Introduction

A large body of research has shown that obesity is associated with impairments in the quality of life.1, 2, 3, 4 More specifically, studies have suggested that both health- and weight-related quality of life are impacted by excess body weight. Given the multidimensional nature of both forms of quality of life, weight-related comorbidities such as type 2 diabetes and hypertension, as well as the physical limitations imposed by excess body weight, have the potential to impact the quality of life. A number of studies have identified a strong relationship between the degree of obesity and impairments in health related quality of life.5, 6, 7, 8 This relationship is often mediated by comorbid medical conditions.1, 8, 9, 10, 11

Numerous studies have suggested that individuals report improvements in psychosocial functioning with weight loss.12, 13, 14, 15, 16, 17 Perhaps the most consistent finding in this area is the association between weight loss and quality of life. This relationship may be strongest among individuals who lose larger amounts of weight.18 Health- and weight-related quality of life appear to improve in the vast majority of studies involving individuals who undergo bariatric surgery.19 For example, in the study by Sarwer et al., 200 men and women undergoing bariatric surgery had to complete measures of quality of life and other psychosocial constructs both before surgery and again 20, 40 and 92 weeks postoperatively. Participants reported significant improvements in all domains of health- and weight-related quality of life within the first 20 weeks of surgery. These changes were well maintained during the first two postoperative years and were correlated with percentage weight loss.

The present study was undertaken to investigate changes in the quality of life of individuals suffering from obesity who presented for weight loss treatment and were part of a weight loss intervention being conducted in their primary care physicians’ offices. We also investigated whether improvements in the quality of life would be associated with the magnitude of weight loss and whether participants from different demographic groups would experience different relationships between weight loss and improvements in the quality of life.

Materials and methods

Study design

This investigation used data from a 2-year randomized controlled trial titled Practice-based Opportunities for Weight Reduction at the University of Pennsylvania (POWER-UP). The design of the study20 and major results21 have been published elsewhere. Participants were 390 obese men and women who also had at least two components of the metabolic syndrome. The questionnaires used in these analyses were collected during the participants’ baseline visit, which took place between January 2008 and February 2009. Participants completed the measures again at months 6, 12 and 24. The trial was approved by the University of Pennsylvania Institutional Review Board, and informed consent was obtained from all participants.

Participants

Participants were recruited from six primary care practices owned by the University of Pennsylvania Health System. They had to be 21 years of age or older, have a body mass index of 30–50 kg m−2, be established patients in the practice and fulfill at least two of five criteria for the metabolic syndrome: elevated waist circumference; elevated triglycerides; reduced high-density lipoprotein cholesterol; elevated blood pressure; and elevated fasting glucose.22

Participants were randomized to one of three interventions of varying intensity, as detailed elsewhere.20, 21 Individuals in the Usual Care condition had quarterly visits with their primary care provider, who provided education on weight management. Those in the Brief Lifestyle Counseling (LC) condition (that is, Brief LC) also had quarterly visits with their primary care physician, in conjunction with brief, monthly sessions with lifestyle coaches who provided behavioral weight control counseling. Participants in the Enhanced Brief LC condition (that is, Enhanced Brief LC) also had quarterly visits with their primary care physician and received Brief LC in combination with meal replacements or weight loss medications (orlistat or sibutramine), selected in consultation with their primary care physician.

Measures

Short Form Health Survey (SF-12) The SF-12 is a 12-question version of the widely used 36-item Short-Form Health Survey. Items are divided into two subscales: physical health and mental health.23 Lower scores indicate a lower health-related quality of life. Good evidence of reliability has been demonstrated between the SF-12 and the SF-36.

Impact of Weight on Quality of Life-Lite (IWQOL-Lite) The IWQOL-Lite is a quality-of-life instrument specifically designed for use in individuals who are overweight or obese.24 It contains 31 items, with each item beginning with the phrase ‘Because of my weight.’ The measure examines five domains: physical function; self-esteem; sexual life; public distress; and work. Responses to the 31 items are combined to calculate a total score that ranges from 0 to 100; higher scores indicate better quality of life.25

EuroQol-5D The EuroQol-5D contains a five-question descriptive system that measures the following domains: mobility; self-care; usual activities; pain/discomfort; and anxiety/depression.26 Each domain has three levels: 1=no problems; 2=some problems; and 3=severe problems. The answers from each domain are combined to create an index score that ranges from −0.11 to 1.0. Lower EuroQol-5D index scores indicate lower health status.

Statistical analysis plan

All 390 participants completed the measures at baseline; 332 individuals returned the set of questionnaires at month 6, 305 at month 12 and 285 at month 24. We initially proposed to compare differences in the quality of life between the three treatment groups. However, the only statistically significant differences were between the Enhanced Brief LC and Usual Care conditions on the Physical Function subscale of the IWQOL-Lite at month 6 (P=0.003) and month 12 (P=0.004) and for the total score for the IWQOL-Lite at month 6 (P=0.004). For all three comparisons, participants in Enhanced Brief LC (who lost significantly more weight than those in Usual Care at all assessments) reported greater improvements in their quality of life, as expected. Given the limited number of differences between the three treatment groups, we elected to focus our analytic plan on the following: changes over time across the three groups; the relationship between weight loss and quality of life; and differences in the quality of life based on demographic variables of interest.

To assess changes in mean scores for the measures over the three time points (months 6, 12 and 24), as well as the mean percentage change in weight, repeated-measures analyses were conducted using Mixed Models. An unstructured variance–covariance matrix was assumed for each outcome. Model-based means (s.e.) are reported for each outcome at each time point. A significant main effect of time indicates significant changes in outcomes across the 2 years. Using a Bonferroni adjustment for the pairwise time comparisons, these mixed model analyses were also used to identify significant within-mean differences between time points for each outcome (Bonferroni adjusted α=0.008). The relationships between behavioral outcomes and gender and ethnicity were examined by adding these covariates to the mixed models. We also used correlation analyses to examine associations between percentage weight loss and changes in the quality of life. All analyses were conducted using SAS, version 9.2.

Results

Participants’ baseline characteristics

Baseline characteristics of the 390 participants are shown in Table 1. The sample had a mean (±s.d.) age of 51.5±11.5 years, weight of 107.7±18.3 kg and body mass index of 38.5±4.7 kg m−2. A total of 311 (79.7%) were women. Approximately 60% of participants were European American, 38.5% were African American and the remaining were of other ethnic origins. Approximately 95% reported completing high school.

Table 1 Participants’ baseline characteristics
Full size table

Changes in weight

On the basis of the mixed model analysis that used all available data at each time point, the mean body weight was seen to decline significantly over time (P<0.0001). The 390 participants lost an average (±s.e.) of 3.9±0.3 kg at month 6, 4.1±0.4 kg at month 12 and 3.1±0.4 kg at month 24. These losses corresponded to reductions in initial body weight of 3.7±0.3%, 4.0±0.4% and 3.0±0.4%, respectively. (Changes in body weight according to treatment condition have been reported previously.21)

Changes in the quality of life

Table 2 displays the mixed model-based means±s.e. of the variables of interest at baseline and months 6, 12 and 24. At month 6, participants reported significant improvements in all domains of weight-related quality of life (IWQOL-Lite), as well as in the Physical Component Score of the SF-12 and the EuroQol-5D. There was no change at month 6 in the Mental Composite Score of the SF-12 or at any subsequent time. Most of these improvements remained significantly different from baseline at month 24 (all P’s<0.008). The exception was the EuroQol-5D, which did not differ from baseline at month 12 or 24.

Table 2 Mean scores on quality-of-life measures at baseline and months 6, 12 and 24, collapsed across the three treatment interventions
Full size table

Correlation with weight loss

As shown in Table 3, percentage weight loss was significantly correlated with a number of changes in the quality of life at months 12 and 24. Larger weight losses were associated with significantly greater improvements on all of the IWQOL-Lite subscales with the exception of the Work subscale. On the SF-12, larger weight losses at month 12 were associated with significantly greater improvements on the Mental Composite Scale (MCS), but not on the Physical Component Scale. At month 24, the Physical Component Scale was significantly associated with weight loss, whereas the MCS was not.

Table 3 Correlations between weight change and psychosocial changes at months 12 and 24
Full size table

Gender, ethnicity and quality of life

Significant associations between weight change and changes in the quality of life by gender and ethnicity were found more often in women than in men and in whites than in non-whites at both months 12 and 24. As shown in Table 4, for women, significant associations were found at month 12 between weight loss and improvements on the IWQOL-Lite subscales that measure physical function, self-esteem and sexual life. Larger weight losses in women were also associated with greater improvement on the IWQOL-Lite total score, as well as on the MCS of the SF-12. At month 24, associations continued to be significant for physical function, self-esteem and the IWQOL-total score but not for the MCS. However, a significant association with the Physical Component Scale of the SF-12 was observed.

Table 4 Correlations between weight change and psychosocial changes by gender at months 12 and 24
Full size table

At month 12, among men, significant associations between weight loss and quality of life were observed only on the self-esteem subscale, the IWQOL-total score and the EuroQol-5D index. However, at month 24, significant associations were found between weight loss and changes in physical function, self-esteem and sexual life subscales, as well as in the IWQOL-total score.

At month 12, among white participants, significant correlations were seen between weight loss and improvements in physical functioning, self-esteem, sexual life and public distress, as well as with the IWQOL-Lite total score (Table 5). Among non-white participants, only physical functioning, self-esteem and the IWQOL-Lite total scores were significantly correlated with weight loss at month 12. At month 24, weight loss was significantly associated with changes in physical functioning, self-esteem, sexual life, public distress, the IWQOL-Lite total score and the EuroQol-5D in white participants. In non-white participants, larger weight losses were significantly associated with changes in physical functioning, self-esteem and the IWQOL-Lite total score.

Table 5 Correlations between weight change and psychosocial changes by ethnicity at months 12 and 24
Full size table

Discussion

Individuals suffering from obesity and having features of the metabolic syndrome reported significant improvements in several domains of the quality of life within the first 6 months of a weight loss intervention program undertaken in the offices of their primary care physicians. Almost all of these improvements were well maintained over 2 years. Most of these changes were also correlated with changes in weight.

The notable exception to this pattern of results was the absence of change on the MCS of the SF-12. The psychosocial burden of obesity is well documented, and psychosocial distress, including impairments in the quality of life, body image and sexuality, likely has an important role in the decision to pursue weight loss.1 It may be that the six items of the SF-12 that comprise the MCS are not specific enough to appropriately capture the psychological aspects of the quality of life in individuals with obesity and the metabolic syndrome.

Larger weight losses were associated with greater improvements in several domains of the quality of life. These included physical aspects of the quality of life (as assessed by the Physical Composite Scale of the SF-12), the EuroQol-5, as well as all subscales of the IWQOL-Lite (with the exception of the Work subscale). This replicates recent results from our group, but with individuals who had undergone bariatric surgery and lost much larger amounts of weight.19 In that study, improvements in the quality of life were also experienced relatively early (by 20 weeks after surgery) and were well maintained through the second postoperative year. As with previous research, the results of the present study suggest that even more modest weight losses, obtained through non-surgical interventions, are associated with improvements in most areas of the quality of life.13, 14, 15, 16, 17, 18

Greater weight loss was associated with greater self-reported improvements in multiple domains of the quality of life. This was true for both women and men, although there were more statistically significant relationships for women than for men. For both genders, the quality-of-life benefits of larger weight losses were seen at month 12, the assessment point at which participants had lost the greatest amount of weight. At month 24, larger weight losses generally continued to be associated with greater improvements in the quality of life, despite participants regaining an average of 1 kg from month 12 to month 24.

Both white and non-white participants had statistically significant correlations between percentage weight loss and the subscales of the IWQOL-Lite. However, there were a greater number of statistically significant relationships for white than for non-white participants, a finding that may be partially attributable to the differences in sample size between the two groups. It may be that white individuals experience a wider range of weight-related quality-of-life benefits with greater weight loss, compared with non-white individuals. However, it is important to note that non-white individuals also reported improvements in weight-related quality of life. The ultimate research and clinical implications of these observations are, however, unknown.

In summary, results of this study add to the literature on the relationship between weight loss and improvements in the quality of life. Numerous studies conducted in weight loss clinics have observed these relationships; we were able to replicate them in a clinical trial involving obese individuals who received one of three interventions of differing intensities delivered in primary care practices. Improvements in several domains of both health- and weight-related quality of life were observed when participants reached their maximum weight loss of 4.0%, suggesting that even relatively modest weight losses can have a significant impact.