Sleep medicine provider perceptions and attitudes regarding consumer sleep technology

Abstract

Study Objectives:

This study assessed perceptions and attitudes of sleep medicine providers regarding consumer sleep technology (CST).

Methods:

A convenience sample of n = 176 practicing sleep medicine and behavioral sleep medicine experts was obtained using social media and the American Academy of Sleep Medicine directory. Providers completed a questionnaire that assessed perceptions and attitudes about patient use of CST in the clinical setting.

Results:

The sample included both adult and pediatric psychologists, physicians, and advanced practice providers from a variety of health settings. Providers reported 36% (3%–95%) of patients used CST, and the most common devices seen by providers were wrist-worn devices followed by smartphone apps. The most common perceived patient motivations for frequent use were to measure sleep and self-discovery. Across sleep disorders, clinicians did not endorse frequent CST use; the highest reported use was for assisting patients in the completion of sleep diaries. Overall devices were rated as somewhat accurate and neutral regarding helpfulness. In qualitative responses, providers associated CST use with increased patient engagement but increased orthosomnia and misperceptions about sleep.

Conclusions:

CST is frequently encountered in the sleep medicine clinic, and providers view CST as somewhat accurate but neither helpful nor unhelpful in clinical practice. Although providers viewed these devices as useful to drive patient engagement/awareness and track sleep patterns, providers also viewed them as a contributor to orthosomnia and misperceptions about sleep.

Citation:

Addison C, Grandner MA, Baron KG. Sleep medicine provider perceptions and attitudes regarding consumer sleep technology. J Clin Sleep Med. 2023;19(8):1457–1463.

BRIEF SUMMARY

Current Knowledge/Study Rationale: Consumer sleep technology (CST) continues to gain popularity and is increasingly becoming a part of the sleep medicine encounter. Despite extensive studies of validity/performance, few studies have examined acceptability and use among clinicians. This study sought to examine the perceptions and attitudes of sleep medicine providers toward CST.

Study Impact: Our findings show a generally cautious stance toward CST by sleep medicine providers. Benefits may include engaging patients in conversations about their sleep, but drawbacks include added time in the clinical visit and low perceptions of helpfulness of the data.

INTRODUCTION

The field of sleep medicine has utilized ambulatory wrist-worn accelerometry devices (actigraphy) since the 1970s to better understand sleep and circadian rhythms.^1,2 Since this innovation, advances in technology coupled with the public embrace of self-quantification have led to a burgeoning of consumer sleep technology (CST) that tracks various health indicators.³ There now exists a surfeit of off-the-shelf, consumer-facing multisensor arrays and software applications with claimed abilities to gather behavioral, cardiopulmonary, neurologic, and metabolic markers of sleep.^4,5 Approximately 85% of Americans own a smartphone capable tracking various health metrics, and one in five use some form of wearable health tracker.^6,7 Global spending on these technologies, including wearables, nearables, and smartphone applications, surpassed $82 billion in 2021 and is projected to grow to $94 billion in 2022.⁸ Sleep is one of the most interesting variables for consumers to monitor, and a recent consumer service estimated that 37% of Americans are using some form of sleep tracking technology.⁹ As a sign of this changing environment, the American Academy of Sleep Medicine recently announced a partnership with Google where the Google Nest Hub device will serve to detect sleep-related biomarkers and direct the user to educational resources.¹⁰

Preliminary literature suggests that these technologies may have utility in monitoring disease and effecting therapy in chronic health conditions such as diabetes mellitus, cardiovascular disease, and chronic obstructive pulmonary disease¹¹; yet the role for consumer technology in clinical sleep medicine is relatively unknown. The availability and affordability of these technologies coupled with their exponentially increasing sets of generated data¹² have the potential to better characterize patients’ sleep and improve health outcomes. Recent studies have focused on the validation/performance of consumer-targeted devices compared to commonly used research grade wrist-worn actigraphy devices used to estimate sleep.^13,14 Progress has been made to standardize evaluation of CST performance against current gold-standard diagnostic tools and interventions,^15,16 and the American Academy of Sleep Medicine has provided a position statement detailing the advantages and disadvantages of CST and guidance for integration into the clinical setting.¹⁷ However, high-quality evidence validating other claimed CST capabilities and adherence, outcomes, and adverse events with CST-mediated interventions across different sleep disorders remains lacking. Furthermore, despite positive patient attitudes toward these technologies and a willingness to both collect and share personal data, the transition from consumer tracker to clinical tool has been slow as it remains unclear which metrics are valuable for clinicians in disease diagnosis/rule-out, monitoring, and management. Ever-increasing patient adoption of different CST may place additional burden on providers to review, interpret, and incorporate generated data into an already busy clinical encounter.^13–17

In this evolving landscape, little is known about how providers perceive these technologies, as investigations regarding provider attitudes surrounding the performance and utility of consumer-targeted sleep technology have been few. As CST becomes ubiquitous and further evaluation of their marketed claims is pursued, these perceptions and attitudes will likely steer the use of CST in the clinical setting moving forward. The purpose of this study is to explore sleep medicine providers’ current perceptions and attitudes surrounding this increasingly available consumer technology.

