International Journal of Obesity (2004) 28, 13741382 & 2004 Nature Publishing Group All rights reserved 0307-0565/04

$30.00 www.nature.com/ijo

PAPER
Overweight and obesity and the burden of disease and disability in elderly men
S Goya Wannamethee1*, A Gerald Shaper1, PH Whincup2 and M Walker1
1 2

Department of Primary Care and Population Science, Royal Free and University College Medical School, London, UK; and Department of Public Health Sciences, St Georges Medical School Hospital, London, UK

OBJECTIVE: To examine the prevalence of disease burden and disability associated with overweight and obesity in men aged 6079 y and to assess whether the current WHO weight guidelines are appropriate in the elderly. DESIGN: Cross-sectional survey 20 y after enrolment. SETTING: General practices in 24 British towns. PARTICIPANTS: In total, 4232 men aged 6079 y (77% of survivors) with measured weight and height. MAIN OUTCOME MEASURES: Cardiovascular (CV) risk factors, prevalence of diabetes, cardiovascular disease, cancer, disability and regular medication. RESULTS: In total, 17% of the men were obese (body mass index (BMI) Z30 kg/m2) and a further 52% were overweight (BMI 2529.9 kg/m2). Prevalence of hypertension, low HDL-cholesterol, high triglycerides and insulin resistance and the prevalence of most disease outcomes increased with increasing degrees of overweight/obesity. Men in the normal weight range (18.5 24.9 kg/m2) had the lowest prevalence of ill health. Compared with normal weight men, obese men showed a two-fold risk of major CVD (odds ratio (OR) 1.96, 95% CI 1.442.67) and locomotor disability (OR 2.26, 95% CI 1.66, 3.09) and were nearly three times as likely to have diabetes, CV interventions or to be on CV medication. Over 60% of the prevalence of high insulin resistance was attributable to overweight and obesity as was over a third of diabetes and hypertension, a quarter of locomotor disability and a fifth of major CVD. CONCLUSION: In elderly men, overweight and obesity are associated with a significantly increased burden of disease, in particular CV-related disorders and disability. The current guidelines for overweight and obesity appear to be appropriate in elderly men. International Journal of Obesity (2004) 28, 13741382. doi:10.1038/sj.ijo.0802775 Published online 31 August 2004 Keywords: overweight; cardiovascular disease; CV risk factors; disability

Introduction
During the past few decades, the prevalence of obesity (BMIZ30 kg/m2) has reached epidemic levels in Western societies.1,2 In the UK, 17% of men and 21% of women are obese.2 Overweight and obesity constitute a major public health problem because of the associated increased risk of

*Correspondence: Dr S Goya Wannamethee, Department of Primary Care and Population Science, Royal Free and University College Medical School, London NW3 2PF, UK. E-mail: goya@pcps.ucl.ac.uk Contributors: SGW and AGS developed the study aim and design. SGW carried out the statistical analyses and wrote the initial draft with AGS. PHW designed the study and contributed to the writing of the report. MW was study coordinator and responsible for the follow-up questionnaire data collection. Received 2 June 2003; revised 5 September 2003; accepted 4 November 2003; published online 31 August 2004

coronary heart disease (CHD), stroke, type 2 diabetes, hypertension, dyslipidemia, musculoskeletal disorders and some cancers.1,2 Mean body weight increases with age; in most studies the increase continues up to about age 60 and then levels off.3 Both the National Heart, Lung and Blood Institute (NHLBI) of the National Institutes of Health (NIH)4 and WHO5 now recommend 25 kg/m2 as the upper limit of ideal weight for all adults regardless of age. However, the importance of overweight and obesity as predictors of mortality in elderly people (465 y) has been controversial, with several studies failing to show any relation between overweight/obesity and mortality.6 A recent review has suggested that the weight guidelines may be overly restrictive as they apply to all cause and cardiovascular mortality in elderly people.6 The impact of obesity on morbidity and disability has shown to be higher than the impact on mortality.7 Reports estimating prevalence of

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1375
disease burden associated with overweight and obesity using the current guidelines have generally been carried out in middle-aged populations or in populations with a broad age range.8,9 We have examined the prevalence of a wide range of outcomes and indicators of ill health related to overweight and obesity, based on the current guidelines and with particular focus on cardiovascular-related disorders and on disability in a large population-based study of men aged 6079 y.

Subjects and methods

The British Regional Heart Study is a prospective study of cardiovascular disease involving 7735 men aged 4059 y selected from the age-sex registers of one general practice in each of 24 British towns, who had their initial examination between 1978 and 1980.10 Research nurses administered a standard questionnaire including questions on lifestyle factors and medical history. All men have been followed up for all cause mortality, cardiovascular morbidity and development of type 2 diabetes from the initial screening.11 In 19982000, all surviving men, now aged 6079 y, were invited for a 20th year re-examination, carried out in a local health center. All men completed a mailed questionnaire (Q20) providing information on their medical history and lifestyle factors, had a physical examination and provided a fasting blood sample. The men were asked to fast for a minimum of 6 h and to attend for measurement at a prespecified time between 0800 and 2000. Of the 5565 surviving subjects, 4252 (77%) attended for examination.

