14 - The Bariatric Quality of Life (BQL) Index_A Measure

Prévia do material em texto

538 Obesity Surgery, 15, 2005 © FD-Communications Inc. 
Obesity Surgery, 15, 538-545
Background: Quality of life (QoL) is considered to be
the true measure for the effectiveness of a surgical pro-
cedure, but there are only a few validated instruments
available for bariatric surgery.Therefore, a new disease-
specific 30-item instrument was created, which was
called Bariatric Quality of Life (BQL) questionnaire.
Methods: To validate the BQL, we studied 133
patients after 4 different types of bariatric surgery.
Initially, mean body mass index (BMI) was 47.2 ±7.6
kg/m2 and mean age was 38.8 ±11.0 years. At base-
line, and 1, 6, and 12 months after surgery, patients
filled in the BQL, the SF-12 (Short Form of SF-36
Health Survey), the GIQLI (Gastrointestinal Quality of
Life Index), and the BAROS (Bariatric Analysis and
Reporting Outcome System).
Results: Internal consistency of the BQL was found
to be good, with Cronbach’s alpha ranging between
0.71 and 0.86. Factor analyses suggested that the BQL
included a highly consistent set of QoL items and a
second part on co-morbidities and gastrointestinal
symptoms. At the 12 months follow-up, the BQL was
closely correlated to SF 12 (Pearson’s r = 0.86), GIQLI
(0.68), BAROS (0.71), and excess weight loss (0.55).
Standardized effect sizes over time were larger for the
BQL (1.39 and 1.58) than for the other instruments.
Conclusions: The BQL questionnaire is a validated
instrument ready for clinical use.
Key words: Quality of life, obesity, morbid obesity, psy-
chology, questionnaires, validation study, bariatric surgery
Introduction
As surgery is the most effective treatment for mor-
bid obesity, there is growing interest on assessing
health-related quality of life (QoL) in these
patients.1,2 Generic instruments, such as the SF-36
(Short Form 36), have difficulties in grasping all
important obesity-related QoL issues.3,4 Currently,
several validated disease-specific instruments are
available to measure QoL in obese persons.5-8 These
instruments were primarily intended to describe
patients who are under no or only conservative treat-
ment. Consequently, the instruments include global
and obesity-related domains, but fail to analyze sur-
gery-related problems. Since gastrointestinal (GI)
side-effects commonly impair QoL after bariatric
surgery, the sole assessment of obesity-related prob-
lems would mean looking only at the benefits of
weight loss, while turning a blind eye to the harms
caused. Accordingly, many obesity surgeons use
symptom specific questionnaires, especially the
Gastrointestinal Quality of Life Index (GIQLI),9,10
in addition to an assessment of obesity-related QoL.
A further important QoL issue in the obese patient
is the ameliorization of co-morbidities. Many of the
available instruments were not specifically designed
for the morbidly obese patient who undergoes
weight loss surgery. In consequence, the standard
instruments did not assess co-morbidities, because
these instruments were validated either in patients
who never had obesity-related co-morbidities or in
patients who had co-morbidities but never received
The Bariatric Quality of Life (BQL) Index: A Measure
of Well-being in Obesity Surgery Patients
Sylvia Weiner, MD1; Stefan Sauerland, MD, MPH2; Martin Fein, MD1;
Rafael Blanco, MD3; Ingmar Pomhoff, MD3; Rudolf A. Weiner, MD3
1Obesity Academy Frankfurt e. V., Frankfurt, Germany; 2Biochemical and Experimental Division,
Medical Faculty, University of Cologne, Germany; 3Department of Surgery, Sachsenhausen
Hospital, Frankfurt, Germany
Reprint requests to: Sylvia Weiner, MD, Obesity Academy
Frankfurt e. V., Eschersheimer Landstr. 248, 60320 Frankfurt a.
M., Germany. Fax: ++49 (0) 69 956 36 41 0;
e-mail: sylvia.weiner@gmx.de
a treatment of sufficient efficacy for improvement of
co-morbidities.
