Influence of BMI on Health-related Quality of Life: Comparison Between an Obese Adult Cohort and Age-matched Population Norms

Authors

  • Ananthila Anandacoomarasamy,

    1. Institute of Bone and Joint Research, Kolling Institute, Department of Rheumatology, Royal North Shore Hospital, University of Sydney, Sydney, New South Wales, Australia
    Search for more papers by this author
  • Ian D. Caterson,

    1. Department of Human Nutrition, Institute of Obesity Nutrition and Exercise, University of Sydney, Sydney, New South Wales, Australia
    Search for more papers by this author
  • Steven Leibman,

    1. Department of Surgery, Royal North Shore Hospital, Sydney, New South Wales, Australia
    Search for more papers by this author
  • Garett S. Smith,

    1. Department of Surgery, Royal North Shore Hospital, Sydney, New South Wales, Australia
    Search for more papers by this author
  • Phillip N. Sambrook,

    1. Institute of Bone and Joint Research, Kolling Institute, Department of Rheumatology, Royal North Shore Hospital, University of Sydney, Sydney, New South Wales, Australia
    Search for more papers by this author
  • Marlene Fransen,

    1. Musculoskeletal Division, The George Institute, University of Sydney, Sydney, New South Wales, Australia
    Search for more papers by this author
  • Lyn M. March

    Corresponding author
    1. Institute of Bone and Joint Research, Kolling Institute, Department of Rheumatology, Royal North Shore Hospital, University of Sydney, Sydney, New South Wales, Australia
    Search for more papers by this author

(lynmar@med.usyd.eu.au)

Abstract

The aim of this study was to determine health-related quality of life and fatigue measures in obese subjects and to compare scores with age- and gender-matched population norms. A total of 163 obese subjects were recruited from laparoscopic-adjustable gastric banding or exercise and diet weight loss programs between March 2006 and December 2007. All subjects completed the Medical Outcomes Study Short Form 36 (SF-36), Assessment of Quality of Life (AQoL), and Multidimensional Assessment of Fatigue (MAF) questionnaires. One-sample t-tests were used to compare transformed scores with age- and gender-matched population norms and controls. Obese subjects have significantly lower SF-36 physical and emotional component scores, significantly lower AQoL utility scores and significantly higher fatigue scores compared to age-matched population norms. Within the study cohort, the SF-36 physical functioning, role physical and bodily pain scores, and AQoL utility index were even lower in subjects with clinical knee osteoarthritis (OA). However, obese individuals without OA still had significantly lower scores compared to population norms. Obesity is associated with impaired health-related quality of life and disability as measured by the SF-36, AQoL, and fatigue score (MAF) compared to matched population norms.

Obesity is a major public health problem and carries the risk of developing significant medical problems such as diabetes and cardiovascular disease (1), and musculoskeletal conditions such as osteoarthritis (OA) and low back pain which are associated with restricted mobility, physical impairments, and disability (2). Obesity increases the risk of disability among people with and without self-reported arthritis (3,4).

Despite there being impaired quality of life in obese individuals, little is known about the comparison to population norms. Such comparisons assess the impact of obesity on the ability to work and participate in community activity and allow meaningful evaluation of the potential benefit of weight loss programs. The aim of this study was to determine health-related quality of life measures in an obese cohort and to compare these assessments with age- and gender-matched population norms.

Patients and Methods

Participants were recruited from patients voluntarily enrolled in nonsurgical (dietary modification and exercise) or surgical (laparoscopic-adjustable gastric banding) weight loss programs. The nonsurgical program was conducted at Metabolism and Obesity Services, Royal Prince Alfred Hospital, Sydney. Surgery was performed by two experienced surgeons at Royal North Shore Hospital, Sydney. All subjects were obese, most being obesity grade 2 or higher, i.e., BMI >35. Patients were screened and offered participation at their initial visit. The American College of Rheumatology clinical classification criteria were used to define knee OA (5) (knee pain and at least 3 of the following: age >50, stiffness <30 min, crepitus, bony tenderness, bony enlargement, and no palpable warmth).

Study participants reported on current (past month) and “ever” non-trauma-related knee joint pain. Pain, physical function, general health status, and fatigue were assessed with the Medical Outcomes Study Short Form 36 questionnaire (SF-36) (ref. 6), Assessment of Quality of Life (AQoL) (7) and Multidimensional Assessment of Fatigue (MAF) self-administered questionnaires (8). The SF-36, a health status measure, comprises 36 items organized into eight scales (physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotion, and mental health) (9). The AQoL, a validated quality-of-life tool, assesses five components (illness, independent living, social relationships, physical senses, and psychological well-being). The MAF measures five dimensions of fatigue: degree, severity, distress, impact on activities of daily living, and timing.

Ethics approval was obtained from the Northern Sydney Central Coast Area Health Service Human Research Ethics Committee and University of Sydney, and informed consent was obtained from all study participants.

Statistical analysis

Results for SF-36, AQoL, and MAF were transformed to obtain scores for each SF-36 dimension (100 = best possible score); AQoL utility score (1.00 = full health-related quality of life); and Global Fatigue Index (GFI) (50 = severe fatigue).

Mean scores were calculated for the whole group and stratified for age. The Mann–Whitney U-test was used to assess the effects of gender and clinical knee OA on each quality of life measure. One-sample t-tests were used to compare transformed scores with published general population norms or controls (10,11,12). A P value <0.05 (two-tailed) or 95% confidence interval not including the null point were considered significant. (SPSS statistical package, standard version 14.0; SPSS, Chicago, IL).

Results

From March 2006 to December 2007, 163 (50%) subjects were recruited from a potential number of 333. Those who declined quoted work and family commitments (25%), time limitations (30%), and lack of interest in participating in a research study (45%). There were 118 females (72%) and 45 males, mean age 50 (±12.1) years and mean BMI 41.8 (±6.7) kg/m2. Of them, 68 patients (42%) underwent laparoscopic gastric banding for weight loss; 55 subjects (34%) had clinical knee OA, 102 (63%) had current knee pain, and 121 (74%) had a prior history of knee pain. The prevalence of hypertension, hyperlipidemia, coronary artery disease, and diabetes was 60, 47, 7, and 27% respectively, and 54% of women had undergone menopause.

Obese subjects had significantly lower SF-36 scores for most age-matched physical component measures (physical functioning, role physical, bodily pain, and general health), and age-matched emotional component measures (vitality, social function, role emotion, and mental health) compared with population norms (Figure 1).

Figure 1.

Mean SF-36 (a) physical and (b) mental component scores compared to age-matched population norms. Filled bars represent obese cohort; open bars represent age-matched general population; *P < 0.05. SF-36, Medical Outcomes Study Short Form 36.

Men had higher physical functioning (P = 0.016), role physical (P = 0.03), and bodily pain scores (P = 0.006). When stratified for gender, obese women had significantly lower SF-36 scores for nearly all age-matched physical component measures compared to population norms (Table 1). As there were far fewer men in each age-group, the differences in physical component scores were less significant (Table 1). When stratified for clinical knee OA, these subjects had lower physical functioning (P < 0.0001), role physical (P = 0.007), and bodily pain scores (P < 0.0001). However, even obese subjects without clinical knee OA had significantly lower physical component scores compared with population norms, especially the 45–54 and 55–64 age-groups.

Table 1.  Quality of life and fatigue outcomes
inline image

The mean AQoL utility score (0.66 ± 0.24) was significantly lower in the obese group than the mean population norm (0.83 ± 0.20) (P < 0.0001) as well as for each age category (Table 1). There were no significant gender differences. Subjects with clinical knee OA had lower AQoL utility scores than those without (P = 0.002). However, obese subjects without clinical knee OA also had significantly lower AQoL utility scores (P < 0.0001) compared to population norms.

Mean fatigue score (GFI) was 24.9 (±12.6). This was significantly higher than scores in a control group (GFI = 17, P < 0.0001) (ref. 12). Age- and gender-matched population norms are not yet available for the GFI. Gender and clinical knee OA did not affect scores.

Discussion

This cross-sectional study of 163 obese adults with and without symptomatic knee OA demonstrates significant impairment in health-related quality of life and fatigue measures related to obesity. No studies so far have examined the magnitude of the deficits in health-related quality of life measures in an obese cohort compared to age- and gender-matched population norms.

In Australia, about 20% of the population is obese (13), and grade III obesity represents 8–10% of the total adult population. Most of the obese adults (58%) were from a nonsurgical program where subjects either self-referred or were referred by various health professionals. The clinic is provided by the health service at no charge and the only criterion for entry was the subject's BMI be >35. As described, a proportion of these had obesity-associated comorbidities, including arthritis. Most of the cohort came from metropolitan areas but there were also subjects from outer metropolitan and regional areas. Because of self referral and the lack of requirement for any specified disease, this nonsurgical sample is reasonably representative of the grade III obesity population in New South Wales, Australia. Subjects in the surgical group required BMI >35 plus a comorbidity or BMI >40 for entry. The wide socioeconomic background of the study cohort is also representative of the target population of obese Australians. Those obese adults who volunteered for this study from either program were more likely to be motivated and less disabled, possibly resulting in higher quality of life. It is also possible that obese patients with more severe comorbidities did not enter the study. These patients are more likely to have lower scores secondary to their comorbidities. Our findings may therefore underestimate the influence of obesity on quality of life.

The self-administered questionnaires used in this study are widely used, validated and easy to administer (6,7,12). As with general population norms (10), the SF-36 scores in the obese cohort, particularly physical subscale, declined with age. We observed lower scores for almost all SF-36 physical components and most emotional components compared to age-matched population norms. The influence of obesity appears to be higher for physical components than emotional components, e.g., mental health scores. Nevertheless, scores for vitality, role emotion, and social function were generally lower than corresponding norms, in contrast to a case–control study showing no significant differences in SF-36 psychological and social dimensions in obese subjects (14).

Men had significantly higher SF-36 physical component scores compared to women but differences compared to age-matched norms were not as obvious as in women. Although this may be related to the smaller numbers of men, it is possible that obese men have a different profile in the perception of physical health or a lower body fat percentage compared with obese women.

As patients with arthritis are more likely to have lower SF-36 scores, particularly physical subscale, the scores were also stratified by the presence of knee OA. Not unexpectedly, obese subjects with clinical knee OA had lower SF-36 physical component scores than those without. However, more tellingly, obese subjects without clinical knee OA also had lower physical component scores compared to age-matched norms. This indicates clearly that obesity itself has a significant impact on physical disability, and that this burden increases with the development of conditions such as knee OA. As an age cutoff (>50 years) was applied to minimize over-estimating clinical knee OA, it is possible that some younger subjects were “missed.” However, only 7 of the 37 subjects in the age-group 40–50 would otherwise have been classified as having clinical knee OA. This did not significantly alter results (data not shown). The SF-36 population norms used in the analysis were published in 1995. However, more recently the population norms for the state of South Australia were published in 2004 (ref. 15). These recent norm data are very similar to or lower than those published in 1995, and therefore are not expected to alter our findings.

The AQoL utility score enables the calculation of quality-adjusted life years for use in economic evaluation (11). The lower mean scores in obese subjects, both with and without knee OA, indicates poorer health-related quality of life as well as increased health-care utilization in the obese population. Measures of change in utility scores with interventions such as weight loss will enable a meaningful assessment of the societal economic impact of such programs.

Fatigue, an important outcome for many chronic conditions, has far-ranging consequences on patients' lives (16). Our obese cohort had fatigue scores closer to a group with rheumatoid arthritis (GFI = 29.2) than controls (GFI = 17.0), and their scores were not affected by the presence of OA (12). Our findings indicate that obese individuals have a high burden of fatigue. There is speculation of a vicious cycle between obesity and fatigue, with obesity per se leading to fatigue and fatigue playing an important role in poor weight management by restricting physical activity levels (17).

Obesity has a significant influence on the perception of health, quality of life, and fatigue. The mean scores for each SF-36 domain in this obese cohort was lower than or similar to mean scores observed in a cohort of patients with type 1 diabetes (18). This has important ramifications for self-esteem, participation in physical activity and ultimately the health system, work force and economy. Primary prevention is critical given the challenge of weight reduction and its costliness (1).

In conclusion, this is the first study to our knowledge to demonstrate significant impairment in physical functioning, health-related quality of life and fatigue measures in an obese cohort compared to age- and gender-matched population norms. The global obesity epidemic has significant consequences for both the individual and the community in terms of direct and indirect health-care costs. Longitudinal studies will enable meaningful assessment of both the impact on quality of life and health-care utilization with weight loss interventions.

Acknowledgmant

We thank Elisia Manson, Metabolism and Obesity Services, Royal Prince Alfred Hospital, and Sarah Fisher, NS Private Hospital for their help in accessing patients' records and J.S. Chen, Rheumatology, Royal North Shore Hospital, for statistical advice. We are also very grateful to the study participants. A.A. holds a National Health and Medical Research Council Medical Postgraduate Research Scholarship (ID number 402901).

Disclosure

The authors declared no conflict of interest.

Ancillary