Two-Year Changes in Health-Related Quality of Life in Gastric Bypass Patients Compared to Two Severely Obese Groups
Background
Few weight loss surgery trials evaluate changes in health-related quality of life (HRQOL) relative to obese individuals not participating in weight loss interventions.
Objectives
In a prospective study we evaluated two-year changes in HRQOL in gastric bypass patients compared to two separate severely obese groups not undergoing weight loss intervention.
Setting
Bariatric surgery practice.
Methods
421 surgery patients (GBP) were compared with 405 individuals who sought but did not have bariatric surgery (No GBP) and 319 population-based obese individuals (Pop OB) on obesity-specific (IWQOL-Lite) and general (SF-36) HRQOL at baseline and two-years.
Results
Weight loss was 34.2% for GBP, 1.4% for No GBP and a gain of 0.5% for Pop OB. Both measures of HRQOL showed greater improvements for the GBP group (p<0.001), even after controlling for baseline differences. Effect sizes for changes in physical and weight-related HRQOL were very large for GBP, but small to medium for the two comparison groups. Effect sizes for changes in psychosocial aspects of HRQOL were moderate to very large for the GBP, but small for comparison groups. Ninety-seven percent of surgery patients reported meaningful improvements on IWQOL-Lite total score, versus 43% of the No GBP group and 30% of the Pop OB group (p<0.001).
Conclusions
Dramatic improvements occurred in both obesity-specific and physical health-related quality of life for gastric bypass surgery patients two-years post-surgery compared to two severely obese groups not enrolled in weight loss intervention. Changes in the psychosocial aspects of HRQOL were medium to large and more variable across domains.
Introduction
Weight loss surgery has been associated with major and durable reductions in excess body weight1, 2, total mortality3, 4, co-morbid conditions1, 5, 6, and improvements in health-related quality of life (HRQOL)7–9. A number of controlled trials have been designed to compare various bariatric surgery procedures with one another10–13, but few have investigated HRQOL outcomes in patients receiving weight loss surgery relative to non-surgically treated obese individuals.
The Swedish Obese Subjects (SOS) study, a prospective nonrandomized intervention trial, compared obese individuals (BMI ≥ 34) undergoing three types of bariatric surgery with non-surgically treated individuals undergoing conventional weight loss treatment14. HRQOL, assessed using a battery of general and obesity-specific measures, improved dramatically in surgical patients, while only minor fluctuations in HRQOL scores were observed in controls. In a randomized controlled trial by O’Brien and colleagues15 patients with BMI = 30–35 were randomly assigned to either laparoscopic adjustable gastric banding or a very-low calorie diet that included pharmacotherapy and lifestyle change. At two-years patients receiving surgery reported improved general HRQOL in all eight domains of HRQOL, whereas non-surgical patients reported improvements in three domains (physical functioning, vitality, and mental health).
The present study was a prospective two-year, cohort study comparing patients who had Roux-en-Y gastric bypass surgery with two groups of severely obese individuals who did not receive weight loss intervention ─ (1) individuals who sought gastric bypass surgery but did not have the surgery, and (2) severely obese community subjects derived from a population study. The objective was to evaluate two-year changes in HRQOL in the gastric bypass patients relative to the two comparison groups, thus adding to a sparse literature of prospective trials investigating HRQOL outcomes in gastric surgery patients versus obese individuals not enrolled in weight loss interventions.
Materials and Methods
Participants
Study participants were recruited for the Utah Obesity Study16, an ongoing, prospective study comparing gastric bypass surgery patients with individuals who sought but did not have gastric bypass surgery as well as severely obese subjects randomly chosen from a population database representing over one million first-degree relatives from 120,000 Utah families17–19. The sample for the current study consisted of 421 gastric bypass surgery patients, 405 individuals who sought but did not have surgery, and 319 severely obese population-based subjects. Patients seeking gastric bypass surgery were recruited from a partnership of bariatric surgeons of the Rocky Mountain Associated Physicians (Salt Lake City, UT). Gastric bypass participants had a reported BMI ≥ 40 kg/m2 or ≥ 35 kg/m2 and two comorbidities, which primarily included cardiovascular, sleep apnea, uncontrolled type 2 diabetes or weight induced physical problems that interfered with daily functioning. Exclusion criteria for all study participants included: previous gastric surgery for weight loss, gastric or duodenal ulcers in the previous six months, active cancer within the past five years (except for non-melanoma skin cancer), myocardial infarction in the previous six months, and history of alcohol or narcotic abuse. Participants included in the current paper were all those who had a valid HRQOL assessment at baseline.
Procedures
This study was approved by the University of Utah IRB. Informed consent was obtained for all participants. Upon initial evaluation and again at two-years, participants’ heights and weights were obtained by the study personnel. Body mass index (BMI) was calculated as kilograms/meters2. Participants also completed questionnaires at baseline and two-year follow-up that included demographic information and two measures of HRQOL.
Measures
Impact of Weight on Quality of Life-Lite (IWQOL-Lite)
The IWQOL-Lite20 is a 31-item measure of weight-related quality of life. There are five domain scores (Physical Function, Self-Esteem, Sexual Life, Public Distress and Work) and a Total score. Scores for all domains and Total score range from 0–100, with lower scores indicating greater impairment. The IWQOL-Lite has demonstrated excellent reliability and validity20, 21.
Medical Outcomes Study Short-Form-36 (SF-36)
The SF-3622 is a 36-item measure of general HRQOL, consisting of eight subscales (Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional, and Mental Health) and two summary scores [Physical Component Summary (PCS) and Mental Component Summary (MCS)]. The two summary scores represent independent (orthogonal) indices based on factor analysis of subscale scores using the Medical Outcomes Study data22. Scores on all subscales and PCS and MCS range from 0 to 100, where 100 represents the best HRQOL. Scores for PCS and MCS are norm-based, with a mean of 50 and a standard deviation of 10. Estimates of internal consistency for the SF-36 typically have exceeded 0.80 for all subscales across diverse patient groups23, 24.
