Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

The relationship between body mass index and health-related quality of life: comparing the EQ-5D, EuroQol VAS and SF-6D

Abstract

Background:

No other studies have compared the relationship between body mass index (BMI) and health-related quality of life (HRQL) on more than one utility measure. Estimating the HRQL effects of obesity on a (common) utility scale enables the relative cost-effectiveness of interventions designed to alleviate obesity to be estimated.

Objective:

To examine the relationship between BMI and HRQL according to the EQ-5D, EuroQol visual analogue scale (EQ-VAS) and SF-6D.

Methods:

Patients aged 45 years at one UK general practice were asked to complete the EQ-5D, EQ-VAS, SF-36 questionnaire (used to derive the SF-6D), and information on their characteristics and co-morbidity. Body mass index was categorized according to the World Health Organization (WHO) recommendations. Regression analysis was used to compare the HRQL of normal BMI patients to the HRQL of patients in other BMI categories, while controlling for patient characteristics and co-morbidity.

Results:

A total of 1865 patients responded (67%), mean BMI 26.0 kg/m2, 16% obese (BMI30). Patients with back pain, hip pain, knee pain, asthma, diabetes or osteoarthritis were also significantly more likely to be obese. After controlling for other factors, compared to normal BMI patients, obese patients had a lower HRQL according to the EQ-5D (P<0.01), EQ-VAS (P<0.001) and SF-6D (P<0.001). Pre-obese patients were not estimated to have a significantly lower HRQL, and underweight patients were only estimated to have a significantly lower HRQL according to the SF-6D. These results arose because, on the EQ-5D, obese patients were found to have significantly more problems with mobility and pain, compared to physical functioning, social functioning and role limitations on the SF-6D. Whereas, according to the SF-6D, underweight patients had significantly more problems on the dimension of role limitation.

Conclusion:

The EQ-5D, EQ-VAS and SF-6D were in agreement that, relative to a normal BMI, obesity is associated with a lower HRQL, even after controlling for patient characteristics and co-morbidity. These three measures are thereby sensitive to the HRQL effects of obesity and can be used to estimate the cost-effectiveness of interventions designed to alleviate obesity.

Introduction

There is increasing evidence that obesity is associated with a loss in health-related quality of life (HRQL).1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 To our knowledge, only one study has shown that utility measures, which are designed to estimate and compare the benefits of many interventions on a common scale (where zero equals death and full health equals one), can detect the loss in HRQL associated with obesity.15 The necessity for such utility measures is demonstrated by the fact that both the United States (US) Public Health Service Panel on Cost-effectiveness in Health and Medicine and the United Kingdom (UK) National Institute for Health and Clinical Excellence (NICE) prefer utility measures to be used when assessing the cost-effectiveness of new, and current, interventions.16, 17

A number of utility measures can be used to assess the benefits of interventions, including the EuroQol EQ-5D18 and the SF-6D19 (derived from responses to the SF-36 questionnaire20). These utility measures are based on different health state descriptions and different underlying valuation systems such that there is the potential for each utility measure to assign a different utility score to the same patient, as has been shown previously.21, 22, 23 This in turn means that the choice of utility measure is likely to be influential in determining estimates of the effectiveness, and thereby cost-effectiveness, of the same intervention. As a result, it has been suggested that further comparative work be undertaken, particularly in previously less well researched areas, in order to understand better the implications of the choice of utility measure.19, 24 Consequently, we compared the relative ability of the EQ-5D and SF-6D utility measures to detect the HRQL loss associated with obesity in a sample of the general population. Responses for the EuroQol visual analogue scale (EQ-VAS), which has previously been converted into utility scores to estimate the relationship between body mass index (BMI) and utility,25 are also presented.

Previous studies1, 2, 6, 7, 9, 10, 11, 12, 13, 14, 15 have controlled, to varying degrees, for potentially confounding variables when assessing the relationship between BMI and HRQL. We seek to build upon these studies by estimating the relationship between obesity and HRQL after adjusting for a number of socio-demographic variables and obesity-related co-morbidities. Thereby, we assess whether obese patients have a lower HRQL than normal BMI patients with similar levels of co-morbidity. This comparison is important as it will enable one to estimate whether obesity is associated with a loss of HRQL, or whether obese patients have a lower HRQL because of the increased rates of co-morbidity that are found in obese patients. Similarly, we will be able to assess whether our results are comparable to others13, 15 who have estimated that being pre-obese (BMI 25 to <30 kg/m2) is not associated with a loss of HRQL after controlling for patient characteristics and co-morbidity.

Materials and methods

Participants and study design

As part of the recruitment for a study designed to assess the effectiveness and cost-effectiveness of different lifestyle interventions for knee pain (LIKP) all patients, aged 45 years and registered at one UK general practice, were sent an ascertainment questionnaire from their general practitioner, with the exception of patients that were deemed (by their general practitioner) to be unable to complete information requested in a questionnaire. The ascertainment questionnaire requested information on patient characteristics (age, sex, height, weight and smoking status), HRQL (according to the SF-36 questionnaire20 and the EuroQol questionnaire18) and co-morbidity. Information requested included 10 co-morbidities that have either a direct or indirect relationship to obesity – knee pain,26 back pain, hip pain,27 heart disease, stroke, cancer, diabetes, osteoarthritis,28 asthma29 and rheumatoid arthritis.30 Patients were asked to report whether they had had the first three of these co-morbidities on most days of the last month, and for the last seven they were asked whether they had ever been diagnosed with them.

Body mass index

A person's BMI is calculated by dividing their weight (in kilograms) by their height (in metres) squared (kg/m2). The World Health Organization (WHO) has categorized BMI scores into three main groups: (i) underweight (<18.5 kg/m2), (ii) normal range (18.5 to <25 kg/m2) and (iii) overweight (25 kg/m2).31 Overweight is further subdivided into four groups: pre-obese (25 to <30 kg/m2), obese class I (obese I) (30 to <35 kg/m2), obese class II (obese II) (35 to <40 kg/m2) and obese class III (obese III) (40 kg/m2). In light of this categorization, and in line with US National Institutes of Health (NIH) clinical guidelines,32 within this paper we refer to those patients who have a BMI30 kg/m2 as obese.

Measures of health-related quality of life

The EQ-5D is made up of five dimensions (mobility, self-care, usual activities, pain or discomfort, and anxiety or depression), each of which can be rated at one of three levels (no problem, some problems, extreme/severe problems).18 These combine to create 243 possible health states. Using the time trade-off technique,33 the relative preference for each of these health states has been estimated by assigning utility scores to each of the 243 health states. EQ-5D utility scores range between a full health score of 1 (where the respondent has no problems on any dimension) and the lowest score of −0.59 (when the respondent reports that they are at the bottom level of each dimension).34 The EuroQol questionnaire also contains a visual analogue scale (EQ-VAS), where patients are asked to rate their current health state on a 0 (worst imaginable health state) to 100 (best imaginable health state) scale.18

The SF-6D is derived from responses to 11 questions in the SF-36 questionnaire, although it is not always necessary for all 11 questions to be fully completed in order for an SF-6D score to be calculated (see Gerard et al.35 for further discussion). It has six dimensions (physical functioning, role limitations, social functioning, pain, mental health, and vitality), each of which has between four and six levels. Utility scores for the SF-6D were derived using a modified standard gamble approach.19 Full health is again assigned a score of 1 and the lowest SF-6D score is 0.296. A review of the EQ-5D and SF-6D has been published elsewhere,36 as is the case for the EQ-VAS.18

Statistical analyses

For the purpose of this paper, the following variables were extracted from the data set: BMI (kg/m2), age (years), smoking status (whether the patient had ever smoked regularly, for a period of at least 3 months), sex, information on the 10 co-morbidities and responses to each of the three HRQL measures (the EQ-5D, EQ-VAS and SF-6D). The following analyses were undertaken. Chi-square (χ2) tests were undertaken to assess whether BMI scores varied according to age, sex, smoking status, and presence of the 10 co-morbidities. The mean and standard deviation scores were estimated for the EQ-5D, EQ-VAS and SF-6D across each of the six BMI categories. One-way analysis of variance (ANOVA) was also performed for each of the three HRQL measures, where the HRQL of patients in each BMI category was compared to the HRQL of normal BMI patients.

Multiple linear regression analyses were subsequently undertaken to estimate how HRQL varied according to BMI, while controlling for the effects of age, sex, smoking status, back pain, hip pain, knee pain, heart disease, stroke, asthma, cancer, diabetes, rheumatoid arthritis and osteoarthritis. Body mass index took the form of a categorical variable, where the HRQL of patients in each BMI category was compared to the HRQL of normal BMI patients. The number of obese class II and III patients was expected to be relatively small, however, so in these and subsequent regressions, all patients in obese class I, II and III were grouped together into one obese group (BMI30 kg/m2). Separate regressions were performed with each of the EQ-5D, EQ-VAS and SF-6D taking the role of the dependent variable.

Finally, binary logistic regression analyses were performed in order to identify which of the five dimensions of the EQ-5D (and six dimensions of the SF-6D), obese patients have more problems with. In each of these regressions the binary dependent variable was made up of (i) those who reported being at level 1 on the dimension (i.e. having no problems) and (ii) those who reported being at other levels on the dimension. The same independent variables, and the same comparisons, were used as those in the multiple linear regression analyses. All statistical analyses were conducted in SPSS version 12, a P-value of <0.05 was deemed significant.

Results

Participants

On July 1 2004, 3122 (46.1%) of the 6765 patients registered with the general practice were aged 45 years. Of these, 2770 were deemed well enough to take part in the study and were sent the ascertainment questionnaire; 1865 (67.3%) returned the questionnaire. The mean age of the 1768 who reported their date of birth was 64.7 years (range 45.9 to 99.7 years), and 55.3% of the 1785 who reported their sex were female. Scores for the EQ-5D, EQ-VAS and SF-6D were calculated for 1737 (93.1%), 1719 (92.2%) and 1612 (86.4%) respondents, respectively. The BMI could be calculated for the 1730 (92.7%) patients who reported both their height and weight. The mean BMI was 25.98 kg/m2 (range 14.25–51.55 kg/m2), and 24 were classified as underweight, 804 as normal, 625 as pre-obese, 210 as obese I, 51 as obese II and 16 as obese III (Table 1).

Table 1 Variation in BMI according to age, sex, smoking status, back pain, hip pain, knee pain and other co-morbidities

Statistical analyses

Results in Table 1 show that BMI scores varied according to age (P=0.014), sex (P<0.001) and smoking status (P=0.013). Patients 75 years had the largest proportion of underweight patients and lower obesity levels than other age groups, whereas the youngest age group (45–54 years) had the greatest proportion of obese patients (I, II and III together). A greater proportion of females, than males, were underweight or obese, and smokers generally had lower BMI scores than non-smokers. With regard to each of the 10 co-morbidities, a greater proportion of those who had each respective co-morbidity were classified as obese compared to those who did not have the co-morbidity. Body mass index was, however, only estimated to vary significantly according to the level of back pain, hip pain, knee pain, asthma, diabetes, and osteoarthritis. The extent of the variation is demonstrated by the fact that 37.3% of patients with diabetes were classified as obese compared to 14.5% of those without diabetes, whereas the obesity rates for those with and without stroke (23.3 and 15.8%, respectively) were more similar.

The mean EQ-5D, EQ-VAS and SF-6D scores across each of the six BMI categories are reported in Table 2. It can be seen that on all three measures normal BMI patients had a significantly higher mean HRQL scores than obese I, II and III patients. Moreover, HRQL tended to deteriorate with increasing levels of obesity (although the mean SF-6D score for the 13 obese III patients was marginally higher than for the 43 obese II patients). In addition, according to the EQ-5D and EQ-VAS pre-obese patients had a significantly lower HRQL, as did underweight patients according to the SF-6D.

Table 2 Mean and standard deviation scores for each of the three HRQL measures (the EQ-5D, EQ-VAS and SF-6D), categorized according to the six BMI groups

The multiple linear regression analyses (Table 3) show that BMI, age, sex, smoking status and the 10 co-morbidities can explain between 24.9% (EQ-VAS) and 34.2% (EQ-5D) of the variation in the three HRQL variables. All three measures estimated that obese patients had a significantly lower HRQL than normal BMI patients after controlling for other factors. Only the SF-6D was, however, able to detect a significant difference between the HRQL of underweight patients and normal BMI patients. None of the three measures estimated there to be a significant difference in HRQL between pre-obese patients and normal BMI patients. When controlling for other factors it should also be noticed that each of the EQ-5D, EQ-VAS and SF-6D estimated that the loss of HRQL associated with obesity was equivalent to approximately half of the original mean difference in HRQL between obese and normal BMI patients. This can be seen by comparing the parameter estimates in the final column of Table 2 to those in the fourth row (obese I–III vs normal BMI) of Table 3.

Table 3 Results of multiple linear regression analyses undertaken to estimate variation in the EQ-5D, EQ-VAS and SF-6D

In each of the three linear regressions obesity was also estimated to have a loss of HRQL comparable to many of the co-morbidities – the mean loss of HRQL associated with obesity was estimated to be larger than asthma and cancer according to the EQ-5D, larger than asthma, stroke and osteoarthritis according to the EQ-VAS, and larger than cancer, diabetes and heart disease according to the SF-6D (results not shown in Table 3, but available from authors).

Results of the binary logistic regressions for the five dimensions of the EQ-5D are shown in Table 4. It can be seen that the obese patients were estimated to be 2.76 times more likely to experience mobility problems than normal BMI patients (P<0.001), and 1.94 times more likely to suffer from pain (P<0.001). Underweight patients were estimated to be 2.48 times more likely to be anxious or depressed (P<0.05). The pre-obese were 1.47 times more likely to experience mobility problems compared to patients with a BMI in the normal range (P<0.05).

Table 4 Results of binary logistic regressions analyses, performed in order to explain variation in the five dimensions of the EQ-5D

When attempting to explain variation in the dimensions of the SF-6D it was estimated that obese patients were 2.70 times more likely to have problems with physical functioning (P<0.001), 1.57 times more likely to experience problems with social functioning (P<0.01) and 1.46 times more likely to have role limitations (P<0.05) (Table 5). In addition, underweight patients were estimated to be 2.56 times more likely to have problems on the role limitations dimension (P<0.05), whereas the pre-obese were 1.53 times more likely to have significant problems with physical functioning (P<0.05).

Table 5 Results of the binary logistic regression analyses, performed in order to explain variation in the six dimensions of the SF-6D

Discussion

We have shown that according to each of the EQ-5D, EQ-VAS and SF-6D, obese patients have a significantly lower HRQL than normal BMI patients, pre-obese patients have a significantly lower HRQL according to the EQ-5D and EQ-VAS, and underweight patients have a significantly lower HRQL according to the SF-6D. After controlling for the effects of age, sex, smoking status and co-morbidity, pre-obese patients are not estimated to have significantly lower HRQL levels, but all other effects remained significant (although often at reduced levels). Thus, we have shown that according to all three HRQL measures obesity is associated with a significant decrease in HRQL and that this does not solely arise because of the increased rates of co-morbidity that are found in obese patients. Controlling for co-morbidities does, however, reduce the mean loss of HRQL associated with obesity to approximately half of the actual mean difference in HRQL between obese and normal BMI patients. The mean loss of HRQL associated with obesity is nevertheless still estimated to be greater than that associated with a number of other co-morbidities. Such effects result from the fact that, relative to normal BMI patients, obese patients were estimated to be significantly more likely to have problems with mobility and pain according to the EQ-5D, and physical functioning, social functioning and role limitations according to the SF-6D. The significantly lower SF-6D score for underweight patients was due to their likelihood of having problems on the role limitations dimension.

Comparisons with other studies

Our results are comparable to others who have shown that BMI scores are related to age, sex and smoking status, and the level of co-morbidity.2, 9, 10, 13, 14, 15, 37 We are also in agreement with others who have found that obese patients have lower levels of HRQL,1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 and that the estimated loss of HRQL associated with obesity is reduced (although not eliminated) when levels of co-morbidity are controlled for.12, 13, 14, 15 Similarly, others have also found that pre-obese individuals do not have a significantly lower HRQL than patients with a normal BMI after other factors are controlled for.13, 15

With regard to the EQ-5D and SF-6D, it has been concluded that healthier states tend to be estimated to have a higher HRQL score according to the EQ-5D, compared to the SF-6D, whereas poorer health states tend to have a higher score according to the SF-6D.21, 36, 38, 39 Such conclusions accord with our findings that normal BMI patients have a higher mean HRQL score on the EQ-5D, and that obese patients have a higher mean score on the SF-6D (Table 2). After controlling for other factors, however, the estimated mean loss of utility associated with obesity was very similar according to the EQ-5D and SF-6D, at 0.040 and 0.038, respectively (Table 3). These findings are consistent with some studies that report that interventions are associated with similar mean change scores on the EQ-5D and SF-6D,40, 41 although other studies have found that the mean change scores on the two measures do not concur.21, 22, 23, 39 Thus, we have shown that, even though the EQ-5D has fewer health states with which to describe a person's HRQL, there is no evidence that it is less sensitive than the SF-6D when measuring the loss of HRQL associated with obesity.

Finally, we can compare our results to those of Jia and Lubetkin15 who also compared the HRQL of obese patients to normal BMI patients using the EQ-5D and EQ-VAS. In a study of >13 000 members of the US general population, they estimated that after controlling for patient characteristics and co-morbidity, relative to normal BMI patients, the mean HRQL of obese class I patients was 0.033 lower on the EQ-5D and 3.23 lower according to the EQ-VAS, and that for obese class II patients the mean loss was 0.073 and 4.84, respectively.15 Using similar analyses, we estimated the mean loss of HRQL associated with obesity (class I, II and III together) to be 0.040 on the EQ-5D and 5.3 on the EQ-VAS (Table 3). Thus, compared to Jia and Lubetkin,15 our estimates of the loss in HRQL associated with obesity are similar according to the EQ-5D, but slightly higher according to the EQ-VAS.

Limitations

This study has several caveats. First, it is possible that we have not controlled for all appropriate co-variants, in particular, we only controlled for 10 co-morbidities. Thus, another co-morbidity (e.g. gallbladder disease,42), which is more prevalent in obese patients and is associated with a lower HRQL in itself, could be the reason why obese patients were found to have a lower HRQL in this study. That said, others12, 13, 14, 15 have also found that obese patients have worse HRQL even after controlling for co-morbidities that have been shown to be associated with obesity. Second, as co-morbidities, height and weight were all self-reported they may be inaccurate. This is demonstrated by evidence which suggests that obese individuals tend to overestimate their height and underestimate their weight, whereas men are more likely than women to overestimate their height.43, 44 Third, BMI alone may not be the best measure of obesity as it does not directly incorporate body fat or fat distribution levels, which are independent predictors of health risk.32 Fourth, all participants in our study were aged 45 years. As such we are unable to comment upon whether BMI has a negative impact on HRQL in younger age groups, a result that has been found elsewhere.11

Strengths

The main strengths of the paper are that we have compared the performance of the three commonly used HRQL measures in a large population sample and focused on an issue (obesity) for which they have not previously been compared. We also advance upon a number of studies3, 4, 8 by controlling for co-morbidities that could potentially act as confounders when estimating the relationship between obesity and HRQL. Finally, we have demonstrated the reliability of our data as we have shown that obese patients in our study were more likely to have each of the 10 co-morbidities that patients were questioned about.

Implications

We have shown that each of the three HRQL measures were sensitive to the expected loss of HRQL associated with obesity. Given that two of these measures (the EQ-5D and SF-6D) can be used to compare the cost-effectiveness of different healthcare interventions on a common cost-utility scale, this provides an argument for at least one of them to be used in future studies that assess the effectiveness, and cost-effectiveness, of interventions designed to alleviate obesity. As only the SF-6D estimated that being underweight was associated with a loss of HRQL, this might suggest that the SF-6D should be favoured when assessing the value of interventions for this population group. Detecting the expected loss of utility associated with a particular health state (often referred to as construct or empirical validity24) is, however, only one criterion to be considered when deciding which utility measure to use (see Brazier et al.24 for a discussion of other criteria used to compare the performance of different utility measures). Additionally, the finding that, after controlling for co-morbidity, people with a normal BMI had a higher HRQL also implies that there are likely to be HRQL benefits associated with having a normal BMI (rather than being underweight or obese) even for those individuals with no co-morbidity.

Conclusion

We have shown that according to each of the EQ-5D, EQ-VAS and SF-6D, obese patients are estimated to have a significantly lower HRQL than normal BMI patients. This relationship remained even after controlling for patient characteristics and co-morbidity. In contrast, after controlling for other factors, none of the measures estimated that pre-obese patients had a significantly lower HRQL than normal BMI patients, although the SF-6D did estimate that underweight patients had a significantly lower HRQL than normal BMI patients. This study thereby provides an argument for utility measures to be used when assessing the cost-effectiveness of interventions designed to alleviate obesity. It also demonstrates that there are likely to be HRQL benefits of having a normal BMI, even for those patients with no co-morbidity.

References

  1. 1

    Barofsky I, Fontaine KR, Cheskin LJ . Pain in the obese: impact on health-related quality-of-life. Ann Behav Med 1997; 19: 408–410.

    CAS  Article  Google Scholar 

  2. 2

    Brown WJ, Dobson AJ, Mishra G . What is a healthy weight for middle aged women? Int J Obes Relat Metab Disord 1998; 22: 520–528.

    CAS  Article  Google Scholar 

  3. 3

    Han TS, Tijhuis MA, Lean ME, Seidell JC . Quality of life in relation to overweight and body fat distribution. Am J Public Health 1998; 88: 1814–1820.

    CAS  Article  Google Scholar 

  4. 4

    Hill AJ, Williams J . Psychological health in a non-clinical sample of obese women. Int J Obes Relat Metab Disord 1998; 22: 578–583.

    CAS  Article  Google Scholar 

  5. 5

    Lean ME, Han TS, Seidell JC . Impairment of health and quality of life in people with large waist circumference. Lancet 1998; 351: 853–856.

    CAS  Article  Google Scholar 

  6. 6

    Le Pen C, Levy E, Loos F, Banzet MN, Basdevant A . ‘Specific’ scale compared with ‘generic’ scale: a double measurement of the quality of life in a French community sample of obese subjects. J Epidemiol Community Health 1998; 52: 445–450.

    CAS  Article  Google Scholar 

  7. 7

    Fine JT, Colditz GA, Coakley EH, Moseley G, Manson JE, Willett WC et al. A propective study of weight change and health-related quality of life in women. JAMA 1999; 282: 2136–2142.

    CAS  Article  Google Scholar 

  8. 8

    Lean ME, Han TS, Seidell JC . Impairment of health and quality of life using new US federal guidelines for the identification of obesity. Arch Intern Med 1999; 159: 837–843.

    CAS  Article  Google Scholar 

  9. 9

    Doll HA, Petersen SEK, Stewart-Brown SL . Obesity and physical and emotional well-being: associations between body mass index, chronic illness, and the physical and mental components of the SF-36 questionnaire. Obes Res 2000; 8: 160–170.

    CAS  Article  Google Scholar 

  10. 10

    Ford ES, Moriarty DG, Zack MM, Mokdad AH, Chapman DP . Self-reported body mass index and health-related quality of life: findings from the Behavioral Risk Factor Surveillance System. Obes Res 2001; 9: 21–31.

    CAS  Article  Google Scholar 

  11. 11

    Larsson U, Karlsson J, Sullivan M . Impact of overweight and obesity on health-related quality of life – a swedish population study. Int J Obes Relat Metab Disord 2002; 26: 417–424.

    CAS  Article  Google Scholar 

  12. 12

    Heo M, Allison DB, Faith MS, Zhu S, Fontaine KR . Obesity and quality of life: mediating effects of pain and comorbidities. Obes Res 2003; 11: 209–216.

    Article  Google Scholar 

  13. 13

    Groessl EJ, Kaplan RM, Barrett-Connor E, Ganiats TG . Body mass index and quality of well-being in a community of older adults. Am J Prev Med 2004; 26: 126–129.

    Article  Google Scholar 

  14. 14

    Yan LL, Daviglus ML, Liu K, Pirzada A, Garside DB, Schiffer L et al. BMI and health-related quality of life in adults 65 years and older. Obes Res 2004; 12: 69–76.

    Article  Google Scholar 

  15. 15

    Jia H, Lubetkin EI . The impact of obesity on health-related quality-of-life in the general adult US population. J Public Health (Oxford) 2005; 27: 156–164.

    Article  Google Scholar 

  16. 16

    Gold MR, Seigel JE, Russell LB, Weinstein MC . Cost-Effectiveness in Health and Medicine. Oxford University Press: New York, 1996.

    Google Scholar 

  17. 17

    National Institute of Health and Clinical Excellence. Guide to the Methods of Technology Appraisal. Abba Litho Sales Limited: London, UK, 2004.

  18. 18

    Brooks R . EuroQol: the current state of play. Health Policy 1996; 37: 53–72.

    CAS  Article  Google Scholar 

  19. 19

    Brazier J, Roberts J, Deverill M . The estimation of a preference-based measure of health from the SF-36. J Health Econ 2002; 21: 271–292.

    Article  Google Scholar 

  20. 20

    Ware JE, Sherbourne C . The MOS 36 item short-form health survey: conceptual framework and item selection. Med Care 1992; 30: 473–483.

    Article  Google Scholar 

  21. 21

    Longworth L, Bryan S . An empirical comparison of EQ-5D and SF-6D in liver transplant patients. Health Econ 2003; 12: 1061–1067.

    Article  Google Scholar 

  22. 22

    Conner-Spady B, Suarez-Almazor ME . Variation in the estimation of quality-adjusted life-years by different preference-based instruments. Med Care 2003; 41: 791–801.

    Article  Google Scholar 

  23. 23

    Pickard AS, Johnson JA, Feeny DH . Responsiveness of generic health-related quality of life measures in stroke. Qual Life Res 2005; 14: 207–219.

    Article  Google Scholar 

  24. 24

    Brazier JE, Deverill M, Green C, Harper R, Booth A . A review of the use of health status measures in economic evaluation. Health Technol Assess 1999; 3: 1–164.

    Google Scholar 

  25. 25

    Hakim Z, Wolf A, Garrison LP . Estimating the effect of body mass index on health state preferences. Pharmacoeconomics 2002; 20: 393–404.

    Article  Google Scholar 

  26. 26

    Spector TD, Hart DJ, Doyle DV . Incidence and progression of osteoarthritis in women with unilateral knee disease in the general population: the effect of obesity. Ann Rheum Dis 1994; 53: 565–568.

    CAS  Article  Google Scholar 

  27. 27

    Andersen RE, Crespo CJ, Bartlett SJ, Bathon JM, Fontaine KR . Relationship between body weight gain and significant knee, hip, and back pain in older Americans. Obes Res 2003; 11: 1159–1162.

    Article  Google Scholar 

  28. 28

    National Task Force on the Prevention and Treatment of Obesity. Overweight, obesity, and health risk. Arch Intern Med 2000; 160: 898–904.

  29. 29

    Ronmark E, Andersson C, Nystrom L, Forsberg B, Jarvholm B, Lundback B . Obesity increases the risk of incident asthma among adults. Eur Respir J 2005; 25: 282–288.

    CAS  Article  Google Scholar 

  30. 30

    Escalante A, Haas RW, del Rincon I . Measurement of global functional performance in patients with rheumatoid arthritis using rheumatology function tests. Arthritis Res Ther 2004; 6: S315–S325.

    Article  Google Scholar 

  31. 31

    WHO. Obesity: Preventing and Managing the Global Epidemic. Report of a WHO consultation on Obesity. WHO: Geneva, 2001.

  32. 32

    National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults – the evidence report. Obes Res 1998; 6 (Suppl 2): 51S–209S.

  33. 33

    Torrance GW . Measurement of health state utilities for economic appraisal. J Health Econ 1986; 5: 1–30.

    CAS  Article  Google Scholar 

  34. 34

    Dolan P, Gudex C, Kind P, Williams A . A Social Tariff for the EuroQol: Results from a UK General Population Survey (Discussion Paper 138). Centre for Health Economics, University of York, 1995.

    Google Scholar 

  35. 35

    Gerard K, Nicholson T, Mullee M, Mehta R, Roderick P . EQ-5D versus SF-6D in an OLDER, CHRONIcally Ill Patient Group. Appl Health Econ Health Policy 2004; 3: 91–102.

    Article  Google Scholar 

  36. 36

    Marra CA, Woolcott JC, Kopec JA, Shojania K, Offer R, Brazier JE et al. A comparison of generic, indirect utility measures (the HUI2, HUI3, SF-6D, and the EQ-5D) and disease-specific instruments (the RAQoL and the HAQ) in rheumatoid arthritis. Soc Sci Med 2005; 60: 1571–1582.

    Article  Google Scholar 

  37. 37

    Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP . The continuing epidemics of obesity and diabetes in the United States. JAMA 2001; 286: 1195–1200.

    CAS  Article  Google Scholar 

  38. 38

    Szende A, Svensson K, Stahl E, Meszaros A, Berta GY . Psychometric and utility-based measures of health status of asthmatic patients with different disease control level. Pharmacoeconomics 2004; 22: 537–547.

    Article  Google Scholar 

  39. 39

    Bryan S, Longworth L . Measuring health-related utility: why the disparity between EQ-5D and SF-6D? Eur J Health Econ 2005; 6: 253–260.

    Article  Google Scholar 

  40. 40

    Barton GR, Bankart J, Davis AC, Summerfield AQ . Comparing utility scores before and after hearing-aid provision: results according to the EQ-5D, HUI3 and SF-6D. Appl Health Econ Health Policy 2004; 3: 103–105.

    Article  Google Scholar 

  41. 41

    Fisk JD, Brown MG, Sketris IS, Metz LM, Murray TJ, Stadnyk KJ . A comparison of health utility measures for the evaluation of multiple sclerosis treatments. J Neurol Neurosurg Psychiatry 2005; 76: 58–63.

    CAS  Article  Google Scholar 

  42. 42

    Colditz GA . Economic costs of obesity and inactivity. Med Sci Sports Exerc 1999; 31: S663–S667.

    CAS  Article  Google Scholar 

  43. 43

    Palta M, Prineas RJ, Berman R, Hannan P . Comparison of self-reported and measured height and weight. Am J Epidemiol 1982; 115: 223–230.

    CAS  Article  Google Scholar 

  44. 44

    Rowland ML . Self-reported weight and height. Am J Clin Nutr 1990; 52: 1125–1133.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We thank all patients who completed the Lifestyle Interventions for Knee Pain (LIKP) study questionnaire. The LIKP study was funded by the UK Arthritis Research Campaign (ARC) (Grant number 13550). PhD funding for Garry Barton was provided by the UK Economic & Social Research Council (ESRC) (PTA-037-2004-00051).

Author information

Affiliations

Authors

Corresponding author

Correspondence to T H Sach.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Sach, T., Barton, G., Doherty, M. et al. The relationship between body mass index and health-related quality of life: comparing the EQ-5D, EuroQol VAS and SF-6D. Int J Obes 31, 189–196 (2007). https://doi.org/10.1038/sj.ijo.0803365

Download citation

Keywords

  • body mass index
  • health-related quality of life
  • EQ-5D
  • SF-6D

Further reading

Search

Quick links