Body mass index and weight change At initial examination (19781980) and at re-examination (19982000) weight and height were measured and body mass index (BMI), calculated as weight/height2 (kg/m2), was used as an index of relative weight. BMI data were not available for 20 men. We used the National Institutes of Health (NIH) and WHO cutoff points to define normal weight (18.524.9 kg/m2) and to define overweight (25 29.9 kg/m2) and obesity (430 kg/m2).4,5 Within the overweight category the men were divided further into two separate groups in order to assess whether there were any differences between men in the lower and upper range of the overweight category. The men were classified into four groups on the basis of their current BMI: 18.524.9 (normal weight), 2527.4, 27.529.9 (overweight) and Z30 kg/m2 (obese). In total, 25 men had a BMIo18.5 kg/m2 and were excluded from the data analysis in Tables 15. The percentage change in body weight since initial examination was calculated for each man19 and a loss or gain of 4% defined a change in weight.19 The men were also asked whether they had lost weight in the 3 y prior to re-examination in 19982000.

Indicators of ill-health at re-examination

Self perception of healthFThe men were asked to describe their present health status as excellent, good, fair or poor.

Table 1 Characteristics of 4207 men aged 6079 y and weight changes over 20 y, according to BMI (kg/m2) at re-examination (19982000) Normal weight 18.524.9 Overweight 2527.4 27.529.9 Obese Z30 735 (17.4) 32.8

Cardiovascular risk factors Details of classification methods and measurements of the cardiovascular risk factors (including social class, physical activity, alcohol intake, smoking) have been described.1013 Systolic blood pressure at Q20 (Dinamap reading) was adjusted by subtracting 8 mmHg from the reading to accord with the Hawksley random zero sphygmanometer readings at baseline.14 Blood pressure was adjusted for observer variation.15 Hypertension is defined as systolic blood pressure Z160 mmHg or diastolic blood pressure Z90 mmHg or on antihypertensive treatment.16 Serum triglycerides, insulin and glucose were adjusted for the effects of fasting duration and time of day.13 High cholesterol was defined as levels Z240 mg/dl (Z6.2 mmol/l), low HDL-C as levels o40 mg/dl (o1.0 mmol/l), and high triglycerides as levels Z200 mg/dl (Z2.3 mmol/l).17 Insulin resistance was estimated according to the homeostasis model assessment (HOMA) as the product of fasting glucose (mmol/l) and insulin (mU/ml) divided by the constant 22.5.18 HOMA scores were not calculated for diabetic men and for those with a fasting glucose of Z7 mmol/l, as the index is not valid in this group.18

No. (%) Mean BMI (Q20) Weight history Initial mean BMI % Overweight at baseline % Obese at baseline Mean weight change (kg/20 y) % Stable % Gained % Lost Current characteristics Mean age % 75+ % Never smokers % Current smokers % Manual % Inactive % Heavier drinkers (Z21 drinks/week)

1263 (30.0) 1272 (30.1) 937 (22.1) 23.1 26.3 28.7

23.0 15.3 0.3 0.94 40.5 25.3 34.2

25.0 49.1 0.6 +2.42 40.2 45.1 14.7

26.6 75.6 4.8 +4.87 30.0 62.7 7.3

29.0 63.0 32.4 +9.68 20.1 78.0 2.9

69.2 19.9 32.1 16.6 49.0 30.8 10.4

68.8 16.6 29.6 11.3 50.1 32.5 13.0

68.4 14.0 27.8 11.4 56.9 34.5 13.0

68.1 13.1 25.1 9.0 62.4 44.9 13.9

Men under 18.5 kg/m2 excluded (n 25).

International Journal of Obesity

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1376
Table 2 Biological risk factors for cardiovascular disease and adjusted odds ratio (OR) (95% CI) for high or low levels of risk factors according to BMI (kg/m2) in 4207 men aged 6079 y. Odds ratio adjusted for age, social class, smoking, alcohol intake and physical activity Normal weight 18.524.9 Systolic BP % Z160 mmHg Adjusted OR Diastolic BP % Z90 mmHg Adjusted OR Hypertension + % Adjusted OR Cholesterol % Z240 mg/dl/6.2 mmol/l Adjusted OR HDL-C % o40 mg/dl/1.0 mmol/l Adjusted OR Triglycerides %Z 200 mg/dl/2.3 mmol/l Adjusted OR HOMA % top quartile Adjusted OR FEV1 % lowest quartile Adjusted OR 2527.4 Overweight 27.529.9 Obese Z30 Test for linear trend P-value

27.6 1.00

30.1 1.19 (0.98,1.45)

32.9 1.44 (1.15,1.81)

33.9 1.57 (1.18,2.11)

o0.0001 0.0003

14.3 1.00

21.7 1.51 (1.20,1.89)

25.7 1.83 (1.41,2.38)