Only two QoL measures are currently well-
accepted: The BAROS (Bariatric Analysis and
Reporting Outcome System)11 and the IWQoL
(Impact of Weight on Quality of Life).12 The
BAROS assigns point scores to medical events (e.g.
EWL, reoperation) and QoL data, as measured with
the Moorehead-Ardelt QoL questionnaire (M-A
QoLQ I).13 Although the first version of the BAROS
has never been validated and can only be adminis-
tered after surgery, it is commonly used.14 Only
recently, therefore, a revised version has been devel-
oped (M-A QoLQ II) and validated to overcome the
shortcomings of the first version.14 The recently
shortened version of the IWQoL with 31 items, cur-
rently appears to be another very good QoL measure
for obesity surgery patients. Its psychometric prop-
erties have been extensively tested,8,15-18 but it fails
to assess GI symptoms which are common after
obesity surgery. We therefore set out to develop a
new short instrument that measures QoL in relation
to weight, weight-related co-morbidity, and surgery-
related GI symptoms.
Patients and Methods
Questionnaire Development
In the pilot phase, 50 patients were given two ques-
tionnaires, the SF-36 and the BAROS.19 All patients
were asked to provide comments on the usefulness of
the different items and suggest further issues. Based
on these suggestions and comments of various sur-
geons on face validity, questionnaire items were for-
mulated to generate a new instrument for the assess-
ment of Bariatric Quality of Life (BQL). Relying on
these data, a new questionnaire was tested on 110
patients and adapted again (Figures 1 and 2).The 19-
item selection was based on the conceptual rationale
that the following domains should be addressed: psy-
chological well-being, social functioning, physical
functioning, and problems and symptoms related to
obesity surgery and obesity-related co-morbidity.
BQL scores can range between 0 and 78 points, with
lower scores denoting worse results.
Patients
With this new instrument, we studied three different
populations. The first group consisted of 133
bariatric surgery patients (Table 1). This cohort was
The Bariatric Quality of Life (BQL) Index
Obesity Surgery, 15, 2005 539
Do you suffer from:
vomiting...................................... Yes o0 No o0.5
sour belching.............................. Yes o0 No o0.5
heartburn.................................... Yes o0 No o0.5
nausea....................................... Yes o0 No o0.5
diarrhea...................................... Yes o0 No o0.5
flatulence (gassing)..................... Yes o0 No o0.5
foul-odor feces........................... Yes o0 No o0.5
bladder problems /
urinary incontinence.................. Yes o0 No o0.5
hair loss..................................... Yes o0 No o0.5
gallstones 
(or gallbladder removed)........... Yes o0 No o0.5
diabetes..................................... Yes o0 No o0.5
high blood pressure / hyper-
tension (also if treated).............. Yes o0 No o0.5
asthma / sleep apnea................ Yes o0 No o0.5
arthrosis / joint pain................... Yes o0 No o0.5
gout............................................ Yes o0 No o0.5
others: ______________________________________
Do you take any medication 
regulary?...................................Yes o0 No o0.5
If yes, what kind of medication do you take?
- antidiabetics.............................. Yes o No o
- insulin....................................... Yes o No o
- antihypertensives...................... Yes o No o
- antidepressants......................... Yes o No o
- appetite suppressants............... Yes o No o
- diuretics.................................... Yes o No o
- pain killers................................. Yes o No o
others:
_______________________________________
Figure 1. The Bariatric Quality of Life (BQL) questionnaire (part 1). Please note that only the German version of the
BQL was validated within this study.
 
Weiner et al
540 Obesity Surgery, 15, 2005
selected from a consecutive series of 321 operated
patients. We excluded those cases in whom BQL
data was missing for more than one of the three
major time points (at baseline and after 6 and 12
months). Missingdata were caused either by loss to
follow-up or by the fact that surgery was <1 year
ago. Mean age was 38.8 (SD ±11.0) years. To
increase the generalizability of results, all four
major types of bariatric surgery were considered,
although this was a monocentric study. After sur-
gery, BMI fell from a baseline mean value of 47.2
kg/m2 (SD ±7.6) down to 36.6 (±6.2) after 6 months
and to 33.4 (±5.7) after 12 months. Average %EWL
(±SD) increased from 17.3 (±10.3) after 1 month
and 30.6 (±12.5) after 3 months to 42.5 (±16.8) and
53.6 (±21.7) after 6 and 12 months.
