Here, we present a protocol to improve the quality of data from home sleep testing by providing a method to enhance instructions through a structured participant visit. This protocol includes the implementation of a step-by-step educational manual with photos to ensure proper placement of equipment.
The gold standard for sleep monitoring is attended in-lab polysomnography; however, this method may be cost-prohibitive and inconvenient for patients and research participants. Home sleep testing has gained momentum in the field of sleep medicine due to its convenience and lower cost, as well as being more naturalistic. The accuracy and quality of home sleep testing, however, may be variable because studies are not monitored by sleep technologists. There has been some success in improving the accuracy of home sleep studies by having trained sleep technicians assist participants inside their homes with putting on the devices, but this can be intrusive and time-consuming for those involved. In this protocol, participants undergo at-home sleep monitoring with multiple devices: 1) a single-channel EEG device; 2) a home sleep test for sleep-disordered breathing and periodic limb movements; 3) actigraphy; and 4) sleep logs. A major challenge of this study is obtaining high-quality sleep monitoring data on the first attempt in order to minimize participant burden. This protocol describes the implementation of educational manuals with step-by-step instructions and photos. The goal is to improve the quality of home sleep testing.
The relationship between sleep and Alzheimer's disease (AD) is a growing area of research with sleep disturbances hypothesized to have a role in both AD pathogenesis and as a biomarker for AD pathology1,2. In order to study the relationship between sleep and AD biomarkers, cognitively normal or very mildly impaired participants aged ≥65 years old are recruited from a longitudinal study of aging at the Knight Alzheimer's Disease Research Center (ADRC) at Washington University School of Medicine. Although this study was focused on AD research, the methods presented here have broad applicability to home sleep testing in older adults. Attended in-lab polysomnography is the gold standard for sleep monitoring3, but such monitoring can be cost-prohibitive and inconvenient for participants. An alternative is home sleep testing. The accuracy of home sleep studies may be improved by having trained sleep technicians assist participants inside their homes with device placement, but this can also be intrusive and time-consuming4. Therefore, this protocol was developed to instruct the participants how to set up sleep monitoring devices at home and still collect reliable data.
Participants were asked to wear a home sleep test (HST) for measuring sleep-disordered breathing (e.g., obstructive sleep apnea) and periodic leg movements. Following HST recording, participants wore a single-channel EEG device for 6 nights to monitor brain waves for EEG-based sleep staging. Previous findings indicate that the single-channel EEG device has a high level of agreement with polysomnography for multiple sleep parameters5. Throughout the single-channel EEG and HST monitoring period, participants completed a sleep log and were asked to wear actigraphy on their nondominant wrist for the entire duration of the sleep study to track activity. Studies were defined as acceptable if there were at least 2 nights recorded by the single-channel EEG device with <10% artifact and at least 1 night recorded on the HST with ≥4 hours of scorable data. Initially, the failure rate due to poor data quality was ~40−50%. Repeat monitoring for participants with poor data quality was deemed too burdensome, therefore, this protocol was developed.
Previous work found that many elderly persons have difficulty adjusting to technological interventions6,7,8,9. This impacts numerous fields from geriatrics to education and is particularly relevant to medical studies in which older adults must use or interact with unfamiliar technologies. In order to reduce in-home study failure rates, education manuals were created that provided pictures and step-by-step directions to set up the HST and single-channel EEG devices. The education manuals were derived from the device user manuals10,11. Additionally, a 24-hour helpline was provided to participants, so they could reach a member of the study team at any time with any questions or concerns.
To analyze the impact of this protocol, a retrospective review was conducted on the success and failure rates for acceptable data quality from the at-home sleep monitoring before and after implementation of the education manuals. The data sources were successful recordings and participant calls to research study staff with questions. Participants were asked to come into the sleep center to learn about the sleep monitoring equipment. During the 2-hour visit, a study team member educated each participant about the equipment using the manuals, which provided step-by-step instructions for device usage. After reviewing the manuals in the office and being guided through the application and use of the devices, participants were given the opportunity to independently practice applying the home sleep monitoring devices using the manuals as a guide. Feedback was provided to participants during the visit and they were given the opportunity to ask questions in addition to reapplying the equipment as needed. Participants then took the equipment home, put the ambulatory equipment on themselves at night, and were encouraged to call a study team member at any time, day or night, for assistance troubleshooting any questions or problems.
