The epidemiology of weight perception: perceived versus self-reported actual weight status among Albertan adults.
Linder, Jordana ; McLaren, Lindsay ; Siou, Geraldine Lo 等
Obesity is a public health issue that has been termed the
"most prevalent nutritional problem in the world". (1)
Currently, 23% of Canadian adults are classified as obese, (1) which is
concerning as extreme levels of excess weight have been linked with
various mental and physical health problems. (1-3)
To understand, prevent, and manage weight-related health issues,
researchers and clinicians rely on the ability to identify those at
risk. Prevention and management strategies may also rely on accurate
self-perception of weight, since discrepancies between actual and
perceived weight status may hinder intervention acceptance and/or
effectiveness. (4)
To date, much research on weight status perception has focused on
body image concerns and dieting practices among convenience samples of
adolescents. (5,6) Furthermore, many studies that compare weight status
perception to actual weight status have used body mass index (BMI) as
the standard reference. (4,6-12) Waist circumference (WC) has been noted
as an effective method of monitoring body weight due to its ease of
measurement and relation to BMI and fat distribution. (13) It has also
been recognized by some as a better marker of health risk than BMI alone
(14-16) and may better predict mortality risk.(17) However, research
examining weight status perception in relation to WC is lacking.
The aim of this study was to examine weight perception across a
large sample of adult Albertans. The specific objective was to examine
perception accuracy/inaccuracy (perceived vs. actual) in men and women
using three indicators of body size: BMI, WC, and a combined (BMI and
WC) risk profile. The study is unique in that it includes multiple
measures of weight status, most notably WC.
METHODS
Design and study population
The Tomorrow Project (http://www.thetomorrowproject.org) is a
prospective cohort study that consists of a sample of Albertans who, at
the time of enrollment, were 35-69 years of age, had never been
diagnosed with cancer (other than non-melanoma skin cancer), were able
to complete written questionnaires in English, and indicated that they
were likely to remain in Alberta for a year following recruitment.
Participants were recruited using a two-stage random sampling design,
where the first stage identified households using telephone random digit
dialing, while the second stage identified an eligible individual from
each household. In households with more than one eligible participant,
the person with the most recent birthday was selected. Eligible adults
who expressed interest in taking part received a consent form and a
Health and Lifestyle Questionnaire (HLQ) by mail. Those who returned
completed consent forms and the HLQ were enrolled in the study. (18,19)
A follow-up survey (Survey 2004) was mailed to all participants who were
enrolled between 2000 and 2003 (n=11,671, 41% male, 59% female).
Outcome measures
Perceived weight status was assessed in Survey 2004 using the
single item, "How would you describe yourself now?" Response
options were: "underweight", "about the right
weight", "overweight", and "don't know".
We defined actual weight status in three ways. First, BMI (kg/[m.sup.2])
was computed from weight and height. To maximize accuracy, the
self-administered survey was accompanied by a 72 inch (183 cm) tape
measure and detailed instructions for measuring and recording height and
weight. Specifically, participants were asked to weigh themselves using
a bathroom scale, without shoes and wearing light indoor clothing.
Participants were classified as underweight (BMI <18.5 kg/[m.sup.2]),
normal weight (18.5 kg/[m.sup.2] [less than or equal to] BMI <25
kg/[m.sup.2]), overweight (25 kg/[m.sup.2] [less than or equal to] BMI
<30 kg/[m.sup.2]), or obese (BMI [greater than or equal to] 30
kg/[m.sup.2]). (20)
Second, participants were asked to measure their WC, using the tape
measure provided. Instructions and a detailed diagram indicating site of
measurement were included. Based on Health Canada's classification
scheme, (20) participants were categorized as being at higher risk for
developing health problems related to excess abdominal fat if their WC
was greater than or equal to 102 cm for men and 88 cm for women, and
were otherwise characterized as lower risk.