Statistical Analyses
Groups were compared on baseline characteristics using chi-square analysis for categorical measures and analysis of variance (ANOVA) for continuous measures using a two-tailed alpha of .05. Post-hoc tests for categorical measures were based upon pairwise Bonferroni-corrected25 chi-square comparisons (alpha =.05/3 = .017) and for continuous measures were based on Tukey’s honestly significant difference (hsd)26 to control for multiple comparisons. Groups were compared on baseline HRQOL scores using analysis of covariance (ANCOVA), controlling for BMI and gender. Groups were compared on changes in HRQOL at two-year follow-up using ANCOVA, controlling for baseline score, gender, and baseline BMI using a Bonferroni-corrected two-tailed alpha of .003 (.05/16) to control for multiple comparisons. Pair-wise post-hoc comparisons were based upon covariate-adjusted Bonferroni-corrected contrasts using a significance of .001 (.003/3). Within group effect sizes were calculated as the difference between scores at endpoint and baseline divided by the baseline standard deviation. Additionally, we computed the number and percent of participants in each group that demonstrated meaningful improvement in IWQOL-Lite total score using the algorithm described by Crosby and colleagues27. Based on this algorithm, scores have shown meaningful improvement if they have increased 7–12 points, depending upon baseline severity. The percent of patients demonstrating meaningful improvement/no change /deterioration was compared across groups using chi-square analysis. All analyses were conducted using SPSS Version 16.0.128.
Participants
Study participants were recruited for the Utah Obesity Study16, an ongoing, prospective study comparing gastric bypass surgery patients with individuals who sought but did not have gastric bypass surgery as well as severely obese subjects randomly chosen from a population database representing over one million first-degree relatives from 120,000 Utah families17–19. The sample for the current study consisted of 421 gastric bypass surgery patients, 405 individuals who sought but did not have surgery, and 319 severely obese population-based subjects. Patients seeking gastric bypass surgery were recruited from a partnership of bariatric surgeons of the Rocky Mountain Associated Physicians (Salt Lake City, UT). Gastric bypass participants had a reported BMI ≥ 40 kg/m2 or ≥ 35 kg/m2 and two comorbidities, which primarily included cardiovascular, sleep apnea, uncontrolled type 2 diabetes or weight induced physical problems that interfered with daily functioning. Exclusion criteria for all study participants included: previous gastric surgery for weight loss, gastric or duodenal ulcers in the previous six months, active cancer within the past five years (except for non-melanoma skin cancer), myocardial infarction in the previous six months, and history of alcohol or narcotic abuse. Participants included in the current paper were all those who had a valid HRQOL assessment at baseline.
Procedures
This study was approved by the University of Utah IRB. Informed consent was obtained for all participants. Upon initial evaluation and again at two-years, participants’ heights and weights were obtained by the study personnel. Body mass index (BMI) was calculated as kilograms/meters2. Participants also completed questionnaires at baseline and two-year follow-up that included demographic information and two measures of HRQOL.
Measures
Impact of Weight on Quality of Life-Lite (IWQOL-Lite)
The IWQOL-Lite20 is a 31-item measure of weight-related quality of life. There are five domain scores (Physical Function, Self-Esteem, Sexual Life, Public Distress and Work) and a Total score. Scores for all domains and Total score range from 0–100, with lower scores indicating greater impairment. The IWQOL-Lite has demonstrated excellent reliability and validity20, 21.
Medical Outcomes Study Short-Form-36 (SF-36)
The SF-3622 is a 36-item measure of general HRQOL, consisting of eight subscales (Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional, and Mental Health) and two summary scores [Physical Component Summary (PCS) and Mental Component Summary (MCS)]. The two summary scores represent independent (orthogonal) indices based on factor analysis of subscale scores using the Medical Outcomes Study data22. Scores on all subscales and PCS and MCS range from 0 to 100, where 100 represents the best HRQOL. Scores for PCS and MCS are norm-based, with a mean of 50 and a standard deviation of 10. Estimates of internal consistency for the SF-36 typically have exceeded 0.80 for all subscales across diverse patient groups23, 24.
Impact of Weight on Quality of Life-Lite (IWQOL-Lite)
The IWQOL-Lite20 is a 31-item measure of weight-related quality of life. There are five domain scores (Physical Function, Self-Esteem, Sexual Life, Public Distress and Work) and a Total score. Scores for all domains and Total score range from 0–100, with lower scores indicating greater impairment. The IWQOL-Lite has demonstrated excellent reliability and validity20, 21.
Medical Outcomes Study Short-Form-36 (SF-36)
The SF-3622 is a 36-item measure of general HRQOL, consisting of eight subscales (Physical Functioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional, and Mental Health) and two summary scores [Physical Component Summary (PCS) and Mental Component Summary (MCS)]. The two summary scores represent independent (orthogonal) indices based on factor analysis of subscale scores using the Medical Outcomes Study data22. Scores on all subscales and PCS and MCS range from 0 to 100, where 100 represents the best HRQOL. Scores for PCS and MCS are norm-based, with a mean of 50 and a standard deviation of 10. Estimates of internal consistency for the SF-36 typically have exceeded 0.80 for all subscales across diverse patient groups23, 24.
Statistical Analyses
Groups were compared on baseline characteristics using chi-square analysis for categorical measures and analysis of variance (ANOVA) for continuous measures using a two-tailed alpha of .05. Post-hoc tests for categorical measures were based upon pairwise Bonferroni-corrected25 chi-square comparisons (alpha =.05/3 = .017) and for continuous measures were based on Tukey’s honestly significant difference (hsd)26 to control for multiple comparisons. Groups were compared on baseline HRQOL scores using analysis of covariance (ANCOVA), controlling for BMI and gender. Groups were compared on changes in HRQOL at two-year follow-up using ANCOVA, controlling for baseline score, gender, and baseline BMI using a Bonferroni-corrected two-tailed alpha of .003 (.05/16) to control for multiple comparisons. Pair-wise post-hoc comparisons were based upon covariate-adjusted Bonferroni-corrected contrasts using a significance of .001 (.003/3). Within group effect sizes were calculated as the difference between scores at endpoint and baseline divided by the baseline standard deviation. Additionally, we computed the number and percent of participants in each group that demonstrated meaningful improvement in IWQOL-Lite total score using the algorithm described by Crosby and colleagues27. Based on this algorithm, scores have shown meaningful improvement if they have increased 7–12 points, depending upon baseline severity. The percent of patients demonstrating meaningful improvement/no change /deterioration was compared across groups using chi-square analysis. All analyses were conducted using SPSS Version 16.0.128.
Results
Demographic and Weight Characteristics
Table 1 presents baseline demographic and weight characteristics by group. Those who sought but did not have surgery were less likely to be married than surgery patients (52.3% vs. 61.8%), but did not differ from surgery patients in terms of other demographic and weight characteristics. In contrast, obese population-based subjects were older, weighed less, and had a higher proportion of males and Caucasians than both surgery patients and those who sought but did not have surgery.
