/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97111-en.cckm

201606168

page

100

UWHC,UWMF,

Tools,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

STOP Questionnaire - A Tool to Screen Patients for Obstructive Sleep Apnea

STOP Questionnaire - A Tool to Screen Patients for Obstructive Sleep Apnea - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


Anesthesiology 2008; 108:812–21 Copyright ? 2008, the American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc.
STOP Questionnaire
A Tool to Screen Patients for Obstructive Sleep Apnea
Frances Chung, F.R.C.P.C.,* Balaji Yegneswaran, M.B.B.S.,† Pu Liao, M.D.,‡ Sharon A. Chung, Ph.D.,§
Santhira Vairavanathan, M.B.B.S.,� Sazzadul Islam, M.Sc.,� Ali Khajehdehi, M.D.,† Colin M. Shapiro, F.R.C.P.C.#
Background: Obstructive sleep apnea (OSA) is a major risk
factor for perioperative adverse events. However, no screening
tool for OSA has been validated in surgical patients. This study
was conducted to develop and validate a concise and easy-to-use
questionnaire for OSA screening in surgical patients.
Methods: After hospital ethics approval, preoperative pa-
tients aged 18 yr or older and without previously diagnosed OSA
were recruited. After a factor analysis, reliability check, and
pilot study; four yes/no questions were used to develop this
screening tool. The four questions were respectively related to
snoring, tiredness during daytime, observed apnea, and high
blood pressure (STOP). For validation, the score from the STOP
questionnaire was evaluated versus the apnea–hypopnea index
from monitored polysomnography.
Results: The STOP questionnaire was given to 2,467 patients,
27.5% classified as being at high risk of OSA. Two hundred
eleven patients underwent polysomnography, 34 for the pilot
test and 177 for validation. In the validation group, the apnea–
hypopnea index was 20 � 6. The sensitivities of the STOP
questionnaire with apnea–hypopnea index greater than 5,
greater than 15, and greater than 30 as cutoffs were 65.6, 74.3,
and 79.5%, respectively. When incorporating body mass index,
age, neck circumference, and gender into the STOP question-
naire, sensitivities were increased to 83.6, 92.9, and 100% with
the same apnea–hypopnea index cutoffs.
Conclusions: The STOP questionnaire is a concise and easy-
to-use screening tool for OSA. It has been developed and vali-
dated in surgical patients at preoperative clinics. Combined
with body mass index, age, neck size, and gender, it had a high
sensitivity, especially for patients with moderate to severe OSA.
OBSTRUCTIVE sleep apnea (OSA) is the most prevalent
breathing disturbance in sleep,
1
affecting 2–26% of the
general population depending on sex, age, and the def-
inition of criteria.
2
OSA is associated with significant
morbidity, including excessive daytime sleepiness, loud
snoring during sleep, refractory hypertension, and im-
paired quality of life. Studies have also shown that OSA
is associated with a high risk for traffic accidents and
cardiovascular disease.
3,4
The prevalence of OSA in the surgical population is
higher than in the general population and varies with
different surgical populations. In particular, approxi-
mately 7 of every 10 patients undergoing bariatric sur-
gery were found to have OSA,
5
presumably because of
the high level of obesity in this surgical population. Of
even greater concern, despite OSA being present in the
majority of patients presenting for bariatric surgery,
5,6
most cases were not diagnosed, and careful screening
was not implemented before surgery.
6
One of the barri-
ers to study the prevalence of OSA in surgical patients is
the difficulty with recruiting patients to undergo poly-
somnography before surgery. Fidan et al.
7
screened 433
surgical patients; only 18 of 41 invited patients agreed to
undergo polysomnographic testing, and 14 patients
(3.2% of all screened patients) were diagnosed with
OSA. In another study conducted by Chung et al.,
8
24%
of 305 surgical patients were classified as being at high
risk of having OSA using the Berlin questionnaire, and 13
patients were confirmed as having OSA by polysomnog-
raphy, 4.2% of the total number of patients screened.
It is estimated that nearly 80% of men and 93% of
women with moderate to severe sleep apnea are undi-
agnosed.
9
Undiagnosed OSA may pose a variety of prob-
lems for anesthesiologists. A number of case reports
have documented an increase in the incidence of post-
operative complications and deaths among patients sus-
pected of having OSA.
10
Untreated OSA patients are
known to have a higher incidence of difficult intubation,
postoperative complications, increased intensive care
unit admissions, and greater duration of hospital stay.
11–13
Identifying patients with OSA is the first step in prevent-
ing postoperative complications due to OSA.
In-laboratory polysomnography is the accepted stan-
dard for diagnosing OSA.
14
However, polysomnography
is a time-consuming and costly procedure. Further, the
growing awareness of sleep apnea has exacerbated the
long waiting list in many sleep laboratories.
15
To deal
with this issue, a number of screening questionnaires
and clinical screening models have been developed to
help identify patients with OSA.
16–25
However, a signif-
icant limitation to the aforementioned studies is that
patients were preselected because most studies were
conducted in the sleep laboratory setting.
16,18,19,26
Fur-
thermore, clinical models designed for OSA screening
usually require the assistance of a computer and may not
be suitable for clinical practice. One of the most widely
This article is featured in “This Month in Anesthesiology.”
Please see this issue of ANESTHESIOLOGY, page 5A.

* Professor, Department of Anesthesia, # Professor, Department of Psychiatry,
University of Toronto, Toronto Western Hospital, University Health Network.
† Research Fellow, ‡ Research Data Analyst, � Research Coordinator, Department
of Anesthesia, § Staff Scientist, Department of Psychiatry, Toronto Western
Hospital, University Health Network.
Received from the Department of Anesthesia, University Health Network,
University of Toronto, Toronto, Ontario, Canada. Submitted for publication
August 28, 2007. Accepted for publication December 14, 2007. Supported by a
grant from Physician Services Incorporated Foundation, Toronto, Ontario, Can-
ada, and University Health Network Foundation, Toronto, Ontario, Canada.
Address correspondence to Dr. Chung: Room 405, 2McL, Department of
Anesthesia, 399 Bathurst Street, Toronto, Ontario, Canada M5T 2S8.
frances.chung@uhn.on.ca. This article may be accessed for personal use at no
charge through the Journal Web site, www.anesthesiology.org.
Anesthesiology, V 108, No 5, May 2008 812

