Apnea Positive Pressure Long-Term Efficacy Study

Conditions

Lung Diseases, Sleep Apnea Syndromes, Sleep

Keywords

Obstructive Sleep Apnea

Brief summary

The purpose of this study is to determine the effectiveness of nasal continuous positive airway pressure (CPAP) therapy for the treatment of obstructive sleep apnea syndrome (OSAS).

Detailed description

BACKGROUND: Nasal CPAP therapy is in widespread use as the primary treatment for OSAS, a sleep-related breathing disorder affecting more than 15 million Americans. The therapeutic effectiveness of CPAP in providing significant, stable, and long-term neurocognitive or other functional benefits to patients with OSAS has not been systematically investigated. DESIGN NARRATIVE: The study is a randomized, blinded, sham-controlled, multi-center trial of CPAP therapy. The principal aims of the study are: 1) to assess the long-term effectiveness of CPAP therapy on neurocognitive function, mood, sleepiness, and quality of life by administering tests of these indices to subjects randomly assigned to active or sham CPAP; 2) to identify specific neurocognitive deficits associated with OSAS in a large, heterogeneous subject population; 3) to determine which deficits in neurocognitive function in OSAS subjects are reversible and most sensitive to the effects of CPAP; 4) to develop a composite multivariate outcome measure from the results of this study that can be used to assess the clinical effectiveness of CPAP in improving neurocognitive function, mood, sleepiness, and quality of life; and 5) to use functional magnetic resonance imaging to compare cortical activation before and after CPAP therapy, and to assess whether this change is associated with improvement in specific neurocognitive task performance. The primary endpoint of the study is the effect of six months of CPAP treatment on neurocognitive function. A total of 1100 subjects (550 per treatment group) will be enrolled from the patient populations at five sites (Stanford University; University of Arizona; Brigham and Women's Hospital; Massachusetts; St. Luke's Hospital, Missouri; St. Mary Medical Center, Washington).

Andrey Zinchuk, Clete A. Kushida, Alexander Walker, Andrew Wellman, Ali Azarbarzin et al. (9 authors)

European Respiratory Journal2024-11-217 citations

10.1183/13993003.01183-2024

Morning Chronotype Is Associated with Improved Adherence to Continuous Positive Airway Pressure among Individuals with Obstructive Sleep Apnea.

Knauert MP, Adekolu O, Xu Z, Deng A, Chu JH, Baldassarri SR, Kushida C, Yaggi HK, Zinchuk A.

2023-08-0111 citations

10.1513/annalsats.202210-885oc

The effect of CPAP on HRQOL as Measured by the quality of Well-Being Self-Administered Questionnaire (QWB-SA)

Salma Batool‐Anwar, Olabimpe Omobomi, Stuart F. Quan

Southwest Journal of Pulmonary and Critical Care2020-01-143 citations

10.13175/swjpcc070-19

Adherence to continuous positive airway pressure improves attention/psychomotor function and sleepiness: a bias-reduction method with further assessment of APPLES.

Holmes TH, Kushida CA.

2017-07-145 citations

10.1016/j.sleep.2017.06.022

Effects of continuous positive airway pressure on neurocognitive function in obstructive sleep apnea patients: The Apnea Positive Pressure Long-term Efficacy Study (APPLES).

Kushida CA, Nichols DA, Holmes TH, Quan SF, Walsh JK, Gottlieb DJ, Simon RD, Guilleminault C, White DP, Goodwin JL, Schweitzer PK, Leary EB, Hyde PR, Hirshkowitz M, Green S, McEvoy LK, Chan C, Gevins A, Kay GG, Bloch DA, Crabtree T, Dement WC.

2012-12-01287 citations

10.5665/sleep.2226

Interventions

DEVICEActive CPAP

Nightly nasal continuous positive airway pressure (CPAP)

DEVICESham CPAP

Sham CPAP machine will be used for participants in the placebo group.

Study design

Allocation

RANDOMIZED

Intervention model

PARALLEL

Primary purpose

TREATMENT

Masking

QUADRUPLE (Subject, Caregiver, Investigator, Outcomes Assessor)

Eligibility

Sex/Gender

ALL

Age

18 Years to No maximum

Healthy volunteers

No

Inclusion criteria

* Male or female adults age 18 years or older with a diagnosis of OSAS using clinical criteria defined by the study protocol * Study participation may require seven or more laboratory visits over six months

