American Thoracic Society Documents
An Official ATS/AASM/ACCP/ERS Workshop Report:
Research Priorities in Ambulatory Management of
Adults with Obstructive Sleep Apnea
Samuel T. Kuna, M. Safwan Badr, R. John Kimoff, Clete Kushida, Teofilo Lee-Chiong, Patrick Levy,
Walter T. McNicholas, Patrick J. Strollo, Jr., on behalf of the ATS/AASM/ACCP/ERS Committee
on Ambulatory Management of Adults with OSA
THIS OFFICIAL WORKSHOP REPORT OF THE AMERICAN THORACIC SOCIETY (ATS), THE AMERICAN ACADEMY OF SLEEP MEDICINE (AASM),
THE AMERICAN COLLEGE OF CHEST PHYSICIANS (ACCP), AND THE EUROPEAN RESPIRATORY SOCIETY (ERS) WAS APPROVED BY THE ATS
BOARD OF DIRECTORS, SEPTEMBER 2010, THE AASM BOARD OF DIRECTORS, JULY 2009, THE ACCP BOARD OF REAGENTS, OCTOBER 2009,
AND THE ERS EXECUTIVE COMMITTEE, APRIL 2010
III. Background on Portable-Monitor Testing and Significance
of the Workshop
Current Practice Parameters to Diagnose Patients with
OSA and Initiate CPAP Therapy
Limited In-Laboratory Resources Driving Use of
IV. Healthcare Management and Insurance-Industry Panel
Perspective on Emerging Portable-Monitor Technology
V. Workshop Findings and Recommendations
1. Establish Adequate and Appropriate Research Net-
works to Conduct Adequately Powered, Multicenter
Clinical Research Studies
2. Develop Disease Management Models Based on
Successful Clinical and Economic Outcomes
3. Define the Appropriate Patient Population for Am-
4. Standardize the Portable Monitor Devices Used to
Diagnose OSA and Initiate Continuous Positive
5. Establish Appropriate Study Designs That Success-
fully Address Clinically Relevant Questions Re-
garding Ambulatory Management of OSA
6. Ensure the Safety of Ambulatory Monitoring, Espe-
cially Identifying Those Patients Who Are Not Candi-
dates for This Management Approach
7. Ensure Adequate Training and Education of Pro-
viders and Patients Regarding Ambulatory Manage-
ment of OSA
8. Identify Funding Sources to Support the Needed
An international workshop was held to determine the research
priorities for incorporating ambulatory management of adults
with obstructive sleep apnea into healthcare systems. The work-
shop identified the barriers preventing incorporation of portable
monitor testing into clinical management pathways and deter-
mined the research and development needed to address those
barriers. The workshop promoted interaction and collaboration
development and evaluation of portable monitor technology and
was that outcomes-based research studies are needed to demon-
strate the efficacy and cost effectiveness of portable monitor
testing. Closely related to this objective is the need to develop
clinicalsleep researchnetworks capable of performing adequately
powered studies. Recommendations were developed regarding
research study design and methodology that includes the need to
ambulatory management ofobstructivesleepapnea,ensurepatient
safety, and identify sources of research funding. The evidence
resulting from high-quality comparative effectiveness studies that
include cost effectiveness as an outcome will allow decision makers
to develop healthcare policies regarding the clinical application of
portable monitor testing for the ambulatory management of pa-
tients with obstructive sleep apnea.
Keywords: polysomnogram; apnea; hypopnea; comparative effectiveness
research; cost effectiveness
An international workshop was held on October 15–16, 2007 in
Arlington, Virginia, to determine the research priorities for
incorporating ambulatory management of adults with obstruc-
tive sleep apnea (OSA) into healthcare systems. The workshop
was sponsored by the American Thoracic Society, the American
Academy of Sleep Medicine, the American College of Chest
Physicians, and the European Respiratory Society. Objectives
of the workshop were to:
1. Identify the barriers preventing incorporation of portable
monitor testing into clinical management pathways for
1R13HS017402-01 and grants from the American Thoracic Society, the American
College of Chest Physicians, the American Academy of Sleep Medicine, and the
European Respiratory Society.
by the Agency for HealthcareResearch andQualitygrant
Proc Am Thorac Soc
Internet address: www.atsjournals.org
Vol 8. pp 1–16, 2011
the evaluation of patients with suspected OSA and de-
termine the research and development needed to address
2. Develop recommendations regarding research study
designs and methodology that will provide needed infor-
mation regarding the potential clinical application of por-
table monitor testing for patients with OSA;
3. Promote interaction and collaboration of representatives
from government, the medical device industry, healthcare
insurers, professional societies, and researchers in aca-
demic medicine who have interest and expertise in the
development and evaluation of portable monitor technol-
ogy and its clinical application.
Workshop participants identified the following eight inter-
related research priorities for the ambulatory management of
patients with OSA listed in descending order of importance:
1. Establish adequate and appropriate research networks to
conduct adequately powered, multicenter clinical re-
2. Develop disease management models based on prospec-
tive clinical and economic outcome studies that provide
evidence for clinical decision making;
3. Define the appropriate patient population for ambulatory
4. Standardize the portable monitor devices used to diag-
nose OSA and initiate continuous positive airway pres-
sure (CPAP) treatment in terms of their sensors, signal
processing, and data analysis;
5. Establish appropriate study designs that successfully
address clinically relevant questions regarding ambulatory
management of OSA;
6. Ensure the safety of ambulatory monitoring, especially
indentifying those patient who are not candidates for this
7. Ensure adequate training and education of providers and
patients regarding ambulatory management of OSA;
8. Identify funding sources to support the needed research.
Participants felt that the most important research priority is
to conduct adequately powered, high quality research studies to
generate the evidence needed to incorporate ambulatory man-
agement into current practice. To achieve this goal, clinical
sleep research networks need to be established to evaluate
ambulatory management of patients with OSA in large numbers
of subjects across multiple sites. In the United States, existing
practice-based research networks and the NIH Clinical Trans-
lational Research Centers may provide the essential infrastruc-
ture for such initiatives.
The advances made in portable monitor technology far
outweigh our knowledge about their use in clinical testing.
Approval of new portable monitor devices by the U.S. Food
and Drug Administration requires their direct comparison
with existing technologies. These studies should be designed
to take into account the night-to-night variability in the
apnea–hyponea index (AHI) on the in-lab polysomnogram
and differences in equipment and testing environment. These
‘‘head-to-head’’ comparisons, however, do not inform us how
to use the devices in clinical management pathways. Sleep
testing, whether by in-lab polysomnography (PSG) or unat-
tended home portable-monitor recording, is but one component
of the clinical management pathway for patients with OSA.
More prospective comparative effectiveness and economic out-
come studies are needed to provide evidence for clinical
decision making. Comparative effectiveness research studies
need to evaluate portable-monitor testing within clinical man-
agement pathways and demonstrate the effectiveness of this
management against the standard in-laboratory management in
terms of clinical outcomes such as improvements in quality of
life, disease-specific functional outcomes, and cardiovascular
health. Investigators should incorporate economic evaluations in
all such clinical trials using outcomes that are recognized bench-
marks of cost effectiveness.
Anticipating that portable-monitor testing will be of greatest
value in patients with a high pretest likelihood for OSA,
recognition strategies that identify these high-risk patients need
to be developed and validated. Future research studies should
include patients with comorbid conditions, especially chronic
obstructive pulmonary disease (COPD) and chronic heart fail-
ure. Patients with these prevalent conditions have been excluded
from previous studies, limiting the number of potential patients
who may benefit from portable monitor testing.
Despite the technological advances in portable-monitor de-
vices, there remains a lack of standardization in terms of signals
recorded, sensors, signal processing, and data analysis. Large
differences can exist between monitors, even those within the
same classification. Greater standardization is required. Com-
parative effectiveness research using different types of portable
monitors may be one method to identify the optimal signals and
other operating characteristics for ambulatory testing within
specific patient populations.
Part of the design and evaluation of ambulatory management
pathways must include careful consideration of patient safety.
Although portable-monitor testing for OSA can be used as a
primary diagnostic strategy, ambulatory management pathways
should define the decision-making parameters for patients with
negative or failed home testing who may need in-laboratory
polysomnography. Ambulatory management pathways, especially
those using Type 4 monitors, should also be able to identify
patients with Cheyne-Stokes respiration, central sleep apnea,
and complex sleep apnea. These conditions might be identified
by a high AHI reported on auto-continuous positive airway
pressure (CPAP) and CPAP downloads. However, the accuracy
of the AHI and respiratory event information reported by
positive airway pressure units still needs to be validated.
A particular challenge confronting not only future research
in ambulatory monitoring but indeed the entire field of Sleep
Medicine is the relatively small number of clinical investigators
in this specialty. This limitation can only be exacerbated by the
current Accreditation Council for Graduate Medical Education
(ACGME) training requirements, which exclude a substantive
research opportunity during Sleep Medicine fellowship training.
Stakeholder medical specialty organizations and the ACGME
urgently need to negotiate innovative solutions to optimize the
feasibility and appeal of sleep clinician-scientist training path-
ways. As new evidence is generated about ambulatory manage-
ment of OSA, its implementation will require trained healthcare
providers, including sleep specialists, primary care providers,
physician extenders, and allied health professionals. Training
programs need to prepare healthcare providers for that future.
Finally, workshop participants recognized the importance of
identifying funding sources for the needed research. The proposed
research studies regarding large multicenter center comparative
effectiveness will require governmental funding. Other stake-
holders, including private foundations, the insurance industry,
manufacturers, and the Center for Medicare and Medicaid Services
should provide adjunctive resources. The high-quality outcomes-
2 PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 8 2011
based research conducted with this investment will demonstrate
the efficacy and cost effectiveness of portable-monitor testing and
generate the data needed by these funding sources for their
decision making with regard to healthcare policies and directions
for future research and development.
An international workshop to determine the research priorities
for incorporating ambulatory management of adults with obstruc-
tive sleep apnea (OSA) into healthcare systems was held on
October 15–16, 2007 in Arlington, Virginia. The overall goal was
to promote further research on portable-monitor testing that will
generate the high-quality empirical evidence needed to determine
the role of portable monitors in the ambulatory management of
patients with OSA. The objectives of the workshop were to:
1. Identify the barriers preventing incorporation of portable-
monitor testing into clinical management pathways for
the evaluation of patients with suspected OSA and de-
termine the research and development needed to address
2. Develop recommendations regarding research study de-
signs and methodology that will provide needed informa-
tion regarding the potential clinical application of porta-
ble-monitor testing for patients with OSA;
3. Promote interaction and collaboration of representatives
from government, the medical device industry, healthcare
insurers, professional societies, and researchers in aca-
demic medicine who have interest and expertise in the
development and evaluation of portable-monitor technol-
ogy and its clinical application.