Table 1
Baseline characteristics by group.
| Characteristic | Gastric Bypass Surgery (n = 421) | Seeking But Did Not Have Surgery (n = 405) | Population-based Obese (n = 319) | Significance |
|---|---|---|---|---|
| Female gender (n, %) | 355 (84.3)a | 344 (84.9)a | 243 (76.2)b | χ(2) = 11.32, p = .003 |
| Age, yrs. (mean, SD) | 42.1 ± 10.8a | 42.5 ± 11.4a | 48.8 ± 10.9b | F(2, 1142) = 39.98, p < .001 |
| Married (n, %) | 260 (61.8)a | 212 (52.3)b | 222 (69.6)a | χ(2) = 22.60, p < .001 |
| Caucasian race (n, %) | 376 (89.3)a | 367 (90.6)a | 310 (97.2)b | χ(2) = 16.74, p < .001 |
| Education, yrs. (mean, SD) | 14.1 ± 2.2 | 13.9 ± 2.4 | 13.8 ± 2.3 | F(2, 1073) = 1.49, p = .227 |
| Weight, lbs. (mean, SD) | 292.6 ± 61.1a | 284.9 ± 57.0a | 269.8 ± 54.8b | F(2, 1142) = 14.04, p < .001 |
| BMI, kg/m (mean, SD) | 47.3 ± 7.8a | 46.5 ± 7.7a | 43.9 ± 6.4b | F(2, 1142) = 20.35, p < .001 |
Cells without common superscripts are different p < .05 based upon Tukey’s hsd (F test) or Bonferroni (chi-square) correction.
Baseline HRQOL
Baseline comparisons between groups in IWQOL-Lite and SF-36 scores are presented in Table 2. Surgery patients were more impaired than those who sought but did not have surgery on physical HRQOL (including Physical Function and Sexual Life from the IWQOL-Lite and Physical Functioning, Role Physical, and PCS from the SF-36) but did not differ on psychosocial components. In contrast, surgery patients were more impaired than population-based obese subjects on all scales and total scores from both the IWQOL-Lite and the SF-36.
Table 2
Baseline IWQOL-Lite and SF-36 scores by group adjusted for BMI and gender.
| IWQOL-Lite Score | Gastric Bypass Surgery (n = 382–416) | Seeking But Did Not Have Surgery (n = 360–400) | Population-based Obese (n = 286–314) | Significance |
|---|---|---|---|---|
| Physical Function | 26.9 ± 18.5a | 31.1 ± 20.8b | 48.2 ± 21.2c | F(2, 1124) = 81.07, p < .001 |
| Self-Esteem | 22.6 ± 20.8a | 24.3 ± 22.0a | 45.7 ± 25.9b | F(2, 1125) = 95.20, p < .001 |
| Sexual Life | 39.5 ± 30.4a | 46.6 ± 32.5b | 65.7 ± 29.7c | F(2, 1023) = 46.75, p < .001 |
| Public Distress | 36.3 ± 23.2a | 39.4 ± 24.9a | 62.0 ± 24.8b | F(2, 1124) = 83.33, p < .001 |
| Work | 46.2 ± 26.5a | 47.5 ± 27.1a | 68.0 ± 23.8b | F(2, 1113) = 56.08, p < .001 |
| IWQOL-Lite Total | 31.5 ± 16.5a | 34.8 ± 18.5a | 54.5 ± 19.6c | F(2, 1125) = 127.02, p < .001 |
| SF-36 Score | ||||
| Physical Functioning | 37.1 ± 21.8a | 41.8 ± 24.5b | 56.6 ± 23.2c | F(2, 1113) = 43.84, p < .001 |
| Role Physical | 32.7 ± 35.8a | 39.7 ± 39.1b | 58.6 ± 38.7c | F(2, 1121) = 32.35, p < .001 |
| Bodily Pain | 41.0 ± 21.6a | 41.7 ± 21.0a | 56.5 ± 22.2b | F(2, 1121) = 41.11, p < .001 |
| General Health | 43.3 ± 14.4a | 44.5 ± 15.2a | 54.9 ± 16.3b | F(2, 1121) = 49.80, p < .001 |
| Vitality | 25.9 ± 17.4a | 28.1 ± 19.0a | 41.6 ± 20.7b | F(2, 1121) = 58.04, p < .001 |
| Social Functioning | 48.8 ± 25.6a | 51.5 ± 26.9a | 72.5 ± 24.5b | F(2, 1121) = 72.47, p < .001 |
| Role Emotional | 47.6 ± 43.1a | 45.3 ± 42.9a | 65.4 ± 40.0b | F(2, 1121) = 20.05, p < .001 |
| Mental Health | 59.2 ± 19.2a | 57.1 ± 20.9a | 70.4 ± 18.9b | F(2, 1121) = 38.31, p < .001 |
| PCS | 31.5 ± 8.8a | 33.7 ± 9.2b | 39.1 ± 9.6c | F(2, 1113) = 42.90, p < .001 |
| MCS | 41.4 ± 11.6a | 40.4 ± 12.1a | 47.8 ± 11.5b | F(2, 1113) = 37.13, p < .001 |
Cells represent unadjusted means ± SD.
Cells without common superscripts are different, p < .017 based upon covariate adjusted Bonferroni contrasts.
Two-Year Follow-up Rates
Two-year valid HRQOL assessments (IWQOL-Lite and/or SF-36) were obtained for 308 (73.2%) surgery patients, 253 (62.5%) individuals who sought but did not have surgery, and 272 (85.3%) obese community participants (χ(2) = 36.42, p < .001). Bonferroni-corrected post hoc comparisons revealed that follow-up rates were significantly higher for obese community participants than surgery patients, which were in turn significantly higher than follow-up rates for those who sought but did not have surgery. Participants who failed to complete the two-year assessment were younger, less likely to be married, more likely to be a minority, had fewer years of education, higher BMI, and reported poorer quality of life at baseline on most IWQOL-Lite (all except Sexual Life) and SF-36 (all except Physical Function, Role Physical, Vitality, and PCS) scales.
Two-Year Weight Loss
The percent weight loss at two years among surgery patients averaged 34.2% (SD = 10.0, range = 65.4% loss to 1.0% gain), compared to 1.4% (SD = 8.6, 31.7% loss to 20.3% gain) for individuals who sought but did not have surgery and a 0.5% gain (SD = 9.3, 57.4% loss to 27.0% gain) for obese population-based participants (F(2,802) = 1235.54, p < .001).
Two-Year Changes in HRQOL
Two-year changes in IWQOL-Lite and SF-36 scores by group are presented in Table 3. Gastric bypass patients showed significantly greater improvement than both those who sought but did not have surgery and population-based obese individuals on all measures. HRQOL changes in the group who sought and did not have surgery were comparable to those in the population-based obese group except for IWQOL-Lite Sexual Life and Work, where greater improvement was observed in the group that sought but did not have surgery. Within-group effect size changes in the surgery group for the IWQOL-Lite ranged from 1.73 (Work) to 3.31 (total score) and for the SF-36 ranged from .60 (Role Emotional) to 2.04 (Physical Functioning).
