Adjunctive bright light therapy to enhance continuous positive airway pressure adherence in patients with comorbid major depressive disorder and obstructive sleep apnoea syndrome: study protocol for a randomised sham-controlled trial

Abstract

Introduction

Obstructive sleep apnoea syndrome (OSAS) co-occurs with major depressive disorder (MDD) in approximately 50% of cases, and this comorbidity is associated with greater severity of depressive symptoms, sleep disturbances and poorer clinical outcomes. Although continuous positive airway pressure (CPAP) therapy is effective in treating OSAS and alleviating symptoms of MDD, poor adherence during the initial weeks of treatment remains a major clinical challenge. Bright light therapy has been shown to rapidly improve sleep, wakefulness, cognitive function and mood in patients with MDD. Given these complementary mechanisms, we propose that combining CPAP with bright light therapy may enhance patient adherence during the critical initial phase of CPAP treatment, ultimately leading to better clinical and sleep-related outcomes.

Methods and analysis

In a single-centre, double-blind, sham-controlled study with two parallel arms, 130 patients with both MDD and OSAS requiring CPAP therapy will be randomly assigned to receive either 14 sessions of 30 min active bright light therapy (n=65, 1200 Lux, peak wavelength at 500 nm) or 14 sessions of 30 min sham bright light therapy (n=65, 33 Lux, peak wavelength at 600 nm) during the first 2 weeks, following CPAP initiation at home. The primary outcome will be adherence to CPAP (in hours per 24 hours during the 14 days of investigation). Secondary clinical outcomes will include changes in depressive and anxiety symptoms. Secondary sleep-related outcomes will include both objective sleep parameters (polysomnography, melatonin and actimetry) and standardised psychometric scales. The persistence of treatment effects at 1-month follow-up will also be evaluated.

Ethics and dissemination

The study was approved by an ethics committee (CPP Nord Ouest IV, Lille, France), and the French National Agency for Medicines and Health Products Safety registration number 2024-A01551-46. The findings will be disseminated through international peer-reviewed publications, presentations at scientific conferences and outreach at public conferences.

Trial registration number

NCT06781593.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• First randomised, sham-controlled trial to evaluate the impact of bright light therapy on continuous positive airway pressure (CPAP) adherence in patients with comorbid major depressive disorder and obstructive sleep apnoea syndrome.

• Multimodal assessment combining objective measures (actimetry, melatonin, polysomnography, CPAP usage logs) and validated psychometric scales ensures robust data collection.

• Blinding procedures for both participants and investigators are reinforced by identical devices and centralised randomisation, minimising bias.

• The specific inclusion criteria (eg, Montgomery-Åsberg Depression Rating Scale ≥15) may limit generalisability to broader populations.

• Data collection is limited to the first month of treatment, which may not reflect long-term adherence or sustained therapeutic effects.

Introduction

Major depressive disorder (MDD) is a frequent and disabling psychiatric condition affecting about 4% of the world’s population, according to the WHO.¹ Despite advances in the management of this condition, it is estimated that approximately one in three patients exhibits only a partial response to currently available treatments,² leading to major disabilities and societal burden.³ MDD is characterised by tiredness, pessimistic thoughts and sometimes suicide and disabling sleep disorders.⁴

Among sleep disorders, obstructive sleep apnoea syndrome (OSAS) is a common comorbidity in people with MDD.⁵ OSAS is a nocturnal ventilation disorder characterised by repeated pauses in breathing during the night, leading to sleep fragmentation and awakenings with or without oxygen desaturation. Considering an apnoea-hypopnoea index (AHI) ≥5 events per hour, the prevalence of OSAS is estimated to range between 9% and 38% in the general population,⁶ and it may affect up to one in two patients with MDD.^{5 7}