used questionnaires, the Berlin questionnaire, has not
been validated as a screening instrument in surgical
patients. The American Society of Anesthesiologists
(ASA) checklist, a screening instrument recommended
by the ASA Task Force on Perioperative Management of
Patients with Obstructive Sleep Apnea, has not been
validated. Besides the fact that no predictive model or
questionnaire for identifying OSA has been validated in
the surgical patient population, most questionnaires
have numerous items with a confusing scoring system.
As a result, they are not suitable for a busy clinical
setting, such as preoperative clinics.
The purpose of the study was to develop and validate
a concise and easy-to-use questionnaire for OSA screen-
ing in surgical patients.
Materials and Methods
Patient Population of the Study
The study was conducted in the preoperative clinics of
Toronto Western Hospital and Mount Sinai Hospital,
Toronto, Ontario, Canada. Ethics approval was obtained
from both institutions. Patients aged 18 yr or older who
had an ASA physical status of I–IV and were scheduled to
undergo elective procedures in general surgery, gynecol-
ogy, orthopedics, urology, plastic surgery, ophthalmol-
ogy, or neurosurgery were included in the study. Pa-
tients who were unwilling or unable to give informed
consent, patients previously diagnosed with OSA or any
other sleep breathing disorder, or patients who were
expected to have abnormal electroencephalographic find-
ings (e.g., brain tumor, epilepsy surgery, patients with deep
brain stimulator) were excluded. All patients who visited
the preoperative clinics for their scheduled surgery and
met the inclusion criteria were approached by the research
staff. After informed consent was obtained, patients were
asked to complete a questionnaire and were invited to
undergo an overnight polysomnographic study.
Development of the STOP Questionnaire
To keep the questionnaire concise and easy to use, the
questions were designed in yes/no format. Based on our
previous work with the Berlin questionnaire,
8
consensus
from a group of anesthesiologists and sleep specialists,
and a literature review, four questions (STOP Q1–4)
related to snoring, tiredness during the daytime, stopped
breathing during sleep, and hypertension were de-
signed. They were phrased in English at a fifth-grade
reading level by using the Flesch-Kincaid reading-level
determination method built into Microsoft Word (Mi-
crosoft, Redmond, WA).
To examine the association of the questions with the
underlying constructs that the questions were designed
to reflect, these four yes/no questions were combined
with items 1–10 (Berlin Q1–10) from the Berlin ques-
tionnaire to make a question list consisting of 14 ques-
tions. The question list was administered to 278 patients
to answer. Of these patients, 254 answered all of the
questions. Factor analysis with the SAS procedure Factor
was based on the responses from these 254 patients.
After a significant level of association was demonstrated,
these four yes/no questions were accepted to form the
STOP questionnaire (appendix 1). The four-item STOP
questionnaire is a self-report, forced-choice (yes/no), pa-
per-and-pencil scale that takes approximately 1 min to
complete. It consists of the following four questions:
S—“Do you snore loudly (louder than talking or loud
enough to be heard through closed doors)?” T—“Do you
often feel tired, fatigued, or sleepy during daytime?”
O—“Has anyone observed you stop breathing during
your sleep?” P—“Do you have or are you being treated
for high blood pressure?”
The STOP questionnaire was given to 592 preoperative
clinic patients as a pilot study. All patients who an-
swered the STOP questionnaire were invited to undergo
an overnight, technician-supervised polysomnographic
study. According to the interim analysis of the data from
pilot study, the cutoff point of the STOP questionnaire
had been decided and the sample size had also been
adjusted.
To check the reliability of the questionnaire, 55 pa-
tients answered the STOP questionnaire twice at differ-
ent time intervals of 1–27 days (median, 8 days). Because
these four questions reflected four different dimensions
of OSA morbidity, internal consistency checking was not
applicable.
Validation of the STOP Questionnaire
After the pilot study, 1,875 patients were screened and
asked to complete the STOP questionnaire. All patients,
regardless of their score on the STOP questionnaire,
were invited to undergo an overnight polysomnographic
study. The data from patients who completed the poly-
somnographic study was used to validate the STOP ques-
tionnaire. The predictive parameters of the STOP ques-
tionnaire versus the apnea–hypopnea index (AHI)
obtained from polysomnography in all patients of the
validation group and in subgroups with different clinical
characteristics—such as body mass index (BMI), age,
neck circumference, and gender—were analyzed. An
alternative scoring model incorporating BMI, age, neck
circumference, and gender into the STOP questionnaire,
termed the STOP-Bang (appendix 2), was also presented.
Sleep Study
A one-night, in-laboratory polysomnographic study
was conducted before surgery at Toronto Western Hos-
pital Sleep Laboratory. The result of polysomnography
was used to evaluate the validity of the STOP question-
naire. During the overnight polysomnographic study,
every patient went to bed at his or her usual bedtime.
813STOP QUESTIONNAIRE: A TOOL TO SCREEN PATIENTS FOR OSA
Anesthesiology, V 108, No 5, May 2008