METHODS

Participants and procedure

We utilized multiple sources for recruitment. In addition to social media and emails from the Society for Behavioral Sleep Medicine, we selected approximately 200 names at random from the American Academy of Sleep Medicine directory and sent emails to those participants with a link to complete the survey. Inclusion criteria included self-reported profession in a clinical sleep setting and willing to complete the survey. We did not have any exclusion criteria. The study was approved and considered exempt by the University of Utah Institutional Review Board (IRB_00151358). The survey was administered via REDCap. Participants first viewed a study cover letter, explaining the purpose of the study, and deidentified data collection, and then if they agreed to complete the survey, they were taken to the survey items.

Survey items

We developed survey items by consensus among the individuals who designed the survey (K.B. and M.G.). The response set is listed in the Methods section and described for each item listed next. We assessed practice characteristics of our sample including practice population, setting, discipline, hours of practice per week and years of practice. Providers were asked the following questions about their own and their patients’ use of CST:

1. “What percentage of your sleep patients use CST (0%–100%)?”

2. “What kinds of CST are your patients using?” Response options included smartphone apps, wrist-worn devices, rings, smart mattresses, contactless sensors, and electroencephalogram headbands. Items were rated 0 (never), 1 (rarely), 2 (sometimes), 3 (often), and 4 (always).

3. “To what degree do you think your patients are using CST for the following reasons?” Reasons included: to measure their sleep, quantified self/self-discovery, to receive sleep-related interventions, and patient is unsure why they are using it. Items were rated 0 (never), 1 (rarely), 2 (sometimes), 3 (often), and 4 (always).

4. “What percentage of new sleep patients ask you to review their CST data (0%–100%)?”

5. “How accurate do you think CST is?” Items were rated 1 (not at all accurate), 2 (somewhat accurate), 3 (moderately accurate), and 4 (very accurate).

6. “Overall, how helpful is CST in your practice?” Items were rated 1 (definitely unhelpful), 2 (somewhat unhelpful), 3 (neutral), 4 (somewhat helpful), and 5 (definitely helpful).

7. “How often do you use CST in the following aspects of treatment?” Items included to diagnose insomnia, to diagnose sleep apnea, to diagnose circadian rhythm disorders, as a substitute for a sleep diary, to help patients complete sleep diaries, to assess paradoxical insomnia, to assess sleep fragmentation, to assess treatment efficacy, to assess behavioral treatment compliance. Items were rated 0 (never), 1 (rarely), 2 (sometimes), 3 (often), and 4 (always).

8. “How often do you use CST?” Items were rated 0 (never), 1 (rarely), 2 (sometimes), 3 (often), and 4 (always).

In addition to these items, two open-ended questions were asked:

1. “How has CST been helpful in your practice?”

2. “How has CST negatively affected your practice?”

Data analysis

Quantitative data were analyzed using SPSS (ver. 24) using descriptive analysis of the survey items. We evaluated the overall means and categorical responses for each of the items and tested differences in perceptions of accuracy and helpfulness between those who used CST most or all of the last year vs those who never used CST in the last year using the Mann-Whitney U test. Qualitative responses were analyzed using thematic content analysis.¹⁸ Responses were reviewed and discussed by two authors (C.A. and K.B.). The initial codes were developed by both authors and all responses coded by C.A. Codes were reviewed and discussed by both authors and revised based on consensus.

RESULTS

Participant characteristics

The sample of 176 providers was primarily composed of psychologists (47%) and physicians (37%) treating adult (77%), pediatric (13%), and mixed (10%) populations in a variety of health settings (Table 1). The average duration of sleep practice was 10.6 (95% confidence interval [CI]: 9.4, 11.8) years with an average of 18.3 (95% CI: 16.5, 20.1) patient contact hours per week. In the year preceding this survey, 57.4% of providers never used CST while only 11.4% endorsed using CST most or all of the past year.