The relationship between MDD and OSAS is complex and bidirectional, despite limited understanding of the underlying mechanisms that connect these two comorbid conditions.⁸ OSAS is characterised by recurrent nocturnal awakenings and reduced sleep quality, resulting in sleep fragmentation which may contribute to the onset or worsening of depressive symptoms.⁹ Kaufman et al reported that patients with OSAS are over 3.11 times more likely to present with MDD than those without OSAS.¹⁰ Data from the literature suggests that treating sleep disturbances in individuals with MDD is associated with higher remission rates and a reduced risk of depressive episode relapse.¹¹ Moreover, the severity of OSAS has been shown to correlate with the severity of depressive symptoms.¹² More specifically, in patients OSAS, continuous positive airway pressure (CPAP) adherence has been associated with better clinical outcomes, including a reduction in major depressive symptoms.¹³

CPAP is the gold standard treatment in severe OSAS and allows for significantly reducing daytime sleepiness and insomnia symptoms.^{14 15} Despite its effectiveness, CPAP treatment is relatively poorly tolerated, with adherence ranging from 30% to 60%¹⁶ and frequent treatment interruptions within the first few weeks after initiation, before the therapeutic benefits can be fully realised.17,19 The therapeutic effects of CPAP tend to emerge gradually, with noticeable clinical benefits often observed within the first 2 weeks and further reinforced over the subsequent months. By the end of the second week, patients often exhibit general neuropsychological improvements; however, specific benefits in cognitive functions remain limited at this early stage.²⁰ The efficacy of CPAP therapy is highly dependent on patients’ adherence, particularly during the early stages of treatment. As previously mentioned, adherence is often compromised.¹⁶ A key factor contributing to treatment intolerance in patients with OSAS is the presence of mood disorders, which represent significant barriers to sustained engagement with therapy. Therefore, it is imperative to explore complementary strategies, such as interventions targeting mood disorders, that can enhance early adherence to CPAP therapy, especially during the initial stages of treatment. Bright light therapy shows promise as an effective intervention for targeting mood disorders. It is a non-invasive intervention that helps reduce fatigue, drowsiness, insomnia and promotes the synchronisation of the circadian timing system²¹ and of the sleep-wake cycle.^{22 23} Bright light therapy has been reported to enhance sleep and mood regulation and was associated with lower depression scores in a large study including more than 6600 participants.²⁴ Evidence suggests that an immediate improvement in depressed mood can be detected after a single session of light therapy.²⁵ Beneficial effects of bright light therapy on alertness and perceived sleep quality can emerge within the first few days, with clearer clinical improvements typically observed after 1–2 weeks of daily exposure to bright light.^{26 27} Early benefits, including enhanced morning wakefulness and reduced depressive symptoms, have been reported in individuals with insomnia and early awakenings.^{22 28} Therefore, because bright light therapy can rapidly improve sleep quality and mood,²⁹ often more quickly than CPAP alone, we hypothesise that, when combined, light therapy and CPAP will exert a synergistic effect in terms of clinical benefits, conditioning a better adherence.

Accordingly, we propose that the adjunction of bright light therapy during CPAP initiation may represent an innovative approach to enhance adherence to CPAP, ultimately leading to improved sleep quality and reduced depressive symptoms in patients with both OSAS and MDD.

Aims

Primary aim

The main objective is to compare the mean CPAP adherence in patients with both OSAS and MDD receiving either 2 weeks of active bright light therapy or sham bright light therapy in combination with CPAP initiation.

Secondary aims

• To assess maintenance of CPAP adherence at 1-month follow-up (D15–D28).

• To evaluate CPAP adherence relative to total sleep time (D1–D28).

• To compare clinical psychiatric, sleep-related and objective parameters measured at D14 and/or D28 with baseline measurement (D1), including:

  • Psychometric clinical parameters: depression, anxiety, fatigue, somnolence, subjective sleep quality, insomnia, cognitive disorders, social jet lag.

  • Objective parameters: actimetry (D-7 to D28) and polysomnography (pre-CPAP performed between D1 and post-CPAP after D14).

• To identify differences in response to bright light therapy and CPAP adherence revealed by stratified analysis based on initial social jetlag, AHI score, daytime sleepiness, as well as depression type (unipolar or bipolar) and time of year (15 October and 15 May: period of low sunlight; 16 May–14 October: period of high sunlight).