In addition, the workshop explored the potential resources
available from the federal government, professional societies,
and medical device industries to fund the research needs identified
by the workshop.
The workshop was sponsored by the American Thoracic
Society (ATS), the American Academy of Sleep Medicine
(AASM), the American College of Chest Physicians (ACCP),
and the European Respiratory Society (ERS). A conference grant
awarded by the Agency for Healthcare Research and Quality
(AHRQ) provided additional financial support. The workshop
did not receive funding from industry. Members of the Steering
Committee were selected by each sponsoring scientific society and
included: M. Safwan Badr, M.D., R. John Kimoff, M.D., and
Samuel T. Kuna, M.D. (Chair), representing the ATS; Teofilo L.
Lee-Chiong, M.D., representing the ACCP; Clete A. Kushida,
M.D., Ph.D. and Patrick J Strollo, Jr., M.D. representing the
AASM; and Patrick Levy, M.D. and Walter T. McNicholas,
M.D. representing the ERS. The invited workshop participants
included researchers in the field of Clinical Sleep Medicine, and
representatives from the American Association of Respiratory
Care, medical device manufacturers, professional societies,
payer organizations, and federal government agencies, includ-
ing representatives of the U.S. Food and Drug Administration,
the AHRQ, the Veterans Health Administration, and the
National Institutes of Health. The names of the workshop
participants are listed at the end of this document.
The workshop consisted of four half-day sessions over the 2-
day meeting period. It was structured to promote highly
interactive discussions with input from all stakeholders. The
initial part of each half-day session began with several short,
invited presentations designed to set the framework for the
session’s particular theme. Participants then broke up into
small groups of seven to eight individuals to discuss preiden-
tified topics of interest. Each group consisted of a heteroge-
neous mix of individuals representing the different stake-
holder organizations. The groups were assigned specific topics
for discussion. A member of the Steering Committee was
assigned to each group to organize and facilitate the discus-
sion. In the last segment of these sessions, participants
reconvened for an open discussion during which each facili-
tator summarized the discussions by their particular group to
share ideas and develop a consensus.
The first session focused on the technological aspects of
portable monitors. Nancy Collop, M.D., from Johns Hopkins
University presented the recommendations of the American
Academy of Sleep Medicine’s Task Force on Portable Monitor
Testing that had just concluded its evaluation (1), and Conor
Heneghan, Ph.D., from the School of Electronic Engineering at
University College Dublin spoke about new applications of
established technology. During the ensuing small-group discus-
sions, the following topics were discussed:
1. Should portable monitors be used to both include and
exclude the diagnosis of OSA, or should they be targeted
to diagnose patients with a high pretest likelihood?
2. What signals are essential for the ‘‘ideal portable monitor’’?
3. What if any standardization is needed of sensors, signal
conditioning, and scoring?
4. Do we need more research studies on unattended auto-
5. What is the best study design to validate the technical
performance of portable monitors versus PSG?
inginfuture research toevaluatethecosteffectivenessand clinical
outcomes of ambulatory management pathways for patients with
OSA. Henry Glick, Ph.D., from the University of Pennsylvania
Trials: an Overview of Design and Analysis,’’ and Sean Tunis,
‘‘Translating Emerging Technology into Clinical Practice.’’ The
following topics were discussed by the small groups:
1. Can we use PSG and portable recording as components of
a disease management model to decide who receives
treatment rather than relying solely on the AHI?
2. If portable monitor technology is used by nonsleep
specialists, what research is needed to guide its applica-
tion outside the sleep center?
3. Do we have adequate clinical prediction algorithms to
assign pretest probability and, if not, how should these be
developed and incorporated into diagnostic strategies?
4. For cost- and outcomes-based comparisons of in-lab versus
ambulatory pathways, what are the critical outcome mea-
5. What are the best research approaches for evaluating
portable monitors in diverse ethnic groups, the elderly
(e.g., Medicare/Medicaid population), and among indi-
American Thoracic Society Documents3
viduals with other cardio-pulmonary disease (COPD,
asthma, heart failure, neuromuscular disease, etc.)?
6. What modifications of the 2003 TriSociety (AASM,
ACCP, and ATS) recommendations of research studies
on portable monitoring for diagnosing OSA are needed to
include guidelines for outcomes-based research studies in
The third session of the workshop focused on identifying the
research opportunities for multicenter trials in home-based
chronic disease management. David Lanier, M.D., from the
Center for Primary Care, Prevention, and Clinical Partnerships
at the Agency for Healthcare Research and Quality spoke about
M.P.H., Ph.D., NIH Deputy Director of the Division of Clinical
Research, National Center for Research Resources, spoke about
‘‘Creating Research Networks within Clinical and Translational
Science Award Centers’’; and Katherine Bent, R.N., Ph.D.,
C.N.S., from the Office of Research and Development in the
Department of Veterans Affairs spoke about ‘‘Cooperative Trials
and Network Opportunities in the Veterans Health Administra-
tion.’’ This was followed by a panel discussion by experts in
healthcare management on emerging portable-monitor technol-
ogy: John Leteria, C.E.O. of Neurocare, Inc., Suresh Ramalin-
gam, M.D., M.B.A., from the University of Pittsburgh, and
Christopher Valerian, D.O., from the University of Medicine
and Dentistry, New Jersey. During the ensuing small-group
discussions, the following topics were discussed:
Iris Obrams, M.D.,
1. Regarding NIH-funded research of disease consequences,
can portable-monitor testing without sleep staging be used
as the only diagnostic sleep test in clinical trials designed to
evaluate effects of CPAP treatment on hypertension and
other cardiovascular outcomes?
2. Are practice-based research networks (PBRN) appropriate
platforms to evaluate clinical prediction algorithms, porta-
ble-monitor testing, and outcomes on CPAP treatment?
3. Are clinical and translational research units in the U.S.
appropriate platforms to evaluate clinical prediction
algorithms, portable-monitor testing, and outcomes on
4. Is the replication of the Spanish Sleep Disordered
Breathing Network model of a multicenter research
network feasible in other countries?
The goal of the fourth and closing session was to set priorities
for future research so that ‘‘within 5 years, optimal research
design(s) will generate adequately powered evidence to prove
or disprove the efficacy and safety of portable monitors for the
diagnosis of OSA.’’ The Hoshin process was used to define the
distant goal, elicit from stakeholders the relevant steps toward
the goal, and to group, title, and stratify those steps by raw
importance and as a function of prior progress (2). Using this
highly interactive process, the workshop participants identified
eight research priorities for the ambulatory management of
patients with OSA, listed in descending order of importance in
Table 1. These priorities are discussed in detail in SECTION V
III. BACKGROUND ON PORTABLE-MONITOR TESTING
AND SIGNIFICANCE OF THE WORKSHOP
Population-based epidemiologic studies estimate the prevalence
of the obstructive sleep apnea–hypopnea syndrome (AHI of 5
events/hr with excessive daytime sleepiness) at 2% of adult
females and 4% of adult males in the U.S. population. A
substantial proportion of these individuals are undiagnosed (3–
6). Existing evidence suggests that OSA is an independent risk
factor for motor vehicle accidents, neurocognitive deficits, and
cardiovascular morbidity and mortality (6, 7). Substantial evi-
dence also supports that appropriate treatment of OSA re-
duces the risk of these consequences (8–11). However, the
need to perform costly and labor-intensive PSG in a sleep
laboratory limits patient access to diagnosis and treatment (12,
13). Commercially available and relatively inexpensive porta-
ble monitors (1, 14–16) might facilitate earlier recognition of
disease and faster initiation of treatment, thereby reducing the
healthcare burden associated with OSA. Interest in the clinical
application of portable-monitor devices is growing rapidly and
TABLE 1. RESEARCH PRIORITIES FOR THE AMBULATORY
MANAGEMENT OF PATIENTS WITH SLEEP APNEA IN
DESCENDING ORDER OF IMPORTANCE
1. Establish adequate and appropriate research networks to conduct
adequately powered, multicenter clinical research studies.
2. Develop disease management models based on prospective clinical
and economic outcome studies that provide evidence for clinical
3. Define the appropriate patient population for ambulatory management.
4. Standardize the portable monitor devices used to diagnose OSA
and initiate continuous positive airway pressure (CPAP) treatment in terms
of their sensors, signal processing, and data analysis.
5. Establish appropriate study designs that successfully address clinically
relevant questions regarding ambulatory management of OSA.
6. Ensure the safety of ambulatory monitoring, especially indentifying
those patient who are not candidates for this management approach.
7. Ensure adequate training and education of providers and patients
regarding ambulatory management of OSA.
8. Identify funding sources to support the needed research.
TABLE 2. CURRENT CLASSIFICATION OF THE DIFFERENT TYPES OF SLEEP STUDIES (16)
Sleep TestDescriptionPersonnel Minimum Signals Required
Type 1 Polysomnography performed
in a sleep laboratory
Attended Minimum of 7 signals, including EEG, EOG, chin EMG, ECG, airflow,
respiratory effort, and oxygen saturation
Type 2Portable polysomnography Unattended Same as Type 1
Type 3 Portable testing limited to sleep
Minimum of 4 signals, including ECG or heart rate, oxygen saturation, and at least
2 channels of respiratory movement, or respiratory movement and airflow
Type 4 Continuous recording of one or
Unattended Usually pulse oximetry
4 PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 82011
is being used as a mainstay approach to the management of
OSA in some settings.
Numerous portable monitors for the diagnosis of OSA are
commercially available. In 1994, a task force on portable-
monitor testing created by the American Sleep Disorders
Association (the current AASM) classified four different levels
of sleep testing (Table 2) (17). Portable monitors are catego-
rized as Types 2 to 4 based on the particular level of study they
record. However, monitors within a given category may vary
widely with regard to the number and type of signals recorded,
the sensors used to record the signals, and signal processing.
Despite the intuitive appeal of portable-monitor testing, high-
quality empiric evidence defining their role in the clinical
management of patients with OSA is limited. Consequently,
portable monitors have failed to gain widespread acceptance in
sleep medicine. Just before the workshop, the AHRQ issued an
outstanding evidence based review on portable-monitor testing
(18). Subsequent to this workshop, the Center for Medicare and
Medicaid Services issued two National Coverage Decisions that
extended coverage for portable monitor testing to diagnose
OSA and justify treatment with CPAP (19, 20). These decisions
have highlighted the urgent need for research to determine the
appropriate role portable monitors should play in the diagnosis
and management of patients with OSA.