Table 1.

Participant characteristics.

Discipline, n (%)
Doctoral level (PsyD, PhD)	82 (46.6)
Physician	66 (37.5)
APRN	8 (4.6)
Masters	7 (4.0)
PA	6 (3.4)
Other	3 (1.7)
RPSGT	3 (1.7)
Doctoral level (PsyD, PhD) student	1 (0.6)
Patient population, n (%)
Adult	136 (77.3)
Pediatric	23 (13.0)
Adult and pediatric	17 (9.7)
Practice setting, n (%)*
Academic medical center	89 (50.6)
Private practice	42 (23.9)
VA	17 (9.7)
Unaffiliated health system	10 (5.7)
Various	15 (8.5)
Counseling center	2 (1.1)
Public hospital (NHS)	1 (0.6)
Patient contact (hours/w), mean [95% CI]	18.3 [16.5, 20.1]
Sleep practice duration (years), mean [95% CI]	10.6 [9.4, 11.8]
Self CST use, n (%)
Never	101 (57.4)
A few days	35 (19.9)
More than 1 month	14 (8.0)
More than 6 months	6 (3.4)
Most/all of the past year	20 (11.4)

Demographic and occupational characteristics of respondents (n = 176). *12 participants reported more than 1 setting; these were characterized by the first listed.

APRN = advance practice registered nurse, CI = confidence interval, CST = consumer sleep technology, NHS = National Health Service, PA = physician’s assistant, RPSGT = registered polysomnographic technologist.

CST use

Providers estimated 36% (range: 3%–95%) of their patients used CST and 23% (range: 0%–94%) of new patients asked to have CST data reviewed by the provider (Table 2). The devices with most frequent patient use seen by providers were wrist-worn devices (often used) followed by smartphone apps (sometimes used). Providers reported their perceptions that measuring sleep (often used) and self-discovery/quantification (often used) were the most common patient motivations for frequent use. CST was not frequently used by providers in the treatment of sleep disorders with highest reported usage in assisting in the completion of patient sleep logs (rarely used).

Table 2.

CST prevalence and motivations for use.

What percentage of your sleep patients use CST?	35.8 ± 3.3
What percentage of new sleep patients ask you to review their CST data?	23.2 ± 3.6
What types of CST are your patients using?*
Smartphone apps	1.5 [1.4, 1.6]
Wrist-worn sleep trackers	2.5 [2.4, 2.6]
Rings/nonwrist wearables	0.5 [0.4, 0.6]
Mattress sensors	0.4 [0.3, 0.5]
Contactless sensors (eg, bedside trackers)	0.5 [0.4, 0.6]
Smart beds	0.4 [0.3, 0.5]
Consumer EEG headband	0.2 [0.1, 0.3]
To what degree do you think your patients are using CST for the following reasons?*
To measure their sleep	2.9 [2.8, 3.0]
To gain insight about their sleep patterns, self-discovery/quantified self	2.7 [2.6, 2.8]
To receive sleep related interventions from the device	1.3 [1.2, 1.4]
Patient is unsure why they are using it	1.3 [1.1, 1.5]
How often do you use CST in the following aspects of treatment?*
Diagnose insomnia	0.5 [0.4, 0.6]
Diagnose sleep apnea	0.3 [0.2, 0.4]
Diagnose circadian rhythm sleep disorders	0.8 [0.6, 1.0]
Substitute for sleep diary	1.0 [0.8, 1.2]
Help patients complete sleep diaries	1.1 [0.9, 1.3]
Assess paradoxical insomnia	0.8 [0.7, 0.9]
Assess sleep fragmentation	0.9 [0.7, 1.1]
Assess treatment efficacy	0.8 [0.7, 0.9]
Assess behavioral treatment compliance	0.8 [0.6, 1.0]

Values reported are mean ± SD or mean [95% confidence interval]. *Items were rated 0 (never), 1 (rarely), 2 (sometimes), 3 (often), and 4 (always).

EEG = electroencephalogram.

Provider ratings of CST use

The majority of providers rated CST as “somewhat accurate” (2.1; 95% CI: 2.0, 2.2) (Figure 1) and “neutral” with regards to helpfulness (2.7; 95% CI, 2.5, 2.9) (Figure 2). Analyses showed providers using CST most/all of the past year (n = 20) found CST more accurate (U = 621.5, P = .001) and helpful (U = 646.5, P = .01) in their practice setting than providers who never used CST in the last year (n = 101) (Figure 3 and Figure 4).