36.7 2.68 (1.95,3.68)

o0.0001 o0.0001

43.7 1.00

53.3 1.54 (1.29,1.83)

55.0 1.72 (1.39,2.12)

62.9 2.50 (1.40,3.28)

o0.0001 o0.0001

37.5 1.00

41.0 1.10 (0.92,1.32)

37.4 0.91 (0.73,1.13)

38.7 1.01 (0.76,1.33)

0.82 0.95

14.5 1.00

23.3 1.75 (1.39,2.19)

29.9 2.38 (1.84,3.08)

41.9 3.74 (2.72,5.13)

o0.0001 o0.0001

11.7 1.00

23.0 1.89 (1.49,2.41)

30.0 2.45 (1.87,3.22)

43.1 3.45 (2.49,4.80)

o0.0001 o0.0001

9.6 1.00

18.1 1.93 (1.47,2.55)

30.4 3.84 (2.83,5.21)

57.0 10.20 (7.04,14.79)

o0.0001 o0.0001

28.1 1.00

21.5 0.79 (0.63,0.98)

23.5 0.96 (0.74,1.24)

27.0 1.20 (0.87,1.66)

0.91 0.24

+Hypertension Systolic blood pressure Z160 or Diastolic blood pressure Z90 or on antihypertensive treatment; BP blood pressure; DL-C high-density lipoprotein cholesterol; [3]HOMA homeostasis model for assessment of insulin resistance; [4]FEV1 forced expiratory volume in 1 s.

Recall of doctor diagnosisFThe men were asked whether a doctor had ever told them that they had angina or myocardial infarction (heart attack, coronary thrombosis), stroke, other heart trouble, deep vein thrombosis (DVT), intermittent claudication (IC), pulmonary embolism (PE), high blood pressure, diabetes or cancer. Prevalence of cardiovascular outcomes (myocardial infarction, angina, stroke) were based on regular biennial review of general practice (GP) medical records (including hospital and clinic correspondence) and on recall of doctor diagnoses obtained by questionnaires. DiabetesFMen with a diagnosis of diabetes (GP reporting or recall) and those with a fasting glucose of Z7 mmol/l were considered to have prevalent diabetes. MedicationFThe men were asked about regular treatment and were required to bring their medication to the examination session. The medications were coded according to the British National Formulary (BNF) codes.20 Cardiovascular medicine included drugs used in the treatment of diseases of the cardiovascular system (BNF 2.12.13). Respiratory medicine included BNF codes 3.13.10 and musculoskeletal and joint diseases included all drugs used in the
International Journal of Obesity

treatment of rheumatic diseases, neuromuscular disorder and soft tissue inflammation (BNF 10.110.3). Cardiovascular interventionFThis included men who had ever had angioplasty or coronary artery bypass operation for chest pain or heart disease. DisabilityFThey were asked if they had any long-standing illness, disability or infirmity resulting in current difficulty with carrying out any of the following activities on their own: (i) going up or down stairs, (ii) bending down/ straightening up, (iii) falling or difficulty keeping balance, or (iv) walking for a quarter of a mile on the level. Men who responded positively to any of these questions were described as having locomotor disability. Activities of daily livingFThey were asked if they had problems performing usual activities (eg work, housework, family activities) and self-care (washing and dressing).

Statistical analysis Multiple logistic regression was used to estimate the odds ratio (relative odds) and 95% confidence intervals (CI) with adjustment for potential confounding factors. In these

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1377
Table 3 Unadjusted prevalence rates of chronic diseases, medication and disability in 4207 men aged 6079 y according to BMI (kg/m2) at re-examination Normal weight 18.524.9 General health Excellent Poor/fair health Prevalent disease Heart attack Angina Stroke Major CVD (any of the above) DVT/PE/IC High blood pressure (doctor diagnosed) Diabetes Cancer Cardiovascular intervention CABG/angioplasty Medication Any Cardiovascular High BP Musculoskeletal and joints Respiratory diseases Disability Long-standing illness/disability Locomotor disability (any) Problems with carrying out usual activities Washing/dressing 2527.4 Overweight 27.529.9 Obese Z30 Test for linear trend P-value

17.7 23.4

16.0 20.4

12.9 25.8

11.3 37.3

o0.0001 o0.0001

11.0 15.7 5.4 23.0 5.9 20.8 7.5 6.3

11.4 17.6 5.4 24.1 6.8 29.0 9.9 5.0

9.9 17.6 5.7 24.8 6.3 32.0 13.3 6.1

15.7 23.0 8.4 33.2 9.4 41.1 19.7 6.3

0.003 o0.0001 o0.0001 o0.0001 0.006 o0.0001 o0.0001 0.59

3.4

5.0

5.7

6.4

o0.0001

63.0 35.9 15.7 9.0 13.1

64.7 42.9 22.8 11.9 10.6

68.9 45.7 25.6 16.0 8.8

75.9 58.1 34.4 16.1 12.5

o0.0001 o0.0001 o0.0001 o0.0001 0.10

31.1 20.9 20.0 4.6

30.1 22.8 20.0 4.2

32.7 27.6 23.8 5.2

45.1 38.8 31.8 9.5

o0.0001 o0.0001 o0.0001 o0.0001

DVT/PE/IC deep vein thrombosis/pulmonary embolism/intermittent claudication.