Current Practice Parameters to Diagnose Patients with OSA
and Initiate CPAP Therapy
In-lab PSG, a recording of physiologic signals to assess sleep
stage and respiration during sleep, remains the gold standard to
diagnose OSA and initiate CPAP treatment (21–25). This testing
is costly, uses substantial resources, and requires the supervision
of a technologist. To initiate CPAP treatment, the most widely
used treatment for OSA, the current standard is for an attendant
technician to manually titrate CPAP during PSG to identify the
optimal pressure level required for treatment. The patient is then
prescribed CPAP nightly at this pressure setting. Although full-
night diagnostic and manual CPAP titration polysomnograms are
recommended, split-night polysomnograms (one night of testing
that includes both diagnostic testing and manual CPAP titration)
are frequently performed when the AHI on the initial diagnostic
portion of the study is greater than 20 to 40 events per hour (21,
26). The split-night polysomnogram imposes significant time
constraints on the ability to obtain the required information
and has been reported to provide inadequate information re-
garding the prescription of the fixed pressure needed for treat-
ment in approximately 15% of patients (27–29). Despite the
drawbacks of split-night PSG, its wide use is driven by limited
resources and reimbursement policies (30).
Limited In-Laboratory Resources Driving Use
of Portable-Monitor Testing
The high prevalence of OSA presents major logistical difficul-
ties for the clinical evaluation and treatment of affected
patients. It is very likely that the epidemiologic data on the
prevalence of OSA in the middle-aged adult in the United
States population from the Wisconsin Sleep Cohort Study
published in 1993 underestimates the current prevalence of
OSA in the U.S. population given the dramatic increase in
obesity over the past 20 years (5, 31). Nevertheless, applying the
findings in middle-aged adults from the Wisconsin Sleep Cohort
Study, that 9% of males and 4% of females in the U.S. have
moderate to severe OSA (AHI of at least 15 events/hr),
Tachibana and colleagues (32) estimated that 2,310 sleep studies
per year per 100,000 population would be required to adequately
address the demand for diagnosis and treatment of patients with
suspected OSA of at least moderate severity. The capacity for
sleep testing is inadequate to meet even those conservative
demands. Specifically, the total number of sleep laboratories in
2002 was about 1,300 in the private sector and 55 in the Veterans
Health Administration. These laboratories had the capacity to
perform approximately 430 studies per year per 100,000 popula-
tion and 160 studies per year per 100,000 population, respectively.
These limited and costly resources restrict patient access to
testing and are driving the need for alternative management
strategies using portable-monitor testing. Indeed, portable-mon-
itor testing is increasingly used by healthcare providers, particu-
larly those working in capitated and public healthcare systems
(12). In 2005, an editorial on international clinical practices for
the diagnosis of patients with OSA commented that
Faced with the dilemma of how to treat the ‘‘flood’’ of patients
presenting with symptoms suggestive of sleep-disordered breath-
ing, physicians are using nonconventional approaches for di-
agnosis and treatment—approaches not based on solid evidence.
Most surprising ... is the widespread use of ambulatory ap-
proaches to diagnosis rather than full in-laboratory polysomnog-
raphy. With the increased recognition of OSA, systems for
delivering diagnosis and treatment are overwhelmed. Physicians
are trying to cope but, even with creative approaches, waiting
lists for diagnosis and treatment are unacceptably long. There is
a need to rethink current strategies. (13)
IV. HEALTHCARE MANAGEMENT AND INSURANCE-
INDUSTRY PANEL PERSPECTIVE ON EMERGING
A panel discussion was held during the workshop to learn how
healthcare management and the insurance industry evaluate
emerging technologies. The panel included John Leteria, CEO
of Neurocare, Inc., Suresh Ramalingam, M.D., MBA from the
University of Pittsburgh, and Christopher Valerian, DO from the
UMD of New Jersey. The panel addressed four questions:
1. What criteria should be used by healthcare management
and the insurance industry to evaluate acceptance of an
emerging technology into clinical practice?
2. What type of evidence is needed to support the evaluation
of portable monitor testing?
3. In the absence high-quality evidence, what would be the
best way to assess portable monitoring as part of a dis-
4. Once portable monitoring is incorporated into a disease-
management pathway, how can providers be assured that
in-lab PSG can be performed when indicated?
The criteria used by healthcare management and the in-
surance industry to evaluate acceptance of an emerging tech-
nology into clinical practice include data from peer-reviewed
studies, evidence-based reviews (Cochrane, etc.), industry re-
ports, and other studies demonstrating safety and efficacy. The
Food and Drug Administration trial results and approval are
also essential. National guidelines and regulatory directives and
community standards are also considered. Other factors in
decision making include whether quality-control criteria are
present and whether a certification process is required for those
who will deliver the service. The potential for misuse or abuse is
also a consideration. Notably, cost-benefit analysis becomes
a primary incentive for payors once safety and efficacy issues
have been adequately addressed.
American Thoracic Society Documents5
These principles apply to the evaluation of portable-mon-
itor testing. High-level supporting evidence would be ideal, but
in its absence, other available data should be used to formulate
policy until high-level research data are available. The recent
Center for Medicare and Medicaid Services (CMS) National
Coverage Decisions to accept portable-monitor testing for the
diagnosis and coverage of CPAP treatment and the application
of these decisions by local carriers is likely to strongly in-
fluence payors. However, more careful analysis of how porta-
ble monitoring will impact patient access, compliance, and
outcomes, is needed. Cost-benefit analysis must go far beyond
the relative costs of portable monitoring versus in-lab PSG. An
intent-to-treat analysis of clinical outcomes for each treatment
option is critical to this assessment. Patient adherence to
CPAP following ambulatory and in-lab testing should also be
compared in a secondary per-protocol analysis. One must
factor in the associated after-study costs of patients positively
diagnosed with OSA. All costs associated with the continuum
of care (CPAP equipment and disposables, dental appliance or
surgical intervention, physician visits, etc.) must be taken into
As a primary diagnostic technique, ambulatory testing of
patients with OSA will improve access to testing but will require
that patients being evaluated have access to a sleep center when
the portable-monitor testing is unsuccessful or not feasible due
to medical, social, and logistic reasons. Therefore, payor policy
should recognize and accommodate in-lab PSG at the discretion
of the physician and subject to patient choice. There should be
specified guidelines for choosing a portable or in-laboratory
study as the first test, as well as clear definitions for failure of
portable monitoring requiring subsequent in-lab PSG. Review
of clinical practice data should indicate that redundant testing in
the lab is done rarely, and if this is not the case, policies and
procedures should be modified.
Overall, the discussion by the panel members served to un-
derscore the importance of high-quality, outcomes-based re-
search for guiding decision making by health management and
insurance industries on ambulatory monitors in the manage-
ment of OSA.
V. WORKSHOP FINDINGS AND RECOMMENDATIONS
The following eight key research priorities identified by the
Workshop for the ambulatory management of patients with OSA
are presented in descending order of importance (Table 1).
1. Establish Adequate and Appropriate Research Networks
to Conduct Adequately Powered, Multicenter Clinical
Existing multicenter studies and research networks for OSA.
The optimal approach for assessment and management of
patients with OSA in the ambulatory setting remains uncertain.
Although specific technical questions are amenable to testing in
a single specialized center, the development and evaluation of
clinical management strategies and outcomes based on portable-
monitor testing require patient populations that are too large
for single center studies. The sleep research community needs
to follow the approach of other specialties, such as cardiology,
that conduct multicenter studies to evaluate particular clinical
interventions in large numbers of patients.
Some notable multicenter research collaborations have been
established in recent years to investigate the epidemiology,
comorbidities and therapeutic outcomes of OSA. In North
America, these include the NIH-funded Sleep Heart Health
Study, that particularly addressed the cardiovascular comorbid-
ities of the disorder, and the Canadian Medical Research
Council funded CANPAP (Canadian Positive Airway Pressure)
clinical trial that evaluated the potential role of nocturnal
CPAP therapy in patients with chronic heart failure (33, 34).
In Europe, a number of multicenter studies relating to OSA are
ongoing such as the MOSAIC (Multicenter Obstructive Sleep
Apnoea Interventional Cardiovascular Trial) study of mild
OSA, based in the UK, and the ESADA (European Sleep
Apnea Database) study of cardiovascular disease in OSA,
which includes more than 20 sleep centers throughout Europe.
The ESADA project is part of a collaborative network of sleep
centers throughout Europe, funded by the European Union as
part of the COST Action scheme (http://www.costb26.net/). The
above initiatives are best described as extended multicenter
studies established to examine particular research topics. How-
ever, the ESADA study involves the establishment of a common
European database of patients with OSA using standardized
and uniform inclusion criteria and has the potential to expand
into a research network.
The Spanish Sleep Disordered Breathing Network is an
excellent model of a multicenter network that has made impor-
tant scientific contributions relating to the study of OSA. This
network consists of a number of sleep centers throughout Spain
and was established to allow the collaborative study of a variety
of topics relating to OSA, ranging from epidemiology to disease
management. A number of important principles underpin the
success of this network. First, the network was developed as
collaboration among equals and adopts a democratic, nonbur-
eaucratic approach. Second, each specific study undertaken by
the network members has one of the collaborators identified as
the principal investigator and who is thus responsible for the
successful implementation of the protocol. Third, the network
members interact regularly to monitor ongoing activities and to
identify new topics for investigation.
A number of existing networks and programs in the United
States could facilitate the implementation of large-scale collab-
orative sleep research projects. These include the Primary Care
Practice-Based Research Networks (PBRN), the Clinical and
Translational Research Centers (CTRC), and networks within
the Veterans Health Administration (VHA). PBRNs involve
groups of ambulatory care practices throughout the U.S. that
collaborate to investigate clinical questions applicable to pri-
mary care. Such networks typically operate under the auspices
of a professional or academic organization and usually depend
on outside funding to support research activities. PBRNs have
the advantage of being able to evaluate clinical decision making
and provision of services to a large number of individuals in
a primary care setting. The CTRCs are funded by a NIH clinical
research program to promote collaboration between and within
academic centers of excellence. The program is intended to
provide relatively large-scale funding to a small number of
academic health centers and support a range of activities
including protocol development and implementation, educa-
tion, and collaboration between clinical and basic researchers.
The CTRC infrastructure provides an opportunity for sleep
centers located in these institutions to build a clinical sleep
research network. Finally, the VHA operates three collabora-
tive research programs that could potentially serve as funding
and infrastructure platforms for portable monitor research:
PBRNs, the Cooperative Studies Program (CSP), and the
Health Services Research and Development (HSRD) program.
These programs have the advantage of operating within a uni-
formly structured national healthcare network that includes
specialized sleep centers.