In addition, the BQL was administered to a con-
venience sample of 220 healthy volunteers selected
from hospital staff and friends (mean age 36.2 years,
range 17 to 72). A third group consisted of 40 patients
who were hospitalised in the wards of the division of
general surgery at the University of Wuerzburg and
were willing to fill in the questionnaires. The patients
were mainly trauma patients or admitted to hospital
because of general surgical indications (e.g. appen-
dicitis, gallstones, etc). These clinically stable
patients (mean age 50, mean BMI 24.4) were sur-
veyed twice within a 48-hour period in order to inves-
tigate the test-retest reliability of the BQL.
QoL Assessments
As recently recommended,20 we based our QoL
assessments on a combination of generic and specific
instruments. However, we were unable to administer
the IWQoL, because no German version was avail-
able at the start of this study. As collateral QoL meas-
ures we therefore used the SF-12 (a shortened version
of the SF-36),21 a translated version of the BAROS,11
and the GIQLI, which has been validated in English
and German.9,10 All questionnaire results were stan-
dardized on a 0 100 percentage scale.
Statistical Analysis
In compliance with general psychometric stan-
dards,22,23 we examined the conceptual model, reli-
ability (including internal consistency), validity (cri-
terion and construct related), and responsiveness
1. I like my weight.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
2. I can accept my weight.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
3. How is your actual quality of life?
o1 o2 o3 o4 o5
very bad bad OK good very good
4. I exercise regularly.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
5. I am participating in social activities (theatre, etc.).
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
6. I often meet friends or family.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
7. I feel excluded from social life.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
8. I feel under pressure because of my weight.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
9. Sometimes, I feel depressed.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
10. All in all, I feel satisfied in my life.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
11. I feel restricted because of my weight.
a) at home
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
right wrong
a) at work
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
right wrong
a) privately
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
right wrong
12. I feel self-confident.
o1 o2 o3 o4 o5
absolutely wrong half/half true absolutely 
wrong right
Figure 2. The Bariatric Quality of Life (BQL) question-
naire (part 2).
The Bariatric Quality of Life (BQL) Index
Obesity Surgery, 15, 2005 541
(i.e. sensitivity to change) of the BQL.
Standard statistical tests (Student’s t-test and
analysis of variance) were used for groupwise com-
parisons of continuous data. Cronbach's alpha was
calculated to determine the internal consistency of
the BQL total and subscale scores. Pearson's corre-
lation coefficient was used to examine the relation-
ship between BQL scores and other QoL instru-
ments as well as weight related parameters. For the
quantification of test retest reliability, we calculated
relative differences and constructed a Bland-Altman
plot.24 To analyze the structure of the BQL, factor
analyses were performed without rotation. Factor
analyses were repeated for the 6- and 12-month fol-
low up examination. We extracted eigenvalues >1,
in order to describe the different subdomains of the
BQL (i.e. principal component analysis). The size of
an eigenvalue describes how much variation among
the original data that this factor explains.
As measures of internal responsiveness, standard-
ized effect sizes were calculated by dividing the dif-
ference between two means (e.g. at baseline and
after 1 year) by the total group standard deviation.
Effect sizes >0.8 represent large changes.25
Significance was defined by a threshold of P< 0.05.
Results
Reliability
Test retest reliability was good with an average vari-
ation of 8.3 points (14.2%). There was no difference
between the results of the first and second test
administration, with a mean difference of 0.75
points. The Bland-Altman plot failed to show any
problems in test repeatability, although variation
was smaller when BQL scores were nearly normal
(i.e. 78 points). Among obese patients, the mean
(±SD) preoperative BQL score reached only 41.6
(±10.5) points, which was significantly lower than
the scores among healthy volunteers (65.3 ±7.2) and
other hospital patients (60.0 ±9.2). The results on
healthy volunteers are shown in Table 2.
