Intervention Review

You have free access to this content

Educational, supportive and behavioural interventions to improve usage of continuous positive airway pressure machines in adults with obstructive sleep apnoea

  1. Dariusz R Wozniak1,*,
  2. Toby J Lasserson2,
  3. Ian Smith1

Editorial Group: Cochrane Airways Group

Published Online: 8 JAN 2014

Assessed as up-to-date: 17 JAN 2013

DOI: 10.1002/14651858.CD007736.pub2


How to Cite

Wozniak DR, Lasserson TJ, Smith I. Educational, supportive and behavioural interventions to improve usage of continuous positive airway pressure machines in adults with obstructive sleep apnoea. Cochrane Database of Systematic Reviews 2014, Issue 1. Art. No.: CD007736. DOI: 10.1002/14651858.CD007736.pub2.

Author Information

  1. 1

    Papworth Hospital, Respiratory Support and Sleep Centre, Cambridge, UK

  2. 2

    The Cochrane Collaboration, Cochrane Editorial Unit, London, UK

*Dariusz R Wozniak, Respiratory Support and Sleep Centre, Papworth Hospital, Papworth Everard, Cambridge, CB23 3RE, UK. darekwozniak@yahoo.co.uk.

Publication History

  1. Publication Status: New search for studies and content updated (conclusions changed)
  2. Published Online: 8 JAN 2014

SEARCH

 
Characteristics of included studies [ordered by study ID]
Aloia 2001

MethodsRandomised parallel-group trial. All randomly assigned participants accounted for


ParticipantsN = 12

Mean age: 65.5, AHI: 43.5, Desaturation: 77.05 ± 9.47

Inclusion criteria: > 55 years of age, RDI (AHI): > 10, Mini Mental Status Examination: > 25

Exclusion criteria: other ICSD, other treatment for apnoea, claustrophobia

Participants had received prior treatment with CPAP


InterventionsIntervention

Two sessions. Session 1: review of participants' sleep data; symptoms; review of performance of cognitive tests; review of importance of treatment; review of PSG and CPAP; discussion of advantages and disadvantages of treatment; development of goals for therapy. Session 2: examination of compliance data for week one; discussion of noticeable changes with treatment; discussion of changes not apparent (hypertension/cardiac problems); troubleshooting discomfort; discussion of realistic aims of treatment; review of treatment goals

Control

Two sessions: general discussion of sleep architecture and opinions on sleep clinic

Study duration: 12 weeks


Outcomes
  • Machine usage
  • Number of participants who were 'compliant' (> six hours per night of usage)
  • Vigilance testing


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomisation by 'urns', stratification by age, RDI, nadir O2 pretreatment, vigilance

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot done for treatment group assignment

'None of the subjects were told that their CPAP machines were measuring their compliance via internal microprocessors'

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInformation not available

Aloia 2012a

MethodsRandomised parallel-group study


ParticipantsN = 154

Intervention group: Age: 47.0, Male sex: 48%, AHI: 46.1, ESS: 12.6, BMI: 35

Control group: Age: 52, Male sex: 57%, AHI: 48.2, ESS: 11.9, BMI: 35.8

Inclusion criteria: new diagnosis of moderate to severe OSA by full in-lab polysomnography, naive to CPAP

Exclusion criteria: diagnosis by split night polysomnography, severe neurological or unstable psychiatric illness, congestive heart failure, end-stage renal disease


InterventionsIntervention

Two 45-minute face-to-face education sessions delivered by a trained nurse one and two weeks after initiation of PAP treatment. One additional booster phone call at week three Education comprised pathophysiology, medical and behavioural consequences of OSA and benefits of treatment

Control

Standard care consisting of physician discussing the benefits of treatment before and after diagnosis. Regular follow-up visits with physicians, usually eight to 10 weeks after PAP initiation

Study duration: 52 weeks


Outcomes
  • Machine usage (hours/night)
  • Withdrawals
  • Decisional balance
  • Self-efficacy


NotesThe study comprised three treatment arms. We consider the effects of the two treatment arms and the one control arms as separate studies. Interventions were initiated one week after initiation of CPAP

Unpublished study. Currently under review for publication


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskParticipants were urn randomly assigned in a 1:1 ratio

Allocation concealment (selection bias)Unclear riskNo sufficient information provided

Blinding (performance bias and detection bias)
All outcomes
Unclear riskParticipants, physicians and other healthcare providers were blinded to whether participants were enrolled in the study. Research staff who downloaded adherence data were blinded to group membership. Participants were informed that machine would be accessed periodically to determine 'how the device was working at night'. Given the nature of the intervention, it is unlikely that true blinding of participants was achieved. Furthermore, the same nurse delivered two different interventions in the study arms

Incomplete outcome data (attrition bias)
All outcomes
High risk27 of 80 participants in intervention group and 25 of 74 participants in control group dropped out from the study. Non-completers were not included in outcome analysis

Aloia 2012b

MethodsRandomised parallel-group study


ParticipantsN = 147

Intervention group: Age: 52, Male sex: 45%, AHI: 45.7, ESS: 11.6, BMI: 35

Control group: Age: 52, Male sex: 57%, AHI: 48.2, ESS: 11.9, BMI: 35.8

Inclusion criteria: new diagnosis of moderate to severe OSA by full in-lab polysomnography, naive to CPAP

Exclusion criteria: diagnosis by split night polysomnography, severe neurological or unstable psychiatric illness, congestive heart failure, end-stage renal disease


InterventionsIntervention

Two 45-minute face-to-face Motivational Enhancement Therapy (MET) sessions delivered by a trained nurse one and two weeks after initiation of PAP treatment. One additional booster phone call at week three. MET consisted of individually tailored counselling focused on addressing ambivalence regarding consistent use of PAP, participant-specific information on OSA, symptom change, treatment expectations, goal development and refinement and enhancing participant's motivation

Control

Standard care involved the physician discussing the benefits of treatment before and after diagnosis. Regular follow-up visits with physicians, usually eight to 10 weeks after PAP initiation

Study duration: 52 weeks


Outcomes
  • Machine usage (hours/night)
  • Withdrawals
  • Decisional balance
  • Self-efficacy


NotesThe study comprised three treatment arms. We consider the effects of the two treatment arms and the one control arm as separate studies. Interventions were initiated one week after initiation of CPAP

Unpublished study. Currently under review for publication


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskParticipants were urn randomly assigned in a 1:1 ratio

Allocation concealment (selection bias)Unclear riskNo sufficient information provided

Blinding (performance bias and detection bias)
All outcomes
Unclear riskParticipants, physicians and other healthcare providers were blinded to whether participants were enrolled in the study. Research staff who downloaded adherence data were blinded to group membership. Participants were informed that machine would be accessed periodically to determine 'how the device was working at night'. Given the nature of the intervention, it is unlikely that blinding of participants was achieved. Furthermore, the same nurse delivered two different interventions in the study arms

Incomplete outcome data (attrition bias)
All outcomes
High risk26 of 73 participants in intervention group and 25 of 74 participants in control group dropped out from the study. Non-completers were not included in outcome analysis

Basoglu 2011

MethodsRandomised, parallel-group study


ParticipantsN = 133

Intervention group: Age: 53.7, Male sex: 82%, AHI 61, ESS: 10.3, BMI: 33.2

Control group: Age: 54, Male sex: 70%, AHI: 57.4, ESS: 12.4, BMI: 33

Inclusion criteria: newly diagnosed, moderate to severe OSA, CPAP naive

Exclusion criteria: use of sedatives, drug abuse, cardiac co-morbidities, COPD, other sleep disorders


InterventionsIntervention

10-Minute videotape on OSA, its consequences and CPAP therapy. In addition, routine information on diagnosis and treatment of OSA given by physician

Control

Standard information on OSA and CPAP therapy given by the same physician

Study duration: 24 weeks


Outcomes
  • N of adherent participants (CPAP use for at least four hours/night for at least 70% of nights)
  • CPAP usage per night
  • ESS
  • Factors predicting CPAP adherence


NotesUnpublished information on study design and outcomes obtained from study authors


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomisation by a set of numbers prepared and randomly assigned by a clinician not involved in the study

Allocation concealment (selection bias)Low riskRandomisation by a third party

Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe primary investigator and the statistician were blinded to the study group assignment. Participants were aware of machine usage monitoring. Given the nature of the intervention, it is unlikely that participant blinding was achieved

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants completed the study, and no data were missing