6 PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 8 2011
Priority 1: Establish adequate and appropriate research net-
works to conduct adequately powered, multicenter clinical
research studies. Essential components for an effective re-
d A robust organizational structure consisting of a designated
leader and a steering committee is critical for the initial
set-up of the network and in driving the research program
d A broad participation of sleep research centers with
appropriate facilities and expertise to implement the
envisaged research program;
d An infrastructure capable of supporting and monitoring
the planned research;
d Effective communication between all interested parties in
the process including manufacturers, third party payers,
professional bodies, and the public;
d Effective mechanisms for national/global tracking of prog-
ress in portable monitor research.
2. Develop Disease Management Models Based on Successful
Clinical and Economic Outcomes
Clinical outcome-based measures for comparing in-lab versus
ambulatory pathways. Recent evidence reviews have high-
lighted the need for a focus on outcomes-based studies in-
volving ambulatory monitors. It is important to assess the
performance characteristics of specific monitors in the intended
clinical context and patient population. However, studies
should not simply assess the ability of the monitor to reproduce
a PSG-derived AHI. Instead, it is imperative to perform
comparative effectiveness research studies that ascertain the
clinical outcomes of OSA management pathways using porta-
ble-monitor testing versus in-laboratory testing. The workshop
participants strongly endorsed this principle for future studies
of portable monitors.
Study designs evaluating ambulatory management strategies
should take into account that differences in the modality of
testing may influence a patient’s attitudes and perceptions about
OSA and thereby influence their subsequent adherence to
CPAP treatment. Differences between testing methods might
arise due to the greater amount of time healthcare providers
interact with patients during in-laboratory versus home testing
(35). Increased opportunities for patient education and support
have been shown to improve patient adherence to treatment
(36, 37). Administering a self-efficacy questionnaire at several
time points throughout the protocol may help assess the impact
of specific interventions (38–40).
The clinical and cost-related outcomes in comparative
effectiveness research studies must be carefully selected and
clearly defined. Ideally, both short-term and long-term out-
comes should be evaluated. Workshop participants identified
the need to develop a consensus on the most appropriate
outcome measures and to standardize these for use in research
studies. Pending such a consensus, there are categories of
validated outcome measures that should be incorporated into
study protocols. These categories include general and disease-
specific quality of life measures, sleep-related symptoms, and
objective measures of neurocognitive function. Examples of
functional outcome measures include: the Psychomotor Vigi-
lance Task (PVT) for objective assessment of daytime sleepiness,
the Epworth Sleepiness Scale (ESS) for subjective assessment
of daytime sleepiness, disease-specific quality of life question-
naires such as the Functional Outcomes of Sleep Questionnaire
(FOSQ) and the Calgary Sleep Apnea Quality of Life Index
(SAQLI), and general quality of life questionnaires such as the
Short Form-36 and Short Form-12 (41–49).
Cardiovascular outcome measures should also be clearly
defined in study protocols. These may include direct measures
of cardiovascular function (individual or 24-hr blood pressure
measurements, ECG measurements including rhythm and is-
chemic changes, and echocardiographic changes) or documen-
tation of clinical cardiovascular events (ischemic, heart failure,
transient ischemic attack, or stroke). Surrogate measures of
cardiovascular risk (e.g., circulating or tissue biomarkers, mea-
sures of endothelial function, and vascular intimamedia thick-
ness) and assessment of metabolic function (lipid metabolism
and insulin resistance) may also be relevant. Ideally, studies
should seek to identify the impact of portable monitors on long-
term cardiovascular morbidity and mortality. The challenge in
designing randomized studies with hard cardiovascular out-
comes, however, is that of including an ethically acceptable
control group and thereby overcome the problems inherent in
observational study designs (50).
The workshop did not endorse any specific study design for
future comparative effectiveness research. Participants did dis-
cuss two ongoing research projects evaluating the ambulatory
management of OSA that have similar study designs (Figure 1).
The Veterans Health Administration is funding one of the
studies, and the other is funded by the American Sleep Medicine
Foundation. In both protocols, patients with suspected OSA were
randomized to standard in-laboratory testing versus overnight
home unattended testing with a Type 3 portable monitor. Those
participants testing positive for OSA on the home recording
performed a home-unattended autoCPAP titration study for
several nights. Patients diagnosed with OSA in both arms were
initiated on CPAP treatment and reassessed after 3 months of
Comparative effectiveness research studies must take into
consideration that different testing modalities (portable-monitor
testing vs. PSG) are being used to diagnose OSA. For example,
AHI on the baseline sleep study should not be used to de-
termine whether participants randomized to each arm have
a similar severity of OSA insofar as the portable-monitor study
(without sleep staging) tends to underestimate the AHI that
would be obtained with PSG. Another indicator of disease
severity, such as the Multivariable Apnea Prediction Index (51,
52), could be used to assess disease severity at baseline across
the two groups. Studies can attempt to compensate for these
differences in diagnostic accuracy by performing polysomnog-
raphy in those patients with negative home studies. This ap-
proach, however, is also potentially problematic. The option of
a second diagnostic test in the home group and not the in-lab
group may make it more likely, due to night-to-night variability
in AHI, that the diagnosis of OSA will be established in patients
randomized to home testing.
To compare the two management pathways using an intent-
to-treat analysis, these comparative effectiveness research stud-
ies should select a primary outcome that evaluates all participants
randomized to each arm of the study, regardless of whether or
not they were diagnosed with OSA and initiated on CPAP
treatment. CPAP adherence should not be the primary outcome
measure since different percentages of participants randomized
to each arm may be diagnosed with OSA and treated with
CPAP. CPAP adherence is an outcome of interest but is limited
to a per protocol analysis.
Cost effectiveness. During the workshop, there was con-
siderable discussion on the importance of evaluating the cost
American Thoracic Society Documents7
effectiveness of portable monitors. Henry Glick, Ph.D. ad-
dressed the topic ‘‘How to incorporate economic evaluations
into clinical trials: an overview of design and analysis.’’ (53)
Economic data collected as primary or secondary endpoints in
randomized trials are commonly used to evaluate whether
there is ‘‘good value for the cost’’ of medical therapies. Short-
term economic impacts are directly observed in studies, while
longer term impacts can be projected by the use of decision
analysis. The key steps in economic evaluation are to: (1)
quantify the costs of care, (2) quantify outcomes, (3) assess
whether and by how much average costs and outcomes differ
among the treatment groups, (4) compare the magnitude of
difference in costs and outcomes and evaluate ‘‘value for
costs’’ (e.g., by reporting a cost-effectiveness ratio or the pro-
bability that the ratio is acceptable), and (5) perform sensi-
Economic analyses should evaluate health resource use for
the entire clinical management pathway, from diagnosis to
treatment outcomes. Ideally, economic evaluation within a trial
should measure all costs of all participants before randomiza-
tion and for the duration of follow-up. The costs after random-
ization represent the cost outcome, whereas the costs before
randomization are a potential predictor of outcomes. Four
strategic study design issues for cost analysis were addressed:
1. What medical service use should one collect?
2. How naturalistic should the study design be?
3. What is the appropriate sample size?
4. What is the likelihood that the cost-effectiveness ratio
observed in the trial describes longer-term therapy?
Study-specific costs to consider include: sensor/supply
and equipment purchases; maintenance and refurbishment/
replacement due to damage from portable use and/or theft;
laboratory space; and personnel costs, including staff training/
development as well as workload for equipment management,
patient training, data download, and scoring. Other costs arise
from failed or inconclusive studies that need to be repeated, or
for which PSG eventually has to be performed; the costs asso-
ciated with incorrect or missed diagnosis of sleep-disordered
breathing; failure to diagnose concomitant nonrespiratory
sleep disorders, and treatment failures, such as nonacceptance
Several economic analyses of ambulatory management of
patients with OSA have been reported (54–58). Three of the
studies were based on decision analysis, and their impact is
diminished because the inputs used for modeling analysis were
not based on direct observation (55, 56, 58). Two studies
assumed unacceptably high failure rates for home testing (57,
58). There has been only one prospective economic evaluation
of portable-monitor testing (54). Using the change in the total
score on the Epworth Sleepiness Scale questionnaire (44) to
assess functional outcome, this study found that a nurse prac-
titioner–led management of patients with OSA using portable-
monitor testing was more cost effective than physician-led
management using in-laboratory PSG (54). More prospective
randomized controlled trials on cost effectiveness are needed to
collect evidence under naturalized, realistic conditions that can
be used for a modeling analysis. These data can then be used to
perform decision modeling that will help sleep specialists decide
whether to use the more expedient home testing versus accept-
ing delays in obtaining in-lab testing. Modeling on these data
can also determine the utility of home testing in patient
populations with different prevalence of OSA. Only one study
has addressed whether it is better to use home testing sooner or
whether it is better to wait for in-laboratory testing (59). That
study reported that earlier diagnosis and treatment was cost
effective compared with waiting.
What medical-service use should one collect? To assess the
cost impact of a therapy, one should be sure to determine the
use of services arising from differences in treatment between
the arms of the study. It is also important to measure high-cost
services. Minimizing the number of unmeasured services re-
duces the likelihood that differences among them will lead to
biased estimates. In fact, it is prudent to capture as many
services as possible. The general strategy should be to identify
a set of pertinent medical services and assess their utilization
longitudinally, independent of the reason for their use. There
are no a priori guidelines about how much data are enough, nor
Figure 1. Alternative clinical management pathways for
patients with obstructive sleep apnea (OSA). The pathway
on the left represents the standard in-laboratory manage-
ment and the pathway on the right represents a possible
approach using home-unattended portable-monitor test-
ing. PSG, polysomnogram; CPAP, continuous positive
8 PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 8 2011
are there data on the incremental value of specific items in the
economic case-report form. One should consider collecting
costs other than medical service use that may be of use to
medical decision makers, such as time costs (e.g., lost time due
to illness and treatment). Practically, decisions about the
services to measure should take into account the expense of
collecting particular data items.
During trial design, it is important to document potential
service use. Decisions are improved if there is documentation of
the types of services used by patients who are similar to those
who will be enrolled in the trial. This can be achieved through
medical record review, administrative data sets, surveys of
patients and experts about the kinds of care received, or through
patient logs of their healthcare resource use. One should also
evaluate the possibility that a new therapy will induce different
medical service use.
How naturalistic should the study design be? The primary
purpose of cost-effectiveness analysis is to inform real-world
decision-makers how to respond to real-world healthcare needs.
Thus, the more naturalistic the trial—in terms of participants,
analysis based on the intention to treat, and limitation of loss to
follow-up—the more likely the data developed within the trial
will speak directly to the decision question.
While advocating a naturalistic framework to assess cost
effectiveness, the workshop participants also cited important
elements that should be included in studies. Treatment efficacy
and patient satisfaction with the care path should be considered.
