Sparrow 2010.

Methods	Randomised parallel‐group trial
Participants	N = 250 patients undergoing initial set‐up of fixed‐pressure CPAP or BiPAP. Inclusion criteria: age 18 to 80 years, AHI > 10 Exclusion criteria: Not reported Baseline characteristics: 18% female. Median age 55. Median AHI 38.3. Median ESS 10.5. Median BMI 35.1. Country: USA
Interventions	Participants were randomised to control (n = 126) or interactive voice response system, TLC‐CPAP (TLC‐CPAP , n = 124) group. TLC‐CPAP: 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: 12 months
Outcomes	Machine usage (data downloaded from memory cards or by direct interrogation of CPAP devices) at 6* and 12 months. N of adherent participants (usage > 4 hours per night) Self‐efficacy index Decisional balance index Withdrawals
Notes	* Indicates primary outcome analysed in this Review.
Risk of bias
Bias	Authors' judgement	Support 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 risk	No information provided
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Blinding 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 risk	Data 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
Selective reporting (reporting bias)	Unclear risk	Information not available
Other bias	Unclear risk	Information not available
Bias arising from the randomisation process (ROB2, primary outcome)	Unclear risk	Only information provided as to the randomisation procedures used was within Methods: " Participants were then randomised to one of two groups: one group used an educational control TLC system ('attention placebo control group') and the other group used the TLC‐CPAP system. Randomisation was stratified by sex, age and AHI using a randomised block design to ensure balance of these factors in the treatment arms." 250 participants were randomised: 124 to intervention, 126 to control. Key baseline characteristics (age, gender, BMI, AHI) were reported for all randomised participants and differences were insignificant (P values reported), consistent with chance.
Bias due to deviations from intended interventions (ROB2, primary outcome)	Low risk	Analyses were performed by intention‐to‐treat. No deviations documented; none suspected based upon review. Authors report, "Analyses were performed by intention to treat." Results report, "CPAP adherence data were available from either the 6‐ or 12‐month follow‐up visit in 93.6% of participants (figure 1), who were therefore included in the primary analysis." This suggests primary CPAP usage results were based on mITT analyses.
Bias due to missing outcome data (ROB2, primary outcome) All outcomes	Unclear risk	Authors report, "CPAP adherence data were available from either the 6‐ or 12‐month follow‐up visit in 93.6% of participants (figure 1)." Participant flow shows that 110 of 124 (88%) randomised to intervention arm had CPAP adherence data at 6‐month endpoint; 112 of 126 (88%) in control arm. Despite the availability of outcome data for at least one endpoint (6‐ or 12‐month) in 93.6% of participants, each group had < 90% of outcome data at each endpoint. Moreover, authors did not provide information as to the reasons for missing data and there was a slight imbalance in missing data across arms at 12 months (83.8% in intervention arm, 88.1% in control arm). Per Cochrane Handbook, 8.13.2.2, "Even if incomplete outcome data are balanced in numbers across groups, bias can be introduced if the reasons for missing outcomes differ." Since authors provide no information as to the reasons for missing data, the potential for bias cannot be adequately assessed.
Bias in measurement of the outcome (ROB2, primary outcome) All outcomes	Low risk	Outcome measured using objective CPAP usage data. Each intervention group outcome data ascertained via automated CPAP device monitoring. Unable to confirm with study author that distribution of CPAP device makes did not differ between intervention arms. Outcome "assessor" is CPAP device: no knowledge of allocation possible.
Bias in selection of the reported result (ROB2, primary outcome) All outcomes	Low risk	Publication (2010) and ClinicalTrials.gov entry, NCT00232544 (first submitted 30 Sep 2005, last updated 10 Oct 2018) available and reviewed. Actual primary completion date is listed as 'March 2008 (final data collection date for primary outcome measure.' Original primary outcome (submitted 30 Sep 2005) was "Objective CPAP use and disease specific quality of life at 6 and 12 months" and current primary outcome (submitted 31 Dec 2013) is "Objective CPAP Use (Time Frame: 12 months) Mean nightly hours of CPAP use over 12 months." Published report (2010) includes both 6‐ and 12‐month endpoints. Though the trialists do not explicitly state that analysis plan was finalised before unblinded outcome data were available for analysis, the original primary outcomes were submitted well before primary completion date and are identical to outcomes reported in published report. Methods section indicates that multiple time points planned; each planned outcome reported. No evidence that multiple analyses (e.g. variable adherence 'thresholds') were conducted. No threshold‐based adherence outcomes reported.
Overall risk of bias (ROB2, primary outcome) Machine usage	Unclear risk	‐