Table 3
Two-year changes in IWQOL-Lite and SF-36 scores by group.
| IWQOL-Lite Score | Gastric Bypass Surgery (n = 260298) | Seeking But Did Not Have Surgery (n = 203244) | Population-based Obese (n = 226262) | Significance | |
|---|---|---|---|---|---|
| Physical Function | Mean, SD | 58.8 ± 20.6a | 13.8 ± 25.8b | 4.9 ± 17.6b | F(2, 798) = 463.02, p < .001 |
| Effect Size | 3.13 | 0.65 | 0.24 | ||
| Self-Esteem | Mean, SD | 56.7 ± 26.5a | 13.1 ± 23.1b | 8.3 ± 18.4b | F(2, 798) = 340.80, p < .001 |
| Effect Size | 2.75 | 0.58 | 0.34 | ||
| Sexual Life | Mean, SD | 47.8 ± 31.8a | 7.2 ± 29.4b | 5.0 ± 26.4c | F(2, 687) = 148.44, p < .001 |
| Effect Size | 1.62 | 0.22 | 0.17 | ||
| Public Distress | Mean, SD | 52.2 ± 23.4a | 11.6 ± 26.0b | 4.6 ± 19.3b | F(2, 797) = 335.92, p < .001 |
| Effect Size | 2.25 | 0.45 | 0.19 | ||
| Work | Mean, SD | 44.0 ± 26.8a | 9.5 ± 25.9b | 4.6 ± 20.4c | F(2, 789) = 2 1 0.12, p < .001 |
| Effect Size | 1.73 | 0.35 | 0.19 | ||
| IWQOL-Lite Total | Mean, SD | 54.0 ± 19.4a | 12.0 ± 21.7b | 5.5 ± 15.1b | F(2, 798) = 468.95, p < .001 |
| Effect Size | 3.31 | 0.63 | 0.29 | ||
| SF-36 Score | |||||
| Physical Functioning | Mean, SD | 45.4 ± 26.1a | 6.7 ± 22.0b | 0.8 ± 21.2b | F(2 744) = 247.65, p < .001 |
| Effect Size | 2.04 | 0.28 | 0.04 | ||
| Role Physical | Mean, SD | 47.4 ± 45.5a | 11.6 ± 42.8b | 2.9 ± 39.8b | F(2, 753) = 71.37, p < .001 |
| Effect Size | 1.33 | 0.29 | 0.07 | ||
| Bodily Pain | Mean, SD | 27.4 ± 25.4a | 4.3 ± 23.0b | −0.4 ± 20.4b | F(2, 752) = 96.57, p < .001 |
| Effect Size | 1.25 | 0.20 | −0.02 | ||
| General Health | Mean, SD | 22.6 ± 16.9a | 5.4 ± 17.5b | 2.1 ± 13.9b | F(2, 752) = 117.86, p < .001 |
| Effect Size | 1.59 | 0.36 | 0.13 | ||
| Vitality | Mean, SD | 31.6 ± 24.0a | 6.2 ± 20.6b | 4.7 ± 16.4b | F(2, 752) = 122.84, p < .001 |
| Effect Size | 1.84 | 0.32 | 0.23 | ||
| Social Functioning | Mean, SD | 29.5 ± 28.4a | 6.0 ± 29.0b | 1.7 ± 23.9b | F(2, 753) = 60.63, p < .001 |
| Effect Size | 1.17 | 0.22 | 0.07 | ||
| Role Emotional | Mean, SD | 25.5 ± 50.3a | 12.8 ± 51.7b | 6.9 ± 42.4b | F(2, 753) = 1 1.73, p < .001 |
| Effect Size | 0.60 | 0.29 | 0.17 | ||
| Mental Health | Mean, SD | 14.8 ± 19.6a | 2.4 ± 19.8b | 1.3 ± 16.5b | F(2, 753) = 39.87, p < .001 |
| Effect Size | 0.80 | 0.12 | 0.7 | ||
| PCS | Mean, SD | 16.9 ± 10.0a | 2.7 ± 9.0b | 0.4 ± 8.3b | F(2 742) = 232.49, p < .001 |
| Effect Size | 1.88 | 0.29 | 0.04 | ||
| MCS | Mean, SD | 7.0 ± 13.1a | 2.4 ± 12.6b | 1.7 ± 10.5b | F(2, 742) = 13.19, p < .001 |
| Effect Size | 0.62 | 0.20 | 0.15 | ||
Cells without common superscripts are significantly different based upon covariate-adjusted Bonferroni-corrected contrasts with p < .017.
Cell entries represent mean (SD) change from baseline and within-group effect size.
Meaningful Changes in IWQOL-Lite Total Score
Ninety-seven percent (97%) of surgery patients experienced meaningful improvements, compared to only 43% of those who sought but did not have surgery and 30% of population-based obese individuals (χ(4) = 299.20, p < 0.001). This difference remained significant (p < .001) after controlling for baseline BMI. No surgery patients reported meaningful deteriorations in IWQOL-Lite total score over the two-year period, compared to nearly one in five individuals in the group that sought but did not have surgery (18.9%) and population-based obese (17.2%) groups.
Demographic and Weight Characteristics
Table 1 presents baseline demographic and weight characteristics by group. Those who sought but did not have surgery were less likely to be married than surgery patients (52.3% vs. 61.8%), but did not differ from surgery patients in terms of other demographic and weight characteristics. In contrast, obese population-based subjects were older, weighed less, and had a higher proportion of males and Caucasians than both surgery patients and those who sought but did not have surgery.