Internal consistency of the BQL was found to be
good, with Cronbach's alpha ranging between 0.71
and 0.86 at the different time points (Table 3). As
expected, the QoL subscale of the BQL was more
consistent (range 0.80 to 0.93) than the remaining
16 items on co-morbidity and side-effects (range
0.55 to 0.69). The SF-12 displayed perfect consis-
tency with Cronbach's alpha between 0.84 and 0.91,
whereas the GIQLI and the BAROS were less con-
sistent. The QoL subscale of the BAROS was also
Table 1. Patient characteristics
LAGB VBG RYGBP BPD Total P-value
Number of patients 55 6 49 23 133 n.a.
Females (proportion) 48 (87%) 4 (66%) 39 (80%) 17 (74%) 108 (81%) 0.39a
Age (years) 37.6 ±12.2 38.2 ±14.8 41.6 ±10.4 36.1 ±7.0 38.8 ±11.0 0.16b
Preoperative BMI (kg/m2) 44.3 ±6.2 43.9 ±5.2 48.3 ±7.5 52.5 ±7.9 47.2 ±7.6 <0.001b
Data are means ± standard deviations or counts with percentages.
aby chi-square test 
bby analysis of variance
Table 2. Results obtained in a convenience sample of healthy volunteers
BMI <25 BMI 25-30 BMI >30 P-value
Number of persons 159 48 13 n.a.
Females (proportion) 110 (69%) 19 (40%) 9 (69%) 0.001a
Age (years) 34.0 ±12.5 41.7 ±14.3 43.1 ±13.2 <0.001b
BMI (kg/m2) 21.6 ±1.9 26.6 ±1.2 32.6 ±2.9 <0.001b
BQL (%) 84.6 ±9.0 83.1 ±8.9 75.8 ±8.6 0.003b
For explanations, see footnote of Table 1.
Weiner et al
542 Obesity Surgery, 15, 2005
less consistent than its BQL counterpart. Of note,
consistency results for the GIQLI and also for the
non-QoL subscale of the BQL improved in parallel
to increasing lengths of follow-up.
Principal Component Analysis
Analysis of the 6-month data showed that 9 factors
with eigenvalues >1 were necessary to describe
answer patterns. We decided to extract only 3 factors
with eigenvalues of 7.2, 2.6 and 2.0, thus explaining
39.4% of total variance. The factor structure is shown
in Table 4. Similarly, examination of the eigenvalues
and the scree plot for the 12-month data indicated that
a 3-factor solution may be appropriate, because these
factors accounted for 44.3% of the total variance.
Eigenvalues of the three factors were 8.6, 2.5, and
2.1. However, there were seven further factors with
eigenvalues between 1.7 and 1. Factor structure was
stable over time, but three items (biliary symptoms,
loss of hair, and sporting) did not load on any of the
main factors. The first and strongest factor with 13
loading items denotes general QoL, while the second
and third factor quite consistently measure co-mor-
biditiesand GI side-effects.
Construct Validity
Good correlations between the BQL and the other
instruments indicated a high construct validity for
the BQL. At baseline, BQL scores were highly cor-
related to the SF-12 (r= 0.79; P<0.0001), but the cor-
relation with the GIQLI was questionable (r= 0.13;
P=0.28). Six months after surgery, highly significant
correlations between the BQL and the SF-12 (r=
0.79), the GIQLI (r= 0.64), and the BAROS (r= 0.52)
were observed. Similar coefficients were calculated
for the 1-year follow-up data (0.86, 0.68, and 0.71).