Chervin 1997

MethodsRandomised parallel-group trial


ParticipantsN = 40

Mean age: 51.7, Mean AHI: 49.4, ESS: 10.9 ± 5.1, Lowest 02, Sat: 75.6% ± 14.4, MSLT: 6 ± 3.9

Recruited from clinic


InterventionsIntervention I

Telephone call each week during trial (max trial time of two months)

Intervention II

Two printed documents

Control

No additional support

Study duration: eight weeks


Outcomes
  • Machine usage


NotesTwo of 33 used Bi-PAP. Both CPAP-naive users and those who had been on CPAP before trial were studied. Reading done at enrolment and at between 1 to 2 months after enrolment

Difference in AHI between active and control groups at baseline


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandom numbers table

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot done for treatment group assignment

Participants' readout of CPAP machine usage data during telephone call to clinic

Incomplete outcome data (attrition bias)
All outcomes
High riskNon-completers excluded from analysis

DeMolles 2004

MethodsRandomised parallel-group study. Methods of randomisation not reported


ParticipantsN = 30

Mean age: 46, BMI: 38, AHI: 40, Functional Outcomes of Sleep Questionnaire: TLC: 15.3, Control: 13.8

Inclusion criteria: participants starting nasal CPAP therapy; > 18 years; English-speaking; > 15 episodes of apnoea or hypopnoea/h

Exclusion criteria: not described


InterventionsTelephone-linked communications technology (TLC) versus usual care. TLC consisted of a computerised digitised human speech programme. TLC asks questions designed to elicit information from participant regarding adherence, education and reinforcement

Study duration: eight weeks


Outcomes
  • Machine usage
  • Sleep symptoms
  • Functional outcomes of sleep questionnaire
  • Number of calls per participant


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; other information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskParticipants aware of treatment group assignment

Intervention involved communication regarding participant's CPAP machine usage

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll completed

Epstein 2000

MethodsRandomised, parallel-group study


ParticipantsN = 50

No baseline characteristics were reported. Participants recruited after diagnosis of OSA confirmed with polysomnography and before initiation with CPAP treatment. No information on withdrawals were reported

Inclusion criteria: not specified


InterventionsEducation course aim at desensitisation versus standard physician follow-up

Study duration: 24 weeks


Outcomes
  • Machine usage


NotesUnpublished abstract


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNo blinding undertaken

Not enough information available to ascertain awareness of CPAP machine usage

Incomplete outcome data (attrition bias)
All outcomes
High riskNon-completers excluded

Fox 2012

MethodsRandomised parallel-group study


ParticipantsN = 75

Mean age: 53.5, Mean AHI: 41.6, ESS: Control group: 9.7, Intervention group: 9.9

Inclusion criteria: adult (≥ 19 years), moderate to severe OSA (AHI ≥ 15)

Exclusion criteria: active cardiopulmonary or psychiatric disease, previously treated for OSA, no access to telephone line in bedroom, not able to return for follow-up


InterventionsIntervention

Physiological data (PAP adherence, applied PAP, mask leak, residual respiratory events) were downloaded using modem attached to the PAP device and sent across the telephone line each morning. Downloaded information was reviewed every weekday except holidays by the research coordinator, who contacted the participant if poor compliance or other problems with treatment (e.g. mask leak) were detected. Participants were advised over the phone or visited the PAP coordinator. Standard care identical to control group

Control

20-Minute orientation to PAP session and mask fitting. Participants contacted after two days to check adherence and to troubleshoot problems, followed up at four to six weeks and at three months; each time, physiological data downloaded from machines and any problems with treatment addressed. In addition, data downloaded at eight weeks

Study duration: 12 weeks


Outcomes
  • Machine usage (minutes per day)


  • Adherence on nights PAP used
  • % days PAP used
  • Decrease in ESS
  • AHI on treatment
  • Length of time spent with participants
  • Overall sleep quality and side effects measured by visual analogue scales


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk'...sequential numbered envelopes'

Allocation concealment (selection bias)Unclear riskenvelopes were prepared by one of the study investigators

Blinding (performance bias and detection bias)
All outcomes
High riskNo blinding undertaken

Intervention involved communication regarding participant's CPAP machine usage

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk'intention to treat approach', high discontinuation rate (control group: 10/36, telemedicine group: 11/39)

Hoy 1999

MethodsRandomised, parallel study. Method of randomisation not reported. ITT


ParticipantsN = 80

78:2 (M:F), Mean age: 51, Mean AHI: 58, ESS: 13

Inclusion criteria: AHI ≥ 15, plus daytime sleepiness or two other major symptoms of the syndrome; resident within 50 miles of Edinburgh

Exclusion criteria: prior use of CPAP; coexisting COPD, asthma or neurological problems


InterventionsIntervention

Full explanation of need for and benefits of CPAP by sleep physician, 20-minute video education programme, given mask to try for 20 minutes, titration of CPAP pressure overnight with following day discharge, nurses telephoned on days two and 21, reviewed in hospital at one, three and six months. Initial education at home with partner, two extra nights in hospital, sleep nurses' home visits to participant and partner at seven, 14 and 28 days and four months after starting CPAP

Control

Full explanation of need for and benefits of CPAP by sleep physician, 20-minute video education programme, given mask to try for 20 minutes, titration of CPAP pressure overnight with following day discharge, nurses telephoned on days two and 21, reviewed in hospital at one, three and six months

Duration: 24 weeks


Outcomes
  • Machine usage (hours/night) at six/12
  • Cognitive function
  • Simple unprepared reaction time
  • Quality of life
  • Symptom score (in-house questionnaire)
  • Mood
  • Sleep factors
  • Epworth Sleepiness Scale score
  • Maintenance of Wakefulness Test


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskEach participant was randomly assigned with predetermined
balanced blocks generated by tossing a coin

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
Unclear riskSingle-blind: 'Patients were blinded to the group to which they were allocated'

Not enough information available to ascertain awareness of CPAP machine usage

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk'Data were analysed on an intention-to-treat basis'

Hui 2000

MethodsRandomised, parallel-group study


ParticipantsN = 108

Mean age: 45, Mean AHI: 48, All participants newly diagnosed with OSA

Inclusion criteria: diagnosis of OSA (AHI > 10 and subjective daytime sleepiness)


InterventionsIntervention

10-Minute CPAP education programme by respiratory nurse, brochure on OSA and CPAP treatment in Chinese, short trial CPAP therapy with comfortable mask for 30 minutes, CPAP titration on second night of study by AutoSet, nursing support following day, follow-up by nursing staff and physician at 1 and 3 months. Locally produced 15-minute videotape, additional nurse led 15-minute educational session, review by physicians at weeks one and two, respiratory nurse telephone call on days one and two, weeks one, two, four, eight and 12

Control

10-Minute CPAP education programme by respiratory nurse, brochure on OSA and CPAP treatment in Chinese, short trial CPAP therapy with comfortable mask for 30 minutes, CPAP titration on second night of study by AutoSet, nursing support following day, follow-up by nursing staff and physician at 1 and 3 months.

Study duration: 12 weeks


Outcomes
  • Mean pressure required
  • Machine usage (objective and participant reported)
  • At least four hours of CPAP use/night for at least 70% of nights/wk)
  • Quality of life
  • ESS
  • SAQLI
  • Cognitive function


Notes91 participants had to purchase or rent their machines. 17 participants (10 in AS group and seven in BS group) qualified for state support


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; other information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot specified.