Collection of continuous sleep architectural data was
accomplished using a standard electroencephalographic
montage consisting of an electroencephalogram, electro-
oculogram, submental electromyogram, and electrocar-
diogram using surface electrodes. Ancillary channels
were used to specifically record respiratory parameters,
including respiratory effort by thoracoabdominal excur-
sion, respiratory inductive plethysmography, and orona-
sal airflow by nasal airflow pressure. Oxygen saturation
was measured with a pulse oximeter.
One certified polysomnographic technologist with 10
yr of experience scored all of the polysomnographic
recordings. Her scoring was under the supervision of a
sleep physician (C.M.S.). The reports had to be assessed
and approved by the sleep physician (C.M.S.). The cer-
tified technologist was blinded to the results of the STOP
questionnaire (i.e., whether patients were at high or low
risk of having OSA) and clinical information of the pa-
tients. Sleep stages and the AHI were scored according
to standard criteria.
27,28
To avoid bias and inaccuracy
from polysomnographic scoring, the polysomnographic
recording of 10 randomly selected patients was rescored
by another experienced certified polysomnographic tech-
nologist, who was blinded to the scores of other technol-
ogist. The scores from two technologists for the same
patient were almost identical (r � 0.984, P � 0.0001).
The clinical diagnosis of OSA was defined as AHI
greater than 5 with fragmented sleep and daytime sleep-
iness. According to the American Academy of Sleep
Medicine practice guideline, the severity of OSA is de-
termined by the AHI: 5–15, mild; greater than 15–30,
moderate; greater than 30, severe.
27
After polysomnog-
raphy, patients were scheduled to meet with a sleep
specialist (C.M.S.) for follow-up assessment and clinical
management, where necessary.
Data Analysis and Statistics
Sample Size Estimation. The calculation of sample
size was performed according to the method reported
by Obuchowski.
29
Briefly, the two separate calculations
of sample size were performed based on either esti-
mated sensitivity, the precision (potential error) of sen-
sitivity, expected power, a type I error, and estimated
prevalence; or specificity, the precision of specificity,
expected power, type I error, and prevalence. The big-
ger number of the two is chosen as the sample size.
Based on the literature on the Berlin questionnaire
30–32
and the prevalence of OSA,
33,34
a sensitivity of 0.88, a
precision of 0.09, an OSA prevalence rate of 24%, a type
I error of 0.05, and a power of 0.8 were used to calcu-
lated sample size. The result was 208. The number cal-
culated based on a specificity of 0.80 was much smaller
than 208. So 208 was initially chosen as the sample size.
From the pilot study data, the sensitivity was 0.72 and
the prevalence was 0.7. If the other parameters were
kept the same, the sample size would be 137. However,
a prevalence of 0.7 is very high. It may be biased because
of the small number of patients in the pilot study. If, for
safety, 0.55 were taken as the prevalence, the adjusted
sample size would be 170.
For factor analysis, the minimum requirement for sam-
ple size is the bigger of 100 respondents or 5 times the
number of variables. In our study, we had 14 questions
(variables), so we needed at least 100 complete respon-
dents. The list of 14 questions was given to 278 patients;
254 patients who answered all of the questions were
used for the factor analysis.
Data Analysis. Data were entered into a specifically
designed Microsoft Access database and checked for
possible errors. SAS 9.1 for Windows (SAS Institute,
Cary, NC) was used for data analysis. Categorical data
were presented as frequency and percentage with 95%
confidence interval (CI). The statistical significance was
checked by chi-square test or Fisher exact test. Resam-
pling with bootstrap was used to calculate the CI of the
likelihood ratios. A logistic regression procedure was
used to calculate the odds ratio and area under the
receiver operating characteristic curve. Continuous
data were presented as mean � SD, and the Student t
test or analysis of variance was used to calculate the P
value. P � 0.05 was defined as significant. The SAS
procedure Factor was used for factor analysis. The
report from the principal components analysis with
varimax rotations was presented. Factors with an eig-
envalue greater than average were retained. Questions
with factor loading of 0.3 or greater were chosen for
interpretation of factors.
Results
Patient Screening
Over a period of 16 months at Toronto Western Hos-
pital and Mount Sinai Hospital preoperative clinics, a
total of 2,974 patients were willing to complete the
questionnaire. Of these, 2,721 patients (91.5%) an-
swered all of the items on the questionnaire completely
and had complete documentation of gender, age, and
BMI. Only these patients were included in the analysis.
Factor analysis was based on the response of 254
patients who answered all 14 questions from the STOP
and Berlin questionnaires. After the STOP questionnaire
was developed, it was administered to 2,467 patients.
The STOP questionnaire classified 27.5% of patients (679
of 2,467) as being at high risk of having OSA. Of all
patients who were invited to undergo the overnight
monitored polysomnographic testing, 416 of 2,467 pa-
tients (17%) gave consent to participate. In total, 211
patients underwent polysomnography, whereas 205
did not show up at the laboratory (fig. 1). Of 211
patients who underwent polysomnography, the first
34 patients were included in a pilot study and the
814 CHUNG ET AL.
Anesthesiology, V 108, No 5, May 2008

following 177 patients were for the validation of the
STOP questionnaire.
Age, gender, and BMI of the different patient groups
are shown in table 1. The patients who gave consent but
did not actually undergo polysomnographic testing were
younger than the group of patients who underwent the
polysomnographic study. The BMI of patients who gave
consent for polysomnography was significantly greater
than that of the patients who did not give consent for
polysomnography, regardless of whether the patient un-
derwent polysomnographic testing. Compared with the
patients who underwent polysomnographic testing,
there was a higher rate of smoking in patients who gave
consent but did not show up for the polysomnographic
testing (26.8% vs. 14.7%; P � 0.002).
Development of the STOP Questionnaire
Four yes/no questions related to snoring, tiredness
during the daytime, observed apnea during sleep, and
hypertension were combined with 10 questions from
the Berlin questionnaire and administered to 278 pa-
tients; 254 patients answered all of the questions. Demo-
graphic data are shown in table 1. Factor analysis dem-
onstrated that four underlying factors were reflected by
the 14 questions. These factors accounted for more than
95% of the total eigenvalue. Based on the category of
questions with a loading factor greater than 0.3, four
factors were identified: snoring, tiredness during day-
time, cessation of breathing during sleep, and high blood
pressure. The factor loading value for each question in
the corresponding category is shown in table 2. The
factor loading value of two questions related to falling
asleep while driving (Berlin Q8 and Q9) is very low for
all four factors.
Among the five questions related to snoring, al-
though question 1 (STOP Q1) did not have the highest
factor loading value, it still demonstrated a significant
association with snoring. Because we wanted to de-
velop a simple and easy-to-use questionnaire with
questions in yes/no format, we chose question 1 to
reflect snoring in our questionnaire. Using a similar
rationale, question 6 was incorporated to reflect day-
time tiredness. Regarding the cessation of breathing
during sleep and high blood pressure, two questions
in each category had similar factor loading values, so
questions 12 and 14 were acceptable choices to re-
flect breathing cessation during sleep and high blood
pressure in the STOP questionnaire. The final STOP
questionnaire consisted of four yes/no questions: 1, 6,
12, and 14 (appendix 1).
Pilot Study
As a pilot study, the STOP questionnaire was adminis-
tered to 592 preoperative clinic patients, and all patients
were invited to undergo polysomnography. Thirty-four
of these patients underwent the polysomnography
study. The other patients either declined to give consent
or gave consent but did not show up. Of 34 patients, 24
(70.5%) had an AHI greater than 5. According to the
analysis of data from these 34 patients, using answering
yes to two or more questions as the cutoff for the STOP
questionnaire to classify the patients as high or low
risk of having OSA demonstrated the best combination
of sensitivity and specificity. The sensitivity of the
STOP questionnaire was 72% (CI, 54.4–89.6), the
specificity was 33.3% (CI, 2.5–64.1), the positive pre-
dictive value (PPV) was 75.0% (CI, 57.7–92.3), and the
negative predictive value (NPV) was 30% (CI, 6.7–
65.3).
To check the test–retest agreement, 55 patients an-
swered the STOP questionnaire twice at a time interval
of 1–27 days (median: 8 days); 53 (96.4%) patients were
found to have the same score upon retesting with a �
coefficient of 0.923 (CI, 0.82–1.00).
Pilot & Validation: 2467 ( pilot: 592 & validation: 1875 )
Consented for PSG: 416 ( 17 % )
No Shows: 205 ( 49% ) Came for PSG: 211( 51% )
Not consented for PSG: 2051 ( 83 %)
AHI≤ 5: 55 ( 31% ) AHI> 5: 122 ( 69% )
Pilot Study: 34 ( 16% ) Validation: 177 ( 84% )
AHI ≤ 5: 9 ( 27% ) AHI> 5: 25 ( 73% )
Preoperative Patients Screened: 2721
Factor Analysis: 254
Fig. 1. Screening flow chart of surgical patients in preopera-
tive clinic. The number in the figure shows the number and
percentage of patients in the different groups. AHI � apnea–
hypopnea index; PSG � polysomnography.
Table 1. Characteristics of Screened Patients
Total
(n � 2,721)
Factor Analysis
(n � 254)
No Consent
(n � 2,051)
Consented, No Polysomnography
(n � 205)
Polysomnography Done
(n � 211)
Gender, M/F 1,305/1,416 126/128 967/1,084 106/99 106/105
Age, yr 57 � 16 56 � 17 57 � 16 54 � 13* 56 � 13
BMI, kg/m
2
28 �628�628�630� 8* 30 � 7*
Continuous data are presented as mean � SD.
* P � 0.05 compared with Factor Analysis and No Consent groups.
BMI � body mass index.
815STOP QUESTIONNAIRE: A TOOL TO SCREEN PATIENTS FOR OSA
Anesthesiology, V 108, No 5, May 2008

