General obesity or abdominal obesity - what should we be focussing upon in children?

General obesity or abdominal obesity - what should we be focussing

upon in children? Dr. David McCarthy RNutr

Institute for Health Research & Policy

London Metropolitan University

13th February 2008Oxford Obesity Seminars

What is the time bomb?

• Type 2 diabetes• Hypertension• Metabolic Syndrome• CVD• Stroke

Diagnostic Criteria for Metabolic Syndrome in Children

• 3 or more of the following:

– BMI >98th centile– TG>95th centile– HDL <5th centile– SBP +/- DBP>95th centile– Impaired GTT

National Child Measurement Programme• One element of the Government’s work programme

on childhood obesity

• Inform local planning and delivery of services for children; gather population-level surveillance data to allow analysis of trends in growth patterns and obesity.

• http://www.dh.gov.uk/en/Policyandguidance/Healthandsocialcaretopics/Healthyliving/DH_073787

UK 1990 BMI References

Cole et al. Arch Dis Child (1995) 73: 25-29

Some drawbacks of BMI in children

• Age-dependent• Correlates with both

fat mass and fat-free mass

• Low sensitivity• no indication of

body fat distribution

• ‘Although body mass index is simple to measure and has been a valuable tool in monitoring trends in obesity, it also has numerous disadvantages. Principally it does not distinguish between increased mass in the form of fat, lean tissue or bone, and hence can lead to significant misclassification.’McCarthy et al. 2006

• ‘The fact that body mass index represents only a crude proxy for body fat and may produce a significant level of misclassification is universally accepted but widely ignored. This is because, in the absence of alternative measures, the advantages of body mass index have outweighed its disadvantages.’McCarthy et al. 2006

Correlation of BMI with Fat Mass (kg)

r2 = 0.763, P<0.0001

16-18 year olds

Correlation of BMI with Fat Free Mass (kg)

r2 = 0.514, P<0.0001

16-18 year olds

Misclassification by BMI

• Overweight/obese children can be classified as normal

• At least 6.5% overfat/obese children misclassified by BMI*

• *McCarthy et al. unpublished observations

Abdominal fat in adults

• Visceral fat, intra-abdominal fat

• Strongly linked to morbidity

• insulin resistance and hyperinsulinaemia

• Waist circumference – a measure of abdominal, intra-abdominal and visceral fat

Waist Measurement

• Now a common measure in adults

• Risk for diabetes, hypertension and CVD

• Cut-offs identified• Where do you take the

waist measurement?

Central body fat accumulation in children

• Intra-abdominal adipose tissue

• Subcutaneous abdominal adipose tissue

• assessed by girth and skinfold measurements

Intra-abdominal fat and morbidity in children

• Brambilla et al. 1994

• Caprio et al. 1995, 1996

• Owens et al. 1998

• Adverse changes in blood lipids, insulin and blood pressure

Is waist circumference in children linked to risk?

Adverse lipoprotein profile in 12-14 year olds (Flodmark et al. 1994)

Adverse insulin levels in 5-17 year olds Freedman et al. 1999)

Raised systolic blood pressure in 4-5 year olds (Jarrett, McCarthy et al, 2002, unpublished)

WC percentile charts for children

• Cuban (Martinez et al. 1994)

• Italian (Zannolli & Morgese 1996)

• Spanish (Moreno et al. 1999)

• UK (McCarthy et al. 2001)

• Canadian (Katzmarzyk et al. 2004)

• US (Fernandez et al. 2004)

• Australian (Eisenmann et al. 2005)

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Age (years)

40

50

60

70

80

Wai

st c

ircu

mfe

ren

ce (

cm)

Boys

Girls

Waist in children

n, 8355McCarthy et al. 2001

Development of WC centile charts for the UK children

Child Growth Foundation

BMI-WC relationship

Subject Age BMI WC BMI %ile WC %ile (y) (cm)

A 7.7 15.2 48 ~50th <9th

B 7.5 15.3 65 ~50th >98th

C 7.1 19.1 53 >91st ~50th

D 7.1 20.0 67 >98th >99.6th

Has upper body fatness increased in British children?