Figure 1. Provider perceptions of consumer sleep technology accuracy.

Figure 2. Provider perceptions of helpfulness.

Figure 3. Provider perceptions of accuracy relational to self-use of consumer sleep technology.

U = 621.5, P = .001.

Figure 4. Provider perceptions of helpfulness relational to self consumer sleep technology use.

U = 646.5, P = .01.

Qualitative responses

A total of 53 providers provided 135 responses to the open-ended responses. Table 3 lists the common positive and negative themes, with representative quotes. Positive themes (49 responses) included: CST as an opportunity for engagement and education, utility of CST in tracking sleep patterns, and usefulness of CST in treatment and monitoring (Table 3). Negative themes (85 responses) included: CST as a driver of sleep-related anxiety and misperceptions about sleep, concerns about CST reliability and validity, and disrupted relationships and care delivery due to CST (Table 3).

Table 3.

Open-ended responses.

Provider-Perceived Benefits of CST Use
Engagement and education	“A patient used his CST to identify new onset arrhythmia which was later confirmed to be a cause of his relentless fatigue, which had previously been attributed to sleep disordered breathing. They provide me information about how engaged a patient is with his or her health and sleep.”
	“[CST] helps patients better appreciate variability in their night-to-night sleep and the impact their environment has on their sleep.”
Tracking sleep patterns	“[I use] CST to help patients self-identify irregular sleep hygiene, especially among those who won’t complete a diary. It’s reasonable to identify behavioral patterns.”
	“[CST is] helpful in treating patients with intellectual disabilities or cognitive difficulties that are unable to track their own sleep.”
Effecting and monitoring therapy	“[Actigraphy] is teens’ favorite part of the study and seems to be very motivating to them!”
	“Occasionally I have a patient who experiences subjective improvement and is reassured by the CST data that also shows improved sleep in response to treatment.”
Provider-Perceived Harms of CST Use
Sleep-related anxiety and misperception	“The use of CST in families with a child who has a behavioral sleep problem also often increases the focus on sleep, which in turn promotes negative thoughts about sleep and worsens the sleep problem in most cases.”
	“Sometimes [CST] keeps patients actively monitoring, focusing and evaluating on their sleep too much. One client’s precipitating event for insomnia was getting a FitBit for Christmas and becoming worried she wasn’t getting enough ‘deep sleep.’”
Reliability and validity concerns	“The ‘deep vs light sleep’ designation is highly problematic and I wish it would be removed. It doesn’t correlate/distinguish known sleep stages/physiology. […] I think it distracts patients from the right questions.”
	“I think that the data is rarely valid and most often patients will say they need help with sleep because the CST says they don’t sleep well, but when asked, they feel fine and never considered that they had a sleep problem until the CST ‘said so.’ I cringe every time a patient brings up these devices.”
Disrupted relationships and care delivery	“Too much focus on the data ALWAYS slows the treatment process. Patients often over-rely on tracking and can get frustrated if treatment doesn’t meet expectations as indicated via [CST].”
	“I think too many people think wrist and bed sensor devices are actually accurate in measuring sleep stages, so they don’t think they need to do the CBT-I training.”

CST = consumer sleep technology.

DISCUSSION

With the rise of the digital health revolution and the self-quantification movement, consumer technologies provide great potential to collect longitudinal data and enhance patient care; however, there is not agreement in the field on utilization of CST in clinical practice and what benefits and/or harms may result from their use. Our findings of frequent CST use among patients in sleep clinic settings are consistent with the trends reported in consumer polling.⁹ For example, providers estimated that approximately one-third of patients were using some type of device and approximately one in five patients requested the provider review their tracker data in the initial consultation. Results also demonstrated that clinicians viewed the data as somewhat accurate and neutral regarding helpfulness in their practice but also identified multiple concerns with these devices.

Of note, many open-ended responses raised concern about the amount of time dedicated to reviewing and discussing data from these devices in the clinical encounter. Although previous publications have suggested that utilizing patient-generated smartwatch data across various specialties may lower burden,¹⁹ clinicians in our survey reported otherwise. This may be further exacerbated by the fact that sleep practices are increasingly busy and there are currently no remote patient monitoring reimbursement codes attributable to incorporation of patient-generated data from consumer wellness devices into the clinical encounter.^20,21