Table 1
Baseline characteristics by group.
| Characteristic | Gastric Bypass Surgery (n = 421) | Seeking But Did Not Have Surgery (n = 405) | Population-based Obese (n = 319) | Significance |
|---|---|---|---|---|
| Female gender (n, %) | 355 (84.3)a | 344 (84.9)a | 243 (76.2)b | χ(2) = 11.32, p = .003 |
| Age, yrs. (mean, SD) | 42.1 ± 10.8a | 42.5 ± 11.4a | 48.8 ± 10.9b | F(2, 1142) = 39.98, p < .001 |
| Married (n, %) | 260 (61.8)a | 212 (52.3)b | 222 (69.6)a | χ(2) = 22.60, p < .001 |
| Caucasian race (n, %) | 376 (89.3)a | 367 (90.6)a | 310 (97.2)b | χ(2) = 16.74, p < .001 |
| Education, yrs. (mean, SD) | 14.1 ± 2.2 | 13.9 ± 2.4 | 13.8 ± 2.3 | F(2, 1073) = 1.49, p = .227 |
| Weight, lbs. (mean, SD) | 292.6 ± 61.1a | 284.9 ± 57.0a | 269.8 ± 54.8b | F(2, 1142) = 14.04, p < .001 |
| BMI, kg/m (mean, SD) | 47.3 ± 7.8a | 46.5 ± 7.7a | 43.9 ± 6.4b | F(2, 1142) = 20.35, p < .001 |
Cells without common superscripts are different p < .05 based upon Tukey’s hsd (F test) or Bonferroni (chi-square) correction.
Baseline HRQOL
Baseline comparisons between groups in IWQOL-Lite and SF-36 scores are presented in Table 2. Surgery patients were more impaired than those who sought but did not have surgery on physical HRQOL (including Physical Function and Sexual Life from the IWQOL-Lite and Physical Functioning, Role Physical, and PCS from the SF-36) but did not differ on psychosocial components. In contrast, surgery patients were more impaired than population-based obese subjects on all scales and total scores from both the IWQOL-Lite and the SF-36.
Table 2
Baseline IWQOL-Lite and SF-36 scores by group adjusted for BMI and gender.
| IWQOL-Lite Score | Gastric Bypass Surgery (n = 382–416) | Seeking But Did Not Have Surgery (n = 360–400) | Population-based Obese (n = 286–314) | Significance |
|---|---|---|---|---|
| Physical Function | 26.9 ± 18.5a | 31.1 ± 20.8b | 48.2 ± 21.2c | F(2, 1124) = 81.07, p < .001 |
| Self-Esteem | 22.6 ± 20.8a | 24.3 ± 22.0a | 45.7 ± 25.9b | F(2, 1125) = 95.20, p < .001 |
| Sexual Life | 39.5 ± 30.4a | 46.6 ± 32.5b | 65.7 ± 29.7c | F(2, 1023) = 46.75, p < .001 |
| Public Distress | 36.3 ± 23.2a | 39.4 ± 24.9a | 62.0 ± 24.8b | F(2, 1124) = 83.33, p < .001 |
| Work | 46.2 ± 26.5a | 47.5 ± 27.1a | 68.0 ± 23.8b | F(2, 1113) = 56.08, p < .001 |
| IWQOL-Lite Total | 31.5 ± 16.5a | 34.8 ± 18.5a | 54.5 ± 19.6c | F(2, 1125) = 127.02, p < .001 |
| SF-36 Score | ||||
| Physical Functioning | 37.1 ± 21.8a | 41.8 ± 24.5b | 56.6 ± 23.2c | F(2, 1113) = 43.84, p < .001 |
| Role Physical | 32.7 ± 35.8a | 39.7 ± 39.1b | 58.6 ± 38.7c | F(2, 1121) = 32.35, p < .001 |
| Bodily Pain | 41.0 ± 21.6a | 41.7 ± 21.0a | 56.5 ± 22.2b | F(2, 1121) = 41.11, p < .001 |
| General Health | 43.3 ± 14.4a | 44.5 ± 15.2a | 54.9 ± 16.3b | F(2, 1121) = 49.80, p < .001 |
| Vitality | 25.9 ± 17.4a | 28.1 ± 19.0a | 41.6 ± 20.7b | F(2, 1121) = 58.04, p < .001 |
| Social Functioning | 48.8 ± 25.6a | 51.5 ± 26.9a | 72.5 ± 24.5b | F(2, 1121) = 72.47, p < .001 |
| Role Emotional | 47.6 ± 43.1a | 45.3 ± 42.9a | 65.4 ± 40.0b | F(2, 1121) = 20.05, p < .001 |
| Mental Health | 59.2 ± 19.2a | 57.1 ± 20.9a | 70.4 ± 18.9b | F(2, 1121) = 38.31, p < .001 |
| PCS | 31.5 ± 8.8a | 33.7 ± 9.2b | 39.1 ± 9.6c | F(2, 1113) = 42.90, p < .001 |
| MCS | 41.4 ± 11.6a | 40.4 ± 12.1a | 47.8 ± 11.5b | F(2, 1113) = 37.13, p < .001 |
Cells represent unadjusted means ± SD.
Cells without common superscripts are different, p < .017 based upon covariate adjusted Bonferroni contrasts.
Two-Year Follow-up Rates
Two-year valid HRQOL assessments (IWQOL-Lite and/or SF-36) were obtained for 308 (73.2%) surgery patients, 253 (62.5%) individuals who sought but did not have surgery, and 272 (85.3%) obese community participants (χ(2) = 36.42, p < .001). Bonferroni-corrected post hoc comparisons revealed that follow-up rates were significantly higher for obese community participants than surgery patients, which were in turn significantly higher than follow-up rates for those who sought but did not have surgery. Participants who failed to complete the two-year assessment were younger, less likely to be married, more likely to be a minority, had fewer years of education, higher BMI, and reported poorer quality of life at baseline on most IWQOL-Lite (all except Sexual Life) and SF-36 (all except Physical Function, Role Physical, Vitality, and PCS) scales.
Two-Year Weight Loss
The percent weight loss at two years among surgery patients averaged 34.2% (SD = 10.0, range = 65.4% loss to 1.0% gain), compared to 1.4% (SD = 8.6, 31.7% loss to 20.3% gain) for individuals who sought but did not have surgery and a 0.5% gain (SD = 9.3, 57.4% loss to 27.0% gain) for obese population-based participants (F(2,802) = 1235.54, p < .001).
Two-Year Changes in HRQOL
Two-year changes in IWQOL-Lite and SF-36 scores by group are presented in Table 3. Gastric bypass patients showed significantly greater improvement than both those who sought but did not have surgery and population-based obese individuals on all measures. HRQOL changes in the group who sought and did not have surgery were comparable to those in the population-based obese group except for IWQOL-Lite Sexual Life and Work, where greater improvement was observed in the group that sought but did not have surgery. Within-group effect size changes in the surgery group for the IWQOL-Lite ranged from 1.73 (Work) to 3.31 (total score) and for the SF-36 ranged from .60 (Role Emotional) to 2.04 (Physical Functioning).
Table 3
Two-year changes in IWQOL-Lite and SF-36 scores by group.
| IWQOL-Lite Score | Gastric Bypass Surgery (n = 260298) | Seeking But Did Not Have Surgery (n = 203244) | Population-based Obese (n = 226262) | Significance | |
|---|---|---|---|---|---|
| Physical Function | Mean, SD | 58.8 ± 20.6a | 13.8 ± 25.8b | 4.9 ± 17.6b | F(2, 798) = 463.02, p < .001 |
| Effect Size | 3.13 | 0.65 | 0.24 | ||
| Self-Esteem | Mean, SD | 56.7 ± 26.5a | 13.1 ± 23.1b | 8.3 ± 18.4b | F(2, 798) = 340.80, p < .001 |
| Effect Size | 2.75 | 0.58 | 0.34 | ||
| Sexual Life | Mean, SD | 47.8 ± 31.8a | 7.2 ± 29.4b | 5.0 ± 26.4c | F(2, 687) = 148.44, p < .001 |
| Effect Size | 1.62 | 0.22 | 0.17 | ||
| Public Distress | Mean, SD | 52.2 ± 23.4a | 11.6 ± 26.0b | 4.6 ± 19.3b | F(2, 797) = 335.92, p < .001 |
| Effect Size | 2.25 | 0.45 | 0.19 | ||
| Work | Mean, SD | 44.0 ± 26.8a | 9.5 ± 25.9b | 4.6 ± 20.4c | F(2, 789) = 2 1 0.12, p < .001 |
| Effect Size | 1.73 | 0.35 | 0.19 | ||
| IWQOL-Lite Total | Mean, SD | 54.0 ± 19.4a | 12.0 ± 21.7b | 5.5 ± 15.1b | F(2, 798) = 468.95, p < .001 |
| Effect Size | 3.31 | 0.63 | 0.29 | ||
| SF-36 Score | |||||
| Physical Functioning | Mean, SD | 45.4 ± 26.1a | 6.7 ± 22.0b | 0.8 ± 21.2b | F(2 744) = 247.65, p < .001 |
| Effect Size | 2.04 | 0.28 | 0.04 | ||
| Role Physical | Mean, SD | 47.4 ± 45.5a | 11.6 ± 42.8b | 2.9 ± 39.8b | F(2, 753) = 71.37, p < .001 |
| Effect Size | 1.33 | 0.29 | 0.07 | ||
| Bodily Pain | Mean, SD | 27.4 ± 25.4a | 4.3 ± 23.0b | −0.4 ± 20.4b | F(2, 752) = 96.57, p < .001 |
| Effect Size | 1.25 | 0.20 | −0.02 | ||
| General Health | Mean, SD | 22.6 ± 16.9a | 5.4 ± 17.5b | 2.1 ± 13.9b | F(2, 752) = 117.86, p < .001 |
| Effect Size | 1.59 | 0.36 | 0.13 | ||
| Vitality | Mean, SD | 31.6 ± 24.0a | 6.2 ± 20.6b | 4.7 ± 16.4b | F(2, 752) = 122.84, p < .001 |
| Effect Size | 1.84 | 0.32 | 0.23 | ||
| Social Functioning | Mean, SD | 29.5 ± 28.4a | 6.0 ± 29.0b | 1.7 ± 23.9b | F(2, 753) = 60.63, p < .001 |
| Effect Size | 1.17 | 0.22 | 0.07 | ||
| Role Emotional | Mean, SD | 25.5 ± 50.3a | 12.8 ± 51.7b | 6.9 ± 42.4b | F(2, 753) = 1 1.73, p < .001 |
| Effect Size | 0.60 | 0.29 | 0.17 | ||
| Mental Health | Mean, SD | 14.8 ± 19.6a | 2.4 ± 19.8b | 1.3 ± 16.5b | F(2, 753) = 39.87, p < .001 |
| Effect Size | 0.80 | 0.12 | 0.7 | ||
| PCS | Mean, SD | 16.9 ± 10.0a | 2.7 ± 9.0b | 0.4 ± 8.3b | F(2 742) = 232.49, p < .001 |
| Effect Size | 1.88 | 0.29 | 0.04 | ||
| MCS | Mean, SD | 7.0 ± 13.1a | 2.4 ± 12.6b | 1.7 ± 10.5b | F(2, 742) = 13.19, p < .001 |
| Effect Size | 0.62 | 0.20 | 0.15 | ||
Cells without common superscripts are significantly different based upon covariate-adjusted Bonferroni-corrected contrasts with p < .017.
Cell entries represent mean (SD) change from baseline and within-group effect size.
Meaningful Changes in IWQOL-Lite Total Score
Ninety-seven percent (97%) of surgery patients experienced meaningful improvements, compared to only 43% of those who sought but did not have surgery and 30% of population-based obese individuals (χ(4) = 299.20, p < 0.001). This difference remained significant (p < .001) after controlling for baseline BMI. No surgery patients reported meaningful deteriorations in IWQOL-Lite total score over the two-year period, compared to nearly one in five individuals in the group that sought but did not have surgery (18.9%) and population-based obese (17.2%) groups.
Discussion
Quality of life is “an essential parameter in measuring the effectiveness of bariatric surgery and should be assessed objectively as a valid outcome measure in clinical trials29.” The current study adds to the sparse literature on HRQOL outcomes in prospective trials of bariatric surgery versus non-surgically treated obese groups14, 15 and is unique in that two separate severely obese comparison groups were used, both of which did not receive weight loss intervention. The group that sought but did not have surgery is more like the surgical group at baseline (as both groups qualified for and desired surgery) and provides a direct test of the effectiveness of gastric bypass surgery on HRQOL. The severely obese community comparison group was randomly selected from a population study and thus is representative of the general population of severely obese individuals not seeking bariatric surgery. The significant two-year post-surgical differences in HRQOL relative to two comparison groups reinforces the findings of the effectiveness of gastric bypass surgery on improving HRQOL and may have implications for policy development regarding reimbursement. The unique inclusion of the two comparison groups provides an opportunity to test what happens to the HRQOL of severely obese individuals if they are left to their own devices regarding weight loss treatment and directly addresses the beneficial effects of gastric bypass surgery versus no intervention.
Statistically significant improvements were observed in all aspects of HRQOL for the surgery patients at two-years compared to the group that sought but did not have surgery. In addition, 97% of the surgical patients experienced meaningful improvements in IWQOL-Lite total score versus 43% of the group that did not have surgery. Dramatic changes in HRQOL occurred at two-years for the surgical patients. For example, IWQOL-Lite total score changed over three standard deviations, SF-36 PCS changed nearly two standard deviations, and SF-36 MCS changed just over half a standard deviation, whereas improvements in HRQOL were much more modest in the group that did not have surgery (IWQOL-Lite total score changed a little more than half a standard deviation, and SF-36 PCS and MCS scores changed less than a third of a standard deviation). The greater changes observed in the obesity-specific measure (IWQOL-Lite) than in the general measure of HRQOL (SF-36) are consistent with previous reports of greater sensitivity of disease-specific measures of HRQOL30.
We also found large and statistically significant differences between bariatric surgery patients and the population-based severely obese control group. For this group, two-year changes in HRQOL were quite modest (IWQOL-Lite total score changed less than a third of standard deviation, SF-36 PCS and MCS scores showed almost no change from baseline), in stark contrast to the dramatic changes observed in the surgical group. Although the severely obese population-based group did not perfectly match the gastric bypass cases on a number of variables, use of this group allowed us to compare changes in HRQOL two-years after surgery with those reported by a general sample of obese individuals not seeking obesity surgery.
One of the limitations of this study is that groups were not comparable in all variables at baseline. Although the surgical patients and the participants who sought but did not have surgery were drawn from the same population (i.e. seekers of gastric bypass surgery), the group who sought but did not have surgery differed from the surgical group with respect to fewer HRQOL impairments at baseline (particularly with respect to the physical and sexual aspects of HRQOL) and they were less likely to be married. It is possible that better baseline physical HRQOL contributed to their not having gastric bypass surgery (i.e. the more physically impaired individuals were deemed more eligible by insurance companies). When subjects were seeking but did not receive surgery, it was primarily due to the insurance company with which the subject was insured; some companies did not cover gastric bypass surgery as part of their insurance plan. None of the subjects were denied surgery because they were too sick. However, some subjects who initially were denied surgery because their insurance did not cover it paid for it out of their own pockets and later had surgery. Thus, socioeconomic factors may have been different at baseline (although there were no differences between the groups on number of years of education). Furthermore, the groups may have differed with respect to insurance company requirements regarding the necessity of undergoing behavioral or other interventions prior to approval for surgery as well as other variables not assessed in this study (e.g. presence of social support). Unfortunately, we lack the data that would allow us to address these issues.
Both comparison groups experienced modest improvements in HRQOL, perhaps due to their participation in a research study. It is likely that the greater changes observed in the seeking surgery but denied controls versus the community controls were a result of their poorer baseline HRQOL, which allowed more opportunity for improvement. Our finding of better HRQOL at baseline in the obese community controls than in subjects who sought but did not have surgery is consistent with previous research comparing HRQOL in bariatric surgery seekers versus obese community volunteers31. We speculate that the presence of better HRQOL among the obese community participants may account for their lack of interest in seeking bariatric surgery in spite of having clinically severe obesity.
Strengths of the current study included the unique design of comparing patients who received gastric bypass surgery with two different comparison groups. Additionally, the comparison groups were not samples of convenience, but consisted of a naturally occurring group of surgery seekers who did not have surgery (primarily due to denial by insurance providers) as well as severely obese subjects randomly selected from a population study. Furthermore, both general and obesity-specific measures were used to assess HRQOL as recommended in a critical review of controlled weight loss trials32. Of the two prospective, controlled trials of bariatric surgery in the literature, the SOS study also used both types of HRQOL measures14, while the study by O’ Brien et al. used only a general measure15.
Follow-up rates were better in the O’Brien et al. study (98% of 40 surgical patients and 83% of 40 nonsurgical patients at two-years) and the Karlsson et al. SOS study (98% of 487 surgical patients and 84% of 487 nonsurgical patients at two-years) than in the present study (73.2% of surgical patients, 62.5% of denied controls, and 85.3% of obese community controls). Incomplete participation in follow-up assessments may have resulted in bias. Additionally, there were differences in subject characteristics as well as baseline HRQOL scores between participants who completed follow-up HRQOL assessments and those who did not, which created a bias in favor in favor of participants with better baseline HRQOL and lower BMI, as well as those who were older, married, more educated and Caucasian. It is also unknown whether participants in the comparison groups sought nonsurgical weight loss treatment during the course of this study, which if they had, may have contributed to improvement in HRQOL.
In conclusion, at two-year follow-up dramatic improvements in two types of HRQOL were found for patients who received gastric bypass surgery compared to patients who sought but did not have gastric bypass surgery and severely obese volunteers from a population sample. Patients undergoing gastric bypass surgery lost an average of 34.2% of their body weight. The large weight reduction is likely responsible for observed improvements in HRQOL. However, it is possible that similar changes in HRQOL would occur in patients achieving this same degree of weight loss through non-surgical means.
Abstract
Background
Few weight loss surgery trials evaluate changes in health-related quality of life (HRQOL) relative to obese individuals not participating in weight loss interventions.
Objectives
In a prospective study we evaluated two-year changes in HRQOL in gastric bypass patients compared to two separate severely obese groups not undergoing weight loss intervention.
Setting
Bariatric surgery practice.
Methods
421 surgery patients (GBP) were compared with 405 individuals who sought but did not have bariatric surgery (No GBP) and 319 population-based obese individuals (Pop OB) on obesity-specific (IWQOL-Lite) and general (SF-36) HRQOL at baseline and two-years.
Results
Weight loss was 34.2% for GBP, 1.4% for No GBP and a gain of 0.5% for Pop OB. Both measures of HRQOL showed greater improvements for the GBP group (p<0.001), even after controlling for baseline differences. Effect sizes for changes in physical and weight-related HRQOL were very large for GBP, but small to medium for the two comparison groups. Effect sizes for changes in psychosocial aspects of HRQOL were moderate to very large for the GBP, but small for comparison groups. Ninety-seven percent of surgery patients reported meaningful improvements on IWQOL-Lite total score, versus 43% of the No GBP group and 30% of the Pop OB group (p<0.001).
Conclusions
Dramatic improvements occurred in both obesity-specific and physical health-related quality of life for gastric bypass surgery patients two-years post-surgery compared to two severely obese groups not enrolled in weight loss intervention. Changes in the psychosocial aspects of HRQOL were medium to large and more variable across domains.
References
- 1. Buchwald H, Avidor Y, Braunwald E, Jensen MD, Pories W, Fahrbach K, et al Bariatric surgery: a systematic review and meta-analysis. JAMA. 2004;292:1724–37.[PubMed][Google Scholar]
- 2. O’Brien PE, McPhail T, Chaston TB, Dixon JBSystematic review of medium-term weight loss after bariatric operations. Obes Surg. 2006;16:1032–40.[PubMed][Google Scholar]
- 3. Adams TD, Gress RE, Smith SC, Halverson RC, Simper SC, Rosamond WD, et al Long-term mortality after gastric bypass surgery. N Engl J Med. 2007;357:753–61.[PubMed][Google Scholar]
- 4. Sjostrom L, Narbro K, Sjostrom CD, Karason K, Larsson B, Wedel H, et al Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med. 2007;357:741–52.[PubMed][Google Scholar]
- 5. Pories WJ, Swanson MS, MacDonald KG, Long SB, Morris PG, Brown BM, et al Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg. 1995;222:339–50. discussion 50–2. [Google Scholar]
- 6. Sjostrom L, Lindroos AK, Peltonen M, Torgerson J, Bouchard C, Carlsson B, et al Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med. 2004;351:2683–93.[PubMed][Google Scholar]
- 7. Dixon JB, O’Brien PEChanges in comorbidities and improvements in quality of life after LAP-BAND placement. Am J Surg. 2002;184:51S–54S.[PubMed][Google Scholar]
- 8. O’Brien PE, Dixon JB, Brown W, Schachter LM, Chapman L, Burn AJ, et al The laparoscopic adjustable gastric band (Lap-Band): a prospective study of medium-term effects on weight, health and quality of life. Obes Surg. 2002;12:652–60.[PubMed][Google Scholar]
- 9. Wadden TA, Sarwer DB, Fabricatore AN, Jones L, Stack R, Williams NSPsychosocial and behavioral status of patients undergoing bariatric surgery: what to expect before and after surgery. Med Clin North Am. 2007;91:451–69. xi–xii.[PubMed][Google Scholar]
- 10. Sugerman HJ, Starkey JV, Birkenhauer RA randomized prospective trial of gastric bypass versus vertical banded gastroplasty for morbid obesity and their effects on sweets versus non-sweets eaters. Ann Surg. 1987;205:613–24.[Google Scholar]
- 11. Nguyen NT, Goldman C, Rosenquist CJ, Arango A, Cole CJ, Lee SJ, et al Laparoscopic versus open gastric bypass: a randomized study of outcomes, quality of life, and costs. Ann Surg. 2001;234:279–89. discussion 89–91. [Google Scholar]
- 12. Hell E, Miller KA, Moorehead MK, Norman SEvaluation of health status and quality of life after bariatric surgery: comparison of standard Roux-en-Y gastric bypass, vertical banded gastroplasty and laparoscopic adjustable silicone gastric banding. Obes Surg. 2000;10:214–9.[PubMed][Google Scholar]
- 13. Muller MK, Wenger C, Schiesser M, Clavien PA, Weber M. Quality of Life After Bariatric Surgery-A Comparative Study of Laparoscopic Banding vs. Bypass. Obes Surg. 2008[PubMed]
- 14. Karlsson J, Sjostrom L, Sullivan M. Swedish obese subjects (SOS)–an intervention study of obesity. Two-year follow-up of health-related quality of life (HRQL) and eating behavior after gastric surgery for severe obesity. Int J Obes Relat Metab Disord. 1998;22:113–26.[PubMed]
- 15. O’Brien PE, Dixon JB, Laurie C, Skinner S, Proietto J, McNeil J, et al Treatment of mild to moderate obesity with laparoscopic adjustable gastric banding or an intensive medical program: a randomized trial. Ann Intern Med. 2006;144:625–33.[PubMed][Google Scholar]
- 16. Adams TD, Avelar E, Cloward T, Crosby RD, Farney RJ, Gress R, et al. Design and rationale of the Utah obesity study. A study to assess morbidity following gastric bypass surgery. Contemp Clin Trials. 2005;26:534–51.[PubMed]
- 17. Hunt SC, Williams RR, Barlow GKA comparison of positive family history definitions for defining risk of future disease. J Chronic Dis. 1986;39:809–21.[PubMed][Google Scholar]
- 18. Williams RR, Hunt SC, Barlow GK, Chamberlain RM, Weinberg AD, Cooper HP, et al Health family trees: a tool for finding and helping young family members of coronary and cancer prone pedigrees in Texas and Utah. Am J Public Health. 1988;78:1283–86.[Google Scholar]
- 19. Williams RR, Hunt SC, Heiss G, Province MA, Bensen JT, Higgins M, et al Usefulness of cardiovascular family history data for population-based preventive medicine and medical research (the Health Family Tree Study and the NHLBI Family Heart Study) Am J Cardiol. 2001;87:129–35.[PubMed][Google Scholar]
- 20. Kolotkin RL, Crosby RD, Kosloski KD, Williams GRDevelopment of a brief measure to assess quality of life in obesity. Obes Res. 2001;9:102–11.[PubMed][Google Scholar]
- 21. Kolotkin RL, Crosby RDPsychometric 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.[PubMed][Google Scholar]
- 22. Ware J, Snow K, Kosinski M, Gandek B SF-36 Health Survey: Manual and Interpretation Guide. The Health Institute, New England Medical Center; Boston: 1993. [PubMed][Google Scholar]
- 23. McHorney CA, Ware JE, Jr, Lu JF, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care. 1994;32:40–66.[PubMed]
- 24. McHorney CA, Ware JE, Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31:247–63.[PubMed]
- 25. Miller RG Simultaneous statistical inference. Second. Springer Verlag; New York: 1981. [PubMed][Google Scholar]
- 26. Winer BJ Statistical Principles in Experimental Design. Second. McGraw-Hill; New York: p. 1971. [PubMed][Google Scholar]
- 27. Crosby RD, Kolotkin RL, Williams GRAn integrated method to determine meaningful changes in health-related quality of life. J Clin Epidemiol. 2004;57:1153–60.[PubMed][Google Scholar]
- 28. SPSS. Statistical package for the social sciences (Version 16.0.1) SPSS Inc; Chicago: pp. 1989–2007. [PubMed]
- 29. Nguyen N, Varela EJ, Nguyen T, Wilson SEQuality of life assessment in the morbidly obese. Obes Surg. 2006;16:531–3.[PubMed][Google Scholar]
- 30. Guyatt GH, Jaeschke R, Feeny DH, Patrick DL. Measurements in clinical trials: Choosing the right approach. In: Spilker B, editor. Quality of Life and Pharmacoeconomics in Clinical Trials. 2nd. Lippincott-Raven; Philadelphia: 1996. pp. 41–48. [PubMed]
- 31. Kolotkin RL, Crosby RD, Williams GRHealth-related quality of life varies among obese subgroups. Obes Res. 2002;10:748–56.[PubMed][Google Scholar]
- 32. Maciejewski ML, Patrick DL, Williamson DFA structured review of randomized controlled trials of weight loss showed little improvement in health-related quality of life. J Clin Epidemiol. 2005;58:568–78.[PubMed][Google Scholar]