were not available for counties in white. CPAP = continuous positive airway pressure.

The major strength of our study is to provide longitudinal data on CPAP adherence by the same individuals, allowing a comparison before and during lockdown. This enabled us to confirm that nationwide CPAP adherence slightly increased, especially in those under 65 and in men, and that improvement was most marked in subgroups with the lowest adherence before lockdown. Furthermore, this improvement in CPAP adherence did not appear to be related to the burden of COVID-19 in the county in which the patients resided. These data are of great value for understanding the mechanisms leading to poor adherence in patients with CPAP-treated OSA. At present we can only speculate on the possible mechanisms involved. First, a proportion of still-working men under 65 might have had chronic sleep deprivation with reduced sleep time and thus reduced adherence owing to their profession or employment, with additional frequent absences from their place of abode without access to their CPAP device. Lockdown curtailed their activities and allowed these individuals to recover from sleep loss, with increased duration of nightly CPAP usage and thus adherence. Second, women might have had a smaller increase in adherence than men owing to stress and duties related to children being at home. Third, CPAP-treated patients starting therapy in the spring before the pandemic (the reference period) might have increased their adherence over one year independently of lockdown effects. Lastly, OSA has been consistently reported to be significantly associated with severe COVID-19 (9) and this forewarning had been widely disseminated to patients by physicians and home care providers (10). In addition, better adherence might also be linked to a fear of hospitalization alongside patients with COVID-19. These anxieties might have triggered changes in behavior regarding CPAP use, particularly as psychological factors are known to be far greater contributors to adherence to CPAP (11) than technical aspects. In all circumstances we suggest that healthcare providers should dedicate more time to coaching and patient education so as to improve adherence.

A logical next step will be to document the evolution of these changes in CPAP adherence, including any relapse to pre–COVID-19 patterns once the strict lockdown period had ended, and then the impact of successive COVID-19 waves and the different degrees of lockdown enforced by governments.