Comparison of data collected 10 and 20 years apart

BSI and NDN surveys

McCarthy et al. 2003. BMJ 326: 624

NDNS boys

NDNS girls BSI boys

BSI girls

McCarthy et al. 2003. BMJ 326: 624

NDNS girls

NDNS boys

BSI girlsBSI boys

Changes over 10-20 years in mean BMI and waist circumference in British children aged 11-16 years.

Mean SD Score (SD) Mean increaseover time (SE)

BSI 1977/87 NDNS 1997Male Female Male Female Male Female

BMI -0.05 -0.15 0.42 0.38 0.47 0.53(1.02) (0.99) (1.13) (1.09) (0.06) (0.06)

WC 0.00 0.00 0.84 1.02 0.84 1.02(0.99) (1.00) (1.02) (1.33) (0.06) (0.06)

McCarthy et al. 2003. BMJ 326: 624

Changes over 10-20 years in overweight and obesity based on BMI and waist circumference in British children aged 11-16 years. Values are % exceeding 91st centile (98th centile)

% prevalence of overweight % change (obesity) over time

BSI 1977/87 NDNS 1997Male Female Male Female Male Female

BMI 7.7 5.9 20.6 17.3 12.9 11.4(3.3) (1.6) (10.0) (8.3) (6.8) (6.6)

WC 8.7 8.8 28.5 38.1 19.8 29.3(3.3) (3.1) (13.8) (17.1) (10.7) (14.5)

McCarthy et al. 2003. BMJ 326: 624

Are these changes in WC also seen in younger children?

The ALSPAC Study

Avon Longitudinal Study of Parents And Children

• ALSPAC comprises 14,000 children born in the Avon region during 1991 and 1992 (Golding et al. 2001).

• Children in Focus (CIF) is a subset of this cohort (approx. 1000 children)

• For this study, BMI and WC from the CIF cohort were compared with equivalent BSI data between 2.5 and 5 years

BSI BMI CIF BMI

n Age(y)

Mean (SD)(kgm2)

Median(kgm2)

n Age(y)

Mean (SD)(kgm2)

Median(kgm2)

D(kgm2)

% D

Boys

182 2.5 16.4 (1.3) 16.3 528 2.5 16.7* (1.3) 16.6 0.29 1.8

205 3.5 16.3 (1.4) 16.1 580 3.5 16.5* (1.2) 16.4 0.22 1.4

195 4.5 16.0 (1.3) 15.8 563 4 16.2* (1.3) 16.1 0.26 1.6

251 5.5 15.9 (1.2) 15.7 534 5 16.0 (1.4) 15.9 0.09 0.6

Girls

186 2.5 16.3 (1.3) 16.1 429 2.5 16.5* (1.3) 16.4 0.27 1.7

185 3.5 16.0 (1.4) 15.9 467 3.5 16.5* (1.5) 16.3 0.42 2.6

210 4.5 16.0 (1.4) 15.9 454 4 16.2* (1.5) 16.1 0.26 1.6

406 5.5 15.5 (1.6) 15.2 445 5 16.1* (1.6) 15.9 0.63 4.1

McCarthy HD et al. (2005). Increasing waist circumferences in young British children - a comparative study. Int J Obesity 29: 157-162.

Results - BMI

BSI waist circumference CIF waist circumference

n Age (y)

Mean (SD) (cm)

Median (cm)

n Age (y)

Mean (SD) (cm)

Median (cm)

D (cm)

%D

Boys

182 2.5 48.4 (3.0) 48.3 536 2.5 50.4* (3.0) 50.3 1.99 4.1

206 3.5 50.9+ (3.1) 50.4 575 3.5 51.7* (3.0) 51.7 0.76 1.5

195 4.5 51.9+ (3.1) 51.6 559 4.0 52.5* (3.2) 52.4 0.60 1.2

251 5.5 52.3+ (3.2) 52.5 531 5.0 53.1* (3.5) 52.8 0.86 1.6

Girls

186 2.5 47.8 (2.8) 47.6 433 2.5 50.2* (3.1) 50.2 2.50 5.2

185 3.5 49.8 (3.1) 49.7 467 3.5 51.7* (3.6) 51.6 1.88 3.8

210 4.5 51.0 (3.6) 50.6 454 4 52.3* (3.6) 52.0 1.34 2.6

406 5.5 51.3 (3.7) 51.2 444 5 52.9* (4.2) 52.3 1.53 2.9

McCarthy HD et al. (2005). Increasing waist circumferences in young British children - a comparative study. Int J Obesity 29: 157-162.