• To evaluate the association or the correlation between CPAP adherence and other measures.

Exploratory aims

As exploratory objectives, we will analyse data collected as part of standard care in a subgroup of participants. We aim to compare between the two groups and before/after treatment, the effect of 2 weeks of active versus sham bright light therapy in association with the initiation of CPAP treatment on sleep architecture (polysomnography PSG baseline and D28) and melatonin secretion before falling asleep (D1, D14 and D28).

Outcomes

Primary outcome

Mean CPAP adherence (use, in hours per 24 hours), as recorded by the CPAP device, during the first 14 nights of treatment (D1–D14).

Secondary outcomes

Secondary outcomes will include:

• CPAP adherence at 1-month follow-up (mean use).

• Difference (in absolute value) between the groups in the ratio CPAP adherence/mean total sleep time for 1 month.

• Clinical and sleep-related self-questionnaire:

  • Depression will be measured using the 10-item Montgomery-Åsberg Depression Rating Scale (MADRS),³⁰ a standardised clinician-administered tool designed to evaluate depressive symptomatology across 10 items.

  • Self-evaluated depression will be assessed using the 13-item Beck Depression Inventory,³¹ assessing cognitive, affective and somatic symptoms of depression.

  • Anxiety will be measured using the seven-item anxiety subscale of the Hospital Anxiety and Depression Scale, a self-report questionnaire to detect states of anxiety in hospital and outpatient settings.³²

  • Self-perception of general daytime sleepiness using both (1) the eight-item Pichot Fatigue Scale³³ and (2) the eight-item self-administered Epworth Sleepiness Scale.³⁴

  • Self-perception of daytime sleepiness will be assessed using the 19-item Pittsburgh Sleep Quality Index,³⁵ a widely used self-report questionnaire that assesses sleep quality and disturbances over a 1-month period.

  • Self-perception of insomnia will be assessed using the seven-item Insomnia Severity Index,³⁶ a brief self-report instrument measuring the patient’s perception of his or her insomnia.

  • Self-evaluation of the frequency of lapses in attention, memory and action in everyday life using the self-report 25-item Mac Nair Memory Complaint Scale.³⁷

  • Individual’s chronotype based on sleep-wake behaviour on workdays and free days will be measured with the Munich Chronotype,³⁸ which provides estimates of sleep timing, sleep duration and social jetlag. It is widely used in research on circadian rhythmicity.

• Difference between the two groups (active bright light therapy vs sham bright light therapy) concerning the following objective parameters:

  • Actimetry: total sleep time in minutes; time in bed in minutes; sleep efficiency in percentage (TST on TPL); sleep fragmentation in minutes (intrasleep wake duration); mean wake after sleep onset; sleep onset latency; motor activity in absolute value; Daily nap time in minutes; motor activity (psychomotor slowing) and light exposure (sedentary lifestyle), nap time. From the Non-Parametric Circadian Rhythm Analysis of actimetry: intradaily variability (IV), intradaily stability (IS), relative amplitude (Ra), 5 hours of lowest (L5), highest (HS) motor activity.

• Polysomnography: sleep architecture and continuity, including Sleep onset (in minutes); onset of REM sleep (in minutes); amount of REM sleep (in minutes); amount of deep slow wave sleep (in minutes); ratio (REM+SWS)/TST; Arousal/microarousal index per hour.

Exploratory outcomes

Comparative measurements between the two groups and before/after treatment within each group, of the effect of 2 weeks of active light therapy versus sham in association with the implementation of CPAP treatment on sleep architecture (PSG) as well as on melatonin secretion profile during the first 2 weeks of fitting (D0–D14, D28), in patients with a diagnosis of OSA (AASM, 2017) and suffering from an associated major depressive state (Diagnostic and Statistical Manual of Mental Illnesses, DSM-V and MADRS>15).

Methods

Design and participants

This is a single-centre, prospective, double-blind, randomised (1:1 per block) and sham-controlled study, including 130 patients diagnosed with MDD according to the DSM-V criteria and OSAS according to The American Academy of Sleep Medicine (AASM) criteria. These patients require CPAP therapy based on either ventilatory polygraphy or polysomnography measurement.