All single-channel EEG studies were scored manually by registered sleep technologists who were validated, gold-standard scorers using modified American Academy of Sleep Medicine (AASM) scoring criteria5. HST recordings included airflow measured by nasal pressure transducer and thermistor, respiratory effort measured by thoracic and abdominal respiratory inductance plethysmograph belts, body position, pulse oximetry, and leg electromyogram (EMG) using the optional ExG yoke. Lights off and lights on were determined by the time that each participant pressed the event button on the HST device and/or entry in a sleep log. A registered sleep technologist manually scored the HST studies and then a board-certified sleep physician performed an epoch-by-epoch review of each study.
Following the introduction of this protocol, the failure rate was reduced to 19% and reliable data was obtained. The protocol represents a novel, low-cost, and effective way to increase the success rate of at-home sleep studies. While the HST device has been used in some studies, it is primarily used as a diagnostic tool and not for scientific studies12,13. This protocol also provides a method that allows participants to easily use a Positive Airway Pressure (PAP) machine while they are wearing either the HST or single-channel EEG device. The use of the HST and single-channel EEG devices with education manuals is a particularly useful research tool that could be more widely utilized using the method shown in this protocol.
This protocol was approved by the Washington University Human Research Protection Office.
Instructions were written specific to commercially available devices and their related software for data collection (see Table of Materials).
1. Set up of the sleep monitoring devices prior to the participant visit
2. Participant visit for instruction on wearing the sleep monitoring devices
3. Data Download and Processing
4. Quality assurance procedures
NOTE: All recordings are reviewed by a registered sleep technologist and a board-certified sleep physician. Specific markers of quality assurance are reviewed as follows:
Single-channel EEG
At the start of the study, an acceptable overnight recording with the single-channel EEG device was defined as 1) aligning with the sleep period defined by sleep log and/or actigraphy device, and 2) <10% of the recording unscorable due to movement, myogenic, electrode, or other artifacts. Each participant needed at least 2 nights meeting these criteria. Prior to the implementation of the single-channel EEG manual, 14 participants wore the single-channel EEG device. Of those 14 participants, 42% (6) needed to repeat testing due to inadequate collection of acceptable data as defined above. After implementing the protocol with the instruction manual, only 2 of 15 (13%) participants needed to repeat monitoring with the single-channel EEG device due to poor data quality (Table 1). Example hypnograms of an acceptable good and unacceptable poor single-channel EEG recording are shown (Figure 1).
Home Sleep Test
For the HST, an acceptable recording was defined as 4 or more hours of data scorable for respiratory events and periodic leg movements. Poor data quality is usually due to one or more of the sensors failing to record, such as from artifacts or losing contact. In this study, one common indication of unscorable data was breaks in the SpO2 channel. Prior to using the HST instruction manual, 3 of 7 (42%) participants failed to meet this standard for one HST recording. 16 participants wore the HST over 2 months after implementing the instruction manual with only 3 (19%) participants failing to meet the acceptable "good" recording standard (Table 1). Example hypnograms of an acceptable good and unacceptable poor HST recordings are shown and the SpO2 channel is highlighted (Figure 2).
Table 1. Participants with Acceptable Sleep Studies Pre- and Post-Protocol Implementation. Before implementing this study protocol, fourteen participants wore the single-channel EEG device and seven participants wore the home sleep test (HST). For the single-channel EEG and the HST, 43% of participants had an unacceptable study. After implementation of the protocol, this was reduced to 13% of participants with unacceptable single-channel EEG studies and 19% of participants with unacceptable HSTs. Neither of these differences were significant.
Figure 1. Acceptable and unacceptable hypnograms from 1-night recording on the single-channel EEG device. (A) A single-channel EEG hypnogram showing artifact-free data collection. This figure shows data for one night of single-channel EEG recording on an hour-by-hour basis. The x-axis indicates the number of hours starting from the beginning (lights-off) to the ending (lights-on) of the recording. The y-axis indicates sleep stages. Movement (MT) indicates artifact during the recording from movement, muscle, electrodes, or other sources. Artifact-free data was recorded throughout the study (shown in green) and were scored for different sleep stages. (B) A single-channel EEG hypnogram showing unscorable data due to artifact. This figure shows data for one night of single-channel EEG recording on an hour-by-hour basis with the same x-axis and y-axis as in (A). This data represents a hypnogram in during which the single-channel EEG recorded artifact indicated by MT (shown in red). The sensors on the single-channel EEG were unable to record brain wave activity that were scored for during different sleep stages during these time periods. W: wake; R: rapid eye movement sleep; N1: non-rapid eye movement (NREM) stage 1; N2: NREM stage 2; N3: NREM stage 3. Please click here to view a larger version of this figure.