Participants provided subjective CPAP machine usage data

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskData were analysed on an intention-to-treat basis

Lewis 2006

MethodsProspective, single-blinded interventional study


ParticipantsN = 72

M/F: 62/10, Mean age: 51.4, Mean AHI control group: 42, All participants newly diagnosed with OSA

Inclusion criteria: diagnosis of OSA (based on home sleep study) and subjective daytime sleepiness


InterventionsIntervention

20-Minute educational video about SAHS. Telephone interview by research assistant between days two and five after CPAP issued to identify early problems and advise. Extra appointment to see sleep physician within seven to 14 days after being issued CPAP. Further appointment with sleep physician at one, six and 12 months

Control

Participants provided telephone number for support within office hours. Sleep physician reviewed participants at one, six and 12 months

Study duration: 52 weeks


Outcomes
  • Machine usage
  • Withdrawal
  • Side effects
  • Satisfactions


NotesOnly 20/36 participants in the intervention group watched the educational video tape

Eight of the 17 defaulters returned machines at different times of the year and had negligible hours of use


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomly assigned using block tables

Allocation concealment (selection bias)Low risk'The sequence of group assignment was indeed concealed from the investigators undergoing the screening and assessments, especially those recording/analysing machine hours'

Blinding (performance bias and detection bias)
All outcomes
Unclear riskSingle-blinded: participant unaware of what 'intensive' or standard support comprised

'The CPAP clock-timers were hidden with a plastic strip. Patients were not informed about the timers, and all covers were intact at each review; both patients and those recording
clock-timers were unaware of group allocation'

Incomplete outcome data (attrition bias)
All outcomes
High riskNon-completers not included in analysis of usage data

Meurice 2007a

MethodsRandomised parallel-group trial 


ParticipantsN = 57

Mean age: 58, Mean AHI: 58

Inclusion criteria: AHI > 30, no prior treatment for OSA


InterventionsIntervention

Reinforced education by the homecare team: home visit by technician at installation and further visits for explanation at one week, one month and two and three months of treatment for repetition of education and problem solving

Reinforced education by prescriber: written material on CPAP use; explanation of OSA and CPAP with side effects; emphasis on importance of compliance with CPAP and detailed demonstration

Control

Standard education by the homecare network. Homecare visit to supply the CPAP machine, fit the mask and explain the technique of using the apparatus. CPAP mechanism and method of using the machine and mask were explained. Participant was encouraged to ask questions and could phone at any time to resolve problems

Reinforced education by prescriber: written material on CPAP use; explanation of OSA and CPAP with side effects; emphasis on importance of compliance with CPAP and detailed demonstration

Study duration: follow-up to 52 weeks (intervention administered at outset of study). Data extracted at three months: 'During the remaining 9 months following the initial study design, there was no specific follow-up protocol and patients benefited from the standard homecare surveillance recommended in the ANTADIR network, with a review every 3 months'


Outcomes
  • Machine usage
  • ESS
  • Quality of life (SF-36)
  • Withdrawals


NotesThe study comprised four arms. We created four intervention/control comparisons and considered the effects of each as a separate study


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; other information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot done

Incomplete outcome data (attrition bias)
All outcomes
High risk‘One hundred thirty-three patients were initially scheduled. However, complete initial data were obtained in only 112 patients who were definitively included in the study’

Meurice 2007b

MethodsRandomised parallel-group trial 


ParticipantsN = 55

Mean age: 58, Mean AHI: 58

Inclusion criteria: AHI > 30, no prior treatment for OSA


InterventionsIntervention 

Reinforced education by the homecare team: home visit by technician at installation and further visits for explanation at one week, one month and two and three months of treatment for repetition of education and problem solving

Standard education by the prescriber

Control

Standard education by the homecare network. Homecare visit to supply the CPAP machine, fit the mask and explain the technique of using the apparatus. CPAP mechanism and method of using the machine and mask were explained. Participant was encouraged to ask questions and could phone at any time to resolve problems

Standard education by the prescriber

Study duration: follow-up to 52 weeks (intervention administered at outset of study). Data extracted at three months: 'During the remaining 9 months following the initial study design, there was no specific follow-up protocol and patients benefited from the standard homecare surveillance recommended in the ANTADIR network, with a review every 3 months'


Outcomes
  • Machine usage
  • ESS
  • Quality of life (SF-36)
  • Withdrawals


NotesThe study comprised four arms. We created four intervention/control comparisons and considered the effects of each as a separate study


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; other information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot done

Incomplete outcome data (attrition bias)
All outcomes
High risk‘One hundred thirty-three patients were initially scheduled. However, complete initial data were obtained in only 112 patients who were definitively included in the study’

Meurice 2007c

MethodsRandomised parallel-group trial 


ParticipantsN = 55

Mean age: 58, Mean AHI: 58

Inclusion criteria: AHI > 30, no prior treatment for OSA


InterventionsIntervention 

Reinforced education by the homecare team: home visit by technician at installation and further visits for explanation at one week, one month and two and three months of treatment for repetition of education and problem solving

Reinforced education by prescriber: written material on CPAP use; explanation of OSA and CPAP with side effects; emphasis on importance of compliance with CPAP and detailed demonstration

Control

Reinforced education by the homecare team: home visit by technician at installation and further visits for explanation at one week, one month and two and three months of treatment for repetition of education and problem solving

Standard education by the prescriber

Study duration: follow-up to 12 months (intervention administered at outset of study). Data extracted at three months: 'During the remaining 9 months following the initial study design, there was no specific follow-up protocol and patients benefited from the standard homecare surveillance recommended in the ANTADIR network, with a review every 3 months'


Outcomes
  • Machine usage
  • ESS
  • Quality of life (SF-36)
  • Withdrawals


NotesThe study comprised four arms. We created four intervention/control comparisons and considered the effects of each as a separate study


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; other information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot done

Incomplete outcome data (attrition bias)
All outcomes
High risk‘One hundred thirty-three patients were initially scheduled. However, complete initial data were obtained in only 112 patients who were definitively included in the study’

Meurice 2007d

MethodsRandomised parallel-group trial 


ParticipantsN = 57

Mean age: 58, Mean AHI: 58

Inclusion criteria: AHI > 30, no prior treatment for OSA


InterventionsIntervention

Standard education by the homecare network. Homecare visit to supply the CPAP machine, fit the mask and explain the technique of using the apparatus. CPAP mechanism and method of using the machine and mask were explained. Participant was encouraged to ask questions and could phone at any time to resolve problems

Reinforced education by prescriber: written material on CPAP use; explanation of OSA and CPAP with side effects; emphasis on importance of compliance with CPAP and detailed demonstration

Control

Standard education by the homecare network. Homecare visit to supply the CPAP machine, fit the mask and explain the technique of using the apparatus. CPAP mechanism and method of using the machine and mask were explained. Participant was encouraged to ask questions and could phone at any time to resolve problems

Standard education by the prescriber

Study duration: follow-up to 12 months (intervention administered at outset of study). Data extracted at three months: 'During the remaining 9 months following the initial study design, there was no specific follow-up protocol and patients benefited from the standard homecare surveillance recommended in the ANTADIR network, with a review every 3 months'


Outcomes
  • Machine usage
  • ESS
  • Quality of life (SF-36)
  • Withdrawals


NotesThe study comprised four arms. We created four intervention/control comparisons and considered the effects of each as a separate study


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; other information not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot done

Incomplete outcome data (attrition bias)
All outcomes
High risk‘One hundred thirty-three patients were initially scheduled. However, complete initial data were obtained in only 112 patients who were definitively included in the study’

Olsen 2012

MethodsRandomised parallel-group study


ParticipantsN = 100

Intervention group: Age: 55.1, Male: 58.5%, ESS: 10.8, RDI: 36.2

Control group: Age: 57.8, Male: 71.7%, ESS: 11.1, RDI: 32.4

Inclusion criteria: OSA confirmed by polysomnography, age ≥ 18, naive to CPAP

Exclusion criteria: need for bi-level ventilation, failed to complete CPAP titration, severe depression


InterventionsIntervention

Three sessions of CPAP-specific Motivational Interview Nurse Therapy (MINT) one month apart. Each session lasted approximately 30 minutes. In addition, all participants received standard one-on-one 45-minute education session conducted on the day of CPAP titration. Participants were followed up at two to four weeks by physician and at two months by a nurse. A questionnaire and a machine meter data on adherence were obtained at one, three and 12 months

Control

Standard one-on-one 45-minute education session conducted on the day of CPAP titration. Participants were followed up at two to four weeks by physician and at two months by a nurse

Study duration: 52 weeks


Outcomes
  • CPAP acceptance and adherence
  • FOSQ
  • Self-efficacy measure for sleep apnoea
  • ESS


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomly assigned using envelopes with group allocation; no blocking or stratification used

Allocation concealment (selection bias)Low risk'...opaque, unlabelled envelopes...shuffled by a research assistant...placed into an allocation box held in a secured clinic area.' Administrative officers not otherwise involved in the study withdrew an envelope and booked the participant's future appointments accordingly

Blinding (performance bias and detection bias)
All outcomes
High riskParticipants and intervention nurses were not blinded

Incomplete outcome data (attrition bias)
All outcomes
Low risk'The adherence analyses were by intent-to-treat...The multiple imputation method for substitution missing data was used...All univariate and bivariate statistical assumptions were met'

Parthasarathy 2012

MethodsRandomised parallel-group open-label


ParticipantsN = 39

All veterans

Intervention group: N = 22, Age: 53 ± 14, Men: 100%, BMI: 35, AHI: 36.7

Control group: N = 17, Age: 50 ± 14, BMI: 33, Men: 100%, AHI: 37.5

Inclusion criteria: new diagnosis of OSA, AHI > five, full night or split night polysomnography, Age: 21 to 85, no sedative medications used