Participants will continue to receive standard clinical care throughout the study period. Following inclusion, concurrent participation in other interventional clinical trials will not be permitted. Medication regimens must remain unchanged for the duration of the study; participants are expected to maintain their existing pharmacological treatment. Participants assigned to the sham group who express interest in bright light therapy following the study may be offered access to active bright light therapy as part of standard care, depending on clinical indications and availability of the device.

Sample size

The number of subjects required was calculated according to Ledermann and colleagues³⁹ and based on a difference in mean CPAP adherence over the first 14 days of treatment of at least 0.5 hours of total sleep time (over the same period) between the active and sham bright light therapy groups, and a pooled SD of the difference of 0.8, a two-sided alpha 0.05 risk for a power of 0.9. According to these estimates, at least 110 patients, ie, 55 patients per group, are required to complete our study with sufficient power.

With a risk of attrition set at 15%, we estimate that the total number of subjects to be included in this study is 130 patients, that is, 65 patients per group.

Regarding exploratory objectives and polysomnography, due to technical and practical limitations, we have decided that a subsample of 30 participants is feasible to achieve exploratory objectives. Although a formal sample size calculation was not necessary for this exploratory analysis, we estimated that 30 participants (15 in each group) would be sufficient to detect an effect size of 0.53 (Cohen’s d), with an 80% statistical power, using a paired-samples two-tailed t-test, and an alpha value of 0.05. This corresponds to a moderate effect size as defined by conventional standards, and it is consistent with past sleep research investigating changes in REM sleep duration after antidepressants.⁴⁰

Procedure

A schematic illustration of the study design and participant flow is provided in figure 1.

Figure 1. Schematic representation of the study design and participant flow. At day 7, participants undergo a clinical evaluation, installation of the actimetry watch and the first polysomnography. At day 0, treatment initiation according to randomisation, a clinical evaluation and melatonin profiling (four samples) are performed. At day 14, a post-treatment evaluation is conducted, including clinical assessment, actimetry, melatonin measurement and polysomnography. At day 28, the follow-up period ends with clinical and melatonin measurements. CPAP, continuous positive airway pressure; D, day.

Screening visit: at least 2 days before inclusion

The study will take place in Le Vinatier, Hospital center, in Bron, France, specifically in the Michel Jouvet Unit, which specialises in the screening and treatment of sleep disorders in patients with psychiatric conditions.

During the screening visit, the investigator will verify the eligibility according to inclusion and non-inclusion criteria (see table 1) of patients, including a clinical routine care evaluation of the severity of depression using the MADRS. The investigator will also provide them with information about the study.

Table 1. Inclusion and exclusion criteria.

Inclusion criteria

Adults consulting at the Michel JOUVET unit, CH Le Vinatier, Bron, France Adults with a diagnosis of major depressive episode (DSM-V) At least mild severe depression (MADRS score ≥15) Adults with a diagnosis of OSAS based on ventilatory polygraphy or polysomnography requiring CPAP (AASM 2017 criteria)46

Non-inclusion criteria

Nasal obstruction Refusal to participate Pregnant and breastfeeding women Under guardianship or curatorship Unstable psychiatric (eg, hypomania, high suicidal risk) or physical pathology, the last 3 weeks Other psychiatric, neurologic or somatic conditions Contraindication to bright light therapy: retinopathy, retinitis pigmentosa, diabetic retinopathy, macular degeneration, glaucoma, recent eye surgery (less than 3 months)

AASM, American Academy of Sleep Medicine; CPAP, continuous positive airway pressure; DSM, Diagnostic and Statistical Manual of Mental Illnesses; MADRS, Montgomery-Åsberg Depression Rating Scale; OSAS, obstructive sleep apnoea syndrome.