Table 2. Summary of the Principal Components Analysis, Varimax Rotation
Factor Loadings*
Snoring
1. (STOP Q1). Do you snore loudly (louder than talking or loud enough to be heard through closed doors)? 0.596
a. Yes
b. No
2. (Berlin Q1). Do you snore? 0.747
a. Yes
b. No
c. Don’t know
3. (Berlin Q2). Your snoring is: 0.825
a. Slightly louder than breathing
b. As loud as talking
c. Louder than talking
d. Very loud—can be heard in adjacent rooms
4. (Berlin Q3). How often do you snore? 0.795
a. Nearly every day
b. 3–4 times a week
c. 1–2 times a week
d. 1–2 times a month
e. Never or nearly never
5. (Berlin Q4). Has your snoring ever bothered other people? 0.404
a. Yes
b. No
c. Don’t know
Tiredness during daytime
6. (STOP Q2). Do you often feel tired, fatigued, or sleepy during daytime? 0.674
a. Yes
b. No
7. (Berlin Q6). How often do you feel tired or fatigued after your sleep? 0.805
a. Nearly every day
b. 3–4 times a week
c. 1–2 times a week
d. 1–2 times a month
e. Never or nearly never
8. (Berlin Q7). During your waking time, do you feel tired, fatigued, or not up to par? 0.743
a. Nearly every day
b. 3–4 times a week
c. 1–2 times a week
d. 1–2 times a month
e. Never or nearly never
Stop breathing during sleep
11. (Berlin Q5). Has anyone noticed that you quit breathing during your sleep? 0.644
a. Nearly every day
b. 3–4 times a week
c. 1–2 times a week
d. 1–2 times a month
e. Never or nearly never
12. (STOP Q4). Has anyone observed you stop breathing during your sleep? 0.606
a. Yes
b. No
High blood pressure
13. (Berlin Q10). Do you have high blood pressure? 0.947
a. Yes
b. No
c. Don’t know
14. (STOP Q3). Do you have or are you being treated for high blood pressure? 0.945
a. Yes
b. No
Questions with low factor loading for all four factors
9. (Berlin Q8). Have you ever nodded off or fallen asleep while driving a vehicle?
a. Yes
b. No
10. (Berlin Q9). How often does nodding off or falling asleep while driving a vehicle occurs?
a. Nearly every day
b. 3–4 times a week
c. 1–2 times a week
d. 1–2 times a month
e. Never or nearly never
* Factor loadings are correlations between the original questions and their factors. Factor loadings greater than 0.30 in absolute value are considered to be
significant.
816 CHUNG ET AL.
Anesthesiology, V 108, No 5, May 2008

Validation of the STOP Questionnaire
Demographic Data and Sleep Study. Table 3 shows
the demographic data of the patients who participated in
the validation study, i.e., they completed the question-
naires and underwent polysomnography. The patients
classified by the STOP questionnaire as being at high risk
of having OSA had a significantly higher frequency of
hypertension and gastroesophageal reflux disease. They
also had significantly higher average ASA physical status,
larger BMI, larger neck circumference, and higher AHI.
Using an AHI greater than 5 as the cutoff for diagnosis
of OSA, 122 patients (68.9%) were found to have OSA,
52 (29.4%) mild, 31 (17.5%) moderate, and 39 (22.0%)
severe. As shown in table 4, there were clear differences
between patients with an AHI of 5 or less and patients
with an AHI greater than 5. There was a higher percent-
age of male patients with an AHI greater than 5 (57%
male vs. 43% female; P� 0.01). The patients with an AHI
greater than 5 were almost more than 10 yr older than
patients with an AHI of 5 or less. They also had signifi-
cantly higher average ASA physical status and blood
pressure, greater BMI, and larger neck size.
Table 5 summarizes the sleep parameters in validation
patients. Compared with the patients with an AHI of 5 or
less, the patients with an AHI greater than 5 demon-
strated a significantly increased arousal index, signifi-
cantly lower minimum arterial oxygen saturation, and
significantly decreased slow wave sleep, which is con-
sistent with the sleep features of the patients with OSA.
STOP Questionnaire Effectively Identified the Pa-
tients with OSA. Predictive parameters for the STOP
questionnaire at AHI greater than 5, greater than 15, and
greater than 30 cutoff values are presented in table 6.
Using AHI greater than 5 as a cutoff value to evaluate the
STOP questionnaire, the sensitivity was 65.6%, the spec-
ificity was 60.0%, the PPV was 78.4%, and the NPV was
44.0%. The sensitivity and NPV were 74.3% and 76.0% at
AHI greater than 15. They were 79.5% and 89.3% with
AHI greater than 30 as the cutoff. This indicates that the
STOP questionnaire was more sensitive in detecting the
patients with moderate to severe OSA.
Further examination of the predictive parameters of
the STOP questionnaire in the different patient groups
demonstrates that the PPV with AHI greater than 5 as the
cutoff was greatly increased in patients with a certain
demographics: BMI greater than 35 kg/m
2
, age older
Table 4. Characteristics of Patients Grouped by AHI >5 Cutoff
AHI �5
(n � 55)
AHI �5
(n � 122)
Gender, M/F 18/37 70/52*
Age, yr 49 � 14 58 � 12*
BMI, kg/m
2
27 �631� 6*
Blood pressure
Systolic 129 � 21 142 � 18*
Diastolic 78 � 12 83 � 14*
ASA physical status, n (%)
I 7 (13) 4 (3)
II 39 (71) 62 (51)
III 9 (16) 54 (44)
IV 0 2 (2)
Average score 2.0 � 0.5 2.4 � 0.6*
Neck circumference 36 �440� 6*
AHI 3 �227� 24*
Categorical data are presented as frequency (percentage), and continuous
data are presented as mean � SD.
* P � 0.01, apnea hypopnea index (AHI) �5 vs. AHI �5.
ASA � American Society of Anesthesiologists; BMI � body mass index.
Table 3. Demographic Data of Patients for Validation of STOP
Questionnaire
STOP
Total
(n � 177)
Low Risk
(n � 75)
High Risk
(n � 102)
Gender (M/F) 88/89 38/37 50/52
Age, yr 55 � 13 54 � 15 56 � 12
BMI, kg/m
2
30 �628�631� 6*
BMI �35 kg/m
2
,n 34 10 24
Neck circumference, cm 39 �638�540� 7*
ASA physical status, n (%)
I 11 (6) 8 (11) 3 (3)
II 101 (57) 47 (63) 54 (53)
III 63 (36) 18 (24) 45 (44)
IV 2 (1) 2 (3) 0
Average score 2.3 � 0.6 2.2 � 0.7 2.4 � 0.6*
AHI 20 �612� 14 25 � 27*
Minimum SaO
2
82 � 11 84 �980� 10*
Existing conditions, n (%)
Hypertension 72 (41) 22 (29) 50 (49)*
GERD 56 (32) 13 (17) 43 (42)*
Diabetes 32 (18) 9 (12) 23 (23)
Asthma 24 (14) 9 (12) 15 (15)
Depression 11 (6) 5 (7) 6 (6)
Categorical data are presented as frequency (percentage), and continuous
data are presented as mean � SD.
* P � 0.05, high risk vs. low risk.
AHI� apnea–hypopnea index; ASA� American Society of Anesthesiologists;
BMI � body mass index; GERD � gastroesophageal reflux disease; SaO
2