Relative wait times, and the number of lost, technically un-
satisfactory, and/or equivocal studies should be recorded,
together with the criteria for each of these categories. Objective
measures of treatment response should include the change in
AHI/respiratory disturbance index (RDI), oxygenation and,
when available, changes in measures of sleep structure. Objec-
tive adherence to CPAP treatment is an essential outcome
measure, and the software and thresholds used to determine
adequate adherence need to be clearly specified.
The design of clinical trials should include plans for robust
follow-up of participants to minimize missing data due to the
loss to follow-up. It is also important to continue data collection
and follow-up until the end of the study period and not
discontinue data collection because a subject reaches a clinical
or treatment stage such as failure to respond. Given that failure
is often associated with a change in the pattern of costs,
discontinuation of these patients from the study is likely to bias
Trial-based cost-effectiveness analyses should adopt an intent-
to-treat design. Economic questions relate to treatment de-
cisions (e.g., whether to prescribe a therapy) and not to whether
the patient received the prescribed intervention, such as CPAP,
nor to whether, once they started the prescribed intervention,
they were switched to another treatment option. Thus, costs and
effects associated with these later decisions should be attributed
to the initial treatment decision.
What is the appropriate sample size to address economic
questions? Economic sample size calculations are based on
the number of study subjects needed to rule out that the
therapy is unacceptable (equivalently, to ruling out that the
net monetary benefits of the intervention are less than 0). Key
factors in sample size formulae include numbers of subjects,
the standard deviation for costs and treatment effects, the
maximum willingness to pay that one wishes to rule out, and
the correlation of the difference in cost and effect. The
required sample size is less when the therapies have a Win/
Lose (positive) correlation (i.e., as the effectiveness in-
creases, the cost increases, e.g., stroke care). The required
sample size is greater when the therapies have a Win/Win
(negative) correlation (i.e., as the effectiveness increases, the
cost decreases, e.g., asthma care). With respect to maximum
willingness to pay and identification of an appropriate out-
come measure, the sample size calculations assume that there
is some concept of what one would be willing to pay to obtain
a unit of outcome. Many researchers use disease specific out-
comes. Although any outcome can be used to calculate a cost-
effectiveness ratio (e.g., cost/case detected or cost/additional
abstinence day), the outcome must have recognized benchmarks
of cost effectiveness to be convincing that a new, more costly
and more effective therapy has good value. This argues against
the use of too disease-specific an outcome for economic assess-
If portable monitor testing is both clinically equivalent to
and less expensive than in-laboratory testing, a ‘‘disease-specific’’
outcome such as cases detected may be sufficient (depending on
‘‘how equivalent’’ it is). If home testing is less effective and less
expensive, one needs to know the value of the lost effectiveness
so that it can be compared with the cost savings. This requires
that we either know the worth of detecting a case or use a more
general health outcome such as quality adjusted life years
(QALYs). OSA research is at a disadvantage because the cost
effectiveness of diagnosis of OSA has not been well established
What is the likelihood that the cost-effectiveness ratio
observed in the trial describes longer-term therapy? A time-
by-treatment interaction manifests when cost and outcome
follow different time courses. For example, risk reduction from
cholesterol-modifying therapy displays substantial time-by-
treatment interaction given the lag time between initiation of
therapy and improved outcome. Conversely, drug therapy for
heart failure displays less interaction as the treatment and
outcome are incurred together. If a strong treatment-by-time
interaction is expected, a decision model will be required to
ascertain the potential magnitude of the interaction. Substantial
amounts of the data used for the decision model should be
derived from the trial and, when necessary, augmented with
data from epidemiological studies.
In summary, clinical trials are invaluable opportunities for
evaluating the cost effectiveness of a therapy. If complete data
sets are collected and analyzed appropriately, these evaluations
can provide data about uncertainties related to the assessment
of the value for the cost of new therapies that may be used by
policy makers, manufacturers, healthcare providers and patients
when the therapy is introduced in the market.
Priority 2: Develop disease management models based on
successful clinical and economic outcomes. Recommendations:
d Develop and validate recognition strategies to identify
patients with high pretest likelihood of OSA;
d Incorporate ambulatory monitors into disease manage-
ment pathways evaluating clearly specified OSA-related
d Develop consensus on the most appropriate outcome
measures and standardize their use in research studies;
d Include economic evaluations in clinical trials using out-
comes that are recognized benchmarks of cost effective-
d Generate data from high quality studies that can be used to
develop decision analytic models.
American Thoracic Society Documents9
3. Define the Appropriate Patient Population for
Establishing the optimal role of portable sleep monitors.
Another goal of the workshop was to promote research to
determine the role portable monitor testing should play in the
diagnostic evaluation and management of patients with OSA.
Debate continues as to whether portable monitor testing should
be used diagnostically in the general population or in a subgroup
such as diagnosing patients with a high pretest likelihood of the
disorder. In most current clinical applications, Type 3 portable
monitors (monitors that record respiratory-related signals but
do not record the PSG signals for sleep staging) are used for
unattended home recordings to diagnose OSA. It is recom-
mended that those patients with a negative Type 3 recording
have an in-laboratory PSG to exclude the possibility of a false
negative study (1). Using portable monitors to include and
exclude the diagnosis of OSA in the general population would
result in a greater proportion of negative studies increasing the
demand for in-laboratory PSG. Limiting portable monitor
testing to patients with a high-likelihood of OSA would
minimize the number of negative studies and the need for in-
laboratory PSG. Therefore, accurate identification of patients
with a high pretest likelihood of OSA is essential. Clinical
prediction rules, including the Multivariable Apnea Prediction
Index, Sleep Apnea Clinical Score, and Berlin Questionnaire,
have been used for this purpose in research studies but they
have not been adequately tested in clinical management path-
ways (51, 52, 60, 61).
Study populations that need to be included in portable
monitor research. Very few published studies have assessed
portable monitors in specific populations, including diverse
ethnic groups, the elderly, and individuals with cardio-respira-
tory and neurological diseases. Participants at the workshop
discussed the best research approaches for evaluating portable
monitors in these subgroups.
The initial question that needs to be addressed is whether
portable monitor testing is feasible and suitable for the screen-
ing and diagnosis of OSA in technologically challenged, socially
disadvantaged, and medically disabled populations. Are there
any specific sensors that are more suitable for certain sub-
populations? Should diagnostic and/or therapeutic threshold
values for respiratory disturbance indices differ between pop-
ulations? How would portable-monitor testing, compared with
PSG, affect response and adherence to positive airway pressure
therapy? What is the likelihood of repeat testing due to
inadequate or suboptimal data collection with portable monitor
testing or PSG in the different study populations?
Research is needed on adapting portable monitors to
patients with specific comorbid medical and neurological con-
ditions. Persons with COPD, asthma, heart failure, and neuro-
muscular disorders have a higher risk of developing sleep-
related hypoventilation and central sleep apnea. Ideally, porta-
ble-monitor testing in these patients should be able to distin-
guish these respiratory disorders. Appropriate sensors and
bioparameters (e.g., oxygen saturation, airflow limitation, re-
spiratory effort) for each condition need to be defined.
Any study evaluating portable monitors should include a di-
versity of racial and ethnic groups. The impact of differing lan-
guages and cultures on portable monitor testing should be
assessed by engaging existing social and medical networks
working with different ethnic groups. Among the elderly pop-
ulation, many of whom have multiple comorbid conditions (e.g.,
medical or other primary sleep disorders such as periodic limb
movement disorder), it is important to identify the factors that
influence the performance of sleep studies using portable mon-
itors, and develop management pathways that ensure patient
safety during the testing. Use of telemetric monitoring might be
explored as one such approach. Studies on the relative feasibility,
access, and convenience of the portable monitor compared with
in-laboratory PSG among community-dwelling compared with
nursing home adults are also important.
In summary, the participants of the workshop identified
several research priorities related to the study population,
including a focus on high-risk subjects with cardio-pulmonary
and neurological comorbidities, older adults, and persons with
different ethnic groups. Appropriate outcome measures for
each study population will need to be identified.
Priority 3. Define the patient population most appropriate for
ambulatory management. Recommendations:
d Target groups not previously studied to identify those best
suited for portable testing;
d Identify and report the age, sex, and ethnic characteristics
of the study population, as well as the nature and severity
of any underlying cardiopulmonary disease or other
d Consider the characteristics of the study population in all
aspects of study design and analysis.
4. Standardize the Portable Monitor Devices Used to
Diagnose OSA and Initiate Continuous Positive
Airway Pressure (CPAP) Treatment
Current lack of standardization among portable monitors.
The current lack of standardization of commercially available
monitors is a major barrier preventing the incorporation of
portable-monitor testing into routine clinical management
pathways. Although portable monitors for sleep testing are
intended primarily for unattended home recordings, they can
be used under attended or unattended conditions and in
a variety of locations, including the sleep laboratory and
healthcare facilities. The monitors differ widely in the number
and type of signals recorded, the sensors used to record the
signals, and the electronic processing of the signals. Scoring of
the recordings may be totally automated or manual with the
assistance of computer software
As stated in a recent practice parameter report on portable
monitor testing, ‘‘There is no universally accepted platform for
generating simplified studies in the diagnosis of OSA. This means
that results obtained for a particular device are applicable to that
device and cannot be extrapolated to other devices, even those of
the same class’’ (62). This lack of uniformity limits the impact of
evidence-based reviews that evaluate the results of research
studies performed using monitors within a particular category
without consideration of the technological differences that exist
among these monitors. Although further standardization of
portable monitors is needed, important technological questions
remain to be answered before we can determine the ideal
portable monitor for diagnosis of OSA. We still need to de-
termine which signals are essential and how they should be
acquired in terms of sensors used, sampling rate, and filtering.
Innovative signals and approaches to portable monitor
testing. In addition to the standard PSG techniques that have
been adopted for portable monitor testing, novel technologies
have been developed to enhance their performance and appli-
cation (63). For example, actigraphy has been evaluated as
a surrogate marker of sleep and wakefulness to improve the
10 PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 82011
calculation of AHI (64). In one commercially available Type 3
monitor, the sensors that record nasal pressure, oximetry, head
movement, snoring, and respiratory effort (venous pulsations)
are contained in a headband worn around the forehead (65).
Some monitors incorporate other novel sensors that detect
cardiac and autonomic responses to sleep-disordered breathing.
One such device measures peripheral artery tone from a sensor
on the finger that estimates changes in vascular flow, a measure
that reflects variations in breathing and sleep-related arousals
(66). Unfortunately, the advances made in portable monitor
technology far outstrip our knowledge about their utility in
clinical testing. This wide diversity in portable monitors com-
plicates the ability to compare results across monitors and to
generalize results obtained with a particular monitor. The test
for any instrumentation is its role in clinical decision making.