Exclusion criteria

* Prior treatment for OSAS with continuous positive airway pressure or surgery * Potential sleep apnea complications that may affect the health or safety of the participant, including low blood oxygen, recent near-miss or prior automobile accident due to sleepiness, congestive heart failure, history of angina, coronary artery disease, myocardial infarction or stroke, cardiac rhythm disturbance, and chronic neurological disorders affecting neurocognitive abilities or daily function * The use of hypnotics, anxiolytics, sedating antidepressants, anticonvulsants, sedating antihistamines, stimulants or other medications likely to affect neurocognitive function and/or alertness * Respiratory disease requiring medications (unless on stable medications for 2 months) * Cancer, unless in remission for greater than one year and not taking exclusionary medications * Self-reported renal failure * Pregnancy anytime during a subject's participation * Psychiatric illness, as defined by a DSM-IV diagnosis, except for depression or mild anxiety * Narcolepsy, idiopathic hypersomnolence, DSM-IV chronic insomnia, restless legs syndrome, or rapid eye movement (REM) behavior disorder * Current use of diurnal or nocturnal supplemental oxygen * Significant vision, hearing, or coordination problems * Difficulty understanding or speaking English * Currently working night or rotating shifts * Consumption of more than 10 caffeinated beverages per day (approximately 1,000 mg per day) * Smokers whose habit interferes with the overnight polysomnogram or with the battery of testing during the day * Consumption of more than 2 alcoholic beverages per day * Any illicit drug usage or marijuana usage more than once a week * Any individual in the household currently on CPAP or on CPAP in the past * A score of 26 or less on the Mini Mental State Examination (MMSE)

Design outcomes

Primary

Measure	Time frame	Description
Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD	2 months and 6 months post intervention	There are three primary measures of neurocognitive function measured for APPLES, each representing a different domain: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD), Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL), and Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR). This is domain #1: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD) SWMT-OMD is a scaled score that indicates whether the participant scored lower or higher relative to baseline using standard deviation units. It is computed as the mean of three sub-scores, one based on working memory (WM) task performance (behavioral WM sub-score: speed, accuracy), and the other two on electroencephalogram (EEG) (cortical activation sub-score: neural workload, attentional effort during WM task; alertness sub-score: resting alertness).
Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention	There are three primary measures of neurocognitive function measured for APPLES, each representing a different domain: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD), Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL), and Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR). This is domain #2: Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL)
Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention	There are three primary measures of neurocognitive function measured for APPLES, each representing a different domain: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD), Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL), and Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR). This is domain #3: Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR)

Secondary

Measure	Time frame	Description
Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. One of the selected variables came from the domain of Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec).
Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Executive and Frontal-Lobe (E/F) Function: SWMT-BehMD, SWMT-ActMD, and SAT-D-NumRuCh. These data are for variable #1: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD) SWMT-BehMD is a scaled score that indicates whether the participant scored lower or higher relative to baseline using standard deviation units. It is computed as the difference from baseline relative to measures of working memory (WM) task performance accuracy (percent correct) and mean and standard deviation of reaction time (milliseconds). High-load WM tasks receive twice the weight of the low-load WM tasks.
Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Executive and Frontal-Lobe (E/F) Function: SWMT-BehMD, SWMT-ActMD, and SAT-D-NumRuCh. These data are for variable #2: SWMT- Mid-day Activation Index (SWMT-ActMD) SWMT-ActMD is a scaled score that indicates whether the participant scored lower or higher relative to baseline (BL) using standard deviation units. It is computed as the difference from BL relative to EEG power spectral variables (decibels) measured during the easier vs. more difficult working memory (WM) tasks. A positive activation sub-score indicates a larger cortical neuronal population was recruited to perform the more difficult WM task relative to BL, while a negative score indicates a smaller population was recruited.
Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD), SWMT- Mid-day Activation Index (SWMT-ActMD), and Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh). These data are for variable #3: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)
Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention	Objective sleepiness/alertness was measured using the Maintenance of Wakefulness Test (MWT); the outcome variable was MWT Mean Sleep Latency (MWT-MSL). The MWT was administered using four twenty-minute trials where the participant was asked to sit in a chair, in a quiet and dimly lit room, with instructions to stay awake. Trials were performed at 10 AM, Noon, 2 PM and 4 PM. The mean sleep latency was calculated using the 4 trials from a given visit, and required that at least 3 of the 4 trials were performed and validated.
Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention	Subjective sleepiness/alertness was measured using the Epworth Sleepiness Scale (ESS); the outcome variable was ESS Total Score (ESS-TS). The ESS is a validated questionnaire (8 questions) that ask the chances of dozing off in specific situations. Summing the scores produces a scaled total score between 0 and 24, with higher numbers indicating more subjective sleepiness. The ESS was administered the evening before the polysomnogram (PSG), or overnight sleep study. Data reported here include questionnaires collected at the DX, 2M, and 6M visits.
Mood	Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention	—
Quality of Life: Calgary Sleep Apnea Quality of Life Index- Total Score (SAQLI-TS)	diagnostic visit (baseline)	Quality of life was measured using the Calgary Sleep Apnea Quality of Life Index (SAQLI), which is an interview-administered instrument with high internal consistency and reliability. The SAQLI was designed to assess components identified as important to patients including daily functioning, social interactions, emotional functioning, symptoms experienced, and treatment-related symptoms. Items are scored on a seven-point scale, averaged (taking into account treatment-related symptoms), to yield a composite score between 1 and 7, where higher scores represent better quality of life.
Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT), Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT), and PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT). These data are for variable #1: Pathfinder Number- Reaction Time (PN-RT)
Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT), Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT), and PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT). These data are for variable #2: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)
Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2 months and 6 months post intervention	The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT), Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT), and PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT). These data are for variable #3: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)