Results – Waist Circumference

Increase in central fatness in British youths and pre-

school children

Hackney

Hackney Demography• Relatively young

population• 50% from black and

minority ethnic groups• High rates of social

and economic deprivation

• CVD death rate twice national average

• High prevalence of type 2 diabetes

Hackney Children's Obesity Survey

Ethnicity-related variation in upper body fatness in East London schoolchildren. By DIMPLE SAMANI1, LIZ PROSSER2, COLIN ALSTON2, and H. DAVID McCARTHY1, 1Institute for Health Research & Policy, London Metropolitan University, Holloway Rd, London N7 8DB, 2The Learning Trust, 1 Reading Lane, London, E8 1GQ.

Proceedings of the Nutrition Society (2007, in press)

Caucasian Mixed S. Asian Caribbean African Other

M F M F M F M F M F M F

n 437 394 125 113 224 215 199 176 286 291 102 105

Ht SDS

0.12 0.26 -0.05 0.50 0.50 0.12

WC

SDS

0.68 0.81 0.51 0.32 0.02 0.12 0.57 0.81 0.65 0.98 0.58 0.52

WHtR 0.45 0.45 0.45 0.44 0.44 0.44 0.44 0.45 0.45 0.45 0.45 0.45

Hackney Children’s Obesity Survey Measures of central fatness across ethnic groups

Caucasian

Mixed S. Asian Caribbean

African Other

M F M F M F M F M F M F

Overwt %

WC

27.2 25.8 24.0 20.4 17.9 19.5 23.6 33.0 29.7 38.6 28.4 28.6

Obese %

WC

14.4 13.5 9.6 8.8 8.5 8.4 12.1 18.8 12.9 21.6 11.8 9.5

Overwt %

BMI

21.7 19.5 17.6 18.7 14.1 15.7 23.0 27.0 27.0 28.0 11.5 15.9

Obese %

BMI

10.4 9.9 5.6 6.9 5.8 6.0 13.5 16.5 13.3 17.8 6.8 6.5

%>

WHtR

0.50

17.6 17.0 12.7 8.7 10.0 11.2 12.1 16.5 11.5 17.9 13.7 15.2

Prevalence of obesity across ethnic groups using different assessment criteria

Key Findings• Within a socially homogenous group of

children, variation in upper body fatness is evident

• Ethnic variation was striking• Prevalence of upper body obesity based solely

upon WC in South Asian was not as great as in other ethnic groups

• When height is accounted for, ethnicity-related variation is less obvious

• Caution should be exercised when interpreting WC measures between children from different ethnic groups

Problems with waist measurement

• No universally agreed definition of measurement site

• Measure over skin or clothing?

• Difficulty with very obese subjects

• Sensitivity issues

Waist circumference measurement

Midway between the 10th riband the iliac crest

WHO standard method

Used by :

McCarthy et al. 2001Freedman et al. 1999Moreno et al. 1999

Waist circumference measurementcontinued.

At the level of the umbilicus

Used for the waist circumference percentiles in Italian children, Zanolli & Morgese. 1996

Dimple Samani-Radia

Conclusions• BMI can fail to identify all

children at risk of obesity-related morbidity

• Abdominal fatness strongly linked to morbidity

• Waist girth measurement is a useful adjunct to BMI

• Upper body fatness has increased in young children and in youths

• Greater attention should be paid to central obesity in children

Acknowledgements

• Dimple Samani

• Karen Jarrett

• ALSPAC team

• Tanita UK

• NoCLoR

• The Learning Trust

• IHRP

• Tim Cole

Thank you

New children's body fat charts