Our combined survey responses and qualitative data indicate that sleep providers, in general, have a cautious approach with regard to device accuracy and helpfulness in the clinic. Although our study did not capture perceptions of accuracy of individual device categories or metrics, previous studies have reported fairly accurate CST performance in estimating total sleep time with wearable devices²² but lower accuracy in populations with sleep disorders.^23–25 Respondents’ views of CST accuracy likely extend beyond total sleep time, and these views may be more directed at device measures such as oximetry and sleep stage differentiation. Interestingly, a large number of responses revealed concerns about the accuracy and validity of sleep stage data presented by these devices, which agrees with multiple validation studies that have demonstrated moderate to poor accuracy of consumer technologies in discriminating sleep stages.²² Across devices, sleep staging is one of the more prominent variables reported to patients; as misclassified sleep stages are most often called light sleep,²² it is likely this leads to increased patient concern. In addition, patients may not understand that light sleep (non-rapid eye movement stages 1 and 2) is expected to comprise half or more of total sleep time among adults.²⁶ Thus, these concerns over light sleep periods may lead to sleep-related anxiety and orthosomnia accompanying CST use that is seen in prior studies.²⁶ The time required of providers to address these concerns likely compounds frustration with CST as sleep stage distribution is often unrelated to the initial diagnosis and not something that is readily addressed with behavioral or pharmacological interventions.²⁷

Providers did report that CST provided an opportunity for patient education and assessment of patient engagement. Helping patients complete sleep diaries was the most frequent reason responding providers used CST, and a number of qualitative responses commented on the adjunctive role CST plays in tracking sleep, especially in populations where self-reporting is challenging. Adherence to sleep diary completion is often difficult for patients^28,29 and utilizing CST may lower the burden of filling out the questionnaire at the instructed time. An important finding of this study was that although the total number of providers using CST most or all of the past year was low, providers with personal use perceived them to be more accurate and helpful. It is unclear if personal use has an effect on patient adoption or incorporation into the clinical encounter, and it is unknown if usage in sleep medicine providers will increase as CST prevalence increases in the public. Thus, providers should be mindful of emergent bias with personal use or disuse.

Limitations of this study include self-reported data, and retrospective estimation or recall of, for example, percentage of patients who use CST may be subject to bias. Use of a convenience sample may have captured a subgroup more facile with technology, and responses may not be generalizable to the greater population of sleep medicine providers. Additionally, demographic information beyond practice characteristics was not obtained, such as socioeconomic characteristics of the respondents or setting location (eg, urban or rural areas). There is emerging evidence that acceptance of consumer wellness technology in the health care setting may be culturally rooted³⁰ and this sample population may not reflect global trends. Finally, responses were collected prior to the COVID-19 pandemic, and therefore further research is needed to determine the impact of increased telehealth usage during this time period on attitudes toward CST.

In summary, our study showed that providers in the field of sleep medicine found CST somewhat accurate but had neutral views of their helpfulness in clinical practice. Although providers found CST useful to drive patient engagement and track sleep patterns, they are frequently viewed as driver of orthosomnia and misperceptions about sleep. Future research on patient-reported motivations, perceptions, and attitudes surrounding these technologies is needed to understand their evolving role in the clinic as layman perceptions of sleep often differ from that of the provider.³¹ Strategies for optimally incorporating CST-generated data in clinical visits warrant further investigation. Understanding which device characteristics influence clinician attitudes will be paramount to understand the interplay between provider attitudes and resultant patient adoption of CST. Currently companies that develop and manufacture these technologies have little incentive to develop devices that appeal to sleep medicine professionals. Recent literature suggests CST is currently most commonly deployed to direct just-in-time interventions in insomnia and short sleep.³² The finding that sleep providers use these technologies most often as a complementary adjunct to traditional sleep logs may provide an area of mutual interest between industry, researchers, and clinicians to collaboratively develop more relevant devices in the future. These technologies offer tremendous promise in their widespread use and the large datasets they are capable of generating; provider voices must be leveraged if CST is to be a viable tool in the alleviation of the health burden of sleep disorders and their downstream complications.

DISCLOSURE STATEMENT

All authors have seen and approved this manuscript. Institution where work was performed: University of Utah. Conrad Addison, MD, reports no conflicts of interest. Michael Grandner reports that in the last 12 months he has received grant funding from Jazz Pharmaceuticals and CeraZ Technology and has received consulting fees from Fitbit, Natrol, Smartypants Vitamins, and Idorsia. Kelly Baron reports research support by the National Institutes of Health and an unrestricted research gift from Google, and she is a consultant to the National Sleep Foundation. This project was supported by internal funding by the Department of Family and Preventive Medicine and in part by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR002538. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

ABBREVIATION

CST,

consumer sleep technology