Exclusion criteria: central or complex sleep apnoea, requirement of oxygen or Bi-PAP, unstable medical co-morbidities, irregular lifestyle pattern, excess alcohol use


InterventionsIntervention

Peer-driven system (PBS); trained peers with OSA and good CPAP adherence record were paired with newly diagnosed participants over three months. During two face-to-face sessions and eight telephone-based conversations, trained peers shared their experiences on coping strategies with CPAP, knowledge of perceived vulnerabilities of untreated OSA, motivated participants and promoted methods for improving efficacy of CPAP

Control group

Usual care: CPAP initiation and education class, participants were asked to send CPAP adherence 'smart cards' and were followed up at one and three months

Study duration: 12 weeks


Outcomes
  • Participant ratings of acceptability of PBS
  • CPAP adherence
  • Functional Outcomes of Sleep Questionnaire (FOSQ)
  • Vigilance, self-efficacy and participant activation
  • Nasal congestion score


NotesAdditional information on study methods and mean CPAP adherence obtained from the study author

These data were available from a pilot study. Larger trial is currently being undertaken


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomisation was accomplished by computer-generated assignment placed in sealed envelopes that were opened in a predetermined sequence of numbered and sealed envelopes

Allocation concealment (selection bias)Low riskSee above

Blinding (performance bias and detection bias)
All outcomes
Unclear riskObservers who evaluated outcomes and care providers were blinded to group allocation. Participants were not blinded to the intervention and were aware of CPAP adherence monitoring

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskTwo of 17 participants in the control group lost to follow-up versus zero in the intervention group

No information on how this attrition was dealt with

Richards 2007

MethodsRandomised, parallel-group trial  


ParticipantsN = 100

M/F: 86/15, Mean age: 56, RDI: 26, ESS: 10.5

Inclusion criteria: newly diagnosed with OSA

All participants referred for CPAP treatment. 109 screened and nine refused to participate  


InterventionsIntervention

Cognitive-behavioural therapy. Two one-hour group sessions; slide presentation on sleep, OSA and treatment. CPAP machine on display and relaxation techniques in the event of anxiety caused by wearing CPAP mask

Participants also benefited from video presentation with emphasis on perseverance with treatment and educational pamphlet made available   

Control

Treatment as usual: one standardised group education session; explanation of CPAP titration process; familiarisation with equipment used and procedure to be followed on the titration night. Explanation of side effects, all participants strongly encouraged to contact staff to obtain relevant help and support. Participants assessed and fitted with comfortable mask to be worn during titration

Study duration: CBT over course of one week before home treatment with CPAP. Assessment of CPAP made after four weeks


Outcomes
  • Machine usage
  • Withdrawal


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk'...a sequence generated with a blocking factor of 4'

Allocation concealment (selection bias)Low risk'An investigator not involved with recruitment or provision of treatment independently randomised participants using a sequence generated with a blocking factor of 4. Allocation concealment was achieved with sequentially numbered, opaque, sealed envelopes'

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot possible/attempted for participants; assessors and technicians not informed of treatment groups

'Staff members were blinded to which group participants had been allocated and the 3 usual CPAP therapists strictly adhered to a script'

Participants not informed that machine usage would be monitored

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskHigh attrition rate in control group (17/48 refused to take CPAP home)

'Analysis was by intention to treat, and we measured hours
of usage of CPAP at 28 days'

Roecklein 2010

MethodsRandomised parallel-group study


ParticipantsN = 30

Age: 46, Male sex: 30%, African Americans: 66.7%, AHI: 44, RDI: 56, ESS: 11.6, BMI: 42

Inclusion criteria: age 18 to 65, CPAP naive, reported intent to use CPAP; other sleep, psychiatric or health problems were not exclusion criteria


InterventionsIntervention

Written personalised feedback report, including detailed information on severity of the disease, self-reported daytime sleepiness, individually estimated risk of adverse health outcome and risk of motor vehicle accident, all compared with normative data. Feedback addressed barriers to using CPAP, ambivalence about treatment and difficulties of behaviour change and promoted self-efficacy and personal responsibility for choosing to use CPAP

Control

Written information from the American Academy of Sleep Medicine on OSA, Snoring and PAP therapy for OSA

Study duration: three months


Outcomes
  • Objective CPAP usage (total hours, average hours/night, number of sessions)
  • Self-reported CPAP usage


NotesParticipants were not provided machines but obtained them 'naturalistically', most commonly through insurance. Most participants were low-income African Americans


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskInformation not available

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
Unclear risk'Physicians were blind to study participation and participants were blind to their study condition.' Patients were aware that CPAP usage was monitored. Despite intended blinding, it is likely that participants would have been able to distinguish the two interventions

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskOnly two incidents of missing data in each group. However, in addition, participants who took longer to obtain machines (n = 5 in control group and n = 2 in intervention group did not obtain devices by two weeks) were included from the start and had CPAP usage recorded as 0 hours per session. It is possible that financial burden prevented some participants from acquiring CPAP machines in a timely fashion

Schiefelbein 2005

MethodsRandomised, parallel-group trial. The study presented was a secondary analysis on a subset of participants from the parent study (N = 122)


ParticipantsN = 51 (intervention: 32; placebo: 19)

26 M

Inclusion criteria: identified as non-adherent CPAP users from a parent study (N = 122); all used CPAP for < four hours per night


InterventionsIntervention

Internet-based application aimed at encouraging problem solving and preparedness in application of CPAP

Control

Internet-based application similar in format to intervention but directed activities in neutral health topics (vitamin intake)

Study duration: 16 weeks


Outcomes
  • Problem-solving confidence
  • Preparedness for home care
  • Evaluation of website


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated randomisation schedule with gender as a stratification variable

Allocation concealment (selection bias)Unclear riskInformation not provided

Blinding (performance bias and detection bias)
All outcomes
Unclear riskSingle-blind: Participants in the control group were given different content delivered in similar way to intervention

Outcome assessors not aware of assignment of treatment groups

No information on whether participants were aware that CPAP usage was being monitored

Incomplete outcome data (attrition bias)
All outcomes
Low riskData for all participants presented

Smith 2006

MethodsRandomised parallel-group trial


ParticipantsN = 19

Mean age: 63

Inclusion criteria: non-adherent with CPAP for three months, after initial education on CPAP use and supplemental audiotaped/videotaped reinforcement at two and four weeks


InterventionsIntervention

Two-way telehealth sessions mediated by video link-up through phone line. Research nurse emphasised nightly, bedtime routine for CPAP. After standardised protocols, nurse visually assessed participant, guided correct CPAP routine and determined whether the CPAP mask fits properly. Nurse described consequences of non-adherence and managing barriers to CPAP use. Benefits of nightly CPAP use for general health were emphasised

Control

Two-way telehealth sessions mediated by video link-up through phone line. Protocols drawn up to mimic content delivered to intervention group. Instead of CPAP-related information, participants given content on vitamin intake

Study duration: 12 weeks of scheduled telehealth sessions


Outcomes
  • N adhering to CPAP for longer than four hours/night
  • Participant satisfaction
  • Withdrawal


NotesNon-adherence in the study defined as less than four hours of CPAP use per night for fewer than nine of 14 consecutive nights’ use

TJL emailed for details of randomisation and outcome data 12/09/2008. Carol Smith responded 15/09/2008


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk'...randomised and done via computer software generated random assignment’

Allocation concealment (selection bias)Low risk'...allocation sequence and treatment group assignment concealed from investigators conducting the screening and ongoing assessments'

Blinding (performance bias and detection bias)
All outcomes
Unclear riskSingle-blind; nursing interventionist staff aware of different content delivered by video link-up

Machine usage was measured via smart card by blinded sleep lab personnel. Information on participants' awareness of CPAP machine usage was insufficient for us to determine how this might have affected the study

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants finished follow-up and contributed to data on adherence. Two satisfaction surveys were not submitted (one from each group)

Smith 2009

MethodsRandomised parallel-group trial


ParticipantsN = 97

Mean age: 63.4, Male sex: 55%, Mean AHI: Intervention group: 52.3, Control group: 47.3

Inclusion criteria: new diagnosis of OSA, age ≥ 18, AHI ≥ 20

Exclusion criteria: positive screening for drug or alcohol abuse, depression requiring hospitalisation


InterventionsAll participants received usual education on OSA and demonstration of CPAP equipment

Intervention

Audiotaped music along with softly spoken directions on relaxation techniques and habit-promoting instructions for using CPAP nightly. Participants received information packet, which included CPAP use reminder placard, handouts on benefits of CPAP adherence and health consequences of poor compliance, four-week diary for recording experience with CPAP

Control group

Audiotaped music along with spoken information about vitamins. Information packet was the same in format and length as the intervention group, but content was on vitamins

Study duration: 24 weeks


Outcomes
  • N adhering to CPAP (≥ four hours/d and ≥ nine of 14 nights)
  • Self-reported audiotape/diary use
  • Participant satisfaction
  • Withdrawal


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskParticipants randomly assigned using computerised random assignment programme

Allocation concealment (selection bias)Low riskParticipants recruited by 'nurses who had no knowledge of group assignment'

Blinding (performance bias and detection bias)
All outcomes
Unclear riskSingle-blind; '...placebo intervention was used to mimic the daily activities in the experimental treatment...' CPAP usage was measured via smart cards by blinded personnel.

