This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Poor adherence to continuous positive airway pressure (CPAP) treatment by adults with obstructive sleep apnea (OSA) is a common issue. Strategies delivered by means of information and communication technologies (ie, eHealth) can address treatment adherence through patient education, real-time monitoring of apnea symptoms and CPAP adherence in daily life, self-management, and early identification and subsequent intervention when device or treatment problems arise. However, the effectiveness of available eHealth technologies in improving CPAP adherence has not yet been systematically studied.
This meta-analytic review was designed to investigate the effectiveness of a broad range of eHealth interventions in improving CPAP treatment adherence.
We conducted a systematic literature search of the databases of Cochrane Library, PsycINFO, PubMed, and Embase to identify relevant randomized controlled trials in adult OSA populations. The risk of bias in included studies was examined using seven items of the Cochrane Collaboration risk-of-bias tool. The meta-analysis was conducted with comprehensive meta-analysis software that computed differences in mean postintervention adherence (MD), which was defined as the average number of nightly hours of CPAP use.
The meta-analysis ultimately included 18 studies (N=5429 adults with OSA) comprising 22 comparisons between experimental and control conditions. Postintervention data were assessed at 1 to 6 months after baseline, depending on the length of the experimental intervention. eHealth interventions increased the average nightly use of CPAP in hours as compared with care as usual (MD=0.54, 95% CI 0.29-0.79). Subgroup analyses did not reveal significant differences in effects between studies that used eHealth as an add-on or as a replacement to care as usual (
eHealth interventions for adults with OSA can improve adherence to CPAP in the initial months after the start of treatment, increasing the mean nightly duration of use by about half an hour. Uncertainty still exists regarding the timing, duration, intensity, and specific types of eHealth interventions that could be most effectively implemented by health care providers.
Obstructive sleep apnea (OSA) is a clinical sleep disorder characterized by recurrent episodes of partial or complete obstruction of the respiratory passages during sleep [
Unfortunately, acceptance and adherence are often suboptimal in CPAP treatment, thereby jeopardizing the improved health outcomes. It is estimated that 30% to 80% of OSA patients can be classified as nonadherent when operationalized as using CPAP for less than 4 hours per night [
A growing body of research is investigating interventions to promote CPAP adherence [
Strategies delivered by means of information and communication technologies (ie, eHealth) offer strong potential to address the relatively poor rate of CPAP adherence through standardized education, real-time monitoring of symptoms and CPAP adherence in daily life, self-management, and early identification and intervention if device or treatment problems arise [
Our search strategy was part of a broader search performed in a research project on the role of eHealth in treatment adherence in chronic lung diseases. The searches for OSA, asthma, and chronic obstructive pulmonary disease (COPD) were thereby pooled together.
A systematic literature search was conducted in the electronic databases of the Cochrane Library (Wiley), PsycINFO (EBSCO), PubMed, and Embase. The search results were limited to available full-text articles in English or Dutch with publication dates from January 1, 2000, to March 20, 2018. The starting year of 2000 was chosen because technology began greatly advancing around that time. Terms related to eHealth technology, patient adherence, and the target populations were combined, using both free-text and index terms (see
The study inclusion criteria were as follows: (1) The target population comprised patients aged 18 years or older who were undergoing CPAP treatment and whose OSA diagnosis was supported by polysomnographic examination, home sleep apnea testing, or nocturnal pulse oximetry; (2) A major component of the experimental intervention was delivered by eHealth technology or an eHealth component was assessed as an add-on to care as usual (CAU), irrespective of whether it comprised a major part of the experimental intervention. The criteria to qualify as an eHealth intervention were that the intervention was delivered via information and communication technology, such as telephone calls, telemedicine (eg, videoconferencing), websites, smartphone applications, SMS and the intervention was delivered independently of time and place, making distance a critical factor (eg, videos delivered in face-to-face sessions were not considered eHealth interventions); (3) CAU did not include the experimental eHealth intervention or component under investigation, thus excluding any studies comparing similar eHealth interventions with differing contents, such as general versus tailored text messages; (4) Outcomes were assessed in terms of one or more quantitative measures of patient adherence to CPAP treatment; (5) Outcomes were compared statistically between study conditions; (6) Study design was a randomized controlled trial.
Two reviewers (JA and LL) independently screened all titles and abstracts for eligibility. Subsequently, the reviewers independently screened the full text of the selected papers to determine eligibility for inclusion. Disagreements were resolved by discussion. Covidence software [
Data on study reference, design, population, interventions, outcomes, and results were extracted by JA from all eligible studies (
For studies with multiple intervention conditions, the control condition was split into two or more groups corresponding to the number of experimental comparisons, with sample sizes divided by that number, thus enabling separate comparisons of intervention conditions within the same meta-analysis. Since considerable heterogeneity among studies was expected, a random-effects model was chosen [
Subgroup analyses were conducted using a mixed-effects model, pooling the studies within subgroups with a random-effects model and testing for significant differences between subgroups with a fixed-effects model. One subgroup analysis compared CPAP adherence in studies that tested eHealth interventions as an add-on to CAU with adherence in studies that tested them as a replacement of CAU. This was of interest because the context of eHealth delivery could have important implications for how interventions are implemented in the process of care delivery and follow-up, and more generally, for the efficiency of and burden on the health care system. A second subgroup analysis compared interventions delivering eHealth only versus blended approaches combining eHealth and face-to-face strategies. A third analysis compared fully automated versus guided eHealth interventions given that it is often assumed that guided and blended interventions lead to better adherence outcomes.
If included studies did not report the data needed to carry out main or subgroup analyses, we attempted to contact the first or corresponding author to gain the necessary data.
The Cochrane Collaboration risk-of-bias tool [
PRISMA flowchart describing the study identification and selection process.
Most studies (n
Of the 14 add-on studies comparing CAU to the same care supplemented with eHealth, nine studies added eHealth components only, whereas five added a combination of face-to-face and eHealth strategies. Most studies adding eHealth components alone used telemonitoring tools (n=7) to monitor CPAP adherence and efficacy data, and telephone calls (n=7) intended to educate, provide support, promote self-management, or reinforce adherence. One study included a Web-based education portal, as well as automated feedback messages by e-mail, telephone, or SMS, according to CPAP monitoring data [
In the five replacement studies, face-to-face follow-up consultations were replaced by eHealth strategies. More specifically, Fields et al [
All studies, except one [
As shown in
Not a single study was rated as having a low bias risk for all seven assessment dimensions, and this was mainly due to a high bias risk for the blinding of participants and personnel dimension. Most studies had a low bias risk for blinding of outcome assessment because CPAP adherence data were downloaded directly from CPAP devices. A high risk of selective outcome reporting was identified for two studies that failed to adequately report on the types of adherence outcomes specified in their methods sections [
Risk of bias for each individual study included in this meta-analytic review.
Summary of the risk of bias for all included studies in this meta-analytic review.
A visual inspection of the funnel plot did not indicate potential publication bias, but the Egger linear regression test of the intercept was significant (
Among the 19 eligible studies identified, one study [
Because one study [
Results of the main and subgroup analyses at postintervention assessment.
Variables | Studies, n | Comparisons, n | Total, Na | Mean difference |
|||||||||
CPAPc adherenced | 18 | 22 | 5429 | 0.54 (0.29-0.79)e | N/Af | 90.45 (87-93) | |||||||
Outliers excluded | 12 | 14 | 1433 | 0.54 (0.27-0.82)e | N/A | 51.10 (10-73) | |||||||
Subgroup analyses |
|
||||||||||||
|
|
|
|
|
.95 |
|
|||||||
|
Add-on to usual care | 13 | 17 | 4879 | 0.54 (0.20-0.87)e |
|
91.34 (88-94) | ||||||
|
Replacement of usual care | 5 | 5 | 550 | 0.52 (0.13-0.91)e |
|
69.10 (21-88) | ||||||
|
|
|
|
|
.23 |
|
|||||||
|
eHealth only | 11 | 14 | 1690 | 0.38 (0.07-0.70)e |
|
66.35 (41-81) | ||||||
|
Blended: combined eHealth + face-to-face care | 5 | 6 | 3458 | 0.76 (0.23-1.29)e |
|
96.73 (95-98) | ||||||
|
|
|
|
|
.83 |
|
|||||||
|
Fully automated | 4 | 7 | 830 | 0.60 (−0.03 to 1.24) |
|
57.23 (1-82) | ||||||
|
Guided | 14 | 15 | 4599 | 0.53 (0.25-0.81)e |
|
93.19 (90-95) |
aTotal sample analyzed: total randomized N in intent-to-treat analyses and N of completers in completers-only analyses.
bTwo-tailed
cCPAP: continuous positive airway pressure.
dCPAP adherence operationalized as average nightly CPAP use in hours.
e
fNot applicable.
Forest plot of intervention effects on adherence as defined as mean nightly continuous positive airway pressure (CPAP) use in hours.
The results of the subgroup analyses are shown in
Four studies included follow-up assessments subsequent to the postintervention measurement. In view of this limited number of studies and their large variation in follow-up periods, no meta-analysis was conducted. We will now review the follow-up data, distinguishing between short-term follow-up (1-6 months; three studies) and long-term follow-up (≥1 year; two studies).
Regarding studies with short-term follow-up, Lo Bue et al [
Regarding long-term follow-up, Bouloukaki et al [
To our knowledge, this meta-analytic review is the first to systematically assess the influence of eHealth interventions in improving adherence to CPAP treatment among adults with OSA. Nineteen eligible studies were identified, and our meta-analysis included data from 18 studies reporting 22 comparisons. A heterogeneous collection of eHealth interventions, employed either as add-ons or as replacements to CAU, were found to increase the average CPAP adherence by about half an hour a night as compared with CAU alone. No significant differences in effects emerged between eHealth provision supplemented to CAU and eHealth as a replacement of CAU. Additionally, no significant differences were found between other subgroups of approaches (eHealth only versus blended interventions and fully automated versus guided eHealth interventions).
In line with preliminary investigations [
It is difficult to determine the clinical relevance of our meta-analytic finding that eHealth technologies increased average CPAP adherence by half an hour a night. There is no established general cut-off point defining how much adherence leads to clinically meaningful improvement. In contrast to the dose-response relationships noted above, some studies have reported effective treatment of OSA with relatively few hours of CPAP use, whereas others noted little progress at longer durations. Individual variation in CPAP response in terms of indicators, such as sleepiness, may depend on factors such as biological response mechanisms [
As to whether specific characteristics of eHealth adherence interventions could potentially moderate CPAP response, our meta-analysis showed no significant differences in effect sizes for eHealth adherence interventions delivered as (1) replacements to CAU rather than as add-ons, (2) blended versus eHealth-only strategies, or (3) guided versus fully automated interventions
Follow-up data beyond posttreatment measurements were too limited for meta-analytical assessment. Furthermore, the results are limited to adult populations scoring generally well above the threshold for severe OSA; it is unclear whether the results could be generalized to younger populations or those with less severe OSA. Another limitation was the moderate-to-high heterogeneity in the results between the included studies, as well as the high risk of bias in some studies for one or more dimensions. The type and intensity of CAU provided in the control condition varied considerably, potentially biasing the results. The null findings in our subgroup or moderator analyses should be interpreted with caution, as the analyses may have been underpowered. Further limitations lie in the fact that not all studies performed conventional polysomnography to diagnose patients and that CPAP may not have always been manually titrated. We did not search for gray literature, and we searched only for literature published after 2000. Finally, in several studies, routine or as-needed telephone support was part of CAU, whereas in other studies, it was confined to the experimental intervention condition.
Economic evaluations are needed to determine the cost-effectiveness of eHealth adherence interventions in comparison with CAU. To our knowledge, only two studies [
Another future research direction would be to investigate the long-term effectiveness of eHealth interventions in improving adherence to CPAP treatment. What happens when patients are no longer monitored or followed up by visits to the clinic after their first months using the CPAP device?
Currently, little is known about which eHealth strategies or components are most effective in increasing CPAP adherence. Such information could help design the most efficient and effective interventions. Future studies could also investigate the benefits of eHealth adherence interventions for individuals with moderate levels of OSA.
With regard to methodology, future studies should carefully take into account the various risks of bias identified in many studies in this review, that is, outcome measures should be defined a priori and should be adequately reported, an intent-to-treat design should be adopted when analyzing the data, and any baseline imbalances should be adequately accounted.
Finally, an interesting direction for future research would be to examine the potential of incorporating psychological theories and models into eHealth adherence interventions. Promising results have already been reported for interventions based on cognitive-behavioral treatment principles [
The current findings suggest that a broad range of eHealth interventions are effective in increasing adherence to CPAP treatment. Given the literature showing that higher CPAP adherence is generally associated with better outcomes, the potential of eHealth should be further explored and exploited. We therefore recommend assessing personal pathways in more detail to determine who can benefit the most from digitally enabled adherence support. Research is also needed on the cost-effectiveness of interventions and on how they might be implemented on a large scale.
Providing eHealth interventions to adults with OSA during CPAP treatment can improve treatment adherence in the initial months, increasing the mean nightly duration of use by about half an hour. eHealth technologies can also be employed as tools to deliver standardized education and to monitor patients more closely in daily life. This enables the early detection of problems and nonadherence and allows timely and appropriate responses at a distance. More information is still needed about the specific types of eHealth interventions and the timing, duration, and intensity of eHealth interventions that health care providers could effectively implement.
Search string.
An overview of the relevant characteristics of each of the included studies (n=19).
automatically adjusted positive airway pressure
care as usual
chronic obstructive pulmonary disease
continuous positive airway pressure
electronic health
obstructive sleep apnea
This study was funded by the Dutch National Health Care Institute. We would like to thank Caroline Planting for her help in conducting the literature search and Dr Lotte Hermsen for providing content support.
None declared.