models age was entered as a continuous variable, social class (based on longest held occupation) (I, II, III nonmanual vs III manual, IV, V and Armed Forces), physical activity (five levels), alcohol intake (five levels) and smoking (never, long term ex-cigarette smoker, recent ex-cigarette smoker (gave up in 1996) and current cigarette smokers) were entered as categorical variables. Tests for linear trend were carried out fitting BMI as a continuous variable. To test for nonlinear trends a quadratic term was fitted to a model that included a linear term. We have assessed the proportion of the total population who have CV-related disorders and disability attributed to overweight or obesity (ie what proportion of men would not have the disease in the absence of overweight and obesity) by calculating population attributable risk fractions (PARF). Odds ratios (ORs) for each BMI category using normal weight (18.524.9 kg/m2) as the reference group were used to calculate PARF using the equation: PARF Pi(ORi1)/1 Sum[Pi(ORi1)]21 where Pi is the prevalence of the overweight and obesity category i and ORi the odds ratio associated with the specific overweight and obesity category i. The PARF values for the overweight and obese categories were added together to estimate the overall population

attributable risk fraction associated with overweight and obesity.

Results
The mean BMI in the 4232 men with available data on BMI rose from 25.43 kg/m2 (s.d. 2.99) at initial examination (20 y earlier) to 26.94 kg/m2 (s.d. 3.73) on re-examination, an average weight gain of 3.15 kg. Obesity was present in 17.4% of the men (initially 7%) and a further 52% were overweight (initially 47%). Men in the normal weight range were more likely to have been stable in weight. The majority of men who were overweight or obese had gained weight since initial examination. BMI was inversely related to age and current smoking and positively associated with heavier drinking. Both obese and underweight men were more likely to be manual workers and physically inactive than men in the normal weight range.

BMI and cardiovascular risk factors Table 2 shows the unadjusted prevalence rates and the odds ratios of having high/low levels of the biological cardiovascular risk factors, adjusting for age, smoking, physical activity, social class and alcohol intake. Prevalence and
International Journal of Obesity

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1378
Table 4 Adjusted odds ratio (OR) (95% CI) for chronic diseases, medication and disability in 4207 men aged 6079 y according to BMI (kg/m2) at re-examination. Adjusted for age, smoking, social class, alcohol intake and physical activity Normal weight 18.524.9 2527.4 Overweight 27.529.9 Obese 30+ Test for linear trend P-value Test for quadratic trend P-value

General health Excellent Poor/fair health Doctor-diagnosed disease Heart attack Angina Stroke Major CVD (any of the above) DVT/PE/IC High BP Diabetes Cancer Treatment for heart trouble CABG/angioplasty Medication Any Cardiovascular High BP Musculoskeletal and joints Respiratory diseases Disability Long-standing illness/disability Locomotor disability Problems with carrying out usual activities Washing/dressing

1.00 1.00

0.90 (0.71,1.14) 0.95 (0.76,1.19)

0.75 (0.55,1.01) 1.30 (1.00,1.68)

0.77 (0.52,1.14) 2.10 (1.53,2.88)

0.004 o0.0001

0.73 0.06

1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

1.12 1.23 1.00 1.17 1.21 1.55 1.37 0.77

(0.85,1.48) (0.97,1.55) (0.68,1.48) (0.95,1.43) (0.85,1.73) (1.27,1.90) (1.01,1.86) (0.53,1.11)

0.97 1.26 1.03 1.23 1.19 1.79 1.93 0.94

(0.69,1.37) (0.95,1.66) (0.65,1.61) (0.96,1.58) (0.77,1.83) (1.42,2.27) (1.38,2.71) (0.61,1.44)

1.78 1.73 1.73 1.96 1.87 2.47 2.89 0.91

(1.19,2.66) (1.22,2.44) (1.22,2.44) (1.44,2.67) (1.12,3.13) (1.85,3.31) (1.93,4.35) (0.52,1.57)

0.002 0.0007 0.007 o0.0001 0.05 o0.0001 o0.0001 0.96

0.53 0.63 0.36 0.67 0.20 0.01 0.08 0.22

1.00

1.71 (1.11,2.62)

2.20 (1.34,3.61)

2.52 (1.36,4.68)

0.002

0.06

1.00 1.00 1.00 1.00 1.00

1.20 1.40 1.64 1.43 0.91

(1.00,1.44) (1.16,1.67) (1.30,2.02) (1.08,1.91) (0.69,1.21)

1.59 1.60 1.94 2.16 0.82

(1.26,1.99) (1.29,1.99) (1.48,2.47) (1.57,2.98) (0.58,1.16)