Nurses administering experimental or placebo control interventions aware of different content of these interventions. Unclear whether participants were aware of machine usage monitoring. Personnel analysing data on compliance were blind to allocation of treatment

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskIntention-to-treat analysis but imbalanced N of dropouts: Intervention group: 11/55 (20%), Control group: 13/42 (31%) at six months. Unclear whether reasons for dropouts were balanced across groups

Sparrow 2010

MethodsRandomised parallel-group trial


ParticipantsN = 250

Median age: 55.0 years, 82% Men, Median BMI: 35.1

Intervention group: AHI: 36, ESS: 10

Conrol group: AHI: 40.5, ESS: 11

Inclusion criteria: age 18 to 80 years, AHI > 10

Exclusion criteria: not reported


InterventionsIntervention

Automated telephone-linked communication system adapted for CPAP (TLC-CPAP), designed around the concepts of motivational interviewing. Digitised human speech was used, and participants were communicating with it via touch tone keypad of their telephones. The TLC-CPAP content included assessment of the participant's experience with CPAP, self-reported machine use, feedback and counselling to enhance adherence and side effect management. Participants were required to make weekly calls to TLC-CPAP during the first month and monthly thereafter. Printed reports were sent to the participant's physician. Participants were encouraged to contact physician directly if any excessive symptoms or side effects of treatment encountered

Control

'Attention placebo control' group received general education on a variety of health topics via a telephone-linked communication (TLC) system. Participants were required to make calls on the same schedule as the intervention group

Study duration: 52 weeks


Outcomes
  • Machine usage (data downloaded from memory cards or by direct interrogation of CPAP devices)
  • Adherence to CPAP (> four hours/night)
  • Association between CPAP use and FOSQ, sleep symptoms, CES-D, visual reaction time
  • Self-efficacy index
  • Decisional balance index


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk'...randomisation stratified by sex, age and AHI using a randomised block design'

Allocation concealment (selection bias)Unclear riskNo information provided

Blinding (performance bias and detection bias)
All outcomes
Unclear riskBlinding of participants attempted by developing an 'attention placebo control group'. However, given the nature of the intervention, participants may have been aware of group assignment. Participants in the intervention group self-reported frequency and duration of CPAP usage. It is unclear whether participants in the control group were aware of CPAP usage monitoring

'...all data were collected by research assistants blind to group assignment'. Unclear whether the same applied to outcome assessors

Incomplete outcome data (attrition bias)
All outcomes
Low riskData were analysed by intention to treat. Multiple-imputation procedure was implemented to account for missing data in the outcome of CPAP use due to loss to follow-up. 20/124 in the intervention group and 15/126 in the control group lost to follow-up at 12 months

Stepnowsky 2007

MethodsRandomised parallel-group trial        


ParticipantsN = 45 (40 presented as baseline and completed)

Mean age: 59, Male: 98%, AHI: 39, ESS: 12.6

Inclusion criteria: AHI ≥ 15, no prior CPAP treatment, stable sleep environment

Exclusion criteria: allergies/sensitivity to mask or mask material, previous use of any other PAP device (e.g. bi-level PAP, auto-adjusting PAP), current use of prescribed supplemental oxygen or significant comorbid medical conditions that could interfere with daily use of CPAP


InterventionsIntervention

Review of compliance and efficacy data. Monitored information garnered as objective compliance data and subjective reports of usage. Follow-up tailored to how CPAP used by participants. Details on how many total hours the PAP unit was used each night at therapeutic pressure. Efficacy data consisted of the amount of mask leakage (L/s) and the AHI (total number of apnoeas and hypopnoeas per hour of sleep)

Control

Telephone call from staff one week after CPAP initiation and office follow-up visit at one month. Participants encouraged to call clinic any time with problems or concerns 

Study duration: eight weeks


Outcomes
  • Machine usage
  • ESS
  • Withdrawals
  • Depression
  • Quality of life (Functional Outcomes of Sleep Questionnaire)
  • AHI


NotesTJL emailed for randomisation 12/09/2008. Carl Stepnowsky responded 15/09/2008


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk'...we used the uniform random number generator in R to select all sequences of 4 randomly with equal probability so that the occurrence of 3 in a row being assigned to the same group would be extremely rare'

Allocation concealment (selection bias)Low risk'The randomisation scheme was concealed until the time at which the intervention was assigned. The randomisation scheme was generated by the project statistician and carried out by research staff immediately after the informed consent procedure and the completion of the baseline questionnaires'

Blinding (performance bias and detection bias)
All outcomes
Unclear riskParticipants in both groups received a monitoring unit

All participants likely to be aware that CPAP usage was measured

Incomplete outcome data (attrition bias)
All outcomes
High risk‘There were five CPAP “rejectors,” or patients who decided within the first day or two that they did not want to pursue CPAP as the primary treatment for their OSA. Our study did not have a “run-in” period, which could have helped identify these patients prior to the intervention’

Taylor 2006

MethodsRandomised parallel-group trial


ParticipantsN = 133

Mean age: 45, ESS: 14, CPAP pressure: 8.9

All participants were service or ex-service personnel in USA

Inclusion criteria: 18 to 64 years; RDI ≥ five; English speaking

Exclusion criteria: acute illness, hospitalised participants, significant nocturnal hypoxaemia, ESS < eight, disorder interfering with ability to use computer at home (i.e. blindness), major mental illness, physical disability that interfered with optimal use of computer, prior use of CPAP, undergoing concurrent therapy for OSA (MAD, surgery)


InterventionsIntervention

Education on first day; film on OSA and CPAP; instruction in use of CPAP; encouragement to attend sleep clinic. Computer programme (Health Buddy) delivering questions on a daily basis; responses were monitored by sleep medicine practitioners. If persistent high-risk answers given, this prompted a sleep practitioner to contact the participant within 24 hours

Control

Education on first day; film on OSA and CPAP; instruction in use of CPAP; encouragement to attend sleep clinic. Follow-up at one month with a scheduled clinic visit; telephone consultation

Study duration: four weeks


Outcomes
  • Machine usage
  • FOSQ
  • Participant satisfaction questionnaire
  • General perceived self-efficacy
  • Self-efficacy scale to use CPAP
  • Psychometric analysis


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated random numbers. Participants stratified according to age, gender and severity of symptoms

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot undertaken

'Each day, the patient was greeted with three questions regarding reported hours of nasal CPAP use...'