This being the case, it is important to standardize a core set of
features so that one can compare one to another and then show
the effects on clinical decisions.
Issues regarding signal acquisition. The central premise of
portable monitors is that a few signals, extracted from PSG, are
necessary and sufficient for the diagnosis of OSA. An ideal
portable monitor should provide sufficient accuracy for case
finding, should be amenable for self-application, comfortable
for all-night use, and durable enough to withstand nightly
transport and application in the home by inexperienced pa-
tients. Moreover, the monitors should provide full disclosure of
high-quality primary signals for manual review and either
manual scoring or manual editing of automated scoring. In
addition, the device should provide computational algorithms,
allowing them to be used in a large number of patients. In other
words, an ideal portable monitor should contain high-quality
primary signals without requiring excessive labor and perhaps
redundancy of signals to provide complementary data acquisi-
tion and minimize data loss. In addition, there is an unmet need
to ascertain the variables that predict long-term outcomes in
OSA. Longitudinal data linking specific physiologic signals to
neuro-cognitive consequences, daytime hypersomnolence, or
adverse cardiovascular outcomes, are needed. Furthermore,
different variables may predict different outcomes.
Several questions await further evidence to ascertain pri-
mary signals for portable monitoring.
1. What is the cost of repeated studies resulting from failed
data acquisition versus the cost of technical time required
for full PSG?
2. Should the design of portable monitors allow for the
application of the sensors by the patient?
3. What are the limitations of acquired signals under
various clinical settings? For example, noninvasive de-
termination of respiratory effort using respiratory in-
ductance plethysmography is more likely to capture
paradoxical breathing in children with a compliant chest
wall than in a morbidly obese individual with reduced
4. Does the usefulness of individual signals vary by the
population being tested? For example, oxyhemoglobin
desaturation is more likely to occur in obese rather than
Standardizing sensors, signal conditioning, and scoring.
The ideal sensors should be easy to apply and capable of
providing a core group of reliable signals. The output of sensors
should be accurate, reliable, and reproducible. Oximetry, flow
using nasal pressure, ECG, and respiratory effort are the
primary signals requiring standardization and validation. In
contrast, sleep staging is less critical in portable-monitor testing.
Accurate diagnosis also requires scoring standards to minimize
variability. Current AASM PSG scoring standards are applica-
ble to portable recordings and may be amenable to autoscoring.
Given the limitations of current technology in terms of lack of
standardization, variability between devices, and differences in
software interpretation, it seems logical to first resolve these
problems. One possible approach would be to conduct multi-
center comparative effectiveness studies to compare different
portable monitors against each other and evaluate their perfor-
mance in terms of clinical outcomes.
Priority 4. Standardize the portable monitor devices used to
diagnose OSA and initiate continuous positive airway pressure
(CPAP) treatment. Recommendations:
d Standardize equipment and specific signals;
d Standardize therapeutic devices and their outputs;
d Identify the minimal and ideal signals and their biosensors;
d Compare Type 3 to Type 4 portable-monitoring devices;
d Develop a common interpretive taxonomy and data plat-
form for portable-monitoring devices.
5. Establish Appropriate Study Designs That Successfully
Address Clinically Relevant Questions Regarding Ambulatory
Management of OSA
Closely related to the need to develop disease-management
models based on successful clinical and economic outcomes
(Priority 2) is the need to design studies that successfully
evaluate the effectiveness of those models.
Limitations of validation studies directly comparing porta-
ble monitor testing and PSG. Differences in equipment and
testing environments, intrascorer reliability, and the known night-
to-night variability in AHI may explain why direct comparisons
of results from portable monitor testing and PSG are not closely
correlated. Most portable monitors capable of widespread appli-
cation do not include signals that detect whether the patient is
awake or asleep during the recording. The severity of the sleep-
disordered breathing on these recordings is therefore quantified
as the number of apneas and hypopneas per hour of recording,
instead of the number per hour of sleep, and is sometimes
referred to as the respiratory disturbance index (RDI) rather
than the AHI. In patients with delayed sleep onset and low
sleep efficiency, the resulting AHI will underestimate the
‘‘true’’ AHI. Although the correlation between in-laboratory
PSG and Type 3 monitor testing is generally acceptable when
the recordings are performed simultaneously in the sleep
laboratory, all evidence-based reviews comment on the im-
portance of validating portable monitors in the home envi-
ronment, the intended location for their use.
Validating Type 3 portable monitors based on clinical
outcomes. Recognizing the limitations of studies that directly
compare sleep test results between portable monitors and in-
laboratory PSG, investigators are starting to perform studies that
compare participants randomized to these different pathways in
terms of improvements in quality of life and other clinical
outcomes. Assuming that PSG has a higher sensitivity than
portable-monitor testing, patients randomized to home testing
may not be diagnosed with OSA on the portable-monitor
recording, whereas they would have been diagnosed on PSG.
The resulting inequality between groups might influence the
American Thoracic Society Documents11
results of outcome measure(s) such as adherence to CPAP
treatment. Studies can attempt to compensate for these
differences in diagnostic accuracy by performing PSG in those
patients with negative home studies. This approach, however,
may also influence study outcomes. The option of a second
diagnostic test in the home group and not the in-lab group
makes it more likely, due to night-to-night variability in AHI,
that the diagnosis of OSA will be established in patients
randomized to home testing.
Priority 5: Establish appropriate study designs that successfully
address clinically relevant questions regarding ambulatory man-
agement of OSA. Recommendations:
d Research studies to validate performance of portable
monitor equipment against PSG should be designed to
take into account the night-to-night variability in AHI on
the in-lab polysomnogram and differences in equipment
and testing environment.
d Research studies on portable monitor testing should
evaluate the outcomes of integrated ambulatory manage-
ment pathways as well the key components of those
pathways (e.g., diagnostic, therapeutic strategies)
d The role of autoCPAP units in the management of patients
with sleep apnea and the cost effectiveness of ambulatory
pathways needs further investigation.
Use of autoCPAP to titrate the pressure setting needed for
CPAP treatment. The current AASM practice parameter for
portable monitor testing recommends that a manual CPAP PSG
be performed in patients with a positive Type 3 diagnostic study
(62). However, the use of home-unattended portable-monitor
testing to diagnose patients with OSA apnea will only alleviate
the growing demand for in-laboratory testing if those patients can
be initiated on CPAP treatment without requiring PSG to
establish the optimal CPAP setting. AutoCPAP units have been
used successfully to titrate the fixed pressure setting needed for
CPAP treatment in attended and unattended settings (54, 67, 68).
Increasingly, providers are using autoCPAP instead of CPAP for
regular treatment. However, no consensus exists regarding the
optimal role of autoCPAP machines in the clinical management
of patients with OSA. One of the important barriers to the
acceptance of portable-monitor testing for OSA in the United
States is the lack of benefit coverage for home-unattended
autoCPAP titration studies.
6. Ensure the Safety of Ambulatory Monitoring, Especially
Identifying Those Patients Who Are Not Candidates for This
Discussion about safety issues regarding portable-monitor test-
ing related primarily to the appropriate validation of the devices
with respect to diagnostic accuracy in relevant populations, and
the responsible application of diagnostic information from the
monitors to patient management. It was felt that portable-
monitor units that distinguish obstructive and central apneas
will have wider application; whereas units lacking this capability
will require validated clinical correlation to exclude patients with
central sleep apnea.
AutoCPAP features that help ensure patient safety. To
prevent adverse events related to excessive pressure, the
autoCPAP machines are limited to a pressure range from 4
to approximately 20 cm H2O. AutoCPAPs are unable to
distinguish central from obstructive apneas. Therefore, to
avoid the potential problem of increasing pressure in the
presence of central apneas, autoCPAP algorithms uniformly
prevent increases in pressure greater than 10–11 cm H2O in
the presence of persistent apneas. Data uploaded from autoC-
PAP units report not only the pressures delivered but also the
AHI, number of apneas, and amount of air leak. Studies are
needed to document the accuracy of these measurements.
To ascertain the effectiveness of autoCPAP treatment in
restoring oxygen saturation to acceptable levels, some autoC-
PAP devices can interface with pulse oximeter modules to
record oxygen saturation and heart rate. Some autoCPAPs are
also designed to interface with a portable monitor for a verifi-
able documentation of AHI as well as oxygen saturation. The
latest innovations in autoCPAP machines allow remote moni-
toring of their use and performance either by modem or
wireless transmission of recorded data. Although no studies
have investigated the use of this innovative technology, the
ability to remotely track events during the home titration may
enable early intervention that can promote successful titration
and initiation of CPAP treatment.
Priority 6. Ensure the safety of ambulatory monitoring, espe-
cially indentifying those patients who are not candidates for this
management approach. Recommendations:
d Develop validated clinical management pathways using
portable monitor testing that define the decision making
parameters that result in accurate diagnosis andtreatment
d Develop ambulatory management pathways, especially
those using Type 4 monitors, that are capable of identify-
ing patients with Cheyne-Stokes respiration, central sleep
apnea, and complex sleep apnea;
d Evaluate the accuracy of the AHI and respiratory event
information reported by CPAP and autoCPAP units.
7. Ensure Adequate Training and Education of Providers and
Patients Regarding Ambulatory Management of OSA
Discussants highlighted concerns regarding the declining num-
bers of new clinician-scientists entering Sleep Medicine. There
is increasing difficulty with recruitment of trainees through
traditional pathways such as pulmonary medicine given pro-
longed training requirements for pulmonary, critical care, and
sleep medicine. Innovative approaches to training and recruit-
ment must be identified and pursued. Development of research
networks would provide the required infrastructure to support
high-quality training programs and promote recruitment to the
field. Research training for new scientists will require energetic
and dedicated mentors to guide and nurture the careers of
fellows and junior faculty. Academic sleep centers need to
obtain research training grants from governmental and institu-
tional funding sources that will allow trainees to prepare for
careers in sleep research.
The increased use of portable monitoring will elevate the
role of generalists and other nonsleep medicine physicians. The
level of education and training of such practitioners will
critically influence optimum care of patients with OSA. Clinical
practice networks where nonspecialized practitioners operate
in close collaboration with a specialized sleep center offer one
potential solution. More broadly, continuous education and
training of practitioners outside of sleep centers is important;
opportunities through multidisciplinary training programs and
12 PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 82011
clinician and research clerkships at other institutions should be
established. Such programs should be directed not only to
physicians, but to other members of multidisciplinary care
teams including nursing, respiratory therapy, and sleep tech-
Priority 7. Ensure adequate training and education of providers
and patients regarding ambulatory management of sleep apnea.
d Stakeholder medical specialty organizations and the
ACGME urgently need to negotiate innovative solutions
to optimize the feasibility and appeal of sleep clinician-
scientist training pathways;
d Increase investment in research training for new scientists;
d Develop and validate training programs for healthcare
providers, including sleep specialists, primary care pro-
viders, physician extenders and allied health professionals.