2.43 2.62 2.79 2.25 1.32

(1.79,3.28) (1.99,3.46) (2.04,3.83) (1.50,3.36) (0.87,2.00)

o0.0001 o0.0001 o0.0001 0.0007 0.87

0.22 0.02 0.008 0.008 0.19

1.00 1.00 1.00 1.00

1.00 1.17 1.06 1.04

(0.83,1.22) (0.94,1.45) (0.85,1.33) (0.67,1.61)

1.18 1.50 1.33 1.23

(0.94,1.48) (1.16,1.93) (1.03,1.73) (0.75,2.01)

1.95 2.26 1.93 2.19

(1.47,2.59) (1.66,3.09) (1.40,2.63) (1.24,3.88)

o0.0001 o0.0001 o0.0001 o0.0001

0.30 0.47 0.15 0.36

adjusted odds ratio of systolic and diastolic hypertension, low HDL-cholesterol, high triglycerides and high HOMA (measure of insulin resistance) increased significantly and sharply at higher levels of BMI. No significant association was seen with raised blood cholesterol. A U-shaped relationship was seen with lung function with normal weight men showing higher odds ratio of low lung function than men who were overweight (quadratic trend P 0.03).

Indicators of morbidity Table 3 shows the unadjusted prevalence rates of indicators of ill health. The normal weight group had the highest prevalence of excellent health, with obese men having the highest prevalence of poor/fair health. Prevalence rates of doctor diagnosed heart attack, angina and stroke were only increased in obese men. However, prevalence of diabetes, DVT, high blood pressure, locomotor disability, and the use of medication for treating cardiovascular diseases, hypertension and musculoskeletal/joint diseases increased significantly at higher levels of BMI. Normal weight men had higher rates of cancer and use of medication for respiratory diseases than overweight men (2529.9 kg/m2) and rates were similar to obese men.
International Journal of Obesity

Table 4 shows the adjusted odds ratios for the indicators of ill-health using men in the normal weight range as the reference group. Obese men had by far the highest odds ratios for all indicators except cancers and respiratory disease. Overweight men, even at the lower range of overweight (2527.4 kg/m2), tended to have higher odds of CV-related ill health, locomotor disability and musculoskeletal problems than men in the normal weight range. With the exception of diagnosis of high blood pressure and medication for CVD, high blood pressure and musculoskeletal and joints there was no evidence of a significant curvilinear relation or U-shaped relation with these outcome measures after adjustment. The significant quadratic trends seen for these factors represented a curvilinear relationship with risk increasing up to levels of 30 kg/m2 and levelling off thereafter. Exclusion of men with recent weight loss (in the 3 y prior to re-examination) made minor differences to the relationships seen with cancer and medication for respiratory disease but normal weight men now had the lowest odds of poor/fair health (adjusted OR 1.00, 1.07 (0.841.37), 1.37 (1.03,1.82) and 2.23 (1.57,3.11) for the four groups respectively). The positive relationships with cardiovascular diseases, for example heart attacks and angina, strengthened after

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1379
Table 5 Adjusted odds ratio (OR) (95% CI) for CV risk factors, chronic diseases, medication and disability in 1224 never smokers aged 6079 y, according to BMI (kg/m2) at re-examination. Adjusted for age, social class, alcohol intake and physical activity Normal weight 18.524.9 (n 404) CV risk factors High HOMA Hypertension High Trig Low HDL Low FEV1 General health Poor/fair health Doctor-diagnosed disease Heart attack Angina Stroke Major CVD (any of the above) Diabetes Cancer Treatment for heart trouble CABG/angioplasty Medication Any Cardiovascular Musculoskeletal and joints Respiratory diseases Disability Locomotor disability Problems with carrying out usual activities Overweight 2529.9 (n 636) Obese Z 30 (n 184)
HOMA low HDL-C CABG antihyp drug Diabetes Hypertension CVD med disability major CVD

1.00 1.00 1.00 1.00 1.00

1.62 1.92 2.18 2.67 1.19

(1.06,2.49) (1.42,2.60) (1.41,3.37) (1.78,4.00) (0.77,1.82)

5.29 2.08 3.82 5.17 1.76

(2.75,10.16) (1.26,3.43) (2.04,7.14) (2.84,9.40) (0.88,3.52)

10

20 30 40 50 Attributable risk fractions (%)

overweight obesity

60

70

Figure 1 Population attributable risk fractions for overweight and obesity in

relation to prevalence of cardiovascular-related disorders and disability in British men aged 6079 y. HOMA homeostasis model assessment of insulin resistance. HDL-C high-density lipoprotein cholesterol; CABG coronary artery bypass graft; CVD cardiovascular disease.

1.00

1.24 (0.81,1.90)

2.78 (1.46,5.29)

1.00 1.00 1.00 1.00 1.00 1.00

2.43 1.70 0.96 1.71

(1.32,4.49) (1.09,2.66) (0.43,2.13) (1.14,7.28)

5.15 2.64 1.88 3.47

(2.06,12.91) (1.30,5.37) (0.59,6.03) (1.87,6.47)

on CVD-related health outcome, use of medication and disability were also seen in these men (data not shown).

1.77 (1.00,3.11) 0.78 (0.42,1.45)

3.74 (1.69,8.25) 0.61 (0.20,1.83)

1.00

3.20 (1.40,7.28)

5.81 (1.74,19.33)

1.00 1.00 1.00 1.00

1.54 (1.13,2.10) 1.82 (1.32,1.54) 2.24 (1.28,3.91) 0.71 (0.41,1.20)

3.01 (1.77,5.14) 3.74 (2.21,6.32) 3.09 (1.36,7.02) 1.39 (0.58,3.33)

Population attributable risk fractions The population attributable risks fractions for cardiovascular disease, interventional treatment and disability attributable to overweight and obesity are shown in the Figure 1. High insulin resistance and low HDL-cholesterol were most strongly attributed to overweight and obesity (64 and 48% respectively). Over a third of cardiovascular intervention (40%), use of antihypertensive treatment (40%), diabetes (38%), a third of hypertension (33%) and cardiovascular medication use (31%), a quarter of locomotor disability (26%) and a fifth of major CVD (19%) in the population were attributable to overweight and obesity.

1.00 1.00

1.44 (0.95,2.18) 1.31 (0.87,2.00)

3.41 (1.83,6.34) 3.35 (1.80,6.62)

Discussion
In this cohort of British men aged 6079 y, over two-thirds of the men were either overweight (52%) or obese (17%), very similar to the prevalence rates reported in the 1998 Health Survey for England (46% overweight and 17% obese).23 The influence of age on the relationship of BMI to morbidity and mortality has been a subject of much debate. Many studies have failed to find any significant association between BMI and all-cause mortality in the elderly (Z65 y) and some studies based on mortality outcome have even suggested a protective effect of overweight (2529.9 kg/m2) in older subjects.3,6 It has been suggested that the cutoff point of Z25 kg/m2 for overweight might overestimate the risks for older people6 and that this threshold should be raised, as BMI levels between 25.0 and 29.0 kg/m2 may not be associated with excess risk in the elderly.24 We have observed in this cross-sectional study that the prevalence of CV risk factors and morbidity (particularly cardiovascular-related diseases), disability and medication use, increased significantly with increasing degree of overweight and obesity in this age-group. The effects were even stronger in never
International Journal of Obesity

exclusion of men with recent weight loss. The adjusted relative risks for major CVD for the overweight and obese groups were 1.22 (0.97,1.53), 1.36 (1.03,1.79) and 2.12 (1.51,2.97) respectively.

Nonsmokers Since some studies have indicated a stronger relationship between higher BMI and mortality in never smokers,22 we also examined the relationship between BMI and CV risk factors and main health outcome measures in 1224 men who had never smoked. The positive relationships seen between overweight and obesity and CVD-related outcome and disability were even stronger in men who had never smoked (Table 5). There were only 689 men aged 75 y or above at reexamination but the adverse effect of overweight and obesity

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1380
smokers, which is consistent with studies showing stronger positive relations between BMI and mortality in never smokers.22 history is not taken into account. Despite these limitations, it was estimated from the present cross-sectional study that nearly 40% of the burden of diabetes, a third of medication use for cardiovascular-related diseases in the elderly, a quarter of locomotor disability and a fifth of major CVD was attributable to overweight and obesity.

Overweight, obesity and CV risk factors BMI showed strong graded associations with many CV risk factors including hypertension, low HDL-cholesterol, high triglycerides (but not high blood cholesterol) and high insulin resistance, with risk increasing sharply with increasing overweight and obesity. It is evident that cardiovascular risk factors continue to be highly associated with BMI even in the elderly, as has been observed in other studies.25,26 It is estimated that over 60% of raised insulin resistance in this population, 48% of low HDL-cholesterol and about a third of hypertension (33%) is attributable to overweight/obesity.

Morbidity and disability Overweight and obesity is associated with several disease outcomes leading to long-term medication, in particular CVrelated diseases and disability, which is reflected in the increased prevalence of antihypertensive treatment, cardiovascular medication and cardiovascular intervention in overweight and obese men compared to men of normal weight. Although we did not have information on prevalence of noncardiovascular diseases other than cancers, medication for musculoskeletal and joint diseases was more prevalent in the overweight and obese reflecting the known association between obesity and arthritis. Reporting of locomotor disability and difficulty with activities of daily living increased progressively with increasing degrees of overweight and obesity. Men in the upper range of the overweight category (27.529.9 kg/m2) were 50% more likely to report locomotor disability than normal weight men and in obese men the odds were over two-fold. This is consistent with other reports that have reported a two-fold increase in risk of mobility-related disability for men in the top quintile of body weight25 or fat mass27 Although the prevalence data from the present study are based on 77% of survivors, the mean BMI of those who attended re-examination and nonattenders were almost identical, although disabled people were less likely to attend.28 While the prevalence of disability may be underestimated, this is unlikely to affect the BMIdisability relationships.