The correlation between EWL over time and QoL
instruments is shown in Table 5. High BMI before
surgery was associated with high scores on the BQL
(r= 0.26, P=0.01) and the SF-12 (r= 0.26, P<0.01),
but not on the GIQLI (r= 0.10, P=0.38). BMI after
6 months was closely connected to the BQL (r=
0.40) and the BAROS (0.47), rather than to the SF-
12 (0.32) and the GIQLI (0.22). These correlations
to BMI improved over time, with coefficients of
0.48 and 0.46 for BQL and BAROS, and 0.37 and
0.33 for SF-12 and GIQLI after 1 year.
Sensitivity to Change
The comparison of scores between baseline and the
1-year follow-up showed a larger effect size for the
BQL (1.58) than for the SF-12 (1.14) or the GIQLI
(1.00). However, the biggest effect size of surgery
was that on BMI (2.12). At 6 months after surgery,
similar results were already apparent for the BQL
(1.39), the SF-12 (1.04), the GIQLI (0.91), and BMI
(2.24). When comparing BQL scores among patients
with EWL ≥50% (n= 64) versus <50% (n= 39), an
effect size of 1.20 was calculated. This again com-
pares well with the BAROS (1.01), the GIQLI (0.86),
and the SF-12 (0.76). Even in the normal population
(n= 220), mean BQL scores differed between normal
weight, overweight, and obese persons (Table 2).
The frequency distribution for most of the BQL
items showed no evidence of ceiling or floor effects.
Some co-morbidities were found to be quite rare
(i.e. <10% prevalence) at the 6 and 12 month fol-
Table 3. Internal consistency of the difference instruments over time (Data represent Cronbach's alpha or countages)
Number of items at baseline after 1 month after 6 months after 12 months
Number of patients 100 61 122 106
BQL 30 0.790 0.712 0.809 0.857
QoL subscale 14 0.892 0.801 0.898 0.932
non-QoL subscale 16 0.548 0.620 0.690 0.654
SF 12 12 0.882 0.835 0.892 0.913
GIQLI 36 0.686 0.774 0.880 0.854
BAROS 9 n.a. n.a. 0.622 0.710
QoL subscale 5 n.a. n.a. 0.785 0.848
non-QoL subscale 4 n.a. n.a. 0.387 0.466
The Bariatric Quality of Life (BQL) Index
Obesity Surgery, 15, 2005 543
low-up: heartburn (13 and 12 cases, respectively),
urinary incontinence (7 and 5), biliary symptoms (7
and 4), diabetes (8 and 6), pulmonary problems (13
and 9), osteoarthritis (3 and 1). However, they were
not rare at the preoperative assessment, thus justify-
ing their inclusion in the BQL.
Discussion
We have developed a new instrument, the BQL, to
assess QoL together with co-morbidities and GI
problems. We have shown that the BQL question-
naire is valid, reliable, and responsive. Although this
was not formally examined, our experience suggests
that the BQL is also very practical and user-friendly.
Filling out the questionnaire takes about 10 minutes,
and none of the patients refused to fill it out. Our
validation study used the German BQL version, but
from our initial experience with foreign patients, we
also can recommend the English version for clinical
use. Although we also found a significant difference
between obese and normal weight persons in the
control group, it should be noted that the BQL was
Table 4. Results of factor analysis with communalities for the BQL data at 6- and 12-month follow-up
Item 6 months after surgery 12 months after surgery 
Factor 1 Factor 2 Factor 3 Factor 1 Factor 2 Factor 3
Vomiting -.306 .457 .296 -.351 .305 .484
Acid reflux -.257 .510 -.139 -.104 -.126 .637
Heartburn -.320 .282 -.113 -.172 .218 .638
Nausea -.150 .433 .022 -.384 .482 .176
Diarrhea -.095 .555 -.116 -.272 .522 .182
Flatulence (gassing) -.242 .617 -.055 -.370 .447 .163
Foul-odor feces -.346 .447 -.150 -.302 .415 .026
Urinary incontinence -.462 .123 -.053 -.364 .398 -.133
Hair loss -.033 .200 -.223 -.043 .104 .170
Gallstones -.160 .213 -.020 .011 .170 .076
Diabetes -.140 .134 .507 -.253 -.035 -.000
Hypertension -.234 .232 .611 -.224 .172 -.242
Asthma/sleep apnea -.292 .452 -.095 -.226 .555 -.103
Arthrosis -.187 -.014 .495 -.254 .608 -.026
Gout -.412 .084 .111 -.283 .408 -.260
Medication -.431 .085 .519 -.296 .081 -.526
Weight satisfaction .672 .342 -.211 .749 .291 -.274
Acceptance of weight .623 .392 -.265 .696 .295 -.355
Quality of life .751 .009 .023 .809 .090 -.171
Sports .266 -.169 -.261 .351 -.083 .123
Exercise .586 .092 -.202 .718 .032 .282
Social activities .644 -.025 -.120 .622 .008 .359
Feeling excluded .511 .082 .207 .704 .152 .097
Under pressure .653 .363 -.090 .741 .312 -.177
Depressed .668 .007 .257 .688 .135 .119
Life satisfaction .763 -.017 .143 .856 .092 .018
Restriction at home .686 .082 .252 .789 -.014 .089
Restriction at work .701 .140 .360 .766 .172 .081
Restriction in privacy .761 .162 .269 .858 .087 .071
Self-confidence .692 .189 .080 .756 .190 .090
Salient factor loadings (i.e. values >0.40) are marked in bold.
Table 5. Association among different psychometric
indices and excess weight loss (EWL) after 6 and 12
months (All values are Pearson’s correlation coeffi-
cients; all correlations were significant at the 0.001 level)
EWL after 6 months EWL after 12 months
BQL 0.458 (n= 110) 0.553 (n= 103)
BAROS 0.619 (n= 83) 0.586 (n= 86)
GIQLI 0.411 (n= 100) 0.406 (n= 95)
SF-12 0.392 (n= 110) 0.421 (n= 103)
Weiner et al
544 Obesity Surgery, 15, 2005
not intended to examine obesity-related QoL in the
normal population. Furthermore, the BQL can aid
surgeons in selecting patients for bariatric surgery,
but its main purpose is to assess postoperative
patient outcomes.
Internal consistency values for the BQL exceeded
the threshold that is required for good consistency,
but the factor structure of the BQL was heteroge-
neous. Nevertheless, we have to keep in mind that
the BQL contains two parts dealing with different
domains. It was anticipated that some of the co-mor-
bidity items (e.g. loss of hair) were quite independ-
ent of GI problems and weight loss. Therefore, we
decided not to change the instrument, even if inter-
nal consistency for this part of the questionnaire was
clearly lower than for the directly QoL related
items. It also may be helpful for other purposes to
have different self-reported items on co-morbidity.
Our study has several limitations. First, the test-
retest reliability was examined only in a patient
sample that did not undergo obesity surgery.
Secondly, as in any study like this, no true gold stan-
dard was available to define the QoL construct for
bariatric surgery patients. We believe that BQL
includes all major aspects of QoL, but this cannot be
proven. Those researchers who think that a more
detailled assessment of eating behavior or psycho-
logical functioning is necessary should supplement
the BQL with other specific instruments.26-29 One of
the advantages of our study is the large patient sam-
ple, which we surveyed at more than just two time
points. An additional advantage is the inclusion of
all major types of bariatric procedures in the study,
because each of the procedures may lead to specific
problems from the surgical, gastrointestinal, or
nutritional point of view.
When comparing BQL and BAROS, it is no sur-
prise that the BAROS was more closely correlated
to EWL, because an EWL point score is part of the
BAROS. Furthermore, it is very possible that those
operations which produced greater weight loss also
caused more GI side-effects. This inverse relation-
ship between GI symptoms and EWL might have
compromised the results for the BQL, whereas the
BAROS remains unaffected. However, it is essential
also to measure the presence and severity of such
problems. Focussingonly on EWL would be a too
simplistic approach to bariatric surgery.30
Furthermore, it is interesting to compare different
time points, because the impact of GI problems is
probably higher, the longer the follow-up lasts.