Third, a combined risk profile was assigned to each participant,
based on BMI and WC. Participants were categorized into one of four
categories: overweight/obese BMI and higher risk WC; overweight/obese
BMI and lower risk WC; normal BMI and higher risk WC; and normal BMI and
lower risk WC.
Statistical analysis
Descriptive statistics for all variables were examined for the full
sample and for women and men separately. We estimated proportions and
95% confidence intervals (95% CI) of participants with correct and
incorrect perceptions, based on perceived weight status compared with
BMI, WC, and combined BMI and WC profile. Statistical significance of
reporting accuracy by men and women was determined using a 5%
significance level based on non-overlapping confidence intervals, and
with two-sided Z-tests for two proportions when the confidence intervals
overlapped. All analyses were performed using SAS/STAT version 9.1.3
(SAS Institute Inc., Cary, NC).
RESULTS
Of those who had returned at least the Health and Lifestyle
questionnaire at baseline, 9,229 participants (79%) returned Survey
2004. For the purposes of this study, participants were excluded in the
following circumstances: history of cancer prior to enrollment (n=142),
recruited as "second in household" (to avoid within-household
clustering) (n=276), aged 65 years or older in 2004 (n=1138), more than
12 weeks pregnant or less than 6 months postpartum in 2004 (n=14),
missing values on BMI or WC (n=118), and did not answer or answered
"don't know" to the question about perceived weight
status (n=103), or who otherwise had incomplete data (n=268). Analyses
are based on the remaining sample of 7,436 with complete data on all
study variables (n=4,456 women, mean age=50.4 (SD=7.4); n=2,980 men,
mean age=50.4 (SD=7.4)), unless stated otherwise below. The majority of
participants were employed full time (63%), married/living with someone
(80%), and had attained at least some post-secondary education (71%).
Descriptive statistics for BMI, WC, and the combined risk profile
are presented in Table 1. Less than 1% of participants were classified
as underweight according to BMI (<18.5 kg/[m.sup.2]). Approximately
two thirds of participants were overweight or obese, while just over one
third of participants were classified as being at higher health risk
based on WC. Similarly, approximately one third were at higher health
risk based on the combined BMI/WC profile.
Table 2 presents actual versus perceived weight status according to
BMI categories. For men and women combined, among those who were
classified as overweight or obese based on BMI, approximately 83%
demonstrated accurate weight status perception. However, just over one
quarter of overweight/obese men regarded themselves to be 'about
the right weight', whereas only 7% of overweight/obese women
perceived themselves to be 'about the right weight'. In
contrast, almost one quarter of normal-weight women perceived themselves
to be overweight, whereas only 9% of normal-weight men classified
themselves in the overweight category.
Table 3 presents actual versus perceived weight status according to
WC categories. Among those who were classified as at higher risk based
on WC, for men and women combined, approximately 90% demonstrated
accurate weight status perception. Similar to our findings with BMI, we
observed significant gender differences. For example, more men (13.3%)
than women (6.9%) at higher risk for health problems regarded themselves
to be at about the right weight, whereas more women (41.8%) than men
(37.5%) at lower risk regarded themselves to be overweight.
Table 4 presents actual versus perceived weight status according to
combined risk profile categories. Participants who were classified as
both at higher risk according to WC and overweight/obese according to
BMI had the highest proportion of weight perception accuracy
(approximately 92% for men and women combined). However, more men
(12.6%) than women (4.3%) in the higher risk WC and overweight BMI
category classified themselves as about the right weight and more women
(23.9%) than men (9.9%) of normal-weight BMI/lower risk WC perceived
themselves to be overweight.
In those categorized as overweight/obese according to BMI, but at
lower risk according to WC, approximately half of the men perceived
themselves as about the right weight (47.8%), while women in this group
were more likely to perceive themselves as overweight (83.5%).