Age, overweight, obesity and disease burden Studies on the burden of disease associated with overweight or obesity using the current WHO and NIH guidelines have generally been carried out in younger populations or have focused on limited end points.8,9 In an US cross-sectional study, the relationship between overweight and obesity and the comorbidities studied (diabetes, gall bladder disease, CHD, blood cholesterol and blood pressure) was generally stronger among the younger age group (o55 y) than the older age-group (Z55 y), based on prevalence ratios.9 However, given the higher incidence rates of morbidity in the elderly, particularly for diabetes and hypertension, the absolute excess risk of disease associated with overweight and obesity is likely to be substantial. Thus, while overweight (25.029.9 kg/m2) may not be associated with increased mortality in the elderly, it contributes considerably to increased morbidity and disability resulting in diminished quality of life.

Recent weight loss It has been shown that in older people, heavier weight (BMI Z27 kg/m2) is associated with increased risk only when those with recent substantial weight loss were excluded.29 The positive association between overweight and obesity and cardiovascular disease in our study was strengthened after exclusion of men who had lost weight in the 3 y prior to reexamination. The impact of body weight on health status in the elderly is thus likely to be underestimated if weight
International Journal of Obesity

Other measures of adiposity It is well recognized that an indirect measure of anthropometry such as BMI is an inaccurate measure of body fatness in the elderly because of the decrease in skeletal muscle mass and increase in abdominal fat with aging.30 Several studies have reported other measures of body fat such as centrality index (abdominal/femoral fat) and waist circumference to be better correlated with blood pressure, lipid profile and atherogenic index than the BMI in the elderly.31,32 In many studies waist circumference (WC) or waist-to-hip ratio (WHR) have been shown to be better predictors of mortality than BMI in older men3335 and WC has been shown to help identify those at increased risk within the normal weight, overweight and obese BMI categories.36 These findings suggest that WC may be a better indicator of overall body fat and health risk in older subjects than the BMI and it has been suggested that the BMI and WHR should be used in conjunction to characterize subjects at high risk. However, some studies have shown the combination of WC and BMI to make only a minor improvement in prediction of CV risk factors over either the BMI or WC alone.37 Despite its limitations we have shown strong positive relationships between BMI and CV risk factors and have shown overweight and obesity using BMI cutoff points to be associated with increased prevalence of morbidity in the elderly. The main focus of this paper has been to assess whether the WHO guidelines for overweight and obesity using the current BMI definitions are appropriate

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1381
in the elderly. Whether other measures of adiposity such as WC would provide better indication of morbidity than the BMI in the elderly requires further investigation.
14 fasting duration and time of day. Ann Clin Biochem 2002; 39: 493501. Whincup PH, Bruce NG, Cook DG, Shaper AG. The Dinamap 1846SX automated blood pressure recorder: comparison with the Hawksley random zero sphygmomanometer under field conditions. J Epidemiol Comm Health 1992; 46: 164169. Bruce NG, Cook DG, Shaper AG. Differences between observers in blood pressure measurement with an automatic oscillometric recorder. J Hypertens Suppl 1990; 4: S11S13. 1999 World Health Organization International Society of Hypertension Guidelines for the Management of Hypertension. Guidelines Subcommittee. J Hypertens 1999; 17: 151183.15. Third report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation treatment of high blood cholesterol in adults (Adult Treatment Panel III). Final Report. NIH Publication No. 02-5215 2002. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412419. Wannamethee G, Shaper AG. Weight change in middle-aged British men: implications for health. Eur J Clin Nutr 1990; 44: 133142. British National Formulary 1994. Publisher: Joint publication of the British Medical Association and Royal Pharmaceutical Society of Great Britain. English D, Holman D, Milne E, Winter M, Hulse C, Codde G. The quantification of drug-caused mortality and morbidity in Australia 1992. Commonwealth Department of Human Services and Health 1995: Canberra. Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW. Bodymass index and mortality in a prospective cohort of US adults. N Engl J Med 1999; 341: 10971105. Joint Health Surveys Unit on behalf of the Department of Health. 1999 Health Survey for England: Cardiovascular Disease 1998. The Stationary Office: London. Strawbridge WJ, Walhagen MI, Shema SJ. New NHLBI Clinical guidelines for obesity and overweight: will they promote health. Am J Public Health 2000; 90: 340343. Harris TB, Savage PJ, Tell GS, Haan M, Kumanyika S, Lynch JC. Carrying the burden of cardiovascular risk in old age: associations of weight and weight change with prevalent cardiovascular disease, risk factors, and health status in the Cardiovascular Health Study. Am J Clin Nutr 1997; 66: 837844. Masaki KH, Curb JD, Chiu D, Petrovich H, Rodriguez BL. Association of body mass index with blood pressure in elderly Japanese men: The Honolulu Heart Program. Hypertension 1997; 29: 673677. Visser M, Langlois J, Guralnik JM, Cauley JA, Kronmal RA, Robbins J, Williamson JD, Harris TB. High body fatness, but not low fat-free mass, predicts disability in older men and women: the Cardiovascular Health Study. Am J Clin Nutr 1998; 68: 584590. Thomas MC, Walker M, Lennon LT, Thomson AG, Lampe FC, Shaper AG, Whincup PH. Non-attendance at re-examination 20 years after screening in the British Regional Heart Study. J Public Health Med 2002; 24: 285291. Harris TB, Launer LJ, Madans J, Feldman JJ. Cohort study of effect of being overweight and change in weight on risk of coronary heart disease in old age. BMJ 1997; 314: 17911794. Baumgartner RN, Heymsfield SB, Roche AF. Human body composition and the epidemiology of chronic disease. Obes Res 1995; 3: 7395. Wu CH, Yao WJ, Lu FH, Yang YC, Wu JS, Chang CJ. Sex differences of body fat distribution and cardiovascular dysmetabolic factors in old age. Age Aging 2001; 30: 331336. Turcato E, Bosello O, Di Francesco V, Harris TB, Zoico E, Bissoli L, Fracassi E, Zamboni M. Waist circumference and abdominal sagittal diameter as surrogates of body fat distribution in the