Visit 1: inclusion visit (D-7)

The inclusion visit will take place at the Michel Jouvet sleep unit, Le Vinatier psychiatric hospital. During this visit, informed consent will be obtained by specialised psychiatrists, sleep medical physicians from the sleep unit, who serve as study investigators and are not directly involved in the routine psychiatric care of the patients they enrol. Patients will receive full and comprehensible information about the study, including its objectives, procedures (see table 2), potential risks, safety measures and their rights to refuse or withdraw from participation at any time without any consequence to their ongoing care. After the signature of the written informed consent, patients will be randomised to receive either CPAP and active bright light therapy (n=65) or CPAP and sham bright light therapy (n=65). A randomisation list will be drawn up by the study sponsor (1:1, randomisation by block of variable sizes), with an A and a B group. As per double-blind procedures, the investigators will not know which condition (active bright light therapy or sham) has been assigned to group A or B. The randomisation code will be computed by the REDCap software.

Table 2. Assessments at the different time points.

	Inclusion (D-7 to D-2)	Visit D1	Visit D14 (±5 days)	Visit D28 (±5 days)
Eligibility assessment	✓
Informed and written consent	✓
Randomisation	✓
Clinician-rated evaluation (MADRS, HADS)	✓			✓
Booklet with self-questionnaires (clinical, sleep-related)		✓	✓	✓
Actimetry (D-7 to D28)	✓
Sleep agenda	✓
CPAP therapy adherence report		✓ (D1–D28)
Bright light therapy (logbook and nap diary)		✓ (D1–D14)
Only for a subsample of 30 participants
Polysomnography		✓		✓
Melatonin concentration		✓	✓	✓

CPAP, continuous positive airway pressure; D, day; HADS, Hospital Anxiety and Depression Scale; MADRS, Montgomery-Åsberg Depression Rating Scale.

Then, the investigator will collect sociodemographic and clinical characteristics, including weight, height, age, sex, initial AHI, coffee consumption, cigarette, alcohol and other toxic substances, reduction/cessation of substance use in the last 3 weeks, treatments, treatment modification in the last 3 weeks, introduction of psychotherapy in the last 3 weeks, neuromodulation sessions in the last 3 weeks, increase or decrease in physical activity in the last 3 weeks, increase/decrease in sun exposure in the last 3 weeks, bright light therapy conducted in the last 3 weeks, unusual negative life event during the past 3 weeks, associated diagnoses (psychiatric or physical that may affect sleep or limit the use of CPAP), marital status, co-sleeping or not, professional activity or not, chronotype, nap time per day, vulnerability to glare.

Furthermore, as part of the protocol, each patient will be equipped with an actimeter to continuously monitor activity and sleep patterns over a 35-day period (with instructions to press the button at bedtime and on waking, including for naps) and a sleep diary. In the subsample of 30 participants who will accept to participate in the polysomnography exploratory study, a first polysomnography recording using the SOMFIT device (Compumedics, France) will also be conducted prior to CPAP therapy initiation.

Visit 2: initiation of CPAP therapy at patient’s home (D1)

A technician will visit the patient at home between 7 and 14 days after the inclusion visit to instal the CPAP device (as in the routine care) and provide the bright light therapy device (active or sham glasses), its logbook and nap diary, along with a booklet containing all measurement scales/questionnaires. Bright light therapy begins the following morning (D2) and is used every morning for 14 consecutive days. CPAP is used nightly and during daytime naps starting from day 1 (D1), in combination with the light therapy.

Furthermore, as part of routine care for the subsample of participants with polysomnography, four saliva samples will be collected using Salivette (Stardest, France) at 1-hour intervals, beginning 4 hours prior to the participant’s scheduled bedtime. Participants will be instructed to collect these samples under low-light conditions on three separate days: D1, D14 and D28.

Visit 3: end point visit (D14 ±5 days)

As part of routine care, the technician will return to the patient’s home to assess CPAP efficacy and adherence at D14. During this visit, the bright light therapy and its associated completed logbook and nap diary will be collected. In addition, the booklet containing self-assessment scales (previously sent) will also be collected. Technicians responsible for routine home visits for CPAP installation will receive specific training on the study protocol and its methodology.

For the exploratory study, participants will also be provided with salivary samples for subsequent melatonin assay.

Visit 4: follow-up (D28 ±5 days)

The final follow-up visit will be conducted at the Michel Jouvet Unit by the investigator. During this visit, depressive symptom severity will be evaluated using the MADRS. Actimetry data and self-administered questionnaires completed over the study period will be collected. Additionally, patients will return the previously provided booklet containing the completed psychometric assessments.

For participants enrolled in the exploratory study, a third set of salivary samples is collected under low-light conditions to assess melatonin levels and undergo a second polysomnography.

For an overview, table 2 and figure 1 summarise the different assessments planned at each visit.

Intervention

Bright light therapy

Bright light therapy will be administered daily using light therapy glasses (DAYVIA SUNACTIV 2, Lille, France; CE-certified, EAN: 0745844429340, class IIa medical device CE0459). Participants will wear the device for 30 min sessions each morning on waking, over the first 2 weeks of the study, concurrently with the initiation of CPAP therapy. Adherence to bright light therapy will be monitored through self-reported usage logs (usage report).

This device is contraindicated only in rare cases (specific ophthalmic pathologies: retinopathy, retinitis pigmentosa, diabetic retinopathy, macular degeneration, glaucoma). In case of doubt, ophthalmological advice will be sought by the investigator. According to Botanov & Ilardi, there was no significant difference between active and sham regarding the occurrence of side effects associated with the acute use of bright light therapy (eg, headaches 32% vs 27%, eye strain 26% vs 22%, nausea 7% vs 2%) and no serious adverse events were reported.⁴¹ The emitted light spectrum is free of ultraviolet rays. Active bright light therapy emits an intensity of 1200 Lux at the eye level and a cyan-coloured light spectrum (peak wavelength at 500 nm), which corresponds to a retinal exposure of 361.91 µW/cm2. The sham will be delivered using the same model of glasses but set up as a ‘sham’ version, with an emitted intensity of <50 lux and a yellowish spectrum (peak wavelength at 600 nm) such that it does not activate melanopsin cells, which are responsible for the non-visual⁴² and the psychotropic effects of light. Investigators will remain blind to the two light therapy conditions, as both sessions (active and sham) will be conducted by the patient at home, without knowledge from the investigators. Furthermore, patients are only provided with a single device and therefore have no opportunity to compare the active and sham models.^{43 44}

Continuous positive airway pressure

To ensure consistency in the administration and measurement of therapy, a single CPAP device model will be used throughout the study: the S. BOX system (SEFAM, Villers-lès-Nancy, France). All patients will receive the same standardised initial settings, including a CARA nasal mask and a pressure range of 4–14 cm H₂O with humidification. The device will be equipped with integrated software that continuously records compliance and AHI data, as well as the number of days of use. This data will be retrieved from the Linde Homecare cloud through telemonitoring.

Polysomnography

Polysomnography will be conducted using the SOMFIT device (Compumedics, Strasbourg, France), a CE-marked wireless portable device worn at home during sleep. It will record and analyse EEG, EMG, ECG and breath-related signals to assess sleep architecture (Profusion neXus 360). Data will be transmitted via Bluetooth to a smartphone app (iOS/Android) and accessed through a secure, password-protected platform. Recordings will be performed twice, once before the inclusion and at D28.

Actimetry

Motor activity and sleep-wake rhythms will be assessed using a non-invasive actimetry watch (MotionWatch 8, CamNtech, UK), worn on the non-dominant wrist from D-7 to D28. Participants will press the event marker button at bedtime and on awakening, including for naps. Data will be downloaded at the end of the study using MotionWare V.1.3.17 software.

Patient and public involvement

Patients or members of the public were not involved in the design, conduct, reporting or dissemination of this research.

Statistics

Generality

Data analysis will be conducted using the latest available versions of JASP (Amsterdam, Netherlands) and RStudio (Boston, USA) at the time of analysis by an investigator blinded to treatment allocation and with access to the anonymised dataset. Multiple types of data will be gathered using multiple connected devices (polysomnography, actimetry, sleep-related data from the CPAP device), as well as through the administration of various self-report and clinician-rated questionnaires. Analyses will be conducted based on the intention-to-treat (ITT) population. Missing data will undergo specific analysis to determine their nature and underlying causes, to identify an appropriate imputation method. Statistical significance will be set at p<0.05, and p values will be reported to three decimal places. No interim analyses are planned.

Primary analysis

The primary analysis will assess the difference in mean CPAP usage (hours per 24-hour period) during the first 14 days (D1–D14) between participants receiving active vs sham bright light therapy. Group comparisons will be performed on the mean duration of the use during the study period using the Student’s t-test if normality assumptions are met; the Wilcoxon otherwise, or a generalised linear model with the 14 records, depending on the data distribution.

Secondary analysis

Secondary analyses will be conducted similarly, using the Wilcoxon test or Student’s t-test depending on the normality of the data for quantitative variables, and the Fisher’s exact test or the χ² test for binary variables depending on the distribution. Correlations between selected variables will be explored using Pearson’s or Spearman’s correlation tests, depending on the normality of the data. Intergroup comparisons will be performed using Wilcoxon tests or Student’s t-tests, depending on the data distribution. Multiple analyses will be accounted for using false discovery rate correction for all statistical tests. As a first step, exploratory factor analysis will be used to identify the underlying relationships between the large number of measured variables.

The influence of the presence or absence of social jetlag, the type of depression (unipolar or bipolar disorder) and the severity of OSA on CPAP adherence will be carried out using a linear model (continuous explanatory variable) or a logistic model (binary explanatory variable) with the randomisation group as explanatory variables and each of the variables iteratively.

Exploratory results

We will perform a generalised linear regression model to assess the evolution of the percentage of REM sleep before and after bright light therapy, while accounting for the effect of potential confounding variables (eg, age, sex, CPAP adherence, AHI, chronotype, social jetlag). These confounding factors will be previously identified through univariate analysis.

Discussion

The aim of the present study is to investigate whether the combination of bright light therapy with the initiation of CPAP treatment enhances CPAP adherence, thereby reducing depressive symptoms and improving sleep-related abnormalities in patients with MDD and comorbid OSAS requiring CPAP therapy. We hypothesise that by promoting circadian alignment and alleviating depressive symptoms, bright light therapy may enhance key factors essential for optimal CPAP adherence, including both subjective and objective sleep quality as well as patient motivation. This hypothesis is supported by recent studies indicating that bright light therapy not only alleviates depressive symptoms but may also influence sleep architecture and melatonin secretion, both of which are known to be disrupted in individuals with co-occurring MDD and OSAS.^{12 45} To the best of our knowledge, this study is the first to prospectively and comprehensively assess the impact of bright light therapy on CPAP adherence in individuals with comorbid MDD and OSAS, using a double-blind, sham-controlled, parallel-arm trial design. The inclusion of a sham light therapy arm strengthens causal inference by accounting for placebo effects linked to environmental changes.

Following 14 days of concurrent administration of bright light and CPAP therapies, we will evaluate both the acute and sustained effects on sleep-related and psychiatric clinical outcomes. If proven effective, this dual intervention could support the incorporation of bright light therapy as a cost-effective and well-tolerated adjunct to improve CPAP adherence and clinical outcomes in clinical settings. This approach may be especially valuable during the critical initial adaptation period of CPAP and could be readily implemented in routine clinical practice.