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInformation not available

Wang 2011a

MethodsRandomised parallel-group study


ParticipantsN = 76

Intervention group: Male: 89.5%, AHI: 35.7, ESS: 13.4, Control group: Male: 81.6%, AHI: 38.5, ESS: 13.9

Incusion criteria: new diagnosis of OSA, AHI ≥ 10, above elementary school education, 'conscious mind and able to communicate clearly'

Exclusion criteria: personal or family history of mental illness, drug or alcohol abuse, severe cognitive impairment, 'concurrent oncologic or psychiatric diseases'


InterventionsIntervention

Three nights of CPAP titration in the first week, four-hour group education session on OSA and CPAP in the first week, participants were given a brochure describing benefits of CPAP and CD containing a 20-minute video demonstrating how to optimise CPAP treatment, 24-hour consultation telephone line to the sleep nurses was available

Control

One night of CPAP titration in the hospital in the first week

Study duration: 12 weeks


Outcomes
  • N adhering to CPAP (≥ four hours/d and ≥ nine of 14 nights)
  • CPAP usage (hours/d, hours/d used, % days with compliance ≥ four hours)
  • ESS
  • Sleep quality (Pittsburgh)
  • Anxiety (STAI)
  • HADS-D score


NotesThe study comprised four treatment arms: three intervention groups and one control group. We consider the effects of the three intervention arms as separate studies


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk'The patients were randomly assigned...by block randomisation'

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot performed

Information on participants' awareness of CPAP machine usage monitoring not available

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk'The patients' CPAP adherence rates and dropout rates were analysed on an intention-to-treat basis'

Wang 2011b

MethodsRandomised parallel-group study


ParticipantsN = 76

Intervention group: Male: 76.3%, AHI: 41.2, ESS: 14.7

Control group: Male: 81.6%, AHI: 38.5, ESS: 13.9

Incusion criteria: new diagnosis of OSA, AHI ≥ 10, above elementary school education, 'conscious mind and able to communicate clearly'

Exclusion criteria: personal or family history of mental illness, drug or alcohol abuse, severe cognitive impairment, 'concurrent oncologic or psychiatric diseases'


InterventionsIntervention

One night of CPAP titration in the hospital, 12 × 40 minute group Progressive Muscle Relaxation (PMR) practice sessions over 12 weeks, one per week. Self-practice of PMR before each CPAP treatment. Brochure and CD with a guide for PMR practice at home

Control

One night of CPAP titration in the hospital in the first week

Study duration: 12 weeks


Outcomes
  • N adhering to CPAP (≥ four hours/d and ≥ nine of 14 nights)
  • CPAP usage (hours/d, hours/d used, % days with compliance ≥ four hours)
  • ESS
  • Sleep quality (Pittsburgh)
  • Anxiety (STAI)
  • HADS-D score


NotesThe study comprised four treatment arms: three intervention groups and one control group. We consider the effects of the three intervention arms as separate studies


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk'The patients were randomly assigned...by block randomisation'

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot performed

Information on participants' awareness of CPAP machine usage monitoring not available

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk'The patients' CPAP adherence rates and dropout rates were analysed on an intention-to-treat basis'

Wang 2011c

MethodsRandomised parallel-group study


ParticipantsN = 76

Intervention group: Male: 81.6, AHI: 43.1, ESS: 14.5

Control group: Male: 89.5%, AHI-35.7, ESS: 13.4

Incusion criteria: new diagnosis of OSA, AHI ≥ 10, above elementary school education, 'conscious mind and able to communicate clearly'

Exclusion criteria: personal or family history of mental illness, drug or alcohol abuse, severe cognitive impairment, 'concurrent oncologic or psychiatric diseases'


InterventionsIntervention

Three nights of CPAP titration in the hospital. Combination of interventions as in Education and PMR group (see above)

Control

The control for this intervention was the intervention arm of Wang 2011a

Study duration: 12 weeks


Outcomes
  • N adhering to CPAP (≥ four hours/d and ≥ nine of 14 nights)
  • CPAP usage (hours/d, hours/d used, % days with compliance ≥ four hours)
  • ESS
  • Sleep quality (Pittsburgh)
  • Anxiety (STAI)
  • HADS-D score


NotesThe study comprised four treatment arms: three intervention groups and one control group. We consider the effects of the three intervention arms as separate studies. In this study, for the combined intervention, Education + PMR, the control group was Education


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk'The patients were randomly assigned...by block randomisation'

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot performed

Information on participants' awareness of CPAP machine usage monitoring not available

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk'The patients' CPAP adherence rates and dropout rates were analysed on an intention-to-treat basis'

Wang 2011d

MethodsRandomised parallel-group study


ParticipantsN = 76

Intervention group: Male: 81.6, AHI: 43.1, ESS: 14.5

Control group: 76.3%, AHI: 41.2, ESS: 14.7

Inclusion criteria: new diagnosis of OSA, AHI ≥ 10, above elementary school education, 'conscious mind and able to communicate clearly'

Exclusion criteria: personal or family history of mental illness, drug or alcohol abuse, severe cognitive impairment, 'concurrent oncologic or psychiatric diseases'


InterventionsIntervention

Three nights of CPAP titration in the hospital. Combination of interventions as in Education + PMR group (see above)

Control

The control for this intervention was the intervention arm of Wang 2011b

Study duration: 12 weeks


Outcomes
  • N adhering to CPAP (≥ four hours/d and ≥ nine of 14 nights)
  • CPAP usage (hours/d, hours/d used, % days with compliance ≥ four hours)
  • ESS
  • Sleep quality (Pittsburgh)
  • Anxiety (STAI)
  • HADS-D score


NotesThe study comprised four treatment arms: three intervention groups and one control group. We consider the effects of the three intervention arms as separate studies. In this study, for the combined intervention, Education + PMR, the control group was PMR


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk'The patients were randomly assigned...by block randomisation'

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot performed

Information on participants' awareness of CPAP machine usage monitoring not available

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk'The patients' CPAP adherence rates and dropout rates were analysed on an intention-to-treat basis'

Wiese 2005

MethodsRandomised, unblinded parallel-group trial


ParticipantsN = 93

Mean age: 48, BMI: 38, Mean duration of symptoms: 5.4 years, % smokers (treatment: 26%; control: 49%), Mean AHI: 9, ESS: 13

Inclusion criteria: > 20 years; RDI > four; newly diagnosed OSAHS

Exclusion criteria: not reported


InterventionsIntervention

During initial visit, participants received explanations of OSA and CPAP by physician and respiratory therapist. Short instructional video (15-minute tape of interview between two 'blue collar' workers discussing what CPAP felt like and how it helped them)

Control

During initial visit, participants received explanations of OSA and CPAP by physician and respiratory therapist. Control group members were interviewed

Both groups received instruction at outset on using CPAP

Study duration: four weeks


Outcomes
  • Machine usage
  • Attendance at outpatient clinic/withdrawal
  • ESS
  • SAQLI


NotesNot able to assess machine usage, as 13 of the 57 participants who returned for their one-month clinic visit had
unusable machine-recorded data


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; blocks of 10 to ensure balanced group design

Allocation concealment (selection bias)Unclear riskInformation not available

Blinding (performance bias and detection bias)
All outcomes
High riskNot undertaken

Information on participants' awareness of CPAP machine usage was insufficient for us to determine how this might have affected the study

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInformation not available

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Damjanovic 2009Allocation to the standard or intensive support group was based on proximity of participant's address to the sleep centre, and no randomisation occurred. Participants were randomly assigned only to APAP or CPAP treatment. This unpublished information was obtained from the study authors

Fletcher 1991Cross-over study

Gupta 2011Implemented intervention could not be classified as supportive, educational, psychological or behavioural

Hirschowitz 2006Participants randomly assigned to CPAP treatment or no treatment

Kajaste 2004CBT programme given before randomisation to CPAP

Klein 2010Excluded for the same reasons as Damjanovic 2009

Marshall 2003Based on description of the study, it is unlikely that randomisation took place. No contact details of study authors available; therefore not possible to obtain further clarification on the trial design

Shaikh 2009Data for only 19 of 128 enrolled participants analysed and reported

Signes-Costa 2005Clinical review conducted by different practitioners. No systematic intervention that is educational or behavioural in nature

Trupp 2011No control group. Participants randomly assigned to positively or negatively framed education

Wenzel 2008Participants randomly assigned to immediate or delayed CPAP prescription

 
Characteristics of studies awaiting assessment [ordered by study ID]
Bartlett 2010

MethodsRandomised parallel-group study

ParticipantsN = 206

Intervention group: Age: 49, Male sex: 70%, AHI: 30.4, ESS: 12.0, BM: 35.8

Control group: Age: 47, Male sex: 72%, AHI: 39.9, ESS: 12, BMI: 34.4

Exclusion criteria: poor fluency in English and previous use of CPAP

InterventionsIntervention

30-Minute group education session on sleep, OSA, risks of untreated OSA and CPAP treatment. CBT session including slides on health/social benefits of using CPAP and video of real-life successful CPAP users. CBT session was delivered to a group of three or four participants

Control

Same 30-minute group education session on sleep, OSA and CPAP usage. Social reciprocity consisting of afternoon tea served while participants watched a video of a patient undergoing diagnostic and CPAP studies

Study duration: six months

Outcomes
  • CPAP usage (hours/night and % using CPAP ≥ four hours/d)
  • Predictors of CPAP adherence
  • N of withdrawals

NotesInconsistency of preliminary data on CPAP adherence reported in an abstract form (e.g. SE of mean CPAP adherence difference derived from CI differs 100 times from that calculated from SD). Further characteristics of intervention and control  groups required