Conclusion
In this study, overweight even of moderate degree and obesity are both associated with a substantial increase in the burden of CV-related disorders, overall medication and disability and it appears that the guidelines recommended by WHO and the NIH (NHLBI) for overweight and obesity apply to these elderly men. With an increasing life expectancy, an increase in the population prevalence of overweight and obesity will lead to an increase in the population burden of ill health in the future due to increased nonfatal morbidity and disability.

15

16

17

18

19

Acknowledgements The British Regional Heart Study is a British Heart Foundation Research Group and receives support from the Department of Health (England). The views expressed in this publication are those of the authors and not necessarily those of the Department of Health (England).

20

21

22

References
1 British Nutrition Foundations Task Force. Obesity. Blackwell: Oxford, UK; 1999. 2 National Task Force on the Prevention and Treatment of Obesity. Overweight, obesity and health risk. Arch Int Med 2000; 160: 898904. 3 Rossner S. Obesity in the elderlyFa future matter of concern? Obes Rev 2001; 2: 183188. 4 National Institutes of Health. Clinical guidelines on the identification, evaluation and treatment of overweight and obesity in adultsFthe evidence report. Obes Res 1998; 6: 51S209S. 5 World Health Organisation. Report of a WHO Consultation on Obesity. Obesity: preventing and managing the global epidemic. World Health Organisation: Geneva; 1998. 6 Heiat A, Vaccarino V, Krumholz M. An evidence-based assessment of federal guidelines for overweight and obesity as they apply to elderly persons. Arch Intern Med 2001; 161: 11941203. 7 Visscher TLS, Seidell JC. The public health impact of obesity. Annu Rev Public Health 2001; 22: 355375. 8 Lean MEJ, 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: 837843. 9 Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. JAMA 1999; 282: 15231529. 10 Shaper AG, Pocock SJ, Walker M, Cohen NM, Wale CJ, Thomson AG. British Regional Heart Study: cardiovascular risk factors in middle-aged men in 24 towns. BMJ 1981; 283: 179186. 11 Walker M, Shaper AG, Lennon L, Whincup PH. Twenty year follow-up of a cohort based in general practices in 24 British towns. J Public Health Med 2000; 22: 479485. 12 Wannamethee SG, Lowe GDO, Whincup PH, Rumley A, Walker M, Lennon L. Physical activity and hemostatic and inflammatory variables in elderly men. Circulation 2002; 105: 17851790. 13 Emberson J, Whincup PH, Walker M, Thomas M, Alberti KGMM. Biochemical measures in a population based study: the effect of

23

24

25

26

27

28

29

30

31

32

International Journal of Obesity

Overweight and obesity and the burden of disease and disability S Goya Wannamethee et al

1382
elderly: their relation with cardiovascular risk factors. Int J Obes Relat Metab Disord 2000; 24: 10051010. 33 Kalmijn S, Curb JD, Rodriguez BL, Yano K, Abbott RD. The association of body weight and anthropometry with mortality in elderly men: the Honolulu Heart Program. Int J Obes Relat Metab Disord. 1999; 23: 395402. 34 Baik I, Ascherio A, Rimm EB, Giovannucci E, Spiegelman D, Stampfer MJ, Willett WC. Adiposity and mortality in men. Am J Epidemiol 2000; 152: 264271. 35 Visscher TLS, Seidell JC, Molarius A, van der Kuip D, Hofman A, Witteman JCM. A comparison of body mass index, waist-hip ratio and waist circumference as predictors of all-cause mortality among the elderly: the Rotterdam Study. Int J Obesity Relat Metab Disord 2001; 25: 17301735. 36 Janssen I, Katzmarzyk PT, Ross R. Body mass index, waist circumference and health risk. Evidence in support of current National Institutes of Health Guidelines. Arch Intern Med 2002; 162: 20742079. 37 Han TS, van Leer EM, Seidell JC, Lean MEJ. Waist circumference as a screening tool for cardiovascular risk factors: evaluation of receiver operating characteristics (ROC). Obes Res 1996; 4: 533547.

International Journal of Obesity

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.