8. Identify Funding Sources to Support the
Closely related to the need to establish research networks to
conduct adequately powered, multicenter clinical research
studies (Priority 1) is the need to identify funding sources to
support the research. Speakers from three different government
organizations discussed research network infrastructure as well
as funding opportunities for future research studies in portable
monitoring: David Lanier, M.D., Center for Primary Care,
Prevention and Clinical Partnerships, Agency for Healthcare
Research and Quality (AHRQ), who spoke on PBRN’s; G. Iris
Obrams, M.D., M.P.H., Ph.D., National Center for Research
Resources (NCRR), National Institutes of Health (NIH) who
spoke on CTSAs; and Katherine Bent, R.N., Ph.D., C.N.S.,
Office of Research and Development, Department of Veterans
Affairs (VA) who highlighted approaches to research funding
within the VA system.
In the small group discussions on potential funding sources
for portable monitor research, the key issues brought forward
were to identify, cultivate, and amalgamate funding sources for
portable monitoring research. There was also discussion con-
cerning the translation of research findings into policy and
practice. In this regard, it will be essential to eliminate monetary
interest for sleep physicians who are making management
decisions (i.e., eliminate conflict of interest). It will also be
necessary to identify sources of funding for equipment, person-
nel, training, and infrastructure for application of innovative
ambulatory management strategies. As noted in SECTION IV
above, high-quality outcomes-based research demonstrating
efficacy and cost effectiveness should serve as an important
impetus for health management and insurance industries to
provide support for these approaches.
In further discussion on research funding, the Center for
Medicare and Medicaid Services was identified by participants
as a source for research funding through Coverage with
Evidence Development (CED). It is noteworthy that in the
National Coverage Decision on coverage of CPAP therapy
based on ambulatory monitor diagnosis released shortly after
this Workshop was held (19), the Coverage with Evidence
Development provision was made for adult beneficiaries who
do not qualify for CPAP coverage based on standard criteria
but who are enrolled in a study that addresses one of the
following two questions:
(a) In Medicare-aged subjects with clinically identified risk factors
for OSA, how does the diagnostic accuracy of a clinical trial of
CPAP compare with PSG and Type II, III, and IV home sleep test
in identifying subjects with OSA who will respond to CPAP?
(b) In Medicare-aged subjects with clinically identified risk
factors for OSA who have not undergone confirmatory testing
with PSG or Type II, III, and IV home sleep test, does CPAP
cause clinically meaningful harm? (19)
Other aspects of ambulatory management of OSA may also
prove to be appropriate for Coverage with Evidence Develop-
ment funding, and the Center for Medicare and Medicaid
Services should evaluate the indications for such funding on
an ongoing basis.
Although some progress is being made in this area, public
funding agencies and respiratory and sleep societies should
increase funding to target the research priorities identified in
this document. In particular, Workshop participants felt
strongly that the NHLBI/National Centre for Sleep Disorders
Research should prioritize an RFA on ambulatory manage-
ment of OSA.
Priority 8. Identify funding sources to support the needed
d Identify, cultivate and amalgamate funding sources for
portable monitoring research from NIH, AHRQ, CMS,
private foundations, insurance industry, and manufacturers.
d Return the investment by conducting high-quality out-
comes-based research to demonstrate the efficacy and cost
effectiveness of portable monitor testing thereby generat-
ing the data needed by these funding sources for their
decision making regarding healthcare policies and direc-
tions for future research and development.
Pressure for alternative approaches to current recommended
in-laboratory management of patients with OSA will continue
to increase given the cost of PSG and the limited number of
laboratory facilities relative to patient need. In addition,
clinical demand for more rapid access to testing increases with
the growing evidence that treatment of OSA improves func-
tional and cardiovascular outcomes. What role will portable
monitors assume and will this be based on a solid foundation
of evidence or acceptance based on unavailable resources and
familiarity of use? While attempting to validate portable
monitors, we need to understand the significant clinical
limitations of PSG and work to further standardize the
sensors, signal processing, and CPAP titration protocols used
in this so-called ‘‘gold-standard’’ test. Similar efforts are
needed to further standardize portable monitors, especially
to allow study results to be compared across monitors. More
prospective, high-quality clinical trials are needed to compare
home versus in-laboratory testing in terms of treatment out-
comes in diverse patient populations. Cost-effectiveness pro-
tocols should be routinely incorporated into these clinical
trials to collect the data that will allow for the development
of decision-analysis models based on facts rather than assump-
tions. A rational first step would be to target portable monitors
to include but not exclude the diagnosis of OSA. Current
portable-monitor technology seems to be most applicable in
American Thoracic Society Documents13
populations having a high likelihood of OSA. Alternative
approaches should also be made available to underserved
and remote populations that do not have access to gold-
standard testing. Creation of practice-based networks might
be one method for collecting the needed data in healthcare
systems that offer traditional and alternative clinical pathways
(69). Finally, one can predict that the rapid evolution and
expansion of the discipline of Sleep Medicine into a multidis-
ciplinary specialty is another potent force that will drive the
practitioners to alternative testing methods. As physicians in
Family Practice and Otolaryngology join Pulmonologists,
Psychiatrists, and Neurologists to specialize in Sleep Medicine,
the desire to test populations outside of the sleep center will
increase. High-quality research will be needed to guide the
systematic development of these alternative clinical disease-
This statement was written by the Steering Committee of
the ATS/AASM/ACCP/ERS Workshop on Research Priorities in
Ambulatory Management of Adults with Obstructive Sleep
Members of the Steering Committee
SAMUEL T. KUNA, M.D. (Chair) (ATS)
M. SAFWAN BADR, M.D. (ATS)
R. JOHN KIMOFF, M.D. (ATS)
CLETE KUSHIDA, M.D., PH.D. (AASM)
TEOFILO LEE-CHIONG, M.D. (ACCP)
PATRICK LEVY, M.D., PH.D. (ERS)
WALTER T. MCNICHOLAS, M.D. (ERS)
PATRICK J. STROLLO, Jr. M.D. (AASM)
Author Disclosure: S.T.K. reported receiving research support from Respironics.
M.S.B. reported consultancies with Inspiration and Ventus and lecture fees from
Boehringer Ingelheim, GlaxoSmithKline, and Pfizer; he received research support
from GlaxoSmithKline, Inspiration and Ventus; he had a patent pending on
noninvasive assessment of upper airways mechanics. R.J.K. reported receiving
lecture fees from GlaxoSmithKline and VitalAire; he also received research
support from the Canadian Institutes of Health Research, Fonds de la Recherche
en Sante ´ de Quebec, the Multiple Sclerosis Society of Canada, and OSR Medical.
C.K. reported receiving research support from GlaxoSmithKline, the N.I.H.,
Respironics, and XenoPort. T.L.-C. reported consulting for Covidien and research
support from Takeda; he also reported receiving textbook royalties from Oxford
University Press. P.L. reported consultancies with ResMed and Sanofi Aventis; he
holds a patent from SleepInnov Technology for Sensitest (patent #01 06389); his
dependent received research support from Novartis, ResMed, Respironics, Sanofi
Aventis, and Weinmann. W.T.M. reported receiving research support from
AstraZeneca and holding stock options in BiancaMed. P.J.S. reported serving as
an expert witness for ResMed and receiving research support from ResMed and
Respironics. (All disclosures reflect American Thoracic Society disclosure require-
ments at time of workshop and writing of manuscript.)
The other members of the Workshop were:
DONNA L. ARAND, PH.D.
JOHN ATKINSON, B.S., R.P.S.G.T.
CHARLES W. ATWOOD, M.D.
FERRAN BARBE, M.D.
KATE BENT, R.N., PH.D., C.N.S.
ELISE BERLINER, PH.D.
RICHARD BERRY, M.D.
BRIAN BOEHLECKE, M.D.
ALEJANDRO CHEDIAK, M.D.
NANCY COLLOP, M.D.
MICHAEL COPPOLA, M.D.
STEPHEN CRANE, PH.D., M.P.H.
HENRY GLICK, PH.D.
RONALD GRUNSTEIN, M.D., PH.D.
INDIRA GURUBHAGAVATULA, M.D., M.P.H.
CONOR HENEGHAN, PH.D.
THOMAS HURWITZ, M.D.
MEIR KRYGER, M.D.
DAVID LANIER, M.D.
MICHAEL LITTNER, M.D.
GERALDO LORENZI-FILHO, M.D.
KIMBERLY G. LOVE, L.T.J.G.
DOUGLAS R. MCEVOY, M.D.
MERRILL MITLER, PH.D.
ANDREW A. MONJAN, M.D., M.P.H.
MARY MORRELL, M.D.
NAIM MUNIR, M.D.
G. IRIS OBRAMS, M.D., M.P.H., PH.D.
ALLAN I. PACK, M.B., CH.B., PH.D.
THOMAS PENZEL, PH.D.
S. RAMALINGAM, M.D., M.B.A.
SUSAN REDLINE, M.D.
JOHN E. REMMERS, M.D.
CAROL L. ROSEN, M.D.
JEFFREY RUBINS, M.D.
FRANK RYAN, M.D.
PREETAM SCHRAMM, PH.D.
FREDERIC SERIES, M.D.
THOMAS SMALLING, PH.D., R.R.T.
ANN TISTHAMMER, B.S., R.R.T.
SEAN TUNIS, M.D., M.SC.
MICHAEL TWERY, PH.D.
CHRISTOPHER VALERIAN, D.O.
EDWARD WEAVER, M.D.
PHILIP WESTBROOK, M.D.*
WILLIAM WHITELAW, M.D.
* Dr. Westbrook was unable to attend the meeting
Acknowledgments: The authors thank the participants for their involvement in
the workshop. Special thanks to Graham Nelan, Director, Assembly Programs &
External Relations of the American Thoracic Society and Peter C. Johnson, M.D.,
President and CEO of Scintellix LLC.
1. Collop NA, Anderson WM, Boehlecke B, Claman D, Goldberg R,
Gottlieb DJ, Hudgel D, Sateia M, Schwab R. Clinical guidelines for
the use of unattended portable monitors in the diagnosis of obstruc-
tive sleep apnea in adult patients. Portable Monitoring Task Force of
the American Academy of Sleep Medicine. J Clin Sleep Med 2007;15:
2. Kenyon DA. Strategic planning with the Hoshin process. Quality Digest,
3. Kapur V, Strohl KP, Redline S, Iber C, O’Connor GT, Nieto FJ.
Underdiagnosis of sleep apnea syndrome in US communities. Sleep
4. Young T, Evans LK, Finn L, Palta M. Estimation of the clinically
diagnosed proportion of sleep apnea syndrome in middle-aged men
and women. Sleep 1997;20:705–706.
5. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The
occurrence of sleep-disordered breathing among middle-aged adults.
N Engl J Med 1993;328:1230–1235.
6. Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep
apnea; a population health perspective. Am J Respir Crit Care Med
7. McNicholas WT, Bonsigore MR. Sleep apnoea as an independent risk
factor for cardiovascular disease: current evidence, basic mechanisms
and research priorities. Eur Respir J 2007;29:156–178.
8. George C. Reduction in motor vehicle collisions following treatment of
sleep apnoea with nasal CPAP. Thorax 2001;56:508–512.
9. Jenkinson C, Davies RJO, Mullins R, Stradling JR. Comparison of
therapeutic and subtherapeutic nasal continuous positive airway
pressure for obstructive sleep apnoea: a randomised prospective
parallel trial. Lancet 1999;353:2100–2105.
14PROCEEDINGS OF THE AMERICAN THORACIC SOCIETYVOL 8 2011
10. Marin JM, Carrizo SJ, Vicente E, Agusti AGN. Long-term cardiovas-
cular outcomes in men with obstructive sleep apnoea-hypopnoea with
or without treatment with continuous positive airway pressure: an
observational study. Lancet 2005;365:1046–1053.
11. Weaver TE, Chasens ER. Continuous positive airway pressure
treatment for sleep apnea in order adults. Sleep Med Rev 2007;
12. Flemons WW, Douglas NJ, Kuna ST, Rodenstein DO, Wheatley J.
Access to diagnosis and treatment of patients with suspected sleep
apnea. Am J Respir Crit Care Med 2004;169:668–672.
13. Pack AI. Sleep-disordered breathing—Access is the issue. Am J Respir
Crit Care Med 2004;169:666–667.
14. Collop NA, Shepard JW, Strollo PJ. Excecutive summary on the
systematic review and practice parameters for portable monitoring
in the investigation of suspected sleep apnea in adults: co-sponsored
by the American Academy of Sleep Medicine, the American College
of Chest Physicians, and the American Thoracic Society. Am J Respir
Crit Care Med 2004;169:1160–1163.
15. Flemons WW, Littner MR, Rowley JA, Gay PC, Anderson WM, Hudgel
DW, McEvoy RD, Loube DI. Home diagnosis of sleep apnea:
a systematic review of the literature: an evidence review co-sponsored
by the American Academy of Sleep Medicine, the American College
of Chest Physicians, and the American Thoracic Society. Chest 2003;
16. Ross SD, Allen IE, Harrison KJ, Kvasz M, Connelly J, Sheinhait IA.
Systematic review of the literature regarding the diagnosis of sleep
apnea: evidence report/technology assessment No. 1. Rockville, MD:
Agency for Health Care Policy and Research; 1999.
17. Ferber R, Millman RP, Coppola M, Fleetham J, Murray CF, Iber C,
McCall V, Nino-Murcia G, Pressman M, Sanders M, et al. Portable
recording in the assessment of obstructive sleep apnea: ASDA
standards of practice. Sleep 1994;17:378–392.
18. Trikalinos TA, Ip S, Raman G, Cepeda MS, Balk EM, D’Ambrosio C, Lau J.
Technology assessment: home diagnosis of obstructive sleep apnea-hypopnea
syndrome. 2007. Accessed January 15, 2008. Available from: http://
19. Centers for Medicare & Medicaid Services. Decision memo for contin-
uous positive airway pressure (CPAP) therapy for obstructive sleep
apnea (OSA) (CAG-00093R2). 2008. Accessed June 24, 2009. Avail-
able from: http://www.cms.gov/mcd/viewdecisionmemo.asp?id5204.
20. Centers for Medicare & Medicaid Services. Decision memo for sleep
testing for obstructive sleep apnea (OSA) (CAG-00405N). 2009.
Accessed December 20, 2009. Available from: http://www.cms.hhs.
disorders in adults: recommendations for syndrome definitions and
measurement techniques in clinical research. Sleep 1999;22:667–689.
22. American Sleep Disorders Association. Practice parameters for the
indications for polysomnography and related procedures. Sleep
23. American Thoracic Society. Indications and standards for cardiopulmo-
nary sleep studies. Am Rev Respir Dis 1989;139:559–568.
24. American Thoracic Society. Indications and standards for use of nasal
continuous positive airway pressure (CPAP) in sleep apnea syn-
dromes. Am J Respir Crit Care Med 1994;150:1738–1745.
25. Kushida CA, Littner MR, Morgenthaler T, Alessi CA, Bailey D,
Coleman J, Friedman L, Hirshkowitz M, Kapen S, Kramer M, et al.
Practice parameters for indications for polysomnography and related
procedures: an update for 2005. Sleep 2005;28:499–521.
26. Hirshkowitz M, Littner MR, Kuna ST, Berry RB, Norris M, Almenoff P.
Sleep-related breathing disorders: sourcebook. Second edition.
Milwaukee, WI: HAIG; 2003.
27. Fleury B, Rakotonanahary D, Tehindrazanarivelo AD, Hausser-Hauw
C, Lebeau B. Long-term compliance to continuous positive airway
pressure therapy (nCPAP) set up during a split-night polysomno-
gram. Sleep 1994;17:512–515.
28. Strollo P, Sanders MH, Costantino JP, Walsh SK, Stiller RA, Atwood
CW. Split-night studies for the diagnosis and treatment of sleep-
disordered breathing. Sleep 1996;19:S255–S259.
29. Yamashiro Y, Kryger MH. CPAP titration for sleep apnea using a split-
night protocol. Chest 1995;107:62–66.
30. Phillips B, Collop NA, Strollo P. Controversies in sleep disorders
medicine. Pulmonologists respond. Chest 1996;110:1597–1602.
31. Peppard PE, Young T. Estimating the impact of the US obesity
epidemic on trends in prevalence of sleep disordered breathing. Sleep
32. Tachibana N, Ayas NT, White DP. A quantitative assessment of sleep
laboratory activity in the United States. AASM Bulletin 2002;9:26–28.
33. Bradley TD, Logan AG, Floras JS. Rationale and design of the
Canadian Continuous Positive Airway Pressure Trial for congestive
heart failure patients with central sleep apnea–CANPAP. Can J
34. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O’Connor GT,
Rapoport DM, Redline S, Robbins J, Samet JM, et al. The Sleep
Heart Health Study: design, rationale, and methods. Sleep 1997;20:
35. Kreiger J, Sforza E, Petiau C, Weiss T. Simplified diagnostic procedure
for obstructive sleep apnea syndrome: lower subsequent compliance
with CPAP. Eur Respir J 1998;12:776–779.
36. Hoy CJ, Vennelle M, Kingshott RN, Engleman HM, Douglas NJ. Can
intensive support improve continuous positive airway pressure use in
patients with the sleep apnea/hypopnea syndrome? Am J Respir Crit
Care Med 1999;159:1096–1100.
37. Jean Wiese H, Boethel C, Phillips B, Wilson JF, Peters J, Viggiano T.
CPAP compliance: video education may help! Sleep Med 2005;6:171–
38. Aloia MS, Arnedt JT, Stepnowsky C, Hecht J, Borrelli B. Predicting
treatment adherence in obstructive sleep apnea using principles of
behavior change. J Clin Sleep Med 2005;1:346–353.
39. Stepnowsky CJJ, Marler MR, Ancoli-Israel S. Determinants of nasal
CPAP compliance. Sleep Med 2002;3:239–247.
AI. Self-efficacy in sleep apnea; instrument development and patient
to use continuous postive airway pressure. Sleep 2003;27:1219–1232.
41. Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE,
Aptowicz C, Pack AI. Cumulative sleepiness, mood disturbance, and
psychomotor vigilance performance decrements during a week of
sleep restricted to 4–5 hours per night. Sleep 1997;20:267–277.
42. Dinges DF, Powell J. Microcomputer analyses of performance on
a portable, simple visual RT task during sustained operations. Behav
Res Methods Instrum Comput 1985;17:652–655.
43. Flemons WW, Reimer MA. Development of a disease-specific health-
related quality of life questionnaire for sleep apnea. Am J Respir Crit
Care Med 1998;158:494–503.
44. Johns MW. A new method for measuring daytime sleepiness: The
Epworth sleepiness scale. Sleep 1991;14:540–545.
45. Johns MW. Sensitivity and specificity of the multiple sleep latency test
(MSLT), the maintenance of wakefulness test and Epworth sleepiness
scale; failure of the MSLT as a gold standard. J Sleep Res 2000;9:5–11.
46. Luo X, George ML, Kakouras I, Edwards CL, Pietrobon R, Richardson
WS, Hey L. Reliability, validity, and responsiveness of the short form
12-item survey (SF-12) in patients with back pain. Spine (Phila Pa
47. Ware JE Jr, Kosinski M, Keller SD. A 12-item short-form health survey:
construction of scales and preliminary tests of reliability and validity.
Med Care 1996;34:220–233.
48. Ware JE Jr, Snow KK, Kosinski M, Gandek B. SF-36 health survey
manual and interpretation guide. Boston, MA: The Health Institute,
New England Medical Center; 1993.
49. Weaver TE, Laizner AM, Evans LK, Maislin G, Chugh DK, Lyon K,
Smith PL, Schwartz AR, Redline S, Pack AI, Dinges DF. An
instrument to measure functional status outcomes for disorders of
excessive sleepiness. Sleep 1997;20:835–843.
50. Platt AB, Kuna ST. To adhere or not to adhere–patients selectively
decide. Sleep 2009;32:583–584.
51. Maislin G, Gurubhagavatula I, Hachadoorian R, Pack F, O’Brien E,
Staley B, Dinges DF, Pack AI. Operating characteristics of the
multivariable apnea prediction index in non-clinic populations. Sleep
52. Maislin G, Pack AI, Kribbs NB, Smith PL, Schwartz AR, Kline LR,
Schwab RJ, Dinges DF. A survey screen for prediction of apnea.
53. Glick HA, Doshi JA, Sonnad SS, Polsky D. Economic evaluation in
clinical trials. Oxford: Oxford University Press; 2007.
54. Antic NA, Buchan C, Esterman A, Hensley M, Naughton MT, Rowland
S, Williamson B, Windler S, Eckermann S, McEvoy RD. A random-
ized controlled trial of nurse-led care for symptomatic moderate-
severe obstructive sleep apnea. Am J Respir Crit Care Med 2009;179:
55. Ayas NT, Fox J, Epstein L, Ryan CF, Fleetham JA. Initial use of
portable monitoring versus polysomnography to confirm obstructive
American Thoracic Society Documents15