Figure 2. Acceptable and unacceptable hypnograms from 1-night recording on the home sleep test. (A) A home sleep test (HST) hypnogram showing artifact-free data collection over one night. The x-axis indicates the time in hours that the recording was taken, and the y-axis indicates heart rate, oxygen saturation, sleep-disordered breathing, and body position. Note, there are no breaks in the SpO2 channel (shown in green) indicative of data loss, artifact, or otherwise unscorable data. (B) A HST hypnogram showing an unacceptable "poor" recording over one night with a large proportion of artifact during the study. The x-axis and y-axis are the same as in (A). There are breaks or missing data in the SpO2 channel (shown in red) indicative of unscorable data from artifact or other data loss. HR: heart rate; BPM: beats per minute; SpO2: percent oxygen saturation; CA: central apnea; OA: obstructive apnea; MA: mixed apnea; HYPO: hypopnea; S: seconds; Pos: position; B: back; R: right; L: left. Please click here to view a larger version of this figure.
Supplemental File. Selected pages from the Home Sleep Test education manual showing the step-by-step set-up instructions are provided as an example. Please click here to download this file.
This protocol is the novel application of instructional manuals in conjunction with participant education for in-home ambulatory sleep monitoring. Based on the results, the implementation of the protocol with instruction manuals improves the feasibility of in-home sleep testing in older adults. In-lab polysomnography remains the gold standard for sleep monitoring but may be limited by cost as well as disrupted sleep due to the new environment (i.e., the "first night" effect19). This protocol has shown that it is feasible to obtain high quality sleep monitoring at home in an elderly population with a single-channel EEG device for sleep staging and a HST to screen for sleep disorders such as obstructive sleep apnea. Additionally, this protocol provides a method that enables participants to use a PAP machine in conjunction with both the single-channel EEG and HST system.
Participants were educated on the use of all the sleep monitoring devices during an in-person visit with a study team member. Additionally, written instructions were provided for the actigraphy device and sleep log and step-by-step instructional manuals are provided for the single- channel EEG and HST devices. The detailed instructions were tailored to the devices selected (see Table of Materials) and may vary with alternative equipment. The HST instructions were broken down in step-by-step instructions and pictures were inserted to provide a reference for proper sensor placement (Supplemental File).
During the study visit, participants were given the opportunity to apply all study equipment to get feedback on sensor placement and ensure all questions were answered prior to departure. Participants were provided a 24-hour helpline to call with any questions or concerns should they arise during testing. Study team members noted that utilization of the helpline appeared to vary by age, cognitive ability, and participant support system (e.g., if a spouse or partner attended the visit and was available to assist with the devices). Finally, participants were asked to apply the devices at home without a study team member being present.
A strength of the protocol are the novel solutions developed to solve specific problems with home sleep monitoring in older adults. One example is that the HST device has multiple sensors that need to be applied by the participant and this can seem overwhelming initially. Breaking the instructions down into step-by-step directions with pictures and eliminating steps when possible helps to simplify the procedure. For instance, sending participants home with the holster attached to the chest belt eliminates the risk that the participant may open the battery compartment and dislodge the memory card. The protocol is designed to educate the participant about the HST at the office visit by following the steps of the manual while placing the equipment on the participant's body. Afterwards, the HST is removed and participants are given the opportunity to put on the HST unassisted while using the manual. The practice session helps to clarify any steps that the participant does not understand. In addition, the study team member provides feedback about the participant's HST self-hook up afterwards.
All participants were provided a 24-hour contact number to call with any questions. They were encouraged to call as soon as an issue was identified to avoid frustration with the equipment. Participants called for a variety of reasons, but it was common for participants to call for assistance troubleshooting the equipment prior to bedtime. Some participants needed extra guidance with sensor application and others called when a light remained flashing on the HST device after all of the sensors were on the body. The study team member on call would ask questions and walk the participant through any steps needed to resolve the issue. At times, the answers were in the manual and at other times additional information was needed to troubleshoot the problem. If participants had difficulty applying sensors at the office visit, the study team offered to talk through the entire hookup process over the phone at night. The participant was asked to call when ready to apply the sensors and the study team member would talk through each step of the manual as the participant placed each wire on the body.