arterial oxygen saturation.
Table 5. Sleep Parameters of Patients for Validation
Total
(n � 177)
AHI �5
(n � 55)
AHI �5
(n � 122)
Total sleep time, min 351 � 73 356 � 76 348 � 71
Sleep efficiency, % 78 � 14 80 � 12 77 � 15
Wake percent, % 18 � 13 16 � 12 19 � 13†
REM latency, min 124 � 78 111 � 67 131 � 82
REM percent, % 14 �815�613� 8†
Sleep stage 1, % 9 �98�810� 9
Sleep stage 2, % 49 � 13 47 � 13 49 � 13
Slow wave sleep, % 10 �713�89� 7*
AHI 19.5 � 22.9 2.5 � 1.5 27.2 � 23.8*
REM AHI 27.6 � 23.1 8.6 � 8.5 35.9 � 22.6*
Arousal index 29.4 � 18.3 23.2 � 16.7 32.1 � 18.4*
Minimum SaO
2
82 � 11 87 �780� 11*
* P � 0.05, apnea–hypopnea index (AHI) �5 vs. AHI �5. † P � 0.1 vs. AHI
�5.
REM � rapid eye movement; SaO
2
� arterial oxygen saturation.
817STOP QUESTIONNAIRE: A TOOL TO SCREEN PATIENTS FOR OSA
Anesthesiology, V 108, No 5, May 2008

than 50 yr, male gender, and neck circumference greater
than 40 cm (fig. 2). The PPV of those ranked by the STOP
questionnaire as being at high risk of having OSA was
84% in the patients with a BMI greater than 35 kg/m
2
,
86.9% in patients older than 50 yr, 87.5% in male pa-
tients, 89.7% in male patients older than 50 yr, 94.3% in
patients with neck circumference greater than 40 cm,
and 100% in male patients older than 50 yr and with a
BMI greater than 35 kg/m
2
.
STOP-Bang, an Alternative Scoring Model Com-
bining BMI, Age, Neck Circumference, and Gender
with the STOP Questionnaire. To further improve the
sensitivity of the STOP questionnaire to detect most
patients with OSA, especially moderate and severe OSA,
we developed an alternative scoring model, the STOP-
Bang (appendix 2), which incorporated BMI, age, neck
circumference, and gender into the scoring model of the
STOP questionnaire. As shown in table 7, sensitivity and
NPV are significantly increased by using the STOP-Bang.
The sensitivities at AHI cutoffs of greater than 5, greater
than 15, and greater than 30 were 83.6, 92.9, and 100%,
respectively; the corresponding NPVs were 60.8, 90.2,
and 100%.
Discussion
In this study, the STOP questionnaire was developed
and validated as an OSA screening tool for surgical pa-
tients. The STOP questionnaire is a self-administered
screening tool that includes four yes/no questions (ap-
pendix 1). The STOP questionnaire was found to have a
moderately high sensitivity and PPV at AHI greater than
5, greater than 15, and greater than 30 as cutoffs. In
patients with certain clinical characteristics, such as
male gender, age older than 50 yr, BMI greater than 35
kg/m
2
, and neck circumference greater than 40 cm, the
PPV was greatly increased. When incorporating BMI,
age, neck circumference, and gender into the STOP
scoring (STOP-Bang), the sensitivity and NPV signifi-
cantly increased. They were both more than 90% for the
patients with moderate and severe OSA.
Obstructive sleep apnea is known to diminish quality
of life
35
and is associated with many common comorbid
Table 6. Predictive Parameters for STOP Questionnaire
(n � 177)
AHI �5
Sensitivity, % 65.6 (56.4–73.9)
Specificity, % 60.0 (45.9–73.0)
PPV, % 78.4 (69.2–86.0)
NPV, % 44.0 (32.6–56.0)
Likelihood ratio 1.639 (1.172–2.385)
Odds ratio 2.857 (1.482–5.507)
Area under ROC curve 0.703
AHI �15
Sensitivity, % 74.3 (62.4–84.0)
Specificity, % 53.3 (43.4–63.0)
PPV, % 51.0 (41.3–60.7)
NPV, % 76.0 (64.8–85.1)
Likelihood ratio 1.590 (1.280–2.057)
Odds ratio 3.293 (1.707–6.352)
Area under ROC curve 0.722
AHI �30
Sensitivity, % 79.5 (63.5–90.7)
Specificity, % 48.6 (40.0–63.0)
PPV, % 30.4 (21.7–40.3)
NPV, % 89.3 (80.1–95.3)
Likelihood ratio 1.545 (1.261–2.010)
Odds ratio 3.656 (1.636–9.054)
Area under ROC curve 0.769
Data are presented as average (95% confidence interval).
AHI � apnea–hypopnea index; NPV � negative predictive value; PPV �
positive predictive value; ROC � receiver operating characteristic.
78.4
84
86.9 87.5
89.7
94.3
100
0
20
40
60
80
100
P
o
s
i
t
i
v
e

P
r
e
d
i
c
t
i
v
e

V
a
l
u
e

(
%
)
STOP
Age50
Male
BMI35
STOP
NC40
STOP
Age50
Male
STOP
Male
STOP
Age50
STOP
BMI35
STOP
Fig. 2. Positive predictive value for STOP questionnaire in pa-
tients with the different clinical characteristics. The y-axis rep-
resents positive predictive value with 95% confidence interval,
and the x-axis shows high risk of obstructive sleep apnea
ranked by STOP questionnaire in patients with different clin-
ical characteristics. BMI � body mass index; NC � neck
circumference.
Table 7. Predictive Parameters for STOP-Bang (n � 177)
AHI �5
Sensitivity, % 83.6 (75.8–89.7)
Specificity, % 56.4 (42.3–69.7)
PPV, % 81.0 (73.0–87.4)
NPV, % 60.8 (46.1–74.2)
Likelihood ratio 1.9160 (1.416–2.666)
Odds ratio 6.587 (3.217–13.489)
Area under ROC curve 0.806
AHI �15
Sensitivity, % 92.9 (84.1–97.6)
Specificity, % 43.0 (33.5–52.9)
PPV, % 51.6 (42.5–60.6)
NPV, % 90.2 (78.6–96.7)
Likelihood ratio 1.629 (1.401–1.966)
Odds ratio 9.803 (3.654–26.300)
Area under ROC curve 0.782
AHI �30
Sensitivity, % 100 (91.0–100.0)
Specificity, % 37.0 (28.9–45.6)
PPV, % 31.0 (23.0–39.8)
NPV, % 100 (93.0–100.0)
Likelihood ratio 1.586 (1.426–1.838)
Odds ratio �999.999
Area under ROC curve 0.822
Data are presented as average (95% confidence interval).
AHI � apnea–hypopnea index; NPV � negative predictive value; PPV �
positive predictive value; ROC � receiver operating characteristic.
818 CHUNG ET AL.
Anesthesiology, V 108, No 5, May 2008

conditions. Studies have documented an increased inci-
dence of coronary artery diseases, hypertension, cere-
brovascular accidents, gastroesophageal reflux disease,
congestive heart failure, and myocardial infarction in
OSA patients.
36,37
It is estimated that the average life
span of an untreated OSA patient is 58 yr, which is 20 yr
shorter than the average life span of the general popu-
lation (men, 79 yr; women, 83 yr).
38
OSA is also associ-
ated with an increased incidence of postoperative ad-
verse events.
11–13
Undiagnosed OSA in surgical patients
have a serious impact on the postoperative outcome.
Identifying patients with a high risk of OSA is the first
step for the prevention of adverse health events, adverse
perioperative outcomes, and its treatment. Screening
tools work as a filter to separate the patients with a high
risk of OSA from the patients with a low risk of OSA. A
good screening tool should be validated in the target
population against an accepted standard. It should be
easy to use and have a high sensitivity and acceptable
specificity.
Most screening tools for OSA so far have been vali-
dated in patients referred to sleep clinics or sleep labo-
ratories. Seven predictive models, based on the different
combinations of witnessed apneas, snoring, gasping,
BMI, age, gender, and hypertension were developed and
validated in the patients from sleep centers.
16,18,19,21,23,24,39,40
The Sleep Disorders Questionnaire,
41
Apnea Score,
25
and Global Sleep Assessment Questionnaire were all
tested in patients mainly from sleep centers.
42
Patients
referred to sleep centers are suspected of having sleep-
related disorders, especially OSA. They are preselected
patients. Screening tools for OSA developed and vali-
dated in the sleep center patient population cannot be
applied to other patient populations without validation
in the target patient population.
The Berlin questionnaire is one of the few questionnaires
that have been validated in primary care patients.
30
How-
ever, instead of monitored polysomnography in a sleep
laboratory, home portable sleep monitoring was used for
the validation of the Berlin questionnaire. Home portable
sleep monitoring has not been accepted as the standard for
the diagnosis of OSA. The STOP questionnaire is currently
the only questionnaire developed and validated in surgical
patients. Although there was some self-selection from the
patients’ perspective, our study was designed to include all
surgical patients in our preoperative clinics regardless of
their score of the STOP questionnaire to avoid selection
biases.
In most previous studies, reports from monitored poly-
somnography were used to validate models or question-
naires. However, the staff performing the polysomnog-
raphy and scoring the polysomnography were usually
not blinded to the score on the questionnaire.
18,19,23–25,39,40,42
This may have introduced bias into the scoring of poly-
somnography. In our study, in-laboratory polysomnogra-
phy was used to evaluate the accuracy of the STOP
questionnaire. The staff performing and scoring the poly-
somnography was blinded to the score on the STOP
questionnaire. This practice avoided bias during poly-
somnographic scoring.
Ease of use is also very important for a screening tool
in busy clinical settings. Prediction models need calcu-
lation and computer assistance. Most widely used ques-
tionnaires have a long question list with a complicated
scoring system. Although the questions are similar, the
number of questions among the different OSA screening
tools varies. For example, there are 11 multiple-choice
questions organized into three categories on the Berlin
questionnaire,
30
and 14 items under three categories on
the OSA checklist, which is recommended by the ASA.
43
Study has shown that the response rate among patients
decreases with increasing length of the questionnaire.
44
Four questions on the STOP questionnaire were de-
signed in yes/no format, and it takes less than 1 min to
finish. As a result, it had a high completion rate (91.5%)
and test–retest agreement (96.4%). The STOP question-
naire is based on questions referring to snoring, tired-
ness/sleepiness, observed stop of breathing during sleep,
and blood pressure. The alternative scoring model, the
STOP-Bang, is based on eight items including four ques-
tions in STOP questionnaire, BMI, age, neck circumfer-
ence, and gender. This creates the easy mnemonics STOP
and STOP-Bang, which may serve as useful reminders for
clinicians to use these instruments during the preoperative
screening process.
To screen patients for a disease with an important
health impact, a high sensitivity with an acceptable spec-
ificity is a basic requirement for a screening tool. The
sensitivity and specificity of OSA screening tools have
demonstrated considerable variability depending on the
screening tool, the patient population, and the definition
of OSA. For the prediction models tested in sleep center
patients, the sensitivity varied from 76% to 96%, and the
specificity ranged from 13% to 54%.
18,19,40
For the ques-
tionnaire tested in sleep center patients, the sensitivity
varied from 70% to 93%.
25,41,42
The Berlin questionnaire
is the most widely tested screening tool for OSA. The
predictive parameters of the Berlin questionnaire largely
varied in different patient populations. The sensitivity was
86% in primary care patients,
30
62.5% in patients undergo-
ing pulmonary rehabilitation, and 57–68% in sleep labora-
tory patients.
45
The wide variation in the sensitivity of the
Berlin questionnaire also indicates the risk of transferring a
screening tool between different patient populations with-
out validation in the target patient population.
In terms of the predictive parameters, the STOP ques-
tionnaire itself demonstrated a moderately high level of
sensitivity and specificity in surgical patients, and it was
more sensitive to detect the patients with moderate to
severe OSA. In the patients with certain clinical charac-
teristics, such as male gender, age older than 50 yr, BMI
greater than 35 kg/m
2
, and neck circumference greater
819STOP QUESTIONNAIRE: A TOOL TO SCREEN PATIENTS FOR OSA
Anesthesiology, V 108, No 5, May 2008