Preoperatively such problems are usually mild, but
once weight loss has reached a plateau phase, QoL
will be strongly influenced by such symptoms. Our
results strengthen this hypothesis, because we
observed better consistency of BQL and GIQLI
with increasing duration of follow-up, whereas SF-
12 results were similar over time. On the other hand,
the time course of co-morbid conditions must also
be considered, even if some of these co-morbidities
are relatively rare. This heterogeneity of coexistent
illnesses makes it difficult to reduce the number of
questions in the BQL. As we considered it important
to cover all aspects of obesity, we left the first part
of the BQL unchanged. By using a yes-or-no format
for these items rather than a likert scale, the BQL
can still be filled in very quickly. Studies dealing
with side-effects could additionally use the GIQLI,
because this instrument allows a more detailled
assessment of the severity of GI symptoms.31
The clearest advantages of the BQL were found
when looking at responsiveness. It was expected
that the BQL exhibited better sensitivity to change
than the SF-12, because a disease-specific instru-
ment usually is more responsive than a generic
instrument.6,32 High responsiveness is crucial when
selecting an instrument as a outcome measure in a
clinical trial. In this regard, using the BQL in future
studies would allow researchers to find significant
results with smaller sample sizes. As the BQL has
shown good reliability, validity, and responsiveness
in this study, we would recommend our instrument
for clinical usage, either as a research tool or as an
instrument for quality assurance in obesity surgery.
References
1. Weiner S, Weiner R, Pomhoff I. Lebensqualität nach
bariatrischen Eingriffen - ein Überblick. Chir
Gastroenterol 2003; 19: 70-5.
2. Di Gregorio JM, Palkoner R. Quality of life after obesity
surgery, an evidence-based medicine literature review: how
to improve systematic searches for enhanced decision-mak-
ing and clinical outcomes. Obes Surg 2001; 11: 318-26.
3. Hörchner R, Tuinebreijer MW, Kelder PH. Quality-of-
life assessment of morbidly obese patients who have
undergone a Lap-Band® operation: 2-year follow-up
 
The Bariatric Quality of Life (BQL) Index
Obesity Surgery, 15, 2005 545
study. Is the MOS SF-36 a useful instrument to measure
quality of life in morbidly obese patients? Obes Surg
2001; 11: 212-9.
4. Doll HA, Petersen SE, Stewart-Brown SL. Obesity and
physical and emotional well-being: associations between
body mass index, chronic illness, and the physical and
mental components of the SF-36 questionnaire. Obes
Res 2000; 8: 160-70.
5. Mathias SD, Williamson CL, Colwell HH et al.
Assessing health-related quality-of-life and health state
preference in persons with obesity: a validation study.
Qual Life Res 1997; 6: 311-22.
6. Le Pen C, Lévy E, Loos F et al. “Specific” scale compared
with "generic" scale: a double measurement of the quality
of life in a French community sample of obese subjects. J
Epidemiol Community Health 1998; 52: 445-50.
7. Karlsson J, Taft C, Sjöström L et al. Psychosocial func-
tioning in the obese before and after weight reduction:
construct validity and responsiveness of the Obesity-
related Problems scale. Int J Obes 2003; 27: 617-30.
8. Kolotkin RL, Crosby RD, Kosloski KD et al.
Development of a brief measure to assess quality of life
in obesity. Obes Res 2001; 9: 102-11.
9. Eypasch E, Wood-Dauphinée S, Williams JI et al. Der
Gastrointestinale Lebens¬qualitätsindex (GLQI). Ein
klinimetrischer Index zur Befindlichkeitsmessung in der
gastroenterologischen Chirurgie. Chirurg 1993; 64: 264-74.
10. Eypasch E, Williams JI, Wood-Dauphinée S et al.
Gastrointestinal Quality of Life Index: development, val-
idation and application of a new instrument. Br J Surg
1995; 82: 216-22.