Participants were asked to call in the morning if the HST device did not display a complete study. Any issues that arose during the night were discussed and participants were asked if they were willing to wear the device a second night. Troubleshooting measures were reviewed as needed and the participants were encouraged to call with any questions during the hookup process or during the night. It was also common for participants to call in the morning with questions about the button on the actigraphy device. The button does not make a noise when pressed and participants had concerns about inadvertently powering the device off. Participants were reassured that the equipment was recording, and they were following the study protocol.
Another novel aspect is the success in having participants who use PAP to treat obstructive sleep apnea wear both the HST and the single-channel EEG device. This requires an adapter to measure air flow on the HST. For participants using a PAP mask with a forehead attachment, a loaner mask was provided if the participant agreed. Using this protocol allowed for assessment of participants using PAP under their normal sleep conditions both in terms of sleep staging with the single-channel EEG and obstructive sleep apnea severity with the HST.
A major limitation of the protocol is application in the clinical setting. Participant visits to review the equipment, instruction manuals, and practice runs setting up the devices take ~2 hours. In addition, participants are able to call a sleep technologist at night to troubleshoot problems they may experience. It would be challenging to implement this protocol in a clinical setting due to these constraints.
This protocol provides a method to successful use multiple in-home sleep monitoring devices, including a single-channel EEG and HST in older adults. To date, >300 research participants have completed this protocol. While these devices are validated, cost-effective tools, they are not a replacement for attended polysomnography. This protocol does demonstrate that a simple and low-tech introduction of instruction manuals and education can improve the success of in-home sleep studies. This same protocol can be adapted for use with other types of technology and demonstrates the importance of communication and availability for encouraging use of technological research tools in studies, particularly those involving the elderly. This also answers the increasing need for diagnostic tools to address sleep disorders and improve sleep in the general population20. However, this study only considered elderly individuals and was limited to three types of sleep monitoring devices. Further protocol revisions may be needed with different in-home ambulatory equipment are needed.
The authors have nothing to disclose.
We thank the participants for their time.
Funding: This study was supported by the following grants from the National Institutes of Health: P01 AG03991; P01 AG026276; P50 AG05681; K76 AG054863; UL1 TR000448; KL2 TR000450. The Ellison Medical Foundation and a Physician Scientist Training Award from the American Sleep Medicine Foundation also supported this study. The funding sources had no role in the study design, data collection, management, analysis, interpretation of the data, or manuscript preparation.
Actigraph | Phillips Respironics | 1048090 | Actiwatch 2 |
Actiware site license | Phillips Respironics | 1114828 | Actigraphy software |
Actigraphy docking station | Phillips Respironics | 1048092 | Actiwatch charger |
Home Sleep Test (HST) | Phillips Respironics | 1043941 | Alice PDX device |
Sleepware G3 software | Phillips Respironics | 1082462 | Alice PDX software |
CPAP titration kit | Pro-tech | P1391 | Connects PAP to HST |
3-foot SpO2 Extension Cable (Nonin) | Phillips Respironics | 927-3 | Oximeter cable for HST |
SpO2 Sensor (Nonin) | Phillips Respironics | 936 | Oximeter finger clip for HST |
SD Card Reader | Phillips Respironics | 1047300 | card reader for HST |
EXG yoke | Phillips Respironics | 1040808 | HST cable for leg leads |
Dual Snap Leg Lead Wires | MVAP | TLC0048 | leg leads for HST |
Small Foam Electrodes | MVAP | 5000ZT | electrodes for leg leads |
Thermistor Airflow Sensor | Pro-tech | P1388 | HST thermoistor |
Oral/Nasal Air flow Pressure Cannula | Saltar Labs | 5760-7 | Pressure Transducer |
Zrip DuraBelt kit | Phillips Respironics | P1837 | HST Thoracic and Abdominal belt kit |
Single-channel EEG | Advanced Brain Monitoring | SP40-1001 | Sleep Profiler device |
SP Sensor EEG kit | Advanced Brain Monitoring | SP40-4225.1 | Sleep Profiler electrodes |
CPAP masks | Resmed | 62925, 63504, 63506, 63507, 63445, 63446, 63447 |