DISCUSSION
Adults in our sample showed reasonable accuracy in weight
perception. Most participants perceived themselves to be overweight and
were also categorized as such by BMI. The combined percentage of
overweight and obesity in our sample (65.9%) is very similar to national
estimates of overweight/obesity prevalence (66.5%) based on measured
height and weight for those aged 35-64 years in the Canadian Community
Health Survey (Cycle 3.1, sub-sample 2, public use version). This
similarity is notable, considering our use of self-report survey
methods. Our strategy of including measuring aids and detailed
instructions appears to have merit.
Regardless of whether BMI or WC was used, women were more accurate
than men in correctly identifying their weight status. Based on Tables 2
and 3, men appeared to be more accurate in perceiving themselves as
overweight/obese when their WC was elevated, compared to when their BMI
was elevated. Since men tend to gain weight around the midsection, (23)
they may be more focused on this area, whereas women may have more
general body weight awareness. (23)
We observed that women in all categories of BMI were more likely
than men to perceive themselves as overweight. For men, even when they
were classified as overweight/obese based on BMI, almost one third still
perceived themselves as about the right weight. This is consistent with
other studies (4,24) and the view that perceived weight status reflects
the culturally desired body size in Western society: slender for women,
larger/muscular for men. (24)
The combined risk profile data provided insight into the issue of
incongruence between weight status indicators (e.g., elevated risk based
on WC but normal BMI). This issue has been noted in other studies that
observed strong positive correlations between BMI and WC but differing
risk classification (e.g., refs. 25, 26). Data are emerging to suggest
that WC is a stronger prognostic indicator for health outcomes than BMI.
(17,27)
Among those who were of normal weight status based on BMI but
higher risk based on WC, all men and nearly half of women perceived
themselves as about the right weight. This is troubling because these
individuals may have health risks due to their waist size but have
failed to recognize it, and therefore may not be engaging in behaviours
to decrease their WC. Conversely, among those who were overweight/obese
based on BMI but at lower risk based on WC, approximately half of men
and most women (>80%) perceived themselves as overweight. The
Canadian Guidelines on Body Weight Classification in Adults classifies
such individuals as at "increased risk" (20) and from this
point of view, the high proportion of perceived overweight may be
beneficial (trying to lose weight). On the other hand, several studies
have questioned the risk associated with overweight, (28,29) suggesting
that those who have some excess weight but a low-risk WC may not need to
engage in behaviour change. Experts in obesity are investigating
alternate classification systems that may be more sensitive to
identifying those at increased risk of obesity-related disorders. (30)
While we are cautious not to generalize our findings beyond our
study, there is no reason to suspect that within a stratum of actual
weight status, weight status perception would be much different among
participants in this study compared with those in the general
population. Furthermore, since we had asked participants to measure
their weight and waist circumference, their awareness of the weight
category may have been heightened. We also acknowledge that participants
indicated perceived weight status and not perceived health status;
future studies may wish to ask about perceived health or risk status.
Overall, our use of WC, which has not previously been examined in
relation to weight status perception, adds to our understanding of the
public awareness of this indicator of excess weight. Furthermore, we
have described a variety of patterns pertaining to weight perception in
this population, and thereby have considered issues of concern to
researchers interested in a spectrum of weight-related issues.
Conflict of Interest: None to declare.
Received: May 12, 2009
Accepted: October 8, 2009
REFERENCES
(1.) Lau D, Douketis J, Morrison K, Hramiak I, Sharma A, Ur E. 2006
Canadian clinical practice guidelines on the management and prevention
of obesity in adults and children. CMAJ 2007;176(8):1-13.
(2.) Stein J. Fact sheet: Major causes of death in Canada,
1993-1995. Chron Dis Can 1997;18:91-92.
(3.) McLaren L, Beck CA, Patten SB, Fick GH, Adair CE. The
relationship between body mass index and mental health: A
population-based study of the effects of the definition of mental
health. Soc Psychiatry Psychiatr Epidemiol 2008;43:63-71.
(4.) