Abstract
Study objectives:
To examine the usability of positive airway pressure (PAP) devices and its association with PAP adherence among older adults with sleep-disordered breathing.
Methods:
We mailed questionnaires to patients aged ≥65 years prescribed PAP therapy during the prior 36 months from two large healthcare systems. Survey participants completed the Usability of Sleep Apnea Equipment-Positive Airway Pressure (USE-PAP) questionnaire, which assessed the usability of their PAP device. Other questionnaire items included demographics and self-rated health. We also abstracted adherence data (mean nightly hours of PAP use available from one site) and interface type from the electronic health record.
Results:
Five hundred sixty-four patients completed the survey (response rate = 33%). The mean USE-PAP score (0 = best to 100 = worst) was 20 (SD ± 20). Mean duration of PAP use (available in 189 respondents) was 5.2 hours per night (SD ± 2.0). In a nested regression model predicting nightly hours of PAP use, a 10-point (0.5 SD) increase in USE-PAP score corresponded to a 0.37 hour/night reduction in PAP use. The model including the USE-PAP score explained a significant proportion (R2 = 15%) of the variation in nightly hours of PAP use above and beyond demographics, self-reported health, and interface type (∆R2 = 12%).
Conclusions:
Our results demonstrate that PAP usability varies among older patients and is associated with PAP adherence, above and beyond other predictors of adherence. These results support measuring and improving PAP usability to further improve PAP adherence for older patients.
Keywords: sleep apnea, adherence, usability, predictive modeling.
Statement of Significance
We conducted a survey to collect data on patient-reported usability of positive airway pressure (PAP) equipment among older patients, who may be more susceptible to user-device interaction design issues. We found variations in usability issues and showed that usability is significantly associated with PAP adherence. Patient-reported usability is not generally assessed in clinical practice. Because device usability may be a modifiable risk-factor for interventions aimed at improving PAP adherence, we recommend assessing patients’ beliefs about and experiences with PAP usability.
INTRODUCTION
Positive airway pressure (PAP) therapy is the most commonly prescribed treatment for sleep-disordered breathing (SDB).1 Although PAP is an efficacious therapy for reducing the apnea-hypopnea index (AHI), many patients are non-adherent to PAP therapy, with estimates of non-adherence ranging from 28% to 60%.2,3 Research on PAP non-adherence has focused on psychosocial factors (e.g., outcome expectancies, self-efficacy beliefs, socioeconomic characteristics), disease characteristics (e.g., AHI, sleepiness), technical aspects of devices such as interface type (i.e., nasal pillows versus nasal mask versus full face mask), device modality/titration procedure (e.g., auto-titrating, continuous versus bi-level), and side effects (e.g., claustrophobia, irritation).3 One category of barriers that has received little attention is the usability of PAP devices.4–6
Device usability refers to design factors that impact interactions between individuals and devices.7 The International Organization for Standardization defines usability as “the effectiveness, efficiency and satisfaction with which specified users achieve specified goals in particular environments.”8 Usability may impact the accuracy and completeness with which patients achieve PAP therapy goals such as putting on a PAP mask, the resources expended to achieve these goals, the ease of learning how to use the PAP device, the ease of remembering how to use the device, and the degree to which the user finds the device acceptable and pleasant.9 Beneficial outcomes of optimizing usability characteristics of medical devices include minimizing risk of error and ensuring device safety (i.e., promoting use of the device in a manner that does not result in harm to the patient or environment).9 Usability test results are submitted to the Food and Drug Administration (FDA) during the pre-market device approval process.9
In addition to impacting device safety, usability has the potential to impact PAP adherence, which in turn may contribute to the overall impact that PAP has on health outcomes.10,11 Making PAP easier to set up, use, and clean may positively affect adherence to therapy. Large-scale surveys of the usability of PAP devices are lacking, even though interviews with patients about their experiences with these devices suggest variability in PAP usability across patients.5 Older adults may be more likely to experience usability problems due to the higher prevalence of physical and sensory impairments that occur with advanced age.
In this study, we sought to measure PAP usability and to evaluate the associations between PAP usability and PAP adherence among older adult patients prescribed PAP therapy for SDB. We hypothesized there would be a considerable range of PAP usability scores and that usability scores would be associated with adherence to PAP.
METHODS
Conceptual Model
The FDA’s model for medical device usability,9 which is based upon the International Organization for Standardization 9241-10 definition of usability,12 served as the conceptual model for this study. This model posits that the device-user system includes three major components: device users (e.g., role [patient, caregiver, physician], other characteristics), device use environments (e.g., home, hospital), and device-user interfaces (e.g., components and accessories, controls, displays, feedback, labeling, training). These components affect device use, which, in turn, impacts outcomes. As described in the introduction, the FDA focuses on safe and effective use as the main outcome; however, other important outcomes include adherence.
Study Design and Sample
In 2014, we conducted a PAP usability survey of older adults (≥65 years) who received SDB care in the Los Angeles area. We mailed questionnaires to all individuals prescribed PAP therapy during the prior 36 months from one Department of Veterans Affairs (VA) sleep center (N = 1,203) and two sleep clinics affiliated with a university health system in Los Angeles, California (N = 526). A $2 bill was included with the questionnaire as an incentive to return the questionnaire, along with a postage-paid return envelope. The total response rate was 33%: 564 surveys were returned by mail.
In addition to the mailed survey, we conducted a structured medical record abstraction of PAP adherence data in participants recruited from the VA site, where PAP adherence data are routinely documented in the electronic health record (EHR) in the course of usual care; abstraction of adherence data were not conducted at the university site, which uses a different EHR system. PAP adherence data (mean hours of use per night) were available for 189 survey respondents. To evaluate the potential for non-response bias, we abstracted adherence data for a randomly selected subset of VA survey non-responders, which was identified by assigning a random number to each of the non-responders, sorting the observations by that random number, and selecting the top 125 observations. We then compared mean nightly hours of PAP use for survey responders with this randomly selected subset of survey non-responders. Usual care at the VA site consists of hands-on training of all new users of PAP devices. All PAP training is performed by a small core group of experienced certified respiratory therapists who have additional training in sleep devices. Patients are provided detailed information about the device parts, proper care of the device, and comfort settings during the training. At all sleep clinic visits, one to three certified respiratory therapists from this group are available to meet with patients who need new supplies for their PAP devices and to troubleshoot device-related issues. In between clinic visits, support from this core group of certified respiratory therapists is available on a walk-in basis.
All study procedures were approved by the institutional review boards at the Veterans Affairs Greater Los Angeles Healthcare System (#2013–091198) and the University of California, Los Angeles (#13-001132). A waiver of documentation of consent for survey respondents and for EHR review was obtained.
Measures
PAP Usability Measure
An adapted version of the Usability of Sleep Apnea Equipment-Positive Airway Pressure (USE-PAP 18-item scale transformed linearly; 0 [best] to 100 [worst]) questionnaire was included in the mailed survey. It assesses patients’ experience setting up, using, and cleaning their PAP devices.6 This questionnaire was developed using in-depth patient interviews (interviews provided qualitative data on the types of usability issues experienced by older adults), technical advisory panel input (survey items were reviewed and edited by a diverse team of clinicians and researchers), cognitive interviews with patients (participants provided information on the clarity, usefulness, and presentation of survey items), feedback from sleep clinicians, and a pilot survey in sleep clinic. We revised items until patients and clinicians understood the meaning of items such as “what it takes to adjust controls.” These questionnaire development activities have been described previously.6 Total score was used for the main analysis. Two usability subscales were developed to provide information about two specific dimensions of usability, efficiency, and satisfaction. The items comprising these subscales were added and transformed linearly (0 [best] to 100 [worst]). Cronbach’s alpha were calculated for the scales. The three remaining items, which represent other dimensions of usability (effectiveness, learnability, and memorability), were single items and therefore, were not included in subscales.
Adherence Measure
Mean Nightly Hours of PAP Use Documented in the EHR.
for all VA patients who responded to our survey, we performed structured data abstraction from VA EHR sleep center notes, which typically include PAP adherence information since PAP adherence is one of the main topics discussed during most sleep center encounters, and staff are expected to record this information. The sleep center note closest to the date the survey was mailed was selected for abstraction. If no PAP adherence data were available in the selected note, the next sleep center note was reviewed for PAP adherence data, and so forth until the start date of the approved abstraction period (May 1, 2011) was reached. Sleep center notes occurring after the date the survey was received by our research center were also abstracted for PAP adherence data (available in only a limited number of participants) and for 13 participants who did not have any PAP adherence data prior to sending the survey but who had “post-survey” adherence data, these post-survey data were used. Machine-measured data on hours of PAP use/night obtained directly from querying the PAP device or from the PAP device’s memory card were used for this study.
Other Measures
Gender and age were obtained from the medical records during survey mailing preparation. The survey included items on race, ethnicity, and educational level, as well as a single item of self-rated general health from the 12-Item Short Form (SF-12).13,14 Poor self-rated health is associated with worse medication adherence,15 and this item is predictive of hospitalization and mortality.16 Previously validated items on beliefs about PAP effectiveness and importance in sleep apnea management were included in the survey (2 items) to measure how effective individuals believe regular use of PAP is for managing sleep apnea and how important they believe regular use of PAP is for controlling sleep apnea.17
Interface and PAP Device Type.
The brand and model of the interface were abstracted from medical records and were reliably available for participants from the VA site. We categorized the interface by interface type: nasal pillows, nasal mask, and full face mask. Interface types have been found to be associated with adherence and satisfaction.18 The brand and model of the PAP device were also abstracted for participants from the VA site.
Data Analysis
We used standard descriptive statistics (e.g., mean, standard deviations) to describe continuous variables and frequencies to describe categorical variables. We estimated the internal consistency of the USE-PAP scale, using Cronbach’s alpha.19 We used Student’s t-test, Pearson’s chi-squared test, and Fisher’s Exact tests to compare demographic and usability measures between participants from the VA and university sites.
For the VA subset of participants with adherence data, we regressed USE-PAP score on mean nightly hours of PAP use. To assess for nonlinearity, we plotted residuals versus fitted values and ran models that included quadratic and cubic terms. Then, we ran a nested multivariable regression model predicting mean nightly hours of PAP use to measure the change in R2 associated with USE-PAP score, above and beyond demographic variables, self-rated health, and PAP interface type. Covariates were put into three blocks in the nested regression model, and each block was successively added. Block 1 included sociodemographic (i.e., age, race, education) and self-rated health. Block 2 included types of PAP interface. Block 3 included the USE-PAP score. Device model was not entered into the regression model, because all patients were prescribed devices from the same manufacturer and there was little variation in the device model at the VA site. Gender was not included in the model, because the VA sample was predominantly male.
In post hoc analyses, we examined the relationship between the usability subscales (efficiency and satisfaction) and hours/night of PAP use, by substituting the USE-PAP total score in Block 3 with the two subscales and the three other USE-PAP items. In addition, we calculated the number of days that elapsed between the date the adherence data were documented and the date the survey was mailed. Because we observed a large range in the number of elapsed days, we examined whether the relationship between USE-PAP total score and hours/night of PAP use was the same for participants with adherence data that were closer to the survey mail date as those with adherence data farther from the survey mail date, by running a nested multivariable regression model that included an interaction between USE-PAP total score and number of elapsed days.
For these statistical tests, p <.05 was considered statistically significant. All statistical analyses were conducted with Stata/SE 13.1 (College Station, TX: StataCorp LP).
RESULTS
Descriptive Results
A total of 564 individuals participated in the survey (response rate = 33%). Table 1 shows the characteristics of the study participants. The overall mean age of participants was 71 years (SD 5.9; university site = 74 [SD 7.0], VA site = 71 [SD 5.7], t(563) = 5.53, p < .001). Overall, the sample was predominantly male (85%) and non-Hispanic white (60%). The university site had more women than the VA site (44% versus 1%, p < .001), but the two sites were similar in percentage of non-Hispanic white participants (63% versus 61%, χ2 = 0.138, p = 0.711).
Table 1.
Participant Characteristics (n = 564).
| Characteristic | Mean (SD) or frequency (%) |
|---|---|
| Age, in years | 71.4 (5.9) |
| Male | 477 (85%) |
| Non-Hispanic white | 339 (60%) |
| Education | |
| ≤8th grade | 3 (1%) |
| Some high school, but did not graduate | 9 (2%) |
| High school graduate or GED | 52 (10%) |
| Some college or 2-year degree | 215 (39%) |
| Trade or technical school | 33 (6%) |
| 4-year college graduate | 110 (20%) |
| More than 4-year college degree | 127 (23%) |
| Self-rated health | |
| Excellent | 27 (5%) |
| Very good | 131 (25%) |
| Good | 194 (37%) |
| Fair | 133 (26%) |
| Poor | 36 (7%) |
The mean USE-PAP score was 21 (SD = 20; range = 0–100) and was not significantly different between the two sites (p = .29). The median was 15 (interquartile range 4, 31). Cronbach’s alpha for the USE-PAP was 0.94. The USE-PAP efficiency subscale mean was 20 (SD = 20; range 0–100) and median was 14 (interquartile range 0, 89), and the USE-PAP satisfaction subscale mean was 21 (SD = 22) and median was 12.5 (interquartile range 0, 100). Cronbach’s alpha was 0.88 for the USE-PAP efficiency subscale and 0.87 for the USE-PAP satisfaction subscale. Table 2 shows summary statistics of the USE-PAP items. The items with the most favorable usability ratings were related to connecting the tubing (72% strongly agreed that they can easily connect the tubing and 68% were satisfied with what it takes to connect the tubing). The items with the least favorable usability ratings were related to the device controls (11% strongly disagreed that they know how to adjust the device controls) and to the mask/headgear (9% strongly disagreed that they can easily adjust their mask/headgear so it seals).
Table 2.
Usability of Positive Airway Pressure Device.
| USE-PAP item | Strongly agree, N (%) | Agree, N (%) | Neither agree nor disagree, N (%) | Disagree, N (%) | Strongly disagree, N (%) | Sample size, N |
|---|---|---|---|---|---|---|
| Quickly get equipment ready for use | 341 (63.6) | 115 (21.5) | 33 (6.2) | 27 (5.0) | 20 (3.7) | 536 |
| Quickly learned how to get equipment ready for use | 290 (53.7) | 159 (29.4) | 38 (7.0) | 35 (6.5) | 18 (3.3) | 540 |
| Could easily remember how to get equipment ready for use | 318 (59.8) | 117 (22.0) | 52 (9.8) | 24 (4.5) | 21 (3.9) | 532 |
| Know how to adjust controls | 194 (36.1) | 134 (25.0) | 83 (15.5) | 68 (12.7) | 58 (10.8) | 537 |
| Can easily operate controls | 250 (47.0) | 128 (24.1) | 69 (13.0) | 45 (8.5) | 40 (7.5) | 532 |
| Can easily adjust mask/headgear so it seals | 243 (45.1) | 150 (27.8) | 41 (7.6) | 58 (10.8) | 47 (8.7) | 539 |
| Can easily connect tubing | 388 (72.1) | 101 (18.8) | 17 (3.2) | 24 (4.5) | 8 (1.5) | 538 |
| Can easily operate humidifier | 296 (59.1) | 103 (20.6) | 62 (12.4) | 24 (4.8) | 16 (3.2) | 501 |
| Can easily replace filter | 328 (61.4) | 94 (17.6) | 67 (12.5) | 27 (5.1) | 18 (3.4) | 534 |
| Easy to know when working properly | 245 (45.4) | 144 (26.7) | 72 (13.3) | 44 (8.1) | 35 (6.5) | 540 |
| Satisfied with what it takes to adjust controls | 197 (37.0) | 143 (26.8) | 100 (18.8) | 52 (9.8) | 41 (7.7) | 533 |
| Satisfied with what it takes to put on mask/headgear | 269 (50.2) | 129 (24.1) | 53 (9.9) | 48 (9.0) | 37 (6.9) | 536 |
| Satisfied with what it takes to connect tubing | 367 (68.0) | 100 (18.5) | 42 (7.8) | 13 (2.4) | 18 (3.3) | 540 |
| Satisfied with what it takes to prepare humidifier | 279 (55.5) | 107 (21.3) | 61 (12.1) | 32 (6.4) | 24 (4.8) | 503 |
| Satisfied with what it takes to replace filter | 306 (56.9) | 107 (19.9) | 61 (11.3) | 35 (6.5) | 29 (5.4) | 538 |
| Easy to clean equipment | 275 (51.5) | 126 (23.6) | 59 (11.0) | 47 (8.8) | 27 (5.1) | 534 |
| Easy to transport | 273 (50.7) | 108 (20.1) | 64 (11.9) | 54 (10.0) | 39 (7.2) | 538 |
| Would recommend to a friend | 308 (57.1) | 89 (16.5) | 84 (15.6) | 19 (3.5) | 39 (7.2) | 539 |
Note: USE-PAP = Usability of Sleep Apnea Equipment-Positive Airway Pressure.
For the subset of participants with adherence data measured in hours per night (n = 189), the mean nightly hours of PAP use was 5.2 (SD = 2.0). The mean nightly hours of PAP use for a randomly selected subset of survey non-responders (N = 34) was 5.1 (2.1) hours per night. No significant difference in PAP adherence was found between the survey respondents and randomly selected survey non-respondents (t(221) = .27, p = .790.).
Bivariate and Multivariable Results Predicting Adherence
In a bivariate analysis among the subsample with EHR adherence data, worse USE-PAP score was associated with lower mean nightly hours of PAP use (beta coefficient = −.033, p < .001), and additional analyses confirmed the linearity of the relationship (residuals versus fitted values plot showed no relationship, and quadratic and cubic terms were not significant).
In model 1 (F(4,153) = 1.10, p = .360, none of the demographic or self-rated health variables were significantly associated with hours/night of PAP use. In model 2 (F(2, 151) = 0.20, p = .822), no specific interface type was a significant predictor of hour/night of PAP use. As shown in Table 3, in a nested multivariable regression model predicting mean nightly hours of PAP use (see model 3 [full model]), worse USE-PAP score was significantly associated with fewer nightly hours of PAP use (a 10-point increase in USE-PAP total score corresponded to a 0.41 hour/night [24.6 minutes] reduction in PAP use; a 1-standard deviation increase in USE-PAP total score corresponded to a 0.37 hour/night [22.2 minutes] reduction in PAP use). USE-PAP total score increased the model’s R2 by 12% (from 3 to 15%; F(1,150) = 21.43, p < 001). In this full model, age was significantly associated with hours of PAP use/night.
Table 3.
Nested Multivariable Regression Modeling Adherence (Hours of Therapy Use Per Night) to Positive Airway Pressure Therapy, Full Model (n = 158).
| Model 1 sociodemographics | Model 2 + interface | Model 3 +usability | |
|---|---|---|---|
| Beta coefficients (p-value) | Beta coefficients (p-value) | Beta coefficients (p-value) | |
| Block 1 (sociodemographics) | |||
| Age | .133 | .132 | .171 |
| (.100) | (.830) | (.027)* | |
| Non-Hispanic white | .016 | .012 | .022 |
| (.848) | (.830) | (.779) | |
| Education | .027 | .024 | .107 |
| (.747) | (.774) | (.184) | |
| Self-rated health | −.076 | −.076 | −.004 |
| (.347) | (.353) | (.958) | |
| Block 2 (PAP interface type) | |||
| Nasal pillows | .000 | −.005 | |
| (.997) | (.954) | ||
| Nasal mask | .050 | −.004 | |
| (.563) | (.965) | ||
| Full face mask (reference) | — | — | |
| Block 3 (usability) | |||
| USE-PAP total score | −.370 | ||
| (<.001)** | |||
| Observations (N) | 158 | 158 | 158 |
| R 2 | .028 | .030 | 0.15 |
| F | 1.10 | 0.79 | 3.83** |
| (df) | (4,158) | (6,151) | (7,150) |
Note: *p < .05; **p < .01; USE-PAP = Usability of Sleep Apnea Equipment-Positive Airway Pressure. Standardized regression coefficients are presented above.
In a nested multivariable regression model that replaced USE-PAP total score in Block 3 with subscales for efficiency and satisfaction and individual items for learnability, memorability, and effectiveness, the satisfaction subscale was significantly associated with hours/night of PAP use (beta = −482, p = .005), while the efficiency subscale (p = .703), learnability item (p = .405), memorability item (p = .941), and effectiveness item (p = .057) were not statistically significant (data not shown in a table).
The number of days that elapsed between the survey mail date and the date adherence data were documented ranged from 2 days to 1,089 days (mean 381 days [SD 271]). In a nested regression model that included a variable representing the number of days elapsed between these dates and an interaction term between USE-PAP total score and days elapsed (in Block 3), neither the number of days elapsed (p = .922) nor the interaction term (p = .738) were significant (data not shown in a table).
DISCUSSION
In this study, we observed wide variability in usability ratings among older adults prescribed PAP and found that usability ratings were associated with non-adherence, with each 10-point increase in the USE-PAP score corresponding to a 0.41 hour (24.6 minutes) reduction in PAP use per night. Usability ratings help explain the variation in nightly hours of PAP use above and beyond demographic, self-rated health, and interface type. These results, which suggest that a subset of older patients prescribed PAP therapy have challenges using PAP devices, corroborate findings of variation in usability from our previous work testing a prior version of the survey among a smaller sample of adults6 and support the need for assessing and improving PAP usability among older patients.
Our findings suggest that some older adults frequently struggle with PAP usability issues, despite pre-FDA market approval usability testing, which typically includes a sample of 15 representative users for summative testing, and FDA-required mitigation plans for usability issues affecting device safety.9 The post-FDA approval usability issues uncovered in our survey might not impact device safety, but may compromise the impact of PAP on health outcomes by reducing adherence to therapy. For PAP therapy, the relationship between usability and adherence is even more important because adherence rates are so low. Other factors associated with adherence to PAP therapy (e.g., psychosocial) have been studied in many studies,3 but the usability of the device has received less attention.
Systematically surveying patients and other end-users about PAP usability after the devices have been approved by the FDA and creating a registry of PAP usability data could be beneficial to many individuals, including patients/caregivers, prescribers, purchasers, and manufacturers of PAP equipment, especially if these data could be made available to manufacturers and if prescribers/purchasers of the equipment have the data at the point of prescribing/purchase. The FDA has growing interest in usability surveillance studies, requiring surveillance of some devices (e.g., devices posing greater than minimum risk that are life-sustaining).20 Usability may be particularly important when prescribing and fitting devices for individuals at higher risk for usability issues, such as individuals with physical limitations, sensory impairments,5 or cognitive impairment, all of which are common in older adults. For example, information about the types of interfaces most easily used by individuals with severe arthritis or hemiplegia could help other individuals with these comorbidities who are looking for a new interface.
Our study has a number of strengths and several limitations. A strength of our design was the sampling frame, which was comprised all patients who had been prescribed PAP therapy in two large healthcare systems during a three-year period, not just patients who attend clinic appointments. Another strength was the availability of machine-measured PAP adherence data and interface type in a subset of respondents from the VA site, where comprehensive, granular durable medical equipment data are readily available in the EHR. This approach enabled us to collect information from patients who may have had been lost to clinic follow-up. One limitation was the low survey response rate, a finding that is increasingly common in mailed surveys.21 However, we found no differences between survey responders and non-responders in PAP adherence, and using a mailed survey facilitated participation of older patients and those with lower socioeconomic status.22,23 We only had adherence data from the VA site, which may limit the generalizability of our findings to women and non-veterans. The number of days that elapsed between the date adherence data were documented and the survey mail date was large for the majority of participants. Although we did not find that the association between usability and adherence was different for those with more versus fewer days elapsed, future studies should consider using adherence data for the period immediately preceding survey completion (which may be more feasible with the increased availability of PAP modem data and online surveys that provide more accurate estimates of date of survey completion). Another limitation is that we did not have an opportunity to assess health literacy level, to perform cognitive testing, or to measure level of sleepiness objectively. These patient characteristics could confound the relationship between usability and PAP adherence.
In conclusion, we that found that usability ratings help explain variation in PAP adherence. Future large studies are needed to describe PAP usability and its relationship to adherence and health outcomes in samples with more women and racially/culturally diverse patients. In addition, PAP usability may be important to address in future interventions aimed at improving PAP adherence.
FUNDING
Research reported in this publication was supported by the National Institute On Aging of the National Institutes of Health under Award Number K23AG045937, The Beeson Career Development in Aging Research Award Program (supported by NIA, AFAR, The John A. Hartford Foundation, and The Atlantic Philanthropies), and VA Center for the Study of Healthcare Innovation, Implementation & Policy (CSHIIP) Locally Initiated Project (65–155). Ron D. Hays was supported by a grant from NCI (No. 1U2-CCA186878-01). Joseph M Dzierzewski was supported by a grant from the National Institute On Aging (K23AG049955). Preparation of this publication was also supported by the Office of Academic Affiliations, Department of Veterans Affairs. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or Department of Veterans Affairs.
DISCLOSURE STATEMENT
None declared.
ACKNOWLEDGMENTS
We thank Michelle Zeidler, MD, MS and Silverio Santiago, MD from the VA Greater Los Angeles Healthcare System Sleep Center for their support of our study, and Sergio Martinez and Simone Vukelich for their help abstracting data and preparing the survey mailings. We also thank the UCLA Pulmonary/Sleep Clinic administrative staff for their support of our project.
REFERENCES
- 1. Epstein LJ, Kristo D, Strollo PJ, Jr, et al. ; Adult Obstructive Sleep Apnea Task Force of the American Academy of Sleep Medicine Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med. 2009; 5(3): 263–276. [PMC free article] [PubMed] [Google Scholar]
- 2. Collard P, Pieters T, Aubert G, Delguste P, Rodenstein DO. Compliance with nasal CPAP in obstructive sleep apnea patients. Sleep Med Rev. 1997; 1(1): 33–44. [DOI] [PubMed] [Google Scholar]
- 3. Sawyer AM, Gooneratne NS, Marcus CL, Ofer D, Richards KC, Weaver TE. A systematic review of CPAP adherence across age groups: clinical and empiric insights for developing CPAP adherence interventions. Sleep Med Rev. 2011; 15(6): 343–356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Fung CH, Martin JL, Igodan U, Jouldjian S, Alessi C. The association between difficulty using positive airway pressure equipment and adherence to therapy: a pilot study. Sleep Breath. 2013; 17(2): 853–859. [DOI] [PubMed] [Google Scholar]
- 5. Fung CH, Igodan U, Alessi C, et al. Human factors/usability barriers to home medical devices among individuals with disabling conditions: in-depth interviews with positive airway pressure device users. Disabil Health J. 2015; 8(1): 86–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Fung CH, Martin JL, Hays RD, et al. Development of the Usability of Sleep Apnea Equipment-Positive Airway Pressure (USE-PAP) questionnaire. Sleep Med. 2015; 16(5): 645–651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nielsen J. Usability 101: introduction to usability 2010. http://www.useit.com/alertbox/20030825.html Accessed May 22, 2012.
- 8. International Organization for Standardization (ISO). Ergonomics of human-system interaction—Part 20: accessibility guidelines for information/communication technology (ICT) equipment and services 1998. Report No.: ISO 9241-11.
- 9. Food and Drug Administration Center for Devices and Radiological Health. Applying human factors and usability engineering to medical devices http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm259760.pdf Accessed September 14, 2016.
- 10. Bianchi MT, Alameddine Y, Mojica J. Apnea burden: efficacy versus effectiveness in patients using positive airway pressure. Sleep Med. 2014; 15(12): 1579–1581. [DOI] [PubMed] [Google Scholar]
- 11. Weaver TE, Grunstein RR. Adherence to continuous positive airway pressure therapy: the challenge to effective treatment. Proc Am Thorac Soc. 2008; 5(2): 173–178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. International Organization for Standardization (ISO). Ergonomic requirements for office work with visual display terminals (VDTs)—Part 10: Dialogue principles 1996. Report No.: ISO 9241-10.
- 13. Hays RD, Spritzer KL, Thompson WW, Cella D. U.S. General population estimate for “excellent” to “poor” self-rated health item. J Gen Intern Med. 2015; 30(10): 1511–1516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ware J, Jr, Kosinski M, Gandek B. SF-36 Health Survey: Manual & Interpretation Guide. Lincoln, RI: Quality Metric Incorporated; 2000. [Google Scholar]
- 15. Lee GK, Wang HH, Liu KQ, Cheung Y, Morisky DE, Wong MC. Determinants of medication adherence to antihypertensive medications among a Chinese population using Morisky Medication Adherence Scale. PLoS One. 2013; 8(4): e62775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. DeSalvo KB, Fan VS, McDonell MB, Fihn SD. Predicting mortality and healthcare utilization with a single question. Health Serv Res. 2005; 40(4): 1234–1246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Stepnowsky CJ, Jr, Marler MR, Ancoli-Israel S. Determinants of nasal CPAP compliance. Sleep Med. 2002; 3(3): 239–247. [DOI] [PubMed] [Google Scholar]
- 18. Chai CL, Pathinathan A, Smith B. Continuous positive airway pressure delivery interfaces for obstructive sleep apnoea. Cochrane Database Syst Rev. 2006; 4: CD005308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nunnally JC. Psychometric Theory. 2nd ed. New York: McGraw-Hill; 1978. [Google Scholar]
- 20. Food and Drug Administration. 522 Postmarket surveillance studies—frequently asked questions (FAQs) [Internet] http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/PostmarketRequirements/PostmarketSurveillance/ucm134497.htm Accessed September 14, 2016.
- 21. Connelly NA, Brown TL, Decker DJ. Factors affecting response rates to natural resource—focused mail surveys: empirical evidence of declining rates over time. Soci Nat Resour. 2003; 16: 541–549. [Google Scholar]
- 22. Dickinson A, Newell A, Smith M, et al. Introducing the Internet to the over-60s: developing an email system for older novice computer users. Interact Comput. 2005; 17: 621–642. [Google Scholar]
- 23. Lam J, Lee M. Digital inclusiveness—longitudinal study of Internet adoption by older adults. J Manag Infor Syst. 2006; 22: 177–206. [Google Scholar]