than 40 cm, the high risk of OSA ranked by the STOP
questionnaire could have a very high PPV for OSA (fig.
2). On the other hand, when incorporating BMI, age,
neck circumference, and gender (Bang) into the STOP
model (STOP-Bang); we could reach a very high level of
sensitivity and NPV, especially for the moderate and
severe OSA patients (table 6). Therefore, if a patient is
ranked as low risk of OSA by the STOP-Bang scoring
model, we would have a high confidence to exclude the
possibility that the patient would have moderate to se-
vere OSA.
This study has several limitations. In our study, the
refusal rate for polysomnography was high. Self-selec-
tion from patients may exist because patients who had
sleep symptoms might have selectively consented to the
overnight polysomnography. The high refusal rate and
dropout rate (49% of patients did not show up for their
scheduled polysomnographic testing) also indicate the
difficulty that the study faced. This may be due to the
anxiety about surgery and the need to stay one night in
the sleep laboratory. Other factors also played a role in
patient refusal and dropout, e.g., smokers and younger
patients tended not to show up for their scheduled
overnight polysomnography. The high prevalence of
OSA in the group of patients who underwent polysom-
nography may reflect this self-selection. Currently, this
tool has only been tested in surgical (noncancer) pa-
tients. It needs to be validated in the other settings.
In conclusion, the STOP questionnaire is a concise and
easy-to-use screening tool to identify patients with a high
risk of OSA. It has been validated in surgical patients at
preoperative clinics as a screening tool. The STOP-Bang
scoring model, which incorporates BMI, age, neck size,
and gender with the STOP questionnaire, has demon-
strated a higher sensitivity and NPV, especially for pa-
tients with moderate to severe OSA.
The authors thank all of the anesthesiologists at Toronto Western Hospital,
Toronto General Hospital, and Mount Sinai Hospital (Toronto, Ontario, Canada).
References
1. Kryger MH: Diagnosis and management of sleep apnea syndrome. Clin
Cornerstone 2000; 2:39–47
2. Young T, Hutton R, Finn L, Badr S, Palta M: The gender bias in sleep apnea
diagnosis: Are women missed because they have different symptoms? Arch Intern
Med 1996; 156:2445–51
3. Turkington PM, Sircar M, Allgar V, Elliott MW: Relationship between ob-
structive sleep apnoea, driving simulator performance, and risk of road traffic
accidents. Thorax 2001; 56:800–5
4. Shahar E, Whitney CW, Redline S, Lee ET, Newman AB, Javler F, George N,
O’Connor T, Boland LL, Schwartz JE, Samet JM: Sleep-disordered breathing and
cardiovascular disease: Cross-sectional results of the sleep heart health study.
Am J Respir Crit Care Med 2001; 163:19–25
5. Frey WC, Pilcher J: Obstructive sleep-related breathing disorders in patients
evaluated for bariatric surgery. Obes Surg 2003; 13:676–83
6. O’Keeffe T, Patterson EJ: Evidence supporting routine polysomnography
before bariatric surgery. Obes Surg 2004; 14:23–6
7. Fidan H, Fidan F, Unlu M, Ela Y, Ibis A, Tetik L: Prevalence of sleep apnoea
in patients undergoing operation. Sleep Breath 2006; 10:161–5
8. Chung F, Ward B, Ho J, Yuan H, Kayumov L, Shapiro C: Preoperative
identification of sleep apnea risk in elective surgical patients, using the Berlin
questionnaire. J Clin Anesth 2007; 19:130–4
9. Young T, Evans L, Finn L, Palta M: Estimation of the clinically diagnosed
proportion of sleep apnea syndrome in middle-aged men and women. Sleep
1997; 20:705–6
10. Lofsky A: Sleep apnea and narcotic postoperative pain medication: A
morbidity and mortality risk. Anesth Patient Safety Found Newsletter 2002;
17:24–5
11. Gupta RM, Parvizi J, Hanssen AD, Gay PC: Postoperative complications in
patients with obstructive sleep apnea syndrome undergoing hip or knee replace-
ment: A case-control study. Mayo Clin Proc 2001; 76:897–905
12. Gentil B, Delarminat JM, Boucherez C, Lienhart A: Difficult intubation and
obstructive sleep-apnea syndrome. Br J Anaesth 1994; 72:368
13. Liao P, Yegneswaran B, Vairavanathan S, Zaki A, Chung F: Respiratory
complications among obstructive sleep apnea (OSA) patients who underwent
surgery (abstract). Sleep 2007; 30 (suppl):0582
14. Practice parameters for the indications for polysomnography and related
procedures. Polysomnography Task Force, American Sleep Disorders Association
Standards of Practice Committee. Sleep 1997; 20:406–22
15. 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; 169:668–72
16. Deegan PC, McNicholas WT: Predictive value of clinical features for the
obstructive sleep apnoea syndrome. Eur Respir J 1996; 9:117–24
17. Ali NJ, Davies RJO, Fleetham JA, Stradling JR: Periodic movements of the
legs during sleep associated with rises in systemic blood pressure. Sleep 1991;
14:163–5
18. Crocker BD, Olson LG, Saunders NA, Hensley MJ, McKeon JL, Allen KM,
Gyulay SG: Estimation of the probability of disturbed breathing during sleep
before a sleep study. Am Rev Respir Dis 1990; 142:14–8
19. Viner S, Szalai JP, Hoffstein V: Are history and physical examination a good
screening test for sleep apnea? Ann Intern Med 1991; 115:356–9
20. Bliwise DL, Nekich JC, Dement WC: Relative validity of self-reported
snoring as a symptom of sleep apnea in a sleep clinic population. Chest 1991;
99:600–8
21. Hoffstein V, Szalai JP: Predictive value of clinical-features in diagnosing
obstructive sleep-apnea. Sleep 1993; 16:118–22
22. Kump K, Whalen C, Tishler PV, Browner I, Ferrette V, Strohl KP, Rosen-
berg C, Redline S: Assessment of the validity and utility of a sleep-symptom
questionnaire. Am J Respir Crit Care Med 1994; 150:735–41
23. Flemons WW, Whitelaw WA, Brant R, Remmers JE: Likelihood ratios for a
sleep apnea clinical prediction rule. Am J Respir Crit Care Med 1994; 150:
1279–85
24. Dealberto MJ, Ferber C, Garma L, Lemoine P, Alperovitch A: Factors
related to sleep apnea syndrome in sleep clinic patients. Chest 1994; 105:1753–8
25. Kapuniai LE, Andrew DJ, Crowell DH, Pearce JW: Identifying sleep apnea
from self-reports. Sleep 1988; 11:430–6
26. Davies RJ, Ali NJ, Stradling JR: Neck circumference and other clinical
features in the diagnosis of the obstructive sleep apnoea syndrome. Thorax 1992;
47:101–5
27. Sleep-related breathing disorders in adults: Recommendations for syn-
drome definition and measurement techniques in clinical research. The Report of
an American Academy of Sleep Medicine Task Force. Sleep 1999; 22:667–89
28. Meoli AL, Casey KR, Clark RW, Coleman JA Jr, Fayle RW, Troell RJ, Iber C:
Hypopnea in sleep-disordered breathing in adults. Sleep 2001; 24:469–70
29. Obuchowski NA: Sample size calculations in studies of test accuracy. Stat
Methods Med Res 1998; 7:371–92
30. Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP: Using the Berlin
questionnaire to identify patients at risk for the sleep apnea syndrome. Ann
Intern Med 1999; 131:485–91
31. Strauss RS, Browner WS: Risk for obstructive sleep apnea. Ann Intern Med
2000; 132:758–9
32. Gami AS, Pressman G, Caples SM, Kanagala R, Gard JJ, Davison DE, Malouf
JF, Ammash NM, Friedman PA, Somers VK: Association of atrial fibrillation and
obstructive sleep apnea. Circulation 2004; 110:364–7
33. Netzer NC, Hoegel JJ, Loube D, Netzer CM, Hay B, Alvarez-Sala R, Strohl
KP: Prevalence of symptoms and risk of sleep apnea in primary care. Chest 2003;
124:1406–14
34. Kushida CA, Nichols DA, Simon RD, Young T, Grauke JH, Britzmann JB,
Hyde PR, Dement WC: Symptom-based prevalence of sleep disorders in an adult
primary care population. Sleep Breath 2000; 4:9–14
35. Bennett LS, Barbour C, Langford B, Stradling JR, Davies RJO: Health status
in obstructive sleep apnea: Relationship with sleep fragmentation and daytime
sleepiness, and effects of continuous positive airway pressure treatment. Am J
Respir Crit Care Med 1999; 159:1884–90
36. Dincer HE, O’Neill W: Deleterious effects of sleep-disordered breathing on
the heart and vascular system. Respiration 2006; 73:124–30
37. Ing AJ, Ngu MC, Breslin ABX: Obstructive sleep apnea and gastroesopha-
geal reflux. Am J Med 2000; 108:120–5
38. Young T, Finn L: Epidemiological insights into the public health burden of
sleep disordered breathing: Sex differences in survival among sleep clinic pa-
tients. Thorax 1998; 53 (suppl 3):S16–19
39. Maislin G, Pack AI, Kribbs NB, Smith PL, Schwartz AR, Kline LR, Schwab
RJ, Dinges DF: A survey screen for prediction of apnea. Sleep 1995; 18:158–66
820 CHUNG ET AL.
Anesthesiology, V 108, No 5, May 2008

40. Rowley JA, Aboussouan LS, Badr MS: The use of clinical prediction formu-
las in the evaluation of obstructive sleep apnea. Sleep 2000; 23:929–38
41. Douglass AB, Bornstein R, Nino-Murcia G, Keenan S, Miles L, Zarcone VP
Jr, Guilleminault C, Dement WC: The Sleep Disorders Questionnaire: I. Creation
and multivariate structure of SDQ. Sleep 1994; 17:160–7
42. Roth T, Zammit G, Kushida C, Doghramji K, Mathias SD, Wong JM, Buysse
DJ: A new questionnaire to detect sleep disorders. Sleep Med 2002; 3:99–108
43. Gross JB, Bachenberg KL, Benumof JL, Caplan RA, Connis RT, Cote CJ,
Nickinovich DG, Prachand V, Ward DS, Weaver EM, Ydens L, Yu S: Practice
guidelines for the perioperative management of patients with obstructive sleep
apnea: A report by the American Society of Anesthesiologists Task Force on
Perioperative Management of Patients with Obstructive Sleep Apnea. ANESTHESI-
OLOGY 2006; 104:1081–93
44. Iglesias C, Torgerson D: Does length of questionnaire matter? A random-
ised trial of response rates to a mailed questionnaire. J Health Serv Res Policy
2000; 5:219–21
45. Ahmadi N, Chung SA, Gibbs A, Shapiro CM: The Berlin questionnaire for
sleep apnea in a sleep clinic population: Relationship to polysomnographic
measurement of respiratory disturbance. Sleep Breath 2008; 12:39–45
Appendix 1: STOP Questionnaire
Height _____ inches/cm Weight _____ lb/kg
Age _____ Male/Female BMI _____
Collar size of shirt: S, M, L, XL, or _____ inches/cm
Neck circumference* _____ cm
1. Snoring
Do you snore loudly (louder than talking or loud enough to be heard
through closed doors)?
Yes No
2. Tired
Do you often feel tired, fatigued, or sleepy during daytime?
Yes No
3. Observed
Has anyone observed you stop breathing during your sleep?
Yes No
4. Blood pressure
Do you have or are you being treated for high blood pressure?
Yes No
* Neck circumference is measured by staff.
High risk of OSA: answering yes to two or more questions
Low risk of OSA: answering yes to less than two questions
Appendix 2: STOP-Bang Scoring Model
1. Snoring
Do you snore loudly (louder than talking or loud enough to be heard
through closed doors)?
Yes No
2. Tired
Do you often feel tired, fatigued, or sleepy during daytime?
Yes No
3. Observed
Has anyone observed you stop breathing during your sleep?
Yes No
4. Blood pressure
Do you have or are you being treated for high blood pressure?
Yes No
5. BMI
BMI more than 35 kg/m
2
?
Yes No
6. Age
Age over 50 yr old?
Yes No
7. Neck circumference
Neck circumference greater than 40 cm?
Yes No
8. Gender
Gender male?
Yes No
High risk of OSA: answering yes to three or more items
Low risk of OSA: answering yes to less than three items
821STOP QUESTIONNAIRE: A TOOL TO SCREEN PATIENTS FOR OSA
Anesthesiology, V 108, No 5, May 2008