Nevertheless, several limitations warrant consideration. First, although the study design accounts for many confounding variables, factors such as cognitive functioning, ongoing pharmacological treatments, level of treatment resistance, previous exposure to bright light therapy, engagement in psychosocial support and device-related discomfort may still influence CPAP adherence and should be carefully monitored and controlled. Second, the study’s focus on the initial 2 weeks following CPAP initiation and the first month of use, while clinically relevant, limits our ability to draw conclusions about long-term adherence patterns. Future studies with extended follow-up periods will be necessary to assess the long-term sustainability of CPAP adherence and clinical outcomes, as the current study primarily focuses on the early, critical adaptation phase. Moreover, although self-reported logbooks provide valuable daily data to monitor the use of light therapy in our study, future research may benefit from incorporating device-based adherence measures to enhance the objectivity and reliability of adherence assessment. However, depending on the patients, we will verify the correct use of light therapy in some participants by comparing data from the actimetry watch equipped with light sensors to the information recorded in the self-reported logbooks. In addition, a potential ‘healthy adherer effect’ may influence the findings, as participants who are more compliant with bright light therapy might also be inherently more adherent to CPAP. While randomisation is expected to minimise this bias, it cannot be entirely excluded. Sensitivity and subgroup analyses based on adherence profiles will therefore be performed to assess the robustness of the intervention effects.

Lastly, the inclusion of patients with MDD and a MADRS score greater than 15 limits the generalisability of the findings to other psychiatric or neurological populations experiencing low mood and sleep-related disturbances, who might also benefit from the proposed combined therapy.

In conclusion, this study addresses an essential gap in the treatment of comorbid OSAS and MDD by integrating two evidence-based interventions: CPAP and bright light therapies. If successful, it could pave the way for a novel therapeutic paradigm that incorporates chronobiological interventions to optimise adherence and long-term outcomes in patients with complex sleep and mood disorders.

Ethics and dissemination

The study was approved by an independent ethics committee (Comité de Protection des Personnes—CPP Nord Ouest IV, Lille, France) on 4 December 2024, and the French National Agency for Medicines and Health Products Safety registration number 2024-A01551-46. The study protocol was preregistered in a public Clinical Trials Registry (ClinicalTrials.gov; number NCT06781593) on 29 April 2025. Written informed consent will be obtained from all participants. There will be no exclusion period at the conclusion of the trial. The sponsor and investigators commit to conducting the clinical investigation in compliance with French law (Law no 2012-300, 5 March 2012), European Regulation no 2017/745, ISO Standard 14155:2020, and the Declaration of Helsinki. The study will follow the approved protocol, except in case of emergencies. In case of an emergency, unblinding can be performed 24 hours a day on the e-CRF by the investigator or by the clinical trial vigilance team via the on-call pharmacist. Investigators will ensure proper consent collection and adverse event reporting. The sponsor will be in charge of auditing and monitoring. Data will be processed in compliance with French data protection laws (Law no. 78-17 of 1978, amended by law no. 2018-493) and the EU GDPR (Regulation 2016/679). This investigation complies with the ‘Reference Methodology’ (MR-001) per article 54(5) of the amended 1978 law and has authorisation from the National Commission for Information Technology and Civil Liberties (CNIL). The sponsor of the study, the CH le Vinatier, has committed to adhering to this methodology and has obtained dedicated insurance coverage.

Several international publications and communications during national and international congresses are planned. The primary outcome clinical measure will be published in the first article. The results from secondary outcomes and exploratory studies will be published in separate articles. Authorship will be determined according to the ICMJE guidelines, based on each individual’s contributions at the time of publication and their prior involvement in the study design.

Trial status

The current protocol version is 1 (30 August 2024). All potential major changes to the protocol will be reported to the relevant regulatory authorities. The recruitment started in January 2025 and is still ongoing at the time of this manuscript submission.

Availability of data and materials

Data, script of analysis and anonymised participant-level data will be shared on reasonable request by the investigators and may be published in a depository services website with public access after a 3-year embargo following the publication of the principal study (clinical primary outcome), as described in the trial registration.

Data sharing plan (summary)

Deidentified individual participant data underlying future published results will be made available on reasonable request after publication. Requests should be submitted to the corresponding author (JB or at generic address (psyr2.crnl-lyon@inserm.fr)) and will be reviewed to ensure that data are used for scientifically sound purposes. Access will be granted to qualified researchers under a data-sharing agreement that protects participant confidentiality.

Supporting documents (protocol, statistical analysis plan) may also be shared on request.

The sponsor has no role in the conceptualisation, design, data collection, analysis, decision to publish, or preparation of the manuscript.