|Home | About | Journals | Submit | Contact Us | Français|
Metabolic syndrome has been associated with impaired health-related quality of life (HRQoL) in several studies. Many studies used only one HRQoL measure and failed to adjust for important confounding variables including obesity, depression, and comorbid conditions.
To investigate the relationship between metabolic syndrome and HRQoL using multiple measures. We also sought to determine whether increasing body mass index (BMI) or diabetes status further modified this relationship.
This cross-sectional study included 390 obese participants with elevated waist circumference and at least one other criterion for metabolic syndrome. Of these 390, 269 had metabolic syndrome [i.e., met 3 out of 5 criteria specified by the National Cholesterol Education Program (NCEP)] and 121 did not. Participants were enrolled in a primary-care based weight reduction trial. HRQoL was assessed using two generic instruments, the Medical Outcomes Study Short-Form 12 (SF-12) and the EuroQol-5D (EQ-5D), as well as an obesity-specific measure, the Impact of Weight on Quality of Life (IWQoL-Lite). Differences in HRQoL were compared among participants with and without metabolic syndrome. Multivariable linear regression was used to determine how HRQoL varied according to metabolic syndrome status, and whether factors including weight, depression, and burden of comorbid disease modified this relationship.
Metabolic syndrome was not associated with HRQoL as assessed by any of the measures. In univariable analysis, depression, disease burden, and employment status were significantly associated with worse HRQoL on all instruments. In the multivariable models, only depression remained significantly associated with reduced HRQoL on all measures. Increasing obesity and diabetes status did not modify the relationship between metabolic syndrome and HRQoL.
In contrast to previous studies, metabolic syndrome was not associated with impaired HRQoL as assessed by multiple measures. This suggests that metabolic syndrome in itself is not associated with decreased HRQoL, but other factors such as obesity, depression, and greater disease burden may significantly influence quality of life in this population.
The prevalence of metabolic syndrome has increased in tandem with the obesity epidemic. Currently, 34% of U.S. adults have metabolic syndrome (1), which is defined by the National Cholesterol Education Program (NCEP) as meeting any three of the five following criteria: 1) elevated waist circumference (≥40 inches for men; ≥35 inches for women); 2) high triglycerides (≥150 mg/dl); 3) decreased high-density lipoprotein (HDL) cholesterol (<40 mg/dl for men; <50 mg/dl for women); 4) elevated fasting glucose (≥100 mg/dl); and 5) elevated blood pressure (≥130/85 mm Hg) (2). Individuals with metabolic syndrome have an increased risk of developing diabetes and cardiovascular disease (CVD) (3), both of which have been associated with decreased health-related quality of life (HRQoL) (4–7). Impaired HRQoL is particularly important in this population, as it has been associated with a number of adverse outcomes, including poor response to therapy, disease progression, and mortality (8–10).
Several components of the metabolic syndrome including obesity, insulin resistance, and hypertension, have been associated with reduced HRQoL (11–15), leading some investigators to suggest that metabolic syndrome itself may also be associated with impaired quality of life. Several studies have confirmed this finding, but they have been limited by failure to adjust for obesity, burden of co-morbid disease, and depression (16–19). Increasing body mass index (BMI), in particular, may modify the relationship between metabolic syndrome and HRQoL, such that the association is stronger in more obese individuals. This relationship has biologic plausibility, as increasing BMI has consistently been associated with worse physical functioning, a greater number of co-morbid conditions, and a higher prevalence of depression (13,20–26). Thus, increasingly obese individuals with metabolic syndrome may be required to take more medications, visit medical providers more frequently, and experience greater difficulty with mobility, all of which can impair quality of life.
In the present study, we assessed HRQoL in obese individuals with and without metabolic syndrome using two generic quality of life measures, the Medical Outcomes Study Short Form-12 (SF-12) and the EuroQol 5D (EQ-5D), as well as one obesity-specific measure, the Impact of Weight on Quality of Life (IWQoL-Lite). We hypothesized that metabolic syndrome would be associated with decreased quality of life in participants with higher BMI (≥ 40 kg/m2), but this relationship would not be seen in patients with lower BMI. We also examined whether diabetes status modified the relationship between metabolic syndrome and quality of life.
Three-hundred and ninety obese individuals were recruited from six primary care practices within the University of Pennsylvania Health System to participate in the Practice-Based Opportunities for Weight Reduction (POWER) trial, a 2-year primary care-based weight reduction trial. Eligible participants were aged 21 years and older, had a BMI of 30–50 kg/m2, an elevated waist circumference, and at least one other criterion for the metabolic syndrome. Participants were considered to have metabolic syndrome if they met at least three of the five criteria defined by the NCEP (2). (Participants with known diabetes or hypertension were considered to have met the glucose and blood pressure criteria for metabolic syndrome, respectively.) Exclusion criteria included uncontrolled blood pressure, recent cardiovascular events, weight change ≥ 5% over the preceding 6 months, active participation in a weight loss program, prior or planned use of bariatric surgery, serious co-morbid conditions (e.g., severe mental illness, end-stage renal disease), use of medications known to cause significant (≥ 5%) long-term changes in weight, or pregnancy. The study was approved by the Institutional Review Board at the University of Pennsylvania, and all participants provided written informed consent. The questionnaires used in this analysis were collected at the randomization visit before participants received any intervention.
The SF-12 is a 12-item condensed version of the SF-36 (27). Both instruments are validated measures of health-related quality of life (28). The SF-12 includes eight subscales -- four subscales are used to derive a summary score of physical health (physical component summary, PCS-12) and four subscales are used to derive a summary score of mental health (mental component summary, MCS-12) (27). The PCS-12 and MCS-12 scores were designed to have a mean score of 50 and a standard deviation of 10 in a representative sample of the US population (29). Higher scores on the SF-12 are indicative of better functioning. Scores greater than 50 represent above average health status. Both summary scores are highly correlated with those derived from the SF-36 in an obese population (30) and were considered as two separate outcomes.
The EQ-5D is a generic health status instrument that describes five dimensions: mobility; self-care; usual activities; depression/anxiety; and pain/discomfort. Each of the five dimensions is based on a single question with three possible responses (1 = no problems, 2 = some problems, and 3 = extreme problems). Scores from the five dimensions are combined into a single “utility” score. The EQ-5D utility scores range from a full health score of 1 (in which respondents report no problems on any dimension) to the lowest score of −0.59 (when respondents report that they are at the bottom level of each dimension) (31). The EQ-5D has been found to be sensitive to the effects of obesity on HRQoL, even after controlling for comorbiditites, age, and sex (32).
The IWQoL-Lite is a validated 31-item, self-report obesity-specific measure of quality of life (33). It provides a total score, as well as scores on five domains: physical function; self-esteem; sexual function; public distress; and work. Scores are transformed on a 0 to 100 scale, with higher scores indicating better quality of life (34). The IWQoL-Lite has been found to be a reliable and valid instrument for assessing weight-related quality of life in obese persons with type 2 diabetes (35)
Symptoms of depression were assessed with the Patient Health Questionnaire (PHQ-8), a validated eight-item depression scale (36). The PHQ-8 includes eight of the nine criteria for depression according to the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV), but does not contain a question about suicidal ideation contained in the PHQ-9 (37). The PHQ-8 asks the number of days over the past 2 weeks that the respondent has experienced a particular depressive symptom. The scores for each item are summed to produce a total score ranging from 0 to 24 points. Scores of 0 to 4 indicate no depressive symptoms, 5 to 9 represent mild depressive symptoms, and of ≥ 10 are indicative of greater symptoms of depression (37). The PHQ-8 and PHQ-9 scores are highly correlated and have nearly identical operating characteristics (37–38).
Disease burden was assessed using the Functional Co-morbidity Index (FCI). The FCI was developed for use in the general population with physical function as an outcome. The FCI includes 18 diagnoses that are commonly treated in outpatient settings, many of which are more common in obese individuals. It is scored by assigning one point to each condition present (39). Thus, the range is 0–18, with a higher score indicating a greater burden of co-morbid disease. As obesity and diabetes are included in the FCI, these diagnoses were not included in the total FCI score in order to avoid double counting. In addition to the FCI, disease burden was also estimated by summing the number of medications for each participant. The participants’ self-reported medical histories and medications were confirmed by a study physician (MLV or AGT), who reviewed the medical records for each individual enrolled in the study.
Lifestyle habits, including smoking and alcohol intake, were also assessed by questionnaires specifically developed for the POWER study.
Weight and CVD risk factors. BMI was calculated from height and weight, which were obtained at the baseline study visit. Participants were measured in light clothing without shoes on a calibrated scale to the nearest 0.1 kg (Tanita BWB 800, Tanita Corp., Tokyo, Japan). Standing height was measured to the nearest 0.1 cm using a wall-mounted stadiometer (Seca 202, Seca Corp., Hamburg, Germany). All measurements were performed by trained research staff and were repeated twice, with the average measurement used to calculate BMI. Participants were classified according to the BMI categories adopted by the National Institutes of Health (40) and the World Health Organization (41): class I (BMI, 30–34.9 kg/m2), class II (BMI, 35–39.9 kg/m2); and class III (BMI ≥ 40 kg/m2). Waist circumference was measured in duplicate to the nearest 0.1 cm by placing an inelastic tape measure (Gulick II, model 67020, Lafayette Instrument Co, Lafayette, IN) around the abdomen horizontally at the midpoint between the highest point of the iliac crest and the lowest part of the costal margin in the mid-axillary line. The procedure was repeated until two consecutive measurements within 0.5 cm were obtained. Blood pressure was measured with an automated sphygmomanometer (Omron HEM-907-XL,Omron Healthcare Inc, Bannockburn, IL) in triplicate after the participant had been sitting quietly for 5 minutes, with the blood pressure recorded as the mean of the three measurements. Serum glucose and lipids were measured following an overnight fast, and conventional assays were used to measure total triglycerides and HDL cholesterol levels.
Summary statistics for all variables, both continuous and categorical, were examined for range and to assess plausibility of values. All data were assessed for normality prior to analysis. Differences in weight and in other characteristics between participants with and without metabolic syndrome were compared using t-tests for continuous variables and chi-square tests for categorical variables.
Univariable regression, ANOVA, and chi square tests were used to estimate the strength of association between HRQoL and baseline characteristics. Separate models were created for each of the following four outcome variables: PCS-12, MCS-12, EQ-5D, and IWQoL. We then examined the potential for effect modification by stratifying the association of metabolic syndrome and HRQoL according to: 1) obesity class (30–34.9 kg/m2, 35–39.9 kg/m2, and 40–50 kg/m2), and 2) the presence of diabetes.
Based on the strength of association in the univariable model (independent variables were entered if they had a p value of less than 0.2), a multivariable linear regression model was fit to estimate how HRQoL varied according to the presence or absence of metabolic syndrome. Because the a priori hypothesis specified that metabolic syndrome was associated with HRQoL, we included this condition in the multivariable model for each outcome, regardless of its univariable association.
Secondary analyses were performed to determine whether a statistically significant interaction was present between metabolic syndrome and BMI as a continuous variable, as well as metabolic syndrome and diabetes. The association between metabolic syndrome and HRQoL was also examined in a stratified analysis by BMI and by presence or absence of diabetes. We also evaluated whether the number of metabolic syndrome criteria met by the participants (i.e., three versus four versus five criteria) affected the relationship between metabolic syndrome and HRQoL. Lastly, we looked at the association between individual components of metabolic syndrome (excluding elevated waist circumference, as all participants in the trial met this criteria) and HRQoL. All analyses were conducted using Stata, Version 10.1 for Windows (Stata Corporation, College Station, TX). A p value of < 0.05 was considered significant for all analyses.
A total of 269 (68%) of participants met three or more criteria for metabolic syndrome, with a mean (SD) of 3.7 (0.7) criteria. All 390 (100%) participants met the criterion for waist circumference. Of the remaining components of metabolic syndrome, 132 (33.8%) participants had fasting glucose ≥ 100 mg/dl, 184 (47.2%) had low HDL cholesterol, 242 (62.1%) had elevated triglycerides, and 300 (76.9%) had increased blood pressure. The 269 participants with metabolic syndrome were older (p=0.028) and had higher weight (p=0.001) and BMI (p=0.040) compared with the 121 participants without the syndrome (Table 1). The former participants also took significantly more medications (p<0.001), although their FCI score did not differ significantly from that of participants without metabolic syndrome. As expected, participants with metabolic syndrome had significantly higher fasting blood glucose, higher triglycerides, higher blood pressure, and lower HDL-cholesterol than those without the condition (p ≤ 0.001 for all comparisons). Participants with metabolic syndrome had lower low-density-lipoprotein (LDL) cholesterol levels compared to those without the condition, although the difference did not reach statistical significance.
No differences in HRQoL were observed between those with and without metabolic syndrome using the generic measures (PCS-12, MCS-12, and EQ-5D) or the obesity-specific measure (IWQoL-Lite), as shown in Table 2. The mean (SD) score on the PCS-12 was 43.3 (9.7) in participants with metabolic syndrome, compared to 44.2 (9.6) in those without the condition. The MCS-12 scores for the two groups were 49.3 (10.0) and 49.4 (9.7), respectively. Participants with metabolic syndrome had a mean score of 0.820 (0.140), compared to 0.839 (0.141) in those without the condition (p=0.223).
The mean IWQoL-Lite total score in participants with metabolic syndrome was 67.2 (22.0), compared to 69.3 (21.9) in those without. None of the five IWQoL-Lite subscales was associated with metabolic syndrome.
Depression scores did not differ significantly between groups, but were consistent with mild symptoms of depression.
In the unadjusted analyses, metabolic syndrome was not associated with HRQoL on any of the four measures (Table 3). Depression, disease burden (as assessed by the FCI), and employment status were significantly associated with worse HRQoL on all four measures. Body mass index (BMI) was significantly associated with worse quality of life on the PCS-12 and the IWQoL-Lite, but was not associated with the MCS-12 or the EQ-5D. Higher PHQ-8 scores and higher BMI were significantly associated with worse quality of life on all five subscales of the IWQoL-Lite. Educational level, marital status, alcohol use, and medication count were not significantly associated with any of the quality of life measures.
In the adjusted models, only the depression score remained significantly associated with reduced HRQoL on all four outcome measures (Tables 4, ,5,5, ,6,6, and and7).7). Disease burden remained significantly associated with HRQoL on the PCS-12, MCS-12, and EQ-5D, but not the IWQoL-Lite, after controlling for confounding factors. Similarly, BMI remained significantly associated with HRQoL on the PCS-12 and IWQoL. Each of the four outcome measures also detected unique factors that were significantly associated with HRQoL.
There was no evidence of statistical interaction between metabolic syndrome and continuous BMI or between metabolic syndrome and diabetes in the adjusted models. In the analysis stratified by BMI, there were no significant associations between metabolic syndrome and HRQoL on any of the four quality of life measures. The relationship between increasing number of metabolic syndrome components and quality of life was also examined. There was no association between metabolic syndrome and HRQoL on any of the quality of life measures when participants with four or five criteria (n=152) were compared to those who met three or fewer components (n=238). Specific components of the metabolic syndrome were also not significantly associated with HRQoL.
Surprisingly, metabolic syndrome was not associated with impaired quality of life, despite a robust analysis that included two generic measures and an obesity-specific instrument (four outcomes in total). Participants in both groups reported slightly lower than average health status on the PCS-12 and relatively average health status on the MCS-12. [A summary score of 50 on both the physical and mental components of the SF-12 is indicative of average HRQoL (29)]. In contrast, both groups reported relatively high quality of life on the second generic measure, the EQ-5D. Although specific cutoffs for quality of life have not been reported for the IWQoL-Lite (42), the scores for participants with and without metabolic syndrome were intermediate between those reported in similar populations of obese individuals (range of 54.6 to 77.6) (35, 42–44). Mean IWQoL-Lite scores of 91.8 were reported in the nonobese general population (42).
Secondary analyses showed no evidence of an interaction between metabolic syndrome and obesity class or between metabolic syndrome and diabetes. This is notable given that participants with metabolic syndrome were more likely to be female, older, of lower socioeconomic status, and to have a higher BMI and greater medication use, compared to those without metabolic syndrome. Participants with metabolic syndrome were also more likely to have hypertension and hyperlipidemia, although LDL-cholesterol levels were lower in this group. (The latter finding is likely explained by the fact that more participants in the metabolic syndrome group were taking lipid-lowering medications, compared to those without the condition.) Finally, the number of criteria for metabolic syndrome did not influence the null relationship between metabolic syndrome and HRQoL. Because all study participants had to have at least two criteria for metabolic syndrome to be eligible for enrollment, individuals with and without metabolic syndrome had similar baseline characteristics from a clinical perspective. Thus, it was more difficult to detect small differences between groups, and the association between metabolic syndrome and HRQoL was likely attenuated.
Consistent with previous studies, we found that BMI was associated with lower scores on the PCS-12 and IWQoL-Lite, but not on the mental health measures (13,20,45). As the impact of obesity on physical functioning is well established (11–13), the lack of association between HRQoL and metabolic syndrome in more obese subgroups was unexpected. A threshold BMI may exist for which obesity modifies the relationship between metabolic syndrome and HRQoL. Study participants were required to have a BMI between 30 kg/m2 and 50 kg/m2. Significant differences in quality of life may have been detected if individuals with extreme obesity (BMI ≥ 50 kg/m2) and metabolic syndrome had been included.
Mean scores on the PCS-12, MCS-12, and the IWQoL-Lite, for participants with and without metabolic syndrome, were remarkably similar to scores reported in previous studies that evaluated the impact of obesity on quality of life using the same instruments (22,30,35). IWQoL-Lite scores in the present study were also very similar to those reported in a study that used this measure to compare weight-related quality of life in 1197 obese participants with and without diabetes (35).
Findings from this study raise questions about a conceptual model relating obesity, the number of comorbid conditions, depression, and quality of life. Depression was associated with decreased quality of life across measures, while obesity and number of comorbidities were not. One possible explanation is that impaired quality of life is an intermediary which relates these conditions. In one study, for example, increasing BMI was associated with greater reports of physical pain (a component of QOL) which, in turn, were associated with greater symptoms of depression (46). BMI alone was not associated with increased depression scores. Impaired quality of life may also act as a link between comorbidity and depression (47–49). This framework would explain why we found depression to be associated with quality of life on all four measures, while the associations between comorbid conditions, obesity, and HRQoL were more inconsistent. However, the causality and time course of these relationships remain largely unknown. Therefore, prospective longitudinal studies, which include estimates of these factors and quality of life, are needed to elucidate their interrelations.
The present investigation had several strengths. We performed a robust analysis that included important factors that may affect quality of life, including depression and estimates of disease burden. Previous studies have failed to adjust for many of these confounding variables (16–20). Ford and colleagues reported that metabolic syndrome was significantly associated with reduced quality of life in a population of 1,859 U.S. adults. However, only age, sex, ethnicity, educational status and smoking status were considered as covariates in their regression model (16). Similarly, Miettola and colleagues examined the relationship between metabolic syndrome and HRQoL in a Finnish population, but only adjusted for age, gender, marital status, education, employment status, smoking, and physical activity (17). The FCI in the present study indicated that participants had few comorbidities and were in relatively good health. Thus a volunteerism effect may have influenced our findings, in which individuals who elected to participate in the POWER Trial may have had better HRQoL than those who did not volunteer.
An additional strength of the present study was the use of several measures of HRQoL. Although the SF-36 (and the SF-12) and the EQ-5D have been widely used in weight loss studies, few studies have also included a disease-specific measure (50). Obesity-specific instruments may complement generic measures in capturing more subtle differences in health among participants with milder obesity and reductions in aspects of mental or physical health unique to obesity (20,46). Although metabolic syndrome was not associated with decreased quality of life on the IWQoL-Lite, this measure did detect differences in HRQoL that were not observed using the generic instruments.
This study also had several limitations. The particular version of the SF-12 that was used in this study only allowed the calculation of the summary scales but did not allow calculation of the individual subscales. Thus, differences between groups may have been present in the subscales that we were not able to detect. As previously noted, there were few clinically significant differences between participants with and without metabolic syndrome, making it difficult to detect small differences between groups.
In summary, metabolic syndrome was not associated with impaired HRQoL using two generic measures and an obesity-specific instrument. Diabetes and increasing obesity did not modify this relationship. These findings suggest that metabolic syndrome in itself is not associated with a decreased quality of life, but other factors such as obesity, depression, and greater disease burden may have significant effects on quality of life in this population. Larger studies that utilize multiple measures of quality of life and include the important covariates described previously are needed to confirm these findings.
We thank Christopher Vinnard, MD, MPH, MSCE for his editorial assistance.
Grant Support: This study was supported by grants from the National Heart, Lung and Blood Institute (U01HL087072-04) and the National Institute of Diabetes and Digestive and Kidney Diseases (5K24DK065018-07).
Conflict of Interest: The authors have no conflicts of interest to disclose.