Information from the study authors that final withdrawal figures are different from those initially reported

The study has been submitted for journal publication, and further information may be available

Fanfulla 2008

MethodsRandomised parallel-group

ParticipantsN = 20

Obese, OSA patients

Other baseline details not available

Inclusion criteria not available

InterventionsIntervention

Tele-assisted rehabilitation programme consisting of individualised exercise programme to lose weight and monitor CPAP compliance. Regular phone call interview every two weeks to assess OSA symptoms, problems with CPAP, adherence to exercise programme and weight control

Control

Standard care, otherwise not specified

Study duration: six months

Outcomes
  • Machine usage: no data available
  • ESS: groups not matched
  • Reduction of BMI
  • Level of daily exercise

NotesDRW emailed for further study information 30/06/2012

Peach 2003

MethodsRandomised parallel-group

ParticipantsN = not specified

Baseline details not available

Inclusion criteria: newly diagnosed OSA

InterventionsAll participants underwent titration at baseline. Fixed-pressure CPAP was used throughout the study    

Intervention: self-monitoring group

Control: non–self-monitoring group

Study duration: not enough information presented on duration of study

Outcomes
  • AHI
  • Machine usage

NotesTJL emailed for study information 12/09/2008

 
Comparison 1. Increased practical support and encouragement during follow-up + CPAP versus usual care + CPAP

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Machine usage (hours/night)13803Mean Difference (IV, Random, 95% CI)0.82 [0.36, 1.27]

 2 Machine usage, sensitivity analysis: excluding participants aware of machine usage monitoring6378Mean Difference (IV, Fixed, 95% CI)1.07 [0.61, 1.52]

 3 Machine usage, sensitivity analysis: adherence in control group < four hours/night8471Mean Difference (IV, Fixed, 95% CI)1.36 [0.96, 1.76]

 4 N deemed adherent (≥ four hours/night)4268Odds Ratio (M-H, Fixed, 95% CI)2.06 [1.22, 3.47]

 5 Epworth Sleepiness Scale scores8501Mean Difference (IV, Random, 95% CI)-0.60 [-1.81, 0.62]

 6 Quality of life: Functional Outcomes of Sleep Questionnaire270Mean Difference (IV, Fixed, 95% CI)0.98 [-0.84, 2.79]

 7 Quality of life: Sleep Apnoea Quality of Life Index (SAQLI)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 8 Mood3312Mean Difference (IV, Fixed, 95% CI)-0.94 [-1.55, -0.33]

    8.1 HAD Scale for Anxiety
180Mean Difference (IV, Fixed, 95% CI)-1.10 [-2.95, 0.75]

    8.2 HAD Scale for Depression
3232Mean Difference (IV, Fixed, 95% CI)-0.93 [-1.57, -0.28]

 9 Withdrawals12903Odds Ratio (M-H, Fixed, 95% CI)0.65 [0.44, 0.97]

 10 AHI on treatment2115Mean Difference (IV, Fixed, 95% CI)-0.07 [-1.62, 1.48]

 11 Maintenance of Wakefulness Test (MWT)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 
Comparison 2. Educational interventions + CPAP versus usual care + CPAP

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Machine usage (hours/night)7508Mean Difference (IV, Fixed, 95% CI)0.60 [0.27, 0.93]

 2 N deemed adherent (≥ four hours/night)3285Odds Ratio (M-H, Fixed, 95% CI)1.80 [1.09, 2.95]

 3 Epworth Sleepiness Scale scores5336Mean Difference (IV, Fixed, 95% CI)-1.17 [-2.07, -0.26]

 4 Quality of life: Sleep Apnoea Quality of Life Index (SAQLI)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 5 HAD Scale for Depression2152Mean Difference (IV, Fixed, 95% CI)-0.52 [-1.25, 0.22]

 6 Withdrawal8683Odds Ratio (M-H, Fixed, 95% CI)0.67 [0.45, 0.98]

 
Comparison 3. Behavioural therapy + CPAP versus control + CPAP

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Machine usage (hours/night)6584Mean Difference (Random, 95% CI)1.44 [0.43, 2.45]

 2 Sensitivity analysis: excluding participants aware of machine usage monitoring5Mean Difference (Fixed, 95% CI)1.54 [0.99, 2.09]

 3 N deemed adherent (≥ four hours/night)3358Odds Ratio (M-H, Fixed, 95% CI)2.23 [1.45, 3.45]

 4 Epworth Sleepiness Scale score1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 5 Quality of life: Functional Outcomes of Sleep Questionnaire1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 6 Withdrawal5609Odds Ratio (M-H, Fixed, 95% CI)0.85 [0.57, 1.25]

 
Summary of findings for the main comparison. Increased practical support and encouragement for adults with sleep apnoea

Increased practical support and encouragement for adults with sleep apnoea

Patient or population: adults with sleep apnoea
Intervention: increased practical support and encouragement and CPAP

Comparison: CPAP

Settings: community

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

ControlIncreased practical support and encouragement

Machine usage
Hours per night
Follow-up: median 12 weeks
Average CPAP machine usage ranged across control groups from
1.75 to 6.3 hours per night
Mean machine usage in the intervention groups was
0.82 hours higher
(0.36 to 1.27 higher)
803
(13 studies)
⊕⊕⊝⊝
low1,2

N deemed adherent (≥ four hours/night)
Follow-up: median 12 weeks
59 per 10075 per 100
(64 to 83)
OR 2.06
(1.22 to 3.47)
268
(4 studies)
⊕⊕⊝⊝
low1,3

Symptoms of sleepiness
Epworth Scale: zero to 24
Follow-up: median 12 weeks
Average Epworth symptom scores in control groups ranged from

4.5 to 13
Mean symptoms of sleepiness in the intervention groups was
0.6 lower
(1.81 lower to 0.62 higher)
501
(8 studies)
⊕⊝⊝⊝
very low1,4,5

Quality of life
Functional Outcomes of Sleep Questionnaire
Mean quality of life in the intervention groups was
0.98 higher
(0.84 lower to 2.79 higher)
70
(2 studies)
⊕⊕⊝⊝
low1,6

Quality of life
Sleep Apnoea Quality of Life Index (SAQLI)
See commentSee comment108
(1 study)
⊕⊕⊝⊝
low1,6
Single study estimate

Withdrawals
Follow-up: median 12 weeks
17 per 10011 per 100
(eight to 16)
OR 0.65
(0.44 to 0.97)
903
(12 studies)
⊕⊕⊕⊝
moderate1

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1Risk of bias (-1): In the absence of blinding across studies, the study effect estimates are at risk of performance bias.
2Inconsistency (-1): Substantial variation was seen in the direction and magnitude of effect across studies (I2 = 66%). Removal of studies when average machine use in control groups was high yielded a more consistent, larger effect in favour of intervention.
3Imprecision (-1): Low number of participants across studies was seen despite lower limit of the CI favouring intervention.
4Inconsistency (-1): Substantial variation was seen in the direction and magnitude of effect across studies.
5Imprecision (-1): Width of the confidence intervals does not exclude substantial improvement in deterioration of symptoms.
6Imprecision (-1): Low number of participants and very wide confidence intervals were compatible with benefit and harm.
 
Summary of findings 2. Educational interventions for adults with sleep apnoea

Educational interventions for adults with sleep apnoea

Patient or population: adults with sleep apnoea
Intervention: educational interventions and CPAP

Comparison: CPAP

Settings: community

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

ControlEducational interventions

Machine usage
Hours per night
Follow-up: 12 to 24 weeks
Average CPAP machine usage ranged across control groups from

2.6 to 5.7 hours per night
Mean machine usage in the intervention groups was
0.6 higher
(0.27 to 0.93 higher)
508
(7 studies)
⊕⊕⊕⊝
moderate1

N deemed adherent (≥4 hours/night)
Follow-up: 12 to 24 weeks
57 per 10071 per 100
(59 to 80)
OR 1.8
(1.09 to 2.95)
285
(3 studies)
⊕⊕⊝⊝
low1,2

Symptoms of sleepiness
Epworth Scale: 0 to 24
Follow-up: median 12 weeks
Mean Epworth Sleepiness Scale scores across control groups ranged from

5.4 to 10.8
Mean Epworth Sleepiness Scale scores in the intervention groups was
1.17 lower
(2.07 to 0.26 lower)
336
(5 studies)
⊕⊕⊕⊝
moderate1

Quality of life: Sleep Apnoea Quality of Life Index (SAQLI)
Follow-up: 4 weeks
See commentSee commentNot estimable89
(1 study)
⊕⊝⊝⊝
very low1,3
Single study estimate

Withdrawal
Follow-up: 4 to 24 weeks
24 per 10018 per 100
(13 to 24)
OR 0.67
(0.45 to 0.98)
683
(8 studies)
⊕⊕⊝⊝
low1,4

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1Risk of bias (-1): In the absence of blinding across studies, effect estimates may be biased because of performance bias.
2Imprecision (-1): We downgraded because of the low number of participants in the analysis, in spite of the statistically significant increase in the number of participants deemed compliant.
3Imprecision (-2): In view of the very low number of participants and the wide confidence intervals, we downgraded by two points.
4Inconsistency (-1): The direction and magnitude of effect varied across studies.
 