Other

Measure	Time frame
Functional Magnetic Resonance Imaging (fMRI)	Measured at diagnostic visit (baseline) and 6 months post intervention

Countries

United States

Participant flow

Participants by arm

Arm	Count
Active CPAP Active Continuous Positive Airway Pressure (CPAP) is widely used for the treatment of Obstructive Sleep Apnea (OSA)	556
Sham CPAP Sham Continuous Positive Airway Pressure (CPAP) has been modified to be sub-therapeutic, yet closely simulates an Active CPAP device.	542
Total	1,098

Withdrawals & dropouts

Period	Reason	FG000	FG001
Overall Study	Death	2	2
Overall Study	Disqualified Post-Rand for Any Reason	9	11
Overall Study	Dropped Post-Rand for Any Reason	102	126
Overall Study	Excluded Pre-Randomization	2	5

Baseline characteristics

Characteristic	Sham CPAP	Active CPAP	Total
Age, Categorical <=18 years	0 Participants	1 Participants	1 Participants
Age, Categorical >=65 years	78 Participants	96 Participants	174 Participants
Age, Categorical Between 18 and 65 years	464 Participants	459 Participants	923 Participants
Age, Continuous	50.8 years STANDARD_DEVIATION 12.2	52.2 years STANDARD_DEVIATION 12.2	51.5 years STANDARD_DEVIATION 12.2
Region of Enrollment United States	542 participants	556 participants	1098 participants
Sex: Female, Male Female	186 Participants	193 Participants	379 Participants
Sex: Female, Male Male	356 Participants	363 Participants	719 Participants

Adverse events

Event type	EG000 affected / at risk	EG001 affected / at risk
deaths Total, all-cause mortality	— / —	— / —
other Total, other adverse events	390 / 556	373 / 542
serious Total, serious adverse events	32 / 556	31 / 542

Outcome results

Primary

Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL

There are three primary measures of neurocognitive function measured for APPLES, each representing a different domain: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD), Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL), and Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR). This is domain #2: Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL)

Time frame: Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle.

Arm	Measure	Group	Value (MEAN)
Active CPAP	Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	PFN-TOTL Dx	23.32 seconds
Active CPAP	Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	PFN-TOTL 2M	23.56 seconds
Active CPAP	Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	PFN-TOTL 6M	23.48 seconds
Sham CPAP	Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	PFN-TOTL Dx	23.08 seconds
Sham CPAP	Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	PFN-TOTL 2M	22.92 seconds
Sham CPAP	Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL	PFN-TOTL 6M	23.01 seconds

Comparison: Comparison of means (regression estimates) between arms for DX A/P Function- PFN-TOTL.~Data were reciprocal transformed for analysis and back-transformed for reporting.p-value: 0.4538Parametric survival analysis

Comparison: Comparison of means (regression estimates) between arms for 2M A/P Function- PFN-TOTL.~Data were reciprocal transformed for analysis and back-transformed for reporting.p-value: 0.086Parametric survival analysis

Comparison: Comparison of means (regression estimates) between arms for 6M A/P Function- PFN-TOTL.~Data were reciprocal transformed for analysis and back-transformed for reporting.p-value: 0.2103Parametric survival analysis

Primary

Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD

There are three primary measures of neurocognitive function measured for APPLES, each representing a different domain: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD), Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL), and Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR). This is domain #1: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD) SWMT-OMD is a scaled score that indicates whether the participant scored lower or higher relative to baseline using standard deviation units. It is computed as the mean of three sub-scores, one based on working memory (WM) task performance (behavioral WM sub-score: speed, accuracy), and the other two on electroencephalogram (EEG) (cortical activation sub-score: neural workload, attentional effort during WM task; alertness sub-score: resting alertness).

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle.

Arm	Measure	Group	Value (MEAN)
Active CPAP	Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD	SWMT-OMD 2M	0.035 score on a scale
Active CPAP	Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD	SWMT-OMD 6M	0.072 score on a scale
Sham CPAP	Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD	SWMT-OMD 2M	-0.074 score on a scale
Sham CPAP	Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD	SWMT-OMD 6M	0.018 score on a scale

Comparison: Comparison of means (regression estimates) between arms for 2M E/F Function- SWMT-OMD.p-value: 0.0074Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 6M E/F Function- SWMT-OMD.p-value: 0.2254Regression, Linear

Primary

Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR

There are three primary measures of neurocognitive function measured for APPLES, each representing a different domain: Executive and Frontal-lobe (E/F) Function- Sustained Working Memory Test Overall Mid-Day Index (SWMT-OMD), Attention and Psychomotor (A/P) Function- Pathfinder Number Test Total Time (PFN-TOTL), and Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR). This is domain #3: Learning and Memory (L/M) Function- Buschke Selective Reminding Test Sum Recall (BSRT-SR)

Time frame: Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle.

Arm	Measure	Group	Value (MEAN)
Active CPAP	Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	BSRT-SR Dx	49.72 number of words recalled
Active CPAP	Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	BSRT-SR 2M	52.32 number of words recalled
Active CPAP	Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	BSRT-SR 6M	54.09 number of words recalled
Sham CPAP	Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	BSRT-SR Dx	49.86 number of words recalled
Sham CPAP	Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	BSRT-SR 2M	51.95 number of words recalled
Sham CPAP	Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR	BSRT-SR 6M	54.28 number of words recalled

Comparison: Comparison of means (regression estimates) between arms for DX L/M Function- BSRT-SR.p-value: 0.7936Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 2M L/M Function- BSRT-SR.p-value: 0.5444Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 6M L/M Function- BSRT-SR.p-value: 0.7569Regression, Linear

Secondary

Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT), Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT), and PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT). These data are for variable #1: Pathfinder Number- Reaction Time (PN-RT)

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2M PN-RT Mild OSA	0.811 seconds
Active CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2M PN-RT Moderate OSA	0.831 seconds
Active CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2M PN-RT Severe OSA	0.818 seconds
Active CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	6M PN-RT Mild OSA	0.811 seconds
Active CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	6M PN-RT Moderate OSA	0.830 seconds
Active CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	6M PN-RT Severe OSA	0.817 seconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	6M PN-RT Moderate OSA	0.819 seconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2M PN-RT Mild OSA	0.801 seconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	6M PN-RT Mild OSA	0.795 seconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2M PN-RT Moderate OSA	0.825 seconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	6M PN-RT Severe OSA	0.806 seconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)	2M PN-RT Severe OSA	0.812 seconds

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- PN-RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline PN-RT and months since randomization were also included as covariates.p-value: 0.5606Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- PN-RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline PN-RT and months since randomization were also included as covariates.p-value: 0.6487Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- PN-RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline PN-RT and months since randomization were also included as covariates.p-value: 0.5667Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M L/M Function- PN-RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline PN-RT and months since randomization were also included as covariates.p-value: 0.3972Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M L/M Function- PN-RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline PN-RT and months since randomization were also included as covariates.p-value: 0.3973Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M L/M Function- PN-RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline PN-RT and months since randomization were also included as covariates.p-value: 0.3055Mixed Models Analysis

Secondary

Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT), Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT), and PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT). These data are for variable #2: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2M PVT-MedRT Mild OSA	245.31 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2M PVT-MedRT Moderate OSA	248.68 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2M PVT-MedRT Severe OSA	243.25 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	6M PVT-MedRT Mild OSA	245.23 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	6M PVT-MedRT Moderate OSA	248.60 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	6M PVT-MedRT Severe OSA	243.17 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	6M PVT-MedRT Moderate OSA	249.10 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2M PVT-MedRT Mild OSA	253.89 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	6M PVT-MedRT Mild OSA	254.05 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2M PVT-MedRT Moderate OSA	248.94 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	6M PVT-MedRT Severe OSA	247.70 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)	2M PVT-MedRT Severe OSA	247.55 milliseconds

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-MedRT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.3699Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-MedRT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.9673Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-MedRT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.3426Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-MedRT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.3901Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-MedRT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.9464Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-MedRT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.4372Mixed Models Analysis

Secondary

Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT), Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT), and PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT). These data are for variable #3: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2M PVT-Slo10%RT Mild OSA	403.00 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2M PVT-Slo10%RT Moderate OSA	412.44 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2M PVT-Slo10%RT Severe OSA	400.57 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	6M PVT-Slo10%RT Mild OSA	396.87 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	6M PVT-Slo10%RT Moderate OSA	406.17 milliseconds
Active CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	6M PVT-Slo10%RT Severe OSA	394.48 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	6M PVT-Slo10%RT Moderate OSA	406.78 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2M PVT-Slo10%RT Mild OSA	402.32 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	6M PVT-Slo10%RT Mild OSA	401.28 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2M PVT-Slo10%RT Moderate OSA	407.84 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	6M PVT-Slo10%RT Severe OSA	405.04 milliseconds
Sham CPAP	Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)	2M PVT-Slo10%RT Severe OSA	406.11 milliseconds

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-Slo10%RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.9656Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-Slo10%RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.6765Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-Slo10%RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.5288Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-Slo10%RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.7807Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-Slo10%RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.9603Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M A/P Function- PVT-Slo10%RT.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. Months since randomization were also included as covariate.p-value: 0.3075Mixed Models Analysis

Secondary

Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD), SWMT- Mid-day Activation Index (SWMT-ActMD), and Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh). These data are for variable #3: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2M SAT-D-NumRuCh Mild OSA	0.931 number of rule changes (dichotomized)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2M SAT-D-NumRuCh Moderate OSA	0.936 number of rule changes (dichotomized)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2M SAT-D-NumRuCh Severe OSA	0.952 number of rule changes (dichotomized)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	6M SAT-D-NumRuCh Mild	0.897 number of rule changes (dichotomized)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	6M SAT-D-NumRuCh Moderate OSA	0.903 number of rule changes (dichotomized)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	6M SAT-D-NumRuCh Severe OSA	0.927 number of rule changes (dichotomized)
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	6M SAT-D-NumRuCh Moderate OSA	0.935 number of rule changes (dichotomized)
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2M SAT-D-NumRuCh Mild OSA	0.929 number of rule changes (dichotomized)
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	6M SAT-D-NumRuCh Mild	0.907 number of rule changes (dichotomized)
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2M SAT-D-NumRuCh Moderate OSA	0.951 number of rule changes (dichotomized)
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	6M SAT-D-NumRuCh Severe OSA	0.924 number of rule changes (dichotomized)
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)	2M SAT-D-NumRuCh Severe OSA	0.942 number of rule changes (dichotomized)

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SAT-D-NumRuCh.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.9518Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SAT-D-NumRuCh.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.4108Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SAT-D-NumRuCh.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.4528Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SAT-D-NumRuCh.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.8391Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SAT-D-NumRuCh.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.2771Mixed Models Analysis

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SAT-D-NumRuCh.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.8961Mixed Models Analysis

Secondary

Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Executive and Frontal-Lobe (E/F) Function: SWMT-BehMD, SWMT-ActMD, and SAT-D-NumRuCh. These data are for variable #1: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD) SWMT-BehMD is a scaled score that indicates whether the participant scored lower or higher relative to baseline using standard deviation units. It is computed as the difference from baseline relative to measures of working memory (WM) task performance accuracy (percent correct) and mean and standard deviation of reaction time (milliseconds). High-load WM tasks receive twice the weight of the low-load WM tasks.

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2M SWMT-BehMD Mild OSA	0.180 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2M SWMT-BehMD Moderate OSA	0.137 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2M SWMT-BehMD Severe OSA	0.205 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	6M SWMT-BehMD Mild	0.143 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	6M SWMT-BehMD Moderate OSA	0.194 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	6M SWMT-BehMD Severe OSA	0.321 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	6M SWMT-BehMD Moderate OSA	0.314 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2M SWMT-BehMD Mild OSA	0.104 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	6M SWMT-BehMD Mild	0.116 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2M SWMT-BehMD Moderate OSA	0.126 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	6M SWMT-BehMD Severe OSA	0.173 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)	2M SWMT-BehMD Severe OSA	-0.011 score on a scale

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SWMT-BehMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.5419Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SWMT-BehMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.89Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SWMT-BehMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.0031Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M E/F Function- SWMT-BehMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.8703Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M E/F Function- SWMT-BehMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.1838Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M E/F Function- SWMT-BehMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.0739Regression, Linear

Secondary

Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. Three of the selected variables came from the domain of Executive and Frontal-Lobe (E/F) Function: SWMT-BehMD, SWMT-ActMD, and SAT-D-NumRuCh. These data are for variable #2: SWMT- Mid-day Activation Index (SWMT-ActMD) SWMT-ActMD is a scaled score that indicates whether the participant scored lower or higher relative to baseline (BL) using standard deviation units. It is computed as the difference from BL relative to EEG power spectral variables (decibels) measured during the easier vs. more difficult working memory (WM) tasks. A positive activation sub-score indicates a larger cortical neuronal population was recruited to perform the more difficult WM task relative to BL, while a negative score indicates a smaller population was recruited.

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2M SWMT-ActMD Mild	-0.050 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2M SWMT-ActMD Moderate OSA	0.262 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2M SWMT-ActMD Severe OSA	-0.003 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	6M SWMT-ActMD Mild	0.157 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	6M SWMT-ActMD Moderate OSA	0.016 score on a scale
Active CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	6M SWMT-ActMD Severe OSA	0.058 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	6M SWMT-ActMD Moderate OSA	0.014 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2M SWMT-ActMD Mild	0.317 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	6M SWMT-ActMD Mild	0.118 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2M SWMT-ActMD Moderate OSA	0.170 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	6M SWMT-ActMD Severe OSA	0.123 score on a scale
Sham CPAP	Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)	2M SWMT-ActMD Severe OSA	0.033 score on a scale

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SWMT-ActMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.045Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SWMT-ActMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.4512Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M E/F Function- SWMT-ActMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.6672Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M E/F Function- SWMT-ActMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.8197Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M E/F Function- SWMT-ActMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.989Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 6M E/F Function- SWMT-ActMD.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ.p-value: 0.5029Regression, Linear

Secondary

Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)

The APPLES a priori Secondary Neurocognitive Analysis Plan specified a dimension reduction method to reduce twelve secondary neurocognitive variables from three neurocognitive domains to seven variables from three neurocognitive domains. One of the selected variables came from the domain of Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec).

Time frame: 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)
Active CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2M BSRTDR-TotRec Mild OSA	8.54 number of words recalled
Active CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2M BSRTDR-TotRec Moderate OSA	8.49 number of words recalled
Active CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2M BSRTDR-TotRec Severe OSA	8.48 number of words recalled
Active CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	6M BSRTDR-TotRec Mild OSA	9.01 number of words recalled
Active CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	6M BSRTDR-TotRec Moderate OSA	8.56 number of words recalled
Active CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	6M BSRTDR-TotRec Severe OSA	8.87 number of words recalled
Sham CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	6M BSRTDR-TotRec Moderate OSA	8.91 number of words recalled
Sham CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2M BSRTDR-TotRec Mild OSA	8.20 number of words recalled
Sham CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	6M BSRTDR-TotRec Mild OSA	9.44 number of words recalled
Sham CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2M BSRTDR-TotRec Moderate OSA	8.22 number of words recalled
Sham CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	6M BSRTDR-TotRec Severe OSA	8.75 number of words recalled
Sham CPAP	Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)	2M BSRTDR-TotRec Severe OSA	8.21 number of words recalled

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- BSRTDR-TotRec.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline BSRTDR-TotRec was also included as a covariate.p-value: 0.4262Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- BSRTDR-TotRec.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline BSRTDR-TotRec was also included as a covariate.p-value: 0.3161Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- BSRTDR-TotRec.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline BSRTDR-TotRec was also included as a covariate.p-value: 0.1835Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- BSRTDR-TotRec.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline BSRTDR-TotRec was also included as a covariate.p-value: 0.2462Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- BSRTDR-TotRec.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline BSRTDR-TotRec was also included as a covariate.p-value: 0.2069Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for covariate adjusted 2M L/M Function- BSRTDR-TotRec.~Covariates were designated in the a priori secondary analysis plan as being the most likely to explain variation in the outcomes. Covariates included: study arm, OSA severity, sex, race, % of total sleep time oxygen saturation \< 85% (PSG), age \< 60 years, WASI Verbal IQ, and WASI Performance IQ. A pre-randomization baseline BSRTDR-TotRec was also included as a covariate.p-value: 0.5235Regression, Linear

Secondary

Mood

Time frame: Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention

Population: Mood was dropped as a secondary outcome measure for analysis by our Core Team.

Secondary

Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)

Objective sleepiness/alertness was measured using the Maintenance of Wakefulness Test (MWT); the outcome variable was MWT Mean Sleep Latency (MWT-MSL). The MWT was administered using four twenty-minute trials where the participant was asked to sit in a chair, in a quiet and dimly lit room, with instructions to stay awake. Trials were performed at 10 AM, Noon, 2 PM and 4 PM. The mean sleep latency was calculated using the 4 trials from a given visit, and required that at least 3 of the 4 trials were performed and validated.

Time frame: Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)	Dispersion
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL Moderate OSA	17.91 minutes	Standard Deviation 3.39
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL Mild OSA	17.51 minutes	Standard Deviation 3.71
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL Severe OSA	18.10 minutes	Standard Deviation 3.35
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL Severe OSA	16.68 minutes	Standard Deviation 4.05
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL	18.11 minutes	Standard Deviation 3.27
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL	17.13 minutes	Standard Deviation 3.86
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL Mild OSA	17.77 minutes	Standard Deviation 4
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL	17.96 minutes	Standard Deviation 3.4
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL Moderate OSA	17.90 minutes	Standard Deviation 3.41
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL Moderate OSA	17.74 minutes	Standard Deviation 3.5
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL Severe OSA	18.30 minutes	Standard Deviation 2.98
Active CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL Mild OSA	17.52 minutes	Standard Deviation 3.6
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL Severe OSA	16.78 minutes	Standard Deviation 4.1
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL	16.95 minutes	Standard Deviation 4.13
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL Mild OSA	17.62 minutes	Standard Deviation 3.38
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL Moderate OSA	17.76 minutes	Standard Deviation 3.68
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	DX MWT-MSL Severe OSA	16.35 minutes	Standard Deviation 4.43
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL	17.27 minutes	Standard Deviation 3.89
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL Moderate OSA	18.14 minutes	Standard Deviation 2.93
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL Severe OSA	16.63 minutes	Standard Deviation 4.34
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL	17.34 minutes	Standard Deviation 3.82
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL Mild OSA	17.89 minutes	Standard Deviation 3.27
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	6M MWT-MSL Moderate OSA	18.18 minutes	Standard Deviation 3.27
Sham CPAP	Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)	2M MWT-MSL Mild OSA	18.21 minutes	Standard Deviation 2.94

Comparison: Comparison of means (regression estimates) between arms for DX Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.654Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for DX Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.9778Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for DX Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.8314Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for DX Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.5018Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 2M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.0052Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 2M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.2476Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 2M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.752Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 2M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.0002Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 6M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.0022Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 6M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.763Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 6M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.517Chop-lump Wilcoxon

Comparison: Comparison of means (regression estimates) between arms for 6M Objective Sleepiness/Alertness- MWT-MSL.p-value: 0.0002Chop-lump Wilcoxon

Secondary

Quality of Life: Calgary Sleep Apnea Quality of Life Index- Total Score (SAQLI-TS)

Quality of life was measured using the Calgary Sleep Apnea Quality of Life Index (SAQLI), which is an interview-administered instrument with high internal consistency and reliability. The SAQLI was designed to assess components identified as important to patients including daily functioning, social interactions, emotional functioning, symptoms experienced, and treatment-related symptoms. Items are scored on a seven-point scale, averaged (taking into account treatment-related symptoms), to yield a composite score between 1 and 7, where higher scores represent better quality of life.

Time frame: diagnostic visit (baseline)

Population: Analyses performed for group of participants with SAQLI data. Baseline demographics were generally similar (mean age, sex ratio, proportion of white participants, average body mass index), and participants in both groups had similar SAQLI scores at baseline, which are presented below.

Arm	Measure	Value (MEAN)	Dispersion
Active CPAP	Quality of Life: Calgary Sleep Apnea Quality of Life Index- Total Score (SAQLI-TS)	4.7 Units on a scale	Standard Deviation 0.8
Sham CPAP	Quality of Life: Calgary Sleep Apnea Quality of Life Index- Total Score (SAQLI-TS)	4.7 Units on a scale	Standard Deviation 0.8

Secondary

Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)

Subjective sleepiness/alertness was measured using the Epworth Sleepiness Scale (ESS); the outcome variable was ESS Total Score (ESS-TS). The ESS is a validated questionnaire (8 questions) that ask the chances of dozing off in specific situations. Summing the scores produces a scaled total score between 0 and 24, with higher numbers indicating more subjective sleepiness. The ESS was administered the evening before the polysomnogram (PSG), or overnight sleep study. Data reported here include questionnaires collected at the DX, 2M, and 6M visits.

Time frame: Measured at diagnostic visit (baseline) and 2 months and 6 months post intervention

Population: Analyses conducted in accordance with the intention-to-treat principle. Analyses performed by OSA severity level categorized by apnea hypopnea index (AHI: # of respiratory events/hr of sleep) obtained via baseline polysomnography. Categories are: Mild OSA (AHI = 10-14.9; exclusion if AHI \<10), Moderate OSA (AHI = 15-29.9), Severe OSA (AHI \>=30).

Arm	Measure	Group	Value (MEAN)	Dispersion
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS	10.07 scores on a scale	Standard Deviation 4.26
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS Mild OSA	10.10 scores on a scale	Standard Deviation 4.55
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS Moderate OSA	9.57 scores on a scale	Standard Deviation 4.13
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS Severe OSA	10.35 scores on a scale	Standard Deviation 4.24
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS	7.86 scores on a scale	Standard Deviation 4.2
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS Mild OSA	8.59 scores on a scale	Standard Deviation 4.31
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS Moderate OSA	7.25 scores on a scale	Standard Deviation 3.89
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS Severe OSA	8.00 scores on a scale	Standard Deviation 4.31
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS	7.39 scores on a scale	Standard Deviation 4.21
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS Mild OSA	8.37 scores on a scale	Standard Deviation 4.64
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS Moderate OSA	7.07 scores on a scale	Standard Deviation 3.87
Active CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS Severe OSA	7.31 scores on a scale	Standard Deviation 4.25
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS Moderate OSA	8.43 scores on a scale	Standard Deviation 4.55
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS	10.09 scores on a scale	Standard Deviation 4.39
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS Moderate OSA	8.39 scores on a scale	Standard Deviation 4.29
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS Mild OSA	9.73 scores on a scale	Standard Deviation 4.43
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS Mild OSA	7.64 scores on a scale	Standard Deviation 3.98
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS Moderate OSA	9.75 scores on a scale	Standard Deviation 4.56
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS Severe OSA	9.34 scores on a scale	Standard Deviation 4.34
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	DX ESS-TS Severe OSA	10.37 scores on a scale	Standard Deviation 4.28
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS Severe OSA	8.56 scores on a scale	Standard Deviation 4.02
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS	8.89 scores on a scale	Standard Deviation 4.31
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	6M ESS-TS	8.41 scores on a scale	Standard Deviation 4.18
Sham CPAP	Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)	2M ESS-TS Mild OSA	7.90 scores on a scale	Standard Deviation 4.01

Comparison: Comparison of means (regression estimates) between arms for DX Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.9291Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for DX Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.6152Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for DX Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.704Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for DX Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.9537Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 2M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.0004Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 2M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.3886Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 2M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.0236Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 2M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.0005Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 6M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.0005Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 6M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.3796Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 6M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.0106Regression, Linear

Comparison: Comparison of means (regression estimates) between arms for 6M Subjective Sleepiness/Alertness- ESS-TS.p-value: 0.001Regression, Linear

Other Pre-specified

Functional Magnetic Resonance Imaging (fMRI)

Time frame: Measured at diagnostic visit (baseline) and 6 months post intervention

Population: fMRI was dropped as a secondary outcome measure for analysis by our Core Team.

Apnea Positive Pressure Long-Term Efficacy Study

Conditions

Keywords

Brief summary

Detailed description

Related papers

Interventions

Sponsors

Study design

Eligibility

Inclusion criteria

Exclusion criteria

Design outcomes

Primary

Secondary

Other

Countries

Participant flow

Participants by arm

Withdrawals & dropouts

Baseline characteristics

Adverse events

Outcome results

Effect of CPAP on Neurocognitive Function: A/P Function- PFN-TOTL

Effect of CPAP on Neurocognitive Function: E/F Function- SWMT-OMD

Effect of CPAP on Neurocognitive Function: L/M Function- BSRT-SR

Attention and Psychomotor (A/P) Function: Pathfinder Number- Reaction Time (PN-RT)

Attention and Psychomotor (A/P) Function: Psychomotor Vigilance Task- Median Reaction Time (PVT-MedRT)

Attention and Psychomotor (A/P) Function: PVT- Mean Slowest 10% of Reaction Times (PVT-Slo10%RT)

Executive and Frontal-Lobe (E/F) Function: Shifting Attention Test Discovery Condition- Number of Rule Changes (SAT-D-NumRuCh)

Executive and Frontal-Lobe (E/F) Function: Sustained Working Memory Test- Mid-day Behavioral Index (SWMT-BehMD)

Executive and Frontal-Lobe (E/F) Function: SWMT- Mid-day Activation Index (SWMT-ActMD)

Learning and Memory (L/M) Function: Buschke Selective Reminding Test Delayed Recall- Total Recall (BSRTDR-TotRec)

Mood

Objective Sleepiness/Alertness: Maintenance of Wakefulness Test- Mean Sleep Latency (MWT-MSL)

Quality of Life: Calgary Sleep Apnea Quality of Life Index- Total Score (SAQLI-TS)

Subjective Sleepiness/Alertness: Epworth Sleepiness Scale- Total Score (ESS-TS)

Functional Magnetic Resonance Imaging (fMRI)