Ten-year trends in health-related quality of life after surgical and conventional treatment for severe obesity: the SOS intervention study

Abstract

Objective:

To examine trends and effects of weight loss treatment on health-related quality of life (HRQL) in the severely obese over 10 years.

Design:

Swedish obese subjects (SOS) intervention study is a controlled, longitudinal trial of the health effects of weight loss in the severely obese.

Subjects:

A total of 655 of 851 surgically treated and 621 of 852 conventionally treated obese men (body mass index, BMI34) and women (BMI38) who completed 10 years of the study.

Measurements:

HRQL was assessed before treatment and after 0.5, 1, 2, 3, 4, 6, 8 and 10 years.

Results:

HRQL change during the 10-year observation period largely followed phases of weight loss, weight regain and weight stability. Improvements and deteriorations in HRQL were associated with the magnitude of weight loss or regain, except regarding anxiety. Peak improvements in the surgical group were observed during the first year of weight loss, whereas the weight regain phase (mainly between 1- and 6-year follow-up) was accompanied by a gradual decline in HRQL. The period from 6- to 10-year follow-up was characterized by relatively stable observations in both weight and HRQL. At 10 years, net gains were noted in all HRQL domains compared to baseline. Comparisons of treatment effects on HRQL in the surgical vs conventional group after 10 years showed significantly better outcome in the surgical group on current health perceptions, social interaction, psychosocial functioning and depression, whereas no significant differences were found for overall mood and anxiety. Long-term results of the study suggest that a maintained weight loss of about 10% is sufficient for positive long-term effects on HRQL, a limit that was reached in about two-thirds of the surgically treated patients who completed 10 years of the study.

Conclusion:

Long-lasting weight reduction in the severely obese has a general long-standing positive outcome on HRQL. Bariatric surgery is a favorable option for the treatment of severe obesity, resulting in long-term weight loss and HRQL improvements in a majority of patients. However, difficulties among some surgical patients to control and maintain weight loss over time should not be ignored. Future research should study if the long-term efficacy of bariatric surgery may be further enhanced by implementing lifestyle modification techniques in the postoperative management of patients.

Introduction

Severe obesity is a rapidly increasing chronic condition that is associated with a wide range of serious health complications and reduced health-related quality of life (HRQL).1 The harmful effects of obesity influence physical, psychological and social domains of HRQL and tend to be more severe at higher body mass index (BMI) levels.2, 3, 4 Weight reduction in the severely obese is accompanied by improvements in HRQL and some studies indicate that a dose–response relationship exists between the magnitude of weight loss and HRQL benefits.2, 3, 5, 6, 7, 8 However, severe obesity is a refractory condition that is extremely difficult to master in the long run and conventional treatment, including various diet regimens, lifestyle modification programs and pharmaceutical agents, is relatively ineffective in the long-term perspective.1 At present, surgical intervention is the only treatment alternative yielding more enduring weight loss and improvements in health status and HRQL.9 Nevertheless, much of the research reporting improvements in psychosocial areas after bariatric surgery suffers from methodological limitations, such as small study samples, short follow-up periods, large numbers of dropouts and lack of appropriate control subjects.8, 10 Few studies have followed patients for longer periods, and reliable information on the long-term effects of weight reduction treatment on HRQL is lacking. Bariatric surgery produces substantial initial weight loss in the great majority of patients, but after reaching maximum weight loss, many patients start to gradually regain weight over several years.11 A vital question that has not yet been answered is what happens in the long run. The aims of the current paper were to analyze HRQL in relation to weight change in surgically vs. conventionally treated patients followed for 10 years, to examine if long-term weight reduction is associated with long-standing improvements in HRQL, and to evaluate the implications of weight regain on HRQL.

Methods

Design of the SOS study

SOS is an ongoing nationwide project comprising a cross-sectional (registry) study of obese persons, a longitudinal intervention trial of the health effects of weight reduction and a population study. Detailed information on design, recruitment and assessment procedures are reported elsewhere.12, 13, 14, 15 The seven ethics review boards involved in the SOS study approved the protocol. Written informed consent was obtained from all participants.

SOS registry study

A total of 6328 obese persons aged 37–57 years were recruited to the registry study between 1987 and 2000 by advertisements in the media and through information to primary health-care centers. All subjects completed a comprehensive series of mail-out/mail-back questionnaires and were given comprehensive medical examinations. They were invited to participate in the intervention study and asked to specify whether they preferred surgical or nonsurgical treatment.

SOS intervention study

The intervention study is a prospective, controlled clinical trial of the health effects of intentional weight reduction in the severely obese. Inclusion criteria were age (37–60 years at accrual) and BMI (34 kg/m2 for males and 38 kg/m2 for females). Exclusion criteria were minimal and are described elsewhere.12 Between 1987 and 2001, a total of 4047 severely obese subjects were included from the registry study and from existing waiting lists at surgical departments. Of those, 2010 subjects were surgically treated and 2037 matched control subjects were offered conventional treatment at their regular primary health-care center. Surgical treatment included fixed or variable banding (n=377), vertical banded gastroplasty (n=1368) or gastric bypass (n=265), as preferred by the surgeon. All surgical patients were given instructions on nutrition. Treatment for the conventional cases was not standardized and treatment regimens varied according to the local practices. Of the two antiobesity drugs approved in Sweden, orlistat was introduced in 1998 and sibutramin in 2000.

All patients were health-examined before intervention. The surgically treated patient and the control subject started the intervention on the operation day of the surgical case. The two treatment groups are followed in the same manner and follow-up is carried out throughout patient visits and mail-out/mail-back questionnaires at 0.5, 1, 2, 3, 4, 6, 8, 10, 15 and 20 years after treatment start. More extensive examinations, including biochemistry, are performed at inclusion and after 2 and 10 years. A total of 25 surgical departments and 480 primary health care centers in 19 of the 24 Swedish counties are involved in the study.

SOS reference study

A total of 2037 persons aged 37–61 years were randomly selected from a general population registry using a computer-based procedure.15 Between 1994 and 1999, a total of 524 men and 611 women completed mail-out/mail-back questionnaires and were health-examined. The data collection methods were identical to those used in the SOS study. The participation rates were 54% in men and 58% in women. Nonobese (BMI<30 kg/m2) subjects (n=1017) were selected as reference subjects for the present analysis. Of these, a small proportion (0.5%) was underweight, 57.9% were of normal weight and 41.6% were overweight.

Study sample

The present study sample comprised 1703 consecutive subjects included in the SOS intervention study between 1987 and 1994. Subjects with complete data on HRQL at 10-year follow-up were included in the analysis. A total of 1276 (74.9%) subjects were available for analysis, comprising 655 of the 851 surgically and 621 of the 852 conventionally treated cases. Mean (s.d.) age at the inclusion examination was 47.0 (5.7) years in the surgical group and 48.4 (6.7) years in the conventional treatment group.

Weight parameters

Body weight was measured in light clothing without shoes to the nearest 0.1 kg using calibrated balances or electronic scales. Height was measured in a standing position without shoes to the nearest 0.01 m. BMI was calculated as weight divided by height squared (kg/m2).

Health-related quality of life

A battery of generic and condition-specific measures, the SOS Quality of Life Survey,2 was used for HRQL assessment before treatment and at 0.5, 1, 2, 3, 4, 6, 8 and 10 years after intervention.

General health perceptions were measured by the Current Health scale (CH) selected from the general health rating index (GHRI).16 The CH scale includes nine general statements on perceived current health. Item responses are aggregated to a total score ranging between 0 and 100, where a higher score indicates more positive perceived health status. The original five-point response format was modified in the SOS study to a four-point scale with two acceptance and two rejection alternatives. The psychometric properties of the two versions have been examined.17

Mental well-being was measured by the short version of the Mood Adjective Check List (MACL). MACL contains 38 adjectives on a four-point response scale with two acceptance and two rejection categories.18 It measures three major bipolar dimensions of mood: pleasantness/unpleasantness (e.g., satisfied, optimistic/depressed, resigned), activation/deactivation (e.g., alert, active/passive, apathetic) and calmness/tension (e.g., relaxed/tense, distressed). The three mood dimensions are highly correlated and an overall mood score is calculated to describe general mood. A higher score indicates more positive mood states. The MACL has been widely applied in evaluating mental health status in patients with chronic conditions and has proven valid in detecting changes in mental well-being during weight loss and relapse.5, 13, 19

The Hospital Anxiety and Depression scale (HAD) was used to detect mood disorders.20 HAD is a tool for screening of anxiety and depression disorders in somatically ill patients. It contains 14 items on a four-point response scale, which are summed to separate scores on anxiety and depression. Individuals are classified according to clinically tested classification system whereby a scale score <8 is considered in the normal range, a score of 8–10 indicates a possible case and a score >10 indicates a probable case of mood disorder. The validity of the Swedish version has been documented in several HRQL studies in patients with chronic disease/injury.13

The social interaction (SI) category from the Sickness Impact Profile (SIP) was used to assess health-related dysfunction in social life.21 SI is the main psychosocial category of the SIP and contains 20 statements on quality and quantity of social interaction within the family, among friends and in the community. Scores represent the weighted sum of endorsed items and range between 0 and 100, where higher scores indicate more dysfunction in social interaction.

The obesity-related problems scale (OP) is a condition-specific instrument constructed to measure the impact of obesity on psychosocial functioning.7, 13 OP comprises eight items on a four-point scale. Subjects are asked to indicate how bothered they are by their obesity in a broad range of social activities: private gatherings at home or at a friend's or relative's home, going to restaurants, participation in community activities (courses, etc.), holidays away from home, trying on and buying clothes, bathing in public places and intimate relations. Responses are aggregated to a total score that ranges from 0 (no impairment) to 100 (maximum impairment). Construct validity and responsiveness of the OP-scale have been presented elsewhere.7

Statistical methods

Significance testing of differences in HRQL between two groups was performed using Fisher's nonparametric permutation test.22 For comparison of three or more groups, the nonparametric Kruskal–Wallis' ANOVA23 was used and Tukey's range test was utilized for post hoc testing of differences between mean values.24 Correlations between variables were tested using Pitman's nonparametric permutation test22 and Pearson's correlation coefficients were calculated for descriptive purposes. The limit for significance was set at the 5% level. Analysis of longitudinal treatment effects in surgical vs conventional cases was performed using Fisher's nonparametric test. Mantel's pooling technique applied to Fisher's test was used to adjust for baseline differences in the longitudinal analysis.25 Data were analyzed on an intention-to-treat basis and thus 34 controls, who later underwent surgical treatment, were retained in the conventional group.

The magnitude of group differences was further analyzed by means of effect sizes (ES). ES of a between-group difference was estimated by calculating the mean difference, divided by the pooled standard deviation.26 ES of within-group change was calculated as mean change between assessments, divided by the standard deviation of change.27 ES were judged against standard criteria proposed by Cohen:26 trivial (0 to <0.2), small (0.2 to <0.5), moderate (0.5 to <0.8) and large (0.8). This method supplements usual significance testing and provides standardized effect levels regardless of sample size and scaling properties of HRQL instruments.

Results

Comparison between completers and noncompleters

Follow-up rates after 10 years were 77% in the surgically treated group and 72.9% in the conventionally treated group. Testing of baseline differences between completers and noncompleters was conducted separately for surgical and conventional cases. Average age was similar in completers and noncompleters in both treatment groups. In the surgical group, mean (s.d.) body weight was somewhat higher in noncompleters ((122.8 kg (17.4)) than in completers (120.1 kg (16.5)) (P<0.05). No significant difference in body weight was observed between completers and noncompleters in the conventional group. Maximum weight loss was roughly the same in completers and noncompleters in both treatment groups.

No significant differences were found between completers and noncompleters in the surgical group regarding overall mood (MACL), anxiety and depression symptoms (HAD) or psychosocial functioning (OP). However, noncompleters had significantly poorer (P<0.01) current health perceptions (CH) (mean (s.d.) 46.0 (25.5) vs 51.8 (24.1)) and were more limited (P<0.01) in SI (14.7 (14.5) vs 11.7 (11.0)). In the conventional group, noncompleters had generally slightly poorer HRQL ratings than compliers (data not shown).

Weight change in surgically and conventionally treated cases

Weight characteristics before intervention and at 10-year follow-up are shown in Table 1. The average (s.d.) maximum weight loss in the surgical group was 25.3% (9.7) after 1 year. A successive weight regain was observed mainly between 1 and 6 years, and weight reduction at 6-year follow-up was 16.9% (11.6). Thereafter, weight regain leveled off and weight loss after 10 years averaged 16% (12.1). Long-term weight reduction was roughly the same for men (15.8% (12.3)) and women (16.1% (12.1)). Great variation in weight reduction was noted among the surgical cases after 10 years: about 12% lost 30% of their baseline body weight, almost one-fourth lost 20–29.9%, one-third lost 10–19.9%, about one-fourth lost <10% and 9% gained weight. Significant differences in 10-year weight loss were noted for the surgical subgroups: 13.2% (13) for banding (n=161), 16.5% (11) for vertical banded gastroplasty (n=457) and 25.1% (11) for gastric bypass (n=37).

Table 1 Weight characteristics at the time of the inclusion examination and at 10-years follow-up in the surgical and conventional treatment groups

In the conventional group, average (s.d.) maximum weight loss was 1.2% (5.6) after 6 months. The weight loss was regained after 2 years and a weight increase of 1.6% (8.8) was noted after 4 years. Mean weight was more or less stable between 4 and 10 years and the average weight increase from baseline to 10-year follow-up reached 1.5% (9.9). BMI was 39.9 (4.6) kg/m2 at baseline and 40.6 (5.9) kg/m2 after 10 years. The weight increase was 1.9% (12.8) for men and 1.4% (12.1) for women (ns). Variation in 10-year weight change was noted: 15% lost 10% of their baseline body weight, about 12% lost 5–9.9%, about one-fifth were weight stable or had lost less than 5%, and about one-half (54%) had gained weight during the 10-year follow-up period.

Baseline HRQL: surgical vs conventional cases vs nonobese reference subjects

HRQL at baseline was generally poorer in the severely obese than in the age-matched, nonobese reference population. The surgical patients reported worse HRQL than the conventional cases (Table 2). ES estimates indicated small differences between the surgical and conventional group on CH (ES=0.29), SI (ES=0.43), overall mood (ES=0.22) and depression (ES=0.26), whereas the difference on OP was moderate (ES=0.64). Anxiety was slightly higher in the surgical cases; however, ES indicated a trivial group difference (ES=0.15).

Table 2 HRQL before treatment and at 10-years follow-up in the surgical and conventional treatment groups (baseline scores are compared with a general population sample)

Ten-year change in HRQL

The pattern of change in HRQL during the 10-year observation period largely followed phases of weight loss, weight regain and weight stability. In the surgical group, peak HRQL improvements were observed during the first year of weight loss, whereas the weight regain phase (mainly between 1- and 6-year follow-up) was accompanied by a gradual decline in HRQL. The period from 6- to 10-year follow-up was characterized by relatively stable observations in both weight and HRQL. At 10 years, net gains were noted in all HRQL domains compared to baseline. In the conventional group, small initial improvements were mostly lost within 2 years; however, some of the HRQL domains improved at the end of the observation period and the long-term outcome on HRQL was mixed.

Current health perceptions

After 1 year, CH improved by 48% in the surgical group and 7% in the conventional group. Thereafter, a successive deterioration was noted in both groups. In the surgical group, CH improvement was 16% after 6 years and 11% after 10 years. In the conventional group, CH scores after 10 years were significantly worse than at baseline (Table 2). Analysis of treatment outcome in surgical vs conventional cases (adjusted for baseline differences) showed significantly better 10-year change in the surgical group (P<0.0001; Table 2). ES calculation of 10-year change indicated a small positive effect of surgical treatment (ES=0.21), whereas the negative change in the conventional group was trivial (ES=−0.13). The long-term change in CH was roughly the same in men and women. The mean CH level after 10 years did not differ significantly between the surgical and the conventional treatment group (Table 2).

Social interaction

A substantial improvement of about 60% of the baseline value in SI was seen 1 year after surgery. A gradual return of difficulties was then observed and the improvement was 31% after 6 years and 27% after 10 years (Table 2). In the conventional group, a minor improvement (7%) after 1 year eroded and a slight change for the worse was noted after 10 years. Significantly greater 10-year change was observed in the surgical group (P<0.01; Table 2). ES of 10-year change indicated a small treatment effect of surgery (ES=0.25), whereas the effect of conventional treatment was trivial (ES=−0.05). The mean SI level after 10 years was roughly equal in the two treatment groups (Table 2).

Obesity-related psychosocial problems

At 1-year follow-up, OP scores improved by about 63% in the surgical group vs 7% in the conventional group. The improvement decreased to 50% after 6 years and 49% after 10 years in the surgical group (Table 2). ES of 10-year change after surgical treatment was large (ES=1.0). In the conventional group, OP scores were more or less stable from 1- to 4-year follow-up and subsequently improved at 6, 8 and 10 years. The improvement at 10-year follow-up was about 23% of the baseline value and ES showed a small positive effect (ES=0.42). Comparisons between the surgical and conventional groups showed that the surgical group improved significantly more from baseline to 10-year follow-up (P<0.0001; Table 2); however, OP levels at 10-year follow-up were roughly equal in the two treatment groups (Table 2).

Overall mood

At 1-year follow-up, powerful improvements in mood were observed in the surgical group and the score was also significantly better than that of the reference population (P<0.0001). Between 2- and 6-year follow-up, a step-by-step decrease in mood was noted and about one-third of the initial improvement remained after 10 years. In the conventional group, small improvement after 1 year was followed by deterioration; however, mood scores tended to rise at the end of the 10-year period and a minor improvement compared to baseline was noted. Comparisons between the surgical and conventional groups showed that the surgical group improved significantly more from baseline to 10-year follow-up; however, after adjusting for baseline differences the result was nonsignificant (Table 2). ES of 10-year change indicated a small treatment effect in the surgical group (ES=0.25), whereas the effect in the conventional group was trivial (ES=0.10). Comparisons between the two treatment groups showed significantly better mood levels in the surgical cases after 0.5, 1 and 2 years, whereas no significant difference was observed at subsequent follow-ups.

Depression

At 1-year follow-up, depression scores decreased to about one-half of the baseline value in the surgical group. Thereafter, a gradual change for the worse was observed and the improvement after 6 years was 25 and 27% at 10-year follow-up. ES of 10-year change indicated a small positive effect on depression symptoms after surgical treatment (ES=0.35). In the conventional group, depression symptoms were slightly reduced after 10 years; however, the effect was trivial (ES=0.14). Between-group comparisons of 10-year change indicated significantly better outcome in the surgical group (P<0.05; Table 2). Mean depression scores after 10 years were roughly equal in the two treatment groups (Table 2).

According to HAD classifications, 24% of the surgical cases and 16% of the conventional cases had possible or probable morbid depression before treatment vs 6% of the nonobese reference population. At 10-year follow-up, the corresponding values were 15% in the surgical group and 14% in the conventional group.

Anxiety

A substantial short-term reduction of anxiety symptoms (37%) was seen in the surgical group at 1-year follow-up; however, a relapse was noted during the following years, and the improvement after 6 and 10 years was 20 and 23%, respectively. Anxiety scores also improved in the conventional group, especially at the end of the observation period, and a reduction of about 26% was noted after 10 years. ES indicated a small effect in both the surgical (ES=0.33) and conventional (ES=0.35) group.

A total of 34% of the surgical group and 30% of the conventional group had possible or probable anxiety disorder (HAD classification) at baseline vs 17% of the nonobese reference population. At 10-year follow-up, the corresponding values were 24% in the surgical group and 19% in the conventional group.

HRQL change in relation to weight loss and regain

Correlations between weight loss and HRQL improvements after 1, 2, 6 and 10 years are shown in Table 3. In the surgical group, associations after 1 year were significant for OP and SI, but nonsignificant for CH, overall mood, depression and anxiety. The corresponding correlations in the conventional group were significant for all HRQL domains. At 2-year follow-up, improvements in HRQL were significantly associated with weight loss in both treatment groups, except for change in anxiety in the conventional group. At 6-year follow-up, all HRQL improvements, except anxiety, were significantly correlated to weight loss in both treatment groups. Correlations at 10-year follow-up showed roughly the same pattern as after 6 years, except that change in SI was nonsignificant in the conventional group. Improvement in the condition-specific OP scale was most strongly associated with weight loss in both treatment groups at all measurement points and CH was the most responsive of the generic HRQL measures.

Table 3 Pearson correlations between weight change and HRQL change during the 10-year observation period in surgically (n=655) and conventionally (n=621) treated subjects

Correlations between weight regain and HRQL change are shown in Table 3. Weight regain for the surgical cases was calculated as change in weight from maximum weight loss at 1 year to 10-year follow-up, whereas for the conventional cases, weight regain was calculated as change in weight from maximum weight loss at 6 months to 2-year follow-up. Change in HRQL was calculated for the corresponding periods. The magnitude of weight regain was significantly associated with deterioration in all HRQL domains in both treatment groups, except for change in anxiety. Change for the worse in OP was most strongly associated with regain in both groups, whereas somewhat weaker associations were noted for the remaining HRQL measures.

HRQL at 10-year follow-up in surgically treated patients by weight change

Table 4 shows HRQL outcome for surgically treated subjects grouped into five categories of relative weight change after 10 years. There were no significant differences in HRQL among groups at baseline. Comparisons of change in HRQL after 10 years indicated significant differences among groups on all HRQL domains except anxiety (Table 4). ESs of 10-year change was calculated to judge the effect on HRQL in groups differing in long-term weight reduction. According to ES estimates, improvements after 10 years were most favorable in the group with the largest weight loss (30%). ESs were large on OP, moderate on CH and depression, and small on SI, mood and anxiety. Effects in subjects with weight losses of 10–19.9 and 20–29.9% were roughly equal; that is, ESs were small on five of the HRQL measures and large on OP. The outcome in the group with less than 10% weight loss was trivial on four HRQL domains, whereas a small effect on anxiety and a large effect on OP were noted. In the weight gain group, trivial effects were noted for five of the HRQL domains, whereas a small change for the worse was noted for CH.

Table 4 Body weight and HRQL at baseline and 10-year follow-up in surgically treated patients (patients are grouped by relative weight change after 10 years)

The 10-year effect of surgical treatment varied among the different HRQL domains (Table 4). The most positive outcome was noted on psychosocial functioning (OP) with large ESs ranging from 1.6 in the greatest weight loss group to 0.8 in the least weight loss group. ES for OP was trivial in the weight gain group. Small ESs were seen on overall mood and SI in the three groups with weight losses of 10% or more, whereas trivial effects were observed in subjects with less weight loss (0–9.9%) and in the weight gain group. The most differentiated results were noted for CH and depression with moderate ESs in the group with the largest weight loss, small ESs in groups with weight losses of 10–19.9% and 20–29.9%, trivial ESs in the group with less than 10% weight loss and trivial or small negative ESs in the weight gain group. The least discriminative results were found for anxiety with small positive ESs in all four weight loss groups and a trivial effect in the weight gain group.

Ten-year trends in weight loss and HRQL change in surgically treated patients with weight loss 10% vs <10% after 10 years

Ten-year trends in weight loss and HRQL were compared in groups of surgical patients with 10-year weight loss 10% (higher weight loss, HWL; n=447) or <10% (lower weight loss, LWL; n=208). About 68% of the surgical cases had lost 10% or more of their baseline weight after 10 years. As shown in Figure 1a, weight loss (%) was significantly higher in the HWL group even at 1-year follow-up (27.7% (8.9) vs 20.0% (6,3); P<0.0001) and the difference increased for each year of follow-up. After 10 years, average weight loss was 22.1% in the HWL group vs 2.8% in the LWL group (P<0.0001). As shown in Figure 1b–g, trends in HRQL largely followed changes in weight loss in the two groups and HRQL improvements were markedly more positive in the HWL group. After 1 year, improvements in depression, OP and SI were significantly greater in the HWL group and the difference in HRQL change between groups tended to increase during the 10-year observation period. At 10-year follow-up, all HRQL domains except for anxiety had improved significantly more in the HWL group.

Figure 1
figure1

Ten-year trends in body weight and HRQL in surgically treated patients with 10% (HWL) vs <10% (LWL) weight loss after 10 years. Percent change (of the baseline value) in body weight, overall mood, CH, SI, depression, anxiety and OP after 1, 2, 3, 4, 6, 8 and 10 years are shown in Figures 1a–g. Data are for surgical subjects who completed 10 years of the study. P-values indicate significant differences between the HWL and LWL group at 1- and 10-year follow-up (ns=nonsignificant).

Discussion

Treatment effects on HRQL were evaluated in 655 severely obese surgical patients and 621 conventional cases that remained in the study after 10 years. Follow-up rates were satisfactory in the surgical group (77%) and conventional group (73%). In general, changes in HRQL after surgical treatment followed phases of weight loss, weight regain and weight stability. Poor HRQL before treatment was markedly improved after surgery and peak improvements during the 10-year observation period were noted after 6 months and 1 year, which corresponds with the weight loss phase. A maximum weight loss of 25% was registered at 1-year follow-up, and thereafter, a gradual regain of weight was observed, mainly up to 6 years. Weight loss after 6 years was 16.9%, which means that about one-third of the initial weight reduction was gradually regained within 5 years. A slow but steady decline in HRQL was noted during the weight regain phase. Finally, weight regain slowed down between 6 and 10 years and average weight reduction after 10 years reached 16%. Accordingly, quite stable HRQL ratings were observed in the surgical group between 6 and 10 years. All of the six HRQL domains that were examined in the study were significantly improved in the surgical group after 10 years. According to standard criteria for interpreting ESs,26 10-year improvements were in the small range for current health, social interaction, overall mood, depression and anxiety, whereas the improvement in psychosocial functioning was large.

In the conventional group, an average maximum weight loss of 1.2% after 6 months was regained after 2 years and an increase in body weight of 1.5% was noted after 10 years. The long-term outcome on each of the six HRQL domains varied. Small initial improvements in CH and SI during the first year eroded, and scores at 10-year follow-up were slightly worse than at baseline. Minor ameliorations in overall mood and HAD depression scores were noted, but the long-term change was trivial. In contrast, ESs of 10-year change in HAD anxiety and OP indicated small, long-term improvements. In summary, the change in the conventional group after 10 years was trivial in four of six HRQL domains, whereas small improvements were observed in two domains.

Comparisons of treatment effects on HRQL in the surgical vs conventional group after 10 years (adjusted for baseline differences) showed significantly better outcome in the surgical group on CH, SI, OP and HAD depression, whereas no significant differences were found for changes in overall mood and HAD anxiety. It should be noted that the ES of 10-year change in overall mood indicated a small treatment effect after surgery, whereas the effect in the conventional group was trivial. However, as mood scores were lower in the surgical cases at baseline, they had a higher margin for improvement than the conventional cases. Likewise, scores on other HRQL variables were worse in the surgical group at baseline, that is, differences were moderate on psychosocial functioning (OP), small on CH, SI, overall mood and depression and trivial on anxiety. This result is in line with previous findings showing poorer HRQL in surgical patients before treatment than in obese patients enrolled in nonsurgical weight loss treatments. Kolotkin et al.28 have demonstrated that HRQL tends to vary across obese subgroups by treatment-seeking status and treatment modality/intensity.

A peak in both weight loss and HRQL improvements were observed after 1 year in the surgical group. However, the associations between weight reduction and HRQL change after 1 year were quite weak, whereas those after 2 years were considerably stronger. This indicates that short-term reports (up to 1 year) of HRQL effects after bariatric surgery should be interpreted with caution.7 Previously reported results from the SOS study have shown a positive relationship between weight reduction and HRQL improvements up to 4 years after intervention.2, 5, 7 In the present study, we demonstrated that weight reduction after 6 and 10 years was also associated with improvements in HRQL, except for change in anxiety. We also concluded that weight regain after previous weight loss was associated with deteriorations in all HRQL domains, except anxiety. To date, only one earlier study has examined the effect of weight regain on HRQL in persons treated for obesity. Engel et al.29 analyzed the relationship between change in weight and HRQL in 122 participants in a weight reduction program who had been followed for about 1–3 years. Their results suggest that weight loss and regain are equally significant determinants for HRQL change. The outcome of the present study supports their findings and provides additional evidence that HRQL change after bariatric surgery tends to follow the phases of weight loss, regain and weight stability. Another important finding is that long-lasting weight reduction in the severely obese has a general long-term positive outcome on HRQL. However, it is also evident that the impressive early effects on HRQL after bariatric surgery tend to wane with time, most probably because of the continuous regain of weight that takes place over a number of years.

CH scores in the surgical patients improved between baseline and 10-year follow-up, whereas CH in the conventionally treated subjects worsened. CH is the main construct of the GHRI, on which the General Health scale (GH) of the SF-36 health survey is based. Poor general health perceptions have frequently been associated with limitations in functional ability, physical and mental symptoms and number of medical diagnoses.30, 31 Furthermore, poor perceived health status has been linked to different indicators of utilization of health-care services, such as hospitalization and annual visits to doctors, as well as to hospital expenditures. Epidemiological studies have also demonstrated that poor self-rated health is a strong predictor of disease-specific and all-cause mortality.32 The significant improvement in CH scores from baseline to 10-year follow-up in the surgical group is in line with the positive long-term effects of weight reduction observed on a number of medical health indicators in the SOS study.14, 33 On the other hand, the CH level in the surgical group after 10 years was significantly lower than population norms. A possible explanation is that weight reduction after 10 years was less than 10% in nearly one-third (31.8%) of the surgical patients, a cutoff often referred to as a limit for positive effects on weight-related health complications.34 The outcome on CH during the 10-year follow-up varied with the magnitude of weight loss and 10-year change was most favorable in the surgical subgroup with the largest weight loss (30%). Furthermore, no general effect on CH was noted for the group of subjects with less than 10% weight loss, while a change for the worse was seen in the weight gain subgroup.

The condition-specific OP scale was constructed to measure the impact of obesity on psychosocial functioning, which is a key domain in the assessment of HRQL in the severely obese.7 Obese persons are the targets of discrimination and are at high risk for developing disturbances in psychosocial functioning. OP was the most responsive HRQL measure in relation to weight loss and regain during the 10-year observation period. Condition-specific measures are generally more responsive than generic measures to clinical change.2 High levels of weight-related psychosocial distress at baseline substantially improved after weight loss. Previous results from the SOS study have demonstrated a distinct dose–response effect between weight loss and OP improvements up to 4 years after surgery.7 Ten-year change in OP was clearly associated with the magnitude of weight reduction; however, differences among weight loss subgroups were not as distinct as after 4 years. Compared to baseline, OP scores after 10 years had improved on average 75% in the surgical subgroup with the largest weight loss (30%). Long-term improvements in subjects with 10–19.9% and 20–29.9% weight loss were roughly equal, 50% and 55%, respectively, whereas an average reduction of 39% was noted for surgical patients with weight losses of 0–10%. ES of 10-year change for the latter group was in the large range, indicating that even modest weight loss may have a significant positive effect on psychosocial functioning (cf. results on OP also for the conventional group). A similar finding with improvements on OP and physical functioning (SF-36) after modest weight loss was reported in a Finnish 2-year follow-up study of obese outpatients treated with very-low-energy diet and behavior modification.35 In the present study, a slight improvement in 10-year OP (11%) from baseline was noted in surgical patients who had gained weight; however, ES of change for this group was in the trivial range. Similarly, minor improvements in OP and social functioning (SF-36) were reported among those who had gained weight in the Finnish study. This pattern of change suggests that treatment effects, besides maintained weight loss, may to some extent affect the outcome on psychosocial functioning.35

SI is the main psychosocial category of the SIP and measures health-related limitations in social interaction within the family, among friends and in the community. SI improved in the surgical group, whereas no effect was observed for the conventional group after 10 years. A small treatment effect on SI was noted for subgroups of surgical subjects who lost 10% or more of their body weight, whereas no change was seen in surgical subjects who lost less than 10% of their weight. About 85% of the surgical cases reported at least one limitation in SI at baseline, whereas the corresponding figure after 10 years was 59%.

The three measures of mental health status that were used in the present study indicated poor mental well-being in both treatment groups before intervention. Previous research has demonstrated that the prevalence of psychiatric comorbidity is increased in bariatric surgery patients and mood disorders appear to be the most common condition.8 The relationship between obesity and depression is complex and the causal pathway may be bidirectional or reciprocal. Studies of adolescents suggest that depression during adolescence may precede obesity in young adulthood and vice versa.36, 37 On the other hand, studies of adult populations indicate that obesity is a risk factor for the incidence of depression. It has been pointed out that depression should be considered another comorbidity of obesity.38 In the present study, high rates of depression cases were observed in both treatment groups before treatment. The estimated prevalence according to the HAD depression scale was 2.7 and 4 times higher, respectively, in the conventional and surgical group than in the nonobese reference population. Figures at 10-year follow-up were improved, but the prevalence was still 2.3 and 2.5 times higher than the population norm in the conventional and surgical group, respectively. There are a few previous reports on the long-term effects of bariatric surgery on depression symptoms, but no controlled studies have been conducted. In a prospective study, Beck Depression Inventory (BDI) was used to assess symptoms of depression in patients at yearly intervals up to 4 years after gastric banding.38 High scores before treatment decreased with about one-half during the first year postsurgically and most of the positive effect remained in patients available for follow-up after 4 years; however, mean scores tended to increase between 1 and 4 years. In line with the present study, greater weight loss was associated with greater reduction of depression symptoms. A 5-year follow-up study of a small sample of gastric bypass patients also showed substantial improvement in BDI scores.39 Thus, results of the present and previous studies indicate a substantial positive long-term effect of weight reduction on depression symptoms in the severely obese.

The estimated prevalence of anxiety disorder (HAD) before treatment was two times higher in the surgical group and 1.7 times higher in the conventional group compared to the reference population. A long-term decrease in anxiety was observed in both treatment groups and the outcome after 10 years was roughly the same. Alleviation of anxiety symptoms in the surgical group was significantly correlated with weight loss after 2, 3 and 4 years; however, no significant long-term relationship with weight reduction was observed after 6, 8 and 10 years. Similarly, the positive long-term effect on anxiety in the conventional group was not associated with weight reduction. Furthermore, weight regain was not associated with increases in anxiety symptoms. Thus, our results indicate that long-term weight reduction may have different effects on depression and anxiety symptoms, although this finding needs to be confirmed in other studies. The lack of a relationship between weight reduction and anxiety in the long-term was unexpected, as was the long-term improvement of anxiety symptoms in the conventional group. Twin studies have shown that a significant proportion of the risk to develop anxiety disorders is due to heredity.40, 41 Thus, weight reduction may be expected to have little or no effect on anxiety in the long run. On the other hand, we found anxiety symptoms to be significantly alleviated in both treatment groups after 10 years. Further in-depth analysis is necessary to explain this finding and a forthcoming evaluation of long-term trends in mental health within the SOS study will include effects of other treatment options, such as use of psychotropic medication, etc.

Weight loss during the first year after surgery was accompanied by a dramatic positive effect on mood (MACL) and the average mood level in the surgical group after 1 year was significantly higher than the population norm. After reaching maximum weight loss, mood scores slowly decreased and about two-thirds of the initial improvements eroded during the years of weight regain. Comparisons between the two treatment groups after 10 years showed no clear-cut advantage of surgical treatment on overall mood. However, it is noteworthy that effects on mood after 10 years varied in surgical patients by long-term weight reduction. ESs indicated an equally small positive effect in groups of patients with weight losses of 10% or more, whereas no improvement was observed in the group with less than 10% weight loss or in the weight gain group. Also, the average mood level in the group with the largest weight losses (30%) after 10 years was comparable with the mood level in the nonobese reference population. Mood in surgical patients with less than 30% weight reduction was poorer than the reference norm, and markedly poorer mood was observed in subjects with less than 10% weight loss and in the weight gain group.

Ten years after intervention, about 15% of the surgical cases were classified as nonobese (BMI <30 kg/m2). HRQL in this group was roughly equal to levels observed in surgical subjects who reached 30% weight loss or more after 10 years (see Table 4). Mental well-being scores in the nonobese subgroup were similar to levels observed in the population reference group, whereas CH and social interaction were slightly worse than the population norm. Weight-related psychosocial problems (OP) in this group were somewhat worse in women than in men and comparable to levels observed in overweight men and women in the reference population.7

Although weight loss after 10 years was somewhat more positive after vertical banded gastroplasty (16.5%) than after banding (13.2%),14 no significant differences in HRQL were observed. Similarly, 10-year effects on diabetes, cardiovascular risk factors, energy intake and physical activity did not differ between these two surgical subgroups.14 It should be noted though that long-term weight loss was considerably greater in the gastric bypass subgroup (25.1%). As shown in the present paper, long-term improvements in HRQL were positively associated with the magnitude of long-term weight reduction. It is therefore reasonable to expect that 10-year outcome on HRQL may be more favorable after gastric bypass than after purely restrictive operations. However, there are currently no data supporting this assumption and the small number of gastric bypass cases (5%) in the present study sample prohibited separate analysis of HRQL effects in this subgroup. This issue will be further addressed within the SOS intervention study when a sufficient number of gastric bypass cases have passed 10-year follow-up. As gastric bypass is by far the most common surgical procedure today, it is of importance to determine the long-term HRQL effects of this procedure vs other surgical procedures.

In the conventionally treated group, weight reduction after 6 months was considerably less than what is usually reported. This discrepancy is probably explained by the fact that most weight loss studies are carried out at specialized medical centers, whereas conventional treatment in the SOS study was not standardized and carried out across 480 primary health-care centers without extra resources for obesity treatment.

Severe obesity is associated with multiple negative health impacts that affect a variety of HRQL areas. Of domains not taken into account in the present analysis, physical functioning should be particularly noted. Another limitation of the study is that patients were between 47 and 70 years of age at 10-year follow-up. Long-term outcome on HRQL after obesity treatment may be different in younger obese persons.

Results of the study indicate that long-term effects of bariatric surgery on HRQL are weakened by significant weight regain in large groups of patients. Surprisingly little is known about factors that promote or hinder weight maintenance after surgical weight loss treatment and little research attention has been paid to this subject. The simple explanation to the successive weight regain is that it is a result of a continuous imbalance between energy intake and expenditure. A significant increase in energy intake over time has been reported among bariatric surgery patients.42, 43 In the SOS study, self-reported intake was almost halved during the first 6 months postsurgically, from about 2900 to 1500 kcal per day, but increased to about 2000 kcal per day over the next 6 years.14 These data suggest that surgical patients have difficulties to adhere to the postoperative diet. As self-report of eating is systematically underestimated in obese persons, the actual intake may be considerably higher. The other crucial component for long-term weight control is regular physical activity.44, 45, 46 The importance of habitual physical activity for weight maintenance in conventional weight loss treatment has been systematically studied; however, there is little published on exercise patterns among bariatric surgery patients. One study of weight control behaviors in individuals successful at long-term weight maintenance (members of the US National Weight Control Registry) showed that those who had lost weight by surgical means reported much lower levels of physical activity than those who had lost weight by nonsurgical means.47 Although additional research is needed, this finding is thought provoking and has implications for the postoperative management of the bariatric surgery patient. Besides its role for weight control, regular physical activity has a powerful independent, favorable overall effect on health status.45, 48 In particular, there is strong evidence that regular exercise is an efficient component for primary and secondary prevention of several chronic conditions associated with overweight and obesity (cardiovascular disease, diabetes, hypertension, cancer etc.).48, 49 Thus, it seems important to implement treatment strategies that stimulate and facilitate the adoption and maintenance of habitual physical activity among bariatric surgery patients.

Obesity surgery is principally presented as a standalone solution and established standards regarding postoperative management are lacking. However, as disordered eating patterns, psychological difficulties and coping problems are present in subgroups of surgically treated patients,50, 51, 52, 53 it is likely that the efficacy of surgical interventions may be enhanced by postoperative behavioral and nutritional counseling. Experiences from behavioral weight reduction programs indicate that long-term weight maintenance is facilitated by prolonged contact with the patient.44 Various behavioral programs for postoperative management have been implemented in clinical practice. However, few systematic attempts have been made to evaluate these achievements and, thus, it is difficult to ascertain the relative effectiveness of such programs. One study that compared gastric-banding patients with or without psychotherapeutic support showed that the outcome on weight and well-being after 2 years was more favorable in the psychotherapy group, indicating that the positive effects of bariatric surgery may be further improved by postoperative behavioral support.54 Additional studies are needed to evaluate the effects of postoperative behavioral and nutritional support on the long-term outcome of obesity surgery.

Conclusion

Our results demonstrate that long-lasting weight reduction in the severely obese has a general long-standing positive outcome on HRQL. Improvements at follow-ups during the 10-year observation period were associated with the magnitude of weight loss, except regarding anxiety. Early effects of bariatric surgery should be regarded as temporary. After reaching maximum weight loss, large groups of surgically treated patients start to slowly regain considerable amounts of weight, which contributes to a gradual regress in HRQL, especially in mental well-being domains. However, the long-term results of the study also suggest that a maintained weight loss of about 10% is sufficient for positive long-term effects on HRQL, a limit that was reached in about two-thirds of the surgically treated patients who completed 10 years of the study. Thus, bariatric surgery is a favorable option for the treatment of severe obesity, resulting in substantial long-term weight loss accompanied by improvements in health status and quality of life in a majority of patients. Nevertheless, difficulties among surgical patients to control and maintain weight loss over time should not be ignored. Many surgical patients may benefit from behavioral support programs and future research should systematically study if the long-term efficacy of bariatric surgery may be further enhanced by implementing lifestyle-modification techniques in the postoperative management of patients.

References

  1. 1

    Östman J, Britton M, Jonsson E (eds) Treating and Preventing Obesity–An Evidence Based Review. Wiley: Weinheim, 2004.

    Google Scholar 

  2. 2

    Sullivan M, Karlsson J, Sjöström L, Taft C . Why quality-of-life measures should be used in the treatment of patients with obesity. In: Björntorp P. (ed). International Textbook in Obesity. Wiley: Chichester, 2001.

    Google Scholar 

  3. 3

    Sullivan M, Karlsson J . Quality of life. In: Östman J, BrittonM, Jonsson E (eds). Treating and Preventing Obesity – An evidence based review. Wiley: Weinheim, 2004.

    Google Scholar 

  4. 4

    Kolotkin RL, Meter K, Williams GR . Quality of life and obesity. Obes Rev 2001; 2: 219–229.

    CAS  Article  Google Scholar 

  5. 5

    Karlsson J, Sjöström 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; 21: 113–126.

    Article  Google Scholar 

  6. 6

    Kolotkin RL, Crosby RD, Williams GR, Hartley GG, Nicol S . The relationship between health-related quality of life and weight loss. Obes Res 2001; 9: 564–571.

    CAS  Article  Google Scholar 

  7. 7

    Karlsson J, Taft C, Sjöström L, Torgerson JS, Sullivan M . Psychosocial functioning in the obese before and after weight reduction: construct validity and responsiveness of the Obesity-related Problems scale (OP). Int J Obes Relat Metab Disord 2003; 27: 617–630.

    CAS  Article  Google Scholar 

  8. 8

    Sarwer DB, Wadden TA, Fabricatore AN . Psychosocial and behavioral aspects of bariatric surgery. Obes Res 2005; 13: 639–648.

    Article  Google Scholar 

  9. 9

    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: 1724–1737.

    CAS  Article  Google Scholar 

  10. 10

    Maciejewski ML, Patrick DL, Williamson DF . A structured review of randomized controlled trials of weight loss showed little improvement in health-related quality of life. J Clin Epidemiol 2005; 58: 568–578.

    Article  Google Scholar 

  11. 11

    Hsu LK, Benotti PN, Dwyer J, Roberts SB, Saltzman E, Shikora S et al. Nonsurgical factors that influence the outcome of bariatric surgery: a review. Psychosom Med 1998; 60: 338–346.

    CAS  Article  Google Scholar 

  12. 12

    Sjöström L, Larsson B, Backman L, Bengtsson C, Bouchard C, Dahlgren S et al. Swedish Obese Subjects (SOS). Recruitment for an intervention study and a selected description of the obese state. Int J Obes Relat Metab Disord 1992; 16: 465–479.

    PubMed  Google Scholar 

  13. 13

    Sullivan M, Karlsson J, Sjöström L, Backman L, Bengtsson C, Bouchard C et al. Swedish Obese Subjects (SOS). An intervention study of obesity. Baseline evaluation of health and psychosocial functioning in the first 1743 subjects examined. Int J Obes Relat Metab Disord 1993; 17: 503–512.

    CAS  PubMed  Google Scholar 

  14. 14

    Sjöström 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: 2683–2693.

    Article  Google Scholar 

  15. 15

    Larsson I, Berteus Forslund H, Lindroos AK, Lissner L, Naslund I, Peltonen M et al. Body composition in the SOS (Swedish Obese Subjects) reference study. Int J Obes Relat Metab Disord 2004; 28: 1317–1324.

    CAS  Article  Google Scholar 

  16. 16

    Davies AR, Sherbourne CR, Peterson JR, Ware Jr JE . Scoring Manual: Adult Health Status and Patient Satisfaction Measures Used in RAND's Health Insurance Experiment. Santa Monica, CA: The RAND Corporation, (No. N-2190-HHS) 1988.

    Google Scholar 

  17. 17

    Karlsson J, Sjöström L, Sullivan M . Swedish Obese Subjects (SOS) – an intervention study of obesity. Measuring psychosocial factors and health by means of short-form questionnaires. Results from a method study. J Clin Epidemiol 1995; 48: 817–823.

    CAS  Article  Google Scholar 

  18. 18

    Sjöberg L, Svensson E, Persson L . The measurement of mood. Scand J Psychol 1979; 20: 1–18.

    Article  Google Scholar 

  19. 19

    Karlsson J, Hallgren P, Kral J, Lindroos AK, Sjöström L, Sullivan M . Predictors and effects of long-term dieting on mental well-being and weight loss in obese women. Appetite 1994; 23: 15–26.

    CAS  Article  Google Scholar 

  20. 20

    Zigmond AS, Snaith RP . The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67: 361–370.

    CAS  Article  Google Scholar 

  21. 21

    Bergner M, Bobitt RA, Carter WB, Gilson BS . The sickness impact profile: development and final revision of a health status measure. Med Care 1981; 19: 787–805.

    CAS  Article  Google Scholar 

  22. 22

    Bradley JW . Distribution-Free Statistical tests. London: Prentice-Hall, 1968, pp. 68–86.

    Google Scholar 

  23. 23

    Siegel S . Nonparametric Statistics for the Behavioral Sciences. New York: MacGraw-Hill, 1956, pp. 184–194.

    Google Scholar 

  24. 24

    Winer BJ . Statistical Principles in Experimental Design. London: MacGraw-Hill, 1970, pp. 46–104.

    Google Scholar 

  25. 25

    Mantel N . Chi square test with one degree of freedom: extensions of the Mantel-Haenszel procedure. J Am Stat Assoc 1963; 58: 690–700.

    Google Scholar 

  26. 26

    Cohen J . Statistical Power Analysis for the Behavioral Sciences. New York: Academic Press, 1978.

    Google Scholar 

  27. 27

    Katz JN, Larson MG, Phillips CB, Fossel AH, Liang MH . Comparative measurement sensitivity of short and longer health status instruments. Med Care 1992; 30: 917–925.

    CAS  Article  Google Scholar 

  28. 28

    Kolotkin RL, Crosby RD, Williams GR . Health-related quality of life varies among obese subgroups. Obes Res 2002; 10: 748–756.

    Article  Google Scholar 

  29. 29

    Engel SG, Crosby RD, Kolotkin RL, Hartley GG, Williams GR, Wonderlich SA et al. Impact of weight loss and regain on quality of life: mirror image or differential effect? Obes Res 2003; 11: 1207–1213.

    Article  Google Scholar 

  30. 30

    Ware Jr JE, Snow KK, Kosinski M, Gandek B . SF-36 Health Survey Manual and Interpretation Guide. Boston: New England Medical Center: The Health Institute, 1993.

    Google Scholar 

  31. 31

    Persson L-O, Karlsson J, Bengtsson C, Steen B, Sullivan M . The Swedish SF-36 Health Survey. II. Evaluation of clinical validity: Results from population studies of elderly and women in Gothenburg. J Clin Epidemiol 1998; 51: 1095–1103.

    CAS  Article  Google Scholar 

  32. 32

    Björner JB, Sondergaard Kristensen T, Orth-Gomér K, Tibblin G, Sullivan M, Westerholm P . Self-Rated Health – A Useful Concept in Research, Prevention and Clinical Medicine. Swedish council for planning and coordination of research, Report 96:9 Spri: Stockholm 1996.

    Google Scholar 

  33. 33

    Näslund I . Lessons from the Swedish Obese Subjects study: the effects of surgically induced weight loss on obesity comorbidity. Surg Obes Relat Dis 2005; 1: 140–144.

    Article  Google Scholar 

  34. 34

    National Health Lung Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Obes Res 1998; 6: 51S–210S.

    Article  Google Scholar 

  35. 35

    Kaukua J, Pekkarinen T, Sane T, Mustajoki P . Health-related quality of life in obese outpatients losing weight with very-low-energy diet and behaviour modification – a 2-y follow-up study. Int J Obes Rebat Metls Disord 2003; 27: 1233–1241.

    CAS  Article  Google Scholar 

  36. 36

    Fabricatore AN, Wadden TA . Psychological aspects of obesity. Clin Dermatol 2004; 22: 332–337.

    Article  Google Scholar 

  37. 37

    Herva A, Laitinen J, Miettunen J, Veijola J, Karvonen JT, Laksy K, Joukamaa M . Obesity and depression: results from the longitudinal Northern Finland 1966 Birth Cohort Study. Int J Obes Relat Metab Disord 2006; 30: 520–527.

    CAS  Article  Google Scholar 

  38. 38

    Dixon JB, Dixon ME, O'Brien PE . Depression in association with severe obesity: changes with weight loss. Arch Intern Med 2003; 163: 2058–2065.

    Article  Google Scholar 

  39. 39

    Velcu LM, Adolphine R, Mourelo R, Cottam DR, Angus LDG . Weight loss, quality of life and employment status after Roux-en-Y gastric bypass: 5-year analysis. Surg Obes Relat Dis 2005; 1: 413–417.

    Article  Google Scholar 

  40. 40

    Arnold PD, Zai G, Richter MA . Genetics of anxiety disorders. Curr Psychiatry Rep 2004; 6: 243–254.

    Article  Google Scholar 

  41. 41

    Gordon JA, Hen R . Genetic approaches to the study of anxiety. Annu Rev Neurosci 2004; 27: 193–222.

    CAS  Article  Google Scholar 

  42. 42

    Brolin RL, Robertson LB, Kenler HA, Cody RP . Weight loss and dietary intake after vertical banded gastroplasty and Roux-en-Y gastric bypass. Ann Surg 1994; 220: 782–790.

    CAS  Article  Google Scholar 

  43. 43

    Hernandez-Estefania R, Gonzalez-Lamuno D, Garcia-Ribes M, Garcia-Fuentes M, Cagigas JC, Ingelmo A, Escalante C . Variables affecting BMI evolution at 2 and 5 years after vertical banded gastroplasty. Obes Surg 2000; 10: 160–166.

    CAS  Article  Google Scholar 

  44. 44

    Wadden TA, Butryn ML, Byrne KJ . Efficacy of lifestyle modification for long-term weight control. Obes Res 2004; 12: 151S–162S.

    Article  Google Scholar 

  45. 45

    Jakicic JM, Otto AD . Physical activity considerations for the treatment and prevention of obesity. Am J Clin Nutr 2005; 82: 226S–229S.

    CAS  Article  Google Scholar 

  46. 46

    Wing R, Phelan S . Long-term weight loss maintenance. Am J Clin Nutr 2005; 82: 222S–225S.

    CAS  Article  Google Scholar 

  47. 47

    Klem ML, Wing RR, Chang CC, Lang W, McGuire MT, Sugerman HJ et al. A case-control study of successful maintenance of a substantial weight loss: individuals who lost weight through surgery versus those who lost weight through nonsurgical means. Int J Obes Relat Metab Disord 2000; 24: 573–579.

    CAS  Article  Google Scholar 

  48. 48

    Warburton DE, Nicol CW, Bredin SS . Health benefits of physical activity: the evidence. Can Med Assoc J 2006; 174: 801–809.

    Article  Google Scholar 

  49. 49

    Bensimhon DR, Kraus WE, Donahue MP . Obesity and physical activity: a review. Am Heart J 2006; 151: 598–603.

    Article  Google Scholar 

  50. 50

    Hsu LK, Betancourt S, Sullivan SP . Eating disturbances before and after vertical banded gastroplasty: a pilot study. Int J Eat Disord 1996; 19: 23–34.

    CAS  Article  Google Scholar 

  51. 51

    Kalarchian MA, Marcus MD, Wilson GT, Labouvie EW, Brolin RE, LaMarca LB . Binge eating among gastric bypass patients at long-term follow-up. Obes Surg 2002; 12: 270–275.

    Article  Google Scholar 

  52. 52

    Kinzl JF, Trefalt E, Fiala M, Biebl W . Psychotherapeutic treatment of morbidly obese patients after gastric banding. Obes Surg 2002; 12: 292–294.

    Article  Google Scholar 

  53. 53

    Ryden A, Karlsson J, Sullivan M, Torgerson JS, Taft C . Coping and distress: what happens after intervention? A 2-year follow-up from the Swedish Obese Subjects (SOS) study. Psychosom Med 2003; 65: 435–442.

    Article  Google Scholar 

  54. 54

    Nicolai A, Ippoliti C, Petrelli MD . Laparoscopic adjustable gastric banding: essential role of psychological support. Obes Surg 2002; 12: 857–863.

    Article  Google Scholar 

Download references

Acknowledgements

This project has been supported by the Swedish Council for Working Life and Social Research (project 97–0355:1B and 2B; F0140/2000; 2001–1106; 2002–0109), the Swedish Foundation for Health Care Sciences and Allergy Research (project V96–065; V99–046; V2002–172(F)), the Swedish Research Council/Medicine (project K2003–27VX-14685–01A) and the Sahlgrenska Academy at Göteborg University (project ALFGBG-2780), Göteborg.

Author information

Affiliations

Authors

Corresponding author

Correspondence to J Karlsson.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Karlsson, J., Taft, C., Rydén, A. et al. Ten-year trends in health-related quality of life after surgical and conventional treatment for severe obesity: the SOS intervention study. Int J Obes 31, 1248–1261 (2007). https://doi.org/10.1038/sj.ijo.0803573

Download citation

Keywords

  • quality of life
  • severe obesity
  • bariatric surgery
  • controlled clinical trial
  • health assessment

Further reading