11. Oria HE, Moorehead MK. Bariatric Analysis and
Reporting Outcome System (BAROS). Obes Surg 1998;
8: 487-99.
12. Kolotkin RL, Head S, Hamilton M et al. Assessing impact
of weight on quality of life. Obes Res 1995; 3: 49-56.
13. Oria HE. The BAROS and the Moorehead-Ardelt quality
of life questionnaire. Obes Surg 2003; 13: 965.
14. Moorehead MK, Ardelt-Gattinger E, Lechner H et al.
The validation of the Moorehead-Ardelt Quality of Life
Questionnaire II. Obes Surg 2003; 13: 684-92.
15. Kolotkin RL, Crosby RD. Psychometric evaluation of
the impact of weight on quality of life-lite questionnaire
(IWQOL-lite) in a community sample. Qual Life Res
2002; 11: 157-71.
16. Kolotkin RL, Crosby RD, Williams GR et al. The rela-
tionship between health-related quality of life and
weight loss. Obes Res 2001; 9: 564-71.
17. De Zwaan M, Mitchell JE, Howell LM et al. Two meas-
ures of health-related quality of life in morbid obesity.
Obes Res 2002; 10: 1143-51.
18. Crosby RD, Kolotkin RL, Williams GR. An integrated
method to determine meaningful changes in health-
related qualty of life. J Clin Epidemiol 2004; 57: 1153-
60.
19. Weiner R, Datz M, Wagner D et al. Quality-of-life out-
come after laparoscopic adjustable gastric banding for
morbid obesity. Obes Surg 1999; 9: 539-45.
20. Korolija D, Sauerland S, Wood-Dauphinée S et al.
Evaluation of quality of life after laparoscopic surgery:
evidence-based guidelines of the European Association
for Endoscopic Surgery. Surg Endosc 2004; 18: 879-97.
21. Ware J, 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: 220-33.
22. Scientific Advisory Committee of the Medical Outcomes
Trust. Assessing health status and quality-of-life instru-
ments: attributes and review criteria. Qual Life Res
2002; 11: 193-205.
23. Eypasch E, Troidl H, Wood-Dauphinée S et al. Quality
of life and gastrointestinal surgery: a clinimetric
approach to developing an instrument for its measure-
ment. Theor Surg 1990; 5: 3-10.
24. Bland JM, Altman DG. Statistical methods for assessing
agreement between two methods of clinical measure-
ment. Lancet 1986; 1: 307-10.
25. Husted JA, Cook RJ, Farewell VT et al. Methods for
assessing responsiveness: a critical review and recom-
mendations. J Clin Epidemiol 2000; 53: 459-68.
26. Boan J, Kolotkin RL, Westman EC et al. Binge eating,
quality of life and physical activity improve after Roux-
en-Y gastric bypass for morbid obesity. Obes Surg 2004;
14: 341-8.
27. Buddeberg-Fischer B, Klaghofer R, Sigrist S et al.
Impact of psychosocial stress and symptoms on indica-
tion for bariatric surgery and outcome in morbidly obese
patients. Obes Surg 2004; 14: 361-9.
28. Kinzl JF, Traweger C, Trefalt E et al. Psychosocial con-
sequences of weight loss following gastric banding for
morbid obesity. Obes Surg 2003; 13: 105-10.
29. Vage V, Solhaug JH, Viste A et al. Anxiety, depression
and health-related quality of life after jejunoileal bypass:
a 25-year follow-up study of 20 female patients. Obes
Surg 2003; 13: 706-13.
30. Delin CR, Watts JM. Success in surgical intervention for
morbid obesity: is weight loss enough? Obes Surg 1995;
5: 189-91.
31. Arcila D, Velázquez D, Gamino R et al. Quality of life in
bariatric surgery. Obes Surg 2002; 12: 661-5.
32. Wiebe S, Guyatt G, Weaver B et al. Comparative respon-
siveness of generic and specific quality-of-life instru-
ments. J Clin Epidemiol 2003; 56: 52-60.
(Received December 20, 2004; accepted February 2, 2005)