Summary of findings 3. Behavioural therapy for adults with sleep apnoea who are using CPAP

Behavioural therapy for adults with sleep apnoea who are using CPAP

Patient or population: adults with sleep apnoea
Intervention: behavioural therapy and CPAP

Comparison: CPAP

Settings: community

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

ControlBehavioural therapy

Machine usage
Hours per night
See commentAverage machine usage in the intervention groups was
1.44 higher
(0.43 to 2.45 higher)
584
(6 studies)
⊕⊕⊝⊝
low1,2
Data analysed as generic inverse variance

N deemed adherent (≥4 hours/night)
Follow-up: four to 52 weeks
28 per 10047 per 100
(36 to 58)
OR 2.23
(1.45 to 3.45)
358
(3 studies)
⊕⊝⊝⊝
very low1,3

Symptoms
Epworth scores
See commentSee comment100
(1 study)
⊕⊕⊝⊝
low1,4
Single study estimate

Quality of life
Functional Outcomes of Sleep Questionnaire
See commentSee comment100
(1 study)
⊕⊕⊝⊝
low1,4
Single study estimate

Withdrawal
Follow-up: 4 to 52 weeks
23 per 10020 per 100
(15 to 27)
OR 0.85
(0.57 to 1.25)
609
(5 studies)
⊕⊝⊝⊝
very low1,2,5

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1Risk of bias (-1): In the absence of blinding across studies, effect estimates may be biased because of performance bias.
2Inconsistency (-1): Variation was seen in the magnitude and direction of effect across studies.
3Inconsistency (-2): Very substantial variation was seen between direction and results of studies.
4Imprecision (-1): Low number of participants contribute data to this outcome.
5Imprecision (-1): Confidence intervals were compatible with reduction and increase in likelihood of study withdrawal.
 
Table 1. Number screened, entered and completed

StudyN ScreenedEnteredCompleted% Screened% Entered

Aloia 2001NA1212NA100

Aloia 2012a, Aloia 2012b3392271494466

Basoglu 201124613313354100

Chervin 1997NA (75% of those approached agreed to participate)4033NA82.5

DeMolles 2004NA3030NA100

Epstein 2000NA5043NA86

Fox 2012NA7554NA72

Hoy 1999NA8075NA94

Hui 2000NA108108NA100

Lewis 20067472557476

Meurice 2007a, Meurice 2007b, Meurice 2007c, Meurice 2007d133112916881

Olsen 2012132100735573

Parthasarathy 20124939377695

Richards 2007109100797279

Roecklein 2010NA3028NA93

Schiefelbein 2005NA5151NA100

Smith 2006NA1919NA100

Smith 2009NA9773NA75

Sparrow 20104232501152746

Stepnowsky 20079145404488

Taylor 20061601321147186

Wang 2011a, Wang 2011b, Wang 2011c, Wang 2011dNA152130NA86

Wiese 2005NA9356NA60

 
Table 2. Study characteristics

Intervention groupStudyInterventionControlStudy duration

(weeks)

Increased support and reinforcement componentsIncreased educational componentsBehavioural therapy

Increased support and reinforcementChervin 1997Weekly telephone calls to monitor progress and troubleshootWritten information on OSA and CPAPUsual careEight

DeMolles 2004Computer-based telecommunication system allowing for monitoring and reinforcing complianceEducation provided by the computer-based telecommunication systemUsual careEight

Fox 2012Telecomunication system allowing for daily monitoring of CPAP usage, timely detection and troubleshooting of problemsUsual care12

Hoy 19992 additional titration nights in hospital, 4 additional visits at home by sleep nursesInitial education at home with partnerUsual care24

Hui 20002 additional early reviews by sleep physician and frequent telephone calls by sleep nursesVideotape and additional education sessionUsual care12

Lewis 20061 additional early review by sleep physician and 1 early telephone interview with sleep nurseEducational videoUsual care52

Meurice 2007a4 additional home visits in the first 3 months by sleep practitioner for problem solvingWritten information and detailed explanation by the prescriber, additional education during home visitsWritten information and detailed explanation by the prescriber + usual care52

Meurice 2007b4 additional home visits in the first 3 months by sleep practitioner for problem solvingAdditional education during home visitsUsual care52

Parthasarathy 20122 individual sessions and 8 telephone conversations with trained peer CPAP users providing support and sharing their positive experience with CPAPPeers shared their knowledge on CPAP and OSAInterventions delivered by peer contained elements of promoting self-efficacy, risk perception, participant activation and motivationUsual care12

Schiefelbein 2005Internet-based application aimed at encouraging CPAP use and problem solvingInternet-based application similar in format to intervention but directed activities in neutral health topics (vitamin intake)16

Smith 2006Home video-link sessions delivered by nurse, who guided correct CPAP use and provided problem solvingNurse provided education on CPAP and OSAHome video-link sessions similar in form to intervention but directed activities in neutral health topics (vitamin intake)12

Smith 2009Audiotaped music along with softly spoken directions on relaxation techniques and habit-promoting instructions for using CPAP, user reminder placardHandouts on benefits of CPAP adherence and health consequences of poor complianceAudiotaped music along with spoken information about vitamins. Information packet similar in format to intervention, but content was on vitamins24

Stepnowsky 2007Wireless telemonitoring of compliance and treatment efficacy on daily basis and acting on the data via prespecified clinical pathwaysUsual careEight

Taylor 2006Internet-based application aimed at monitoring self-reported compliance, acting on the information in timely fashionUsual careFour

Wang 2011bProgressive muscle relaxationUsual care12

Wang 2011cProgressive muscle relaxation + 2 additional nights of CPAP titration4hour group education session, written information, video CDTwo additional nights of CPAP titration + four-hour group education session, written information, video CD + usual care12

Increased educationAloia 2012aTwo 45-minute individual didactic sessions and one booster phone call by sleep nurseUsual care52

Basoglu 201110-Minute educational video session on OSA and CPAPUsual care24

Epstein 2000Educational and desensitisation courseUsual care24

Meurice 2007c4 additional home visits in the first 3 months by sleep practitioner for problem solvingWritten information and detailed explanation by the prescriber, additional education during home visitsFour additional home visits in the first three months by sleep practitioner for problem solving and additional education + usual care52

Meurice 2007dWritten information and detailed explanation by the prescriberUsual care52

Wang 2011a2 additional nights of CPAP titrationFour-hour group education session, written information, video CDUsual care12

Wang 2011dProgressive muscle relaxation + 2 additional nights of CPAP titrationFour-hour group education session, written information, video CDProgressive muscle relaxation + usual care12

Wiese 200515-Minute educational video addressing misconception about OSA and barriers to effective CPAP treatmentUsual careFour

Behavioural therapyAloia 2001Elements of education on consequences of OSA and efficacy of CPAPTwo 45-minute sessions of cognitive-behavioural therapy interventionsTwo 45-minute sessions involving discussion on sleep architecture and sleep clinic12

Aloia 2012bTwo 45-minute sessions of Motivational Enhancement Therapy, one booster phone callUsual care52

Olsen 201245-Minute individual education sessionThree 30-minute sessions of Motivational Interviewing Therapy45-Minute educational session + usual care52

Richards 2007Slide presentation and written information on OSA and CPAPTwo one-hour group sessions of cognitive-behavioural therapyUsual careFour

Roecklein 2010Written personalised feedback report framed according to Motivational Enhancement TheoryWritten information from the American Academy of Sleep Medicine12

Sparrow 2010Side effects management module incorporated in the automated telephone-linked communication systemInformation exchange on OSA and CPAP incorporated in the automated telephone-linked communication systemAutomated telephone-linked communication system designed around the concept of Motivational Interviewing, which allowed one to assess and enhance CPAP complianceGeneral education on unrelated health topics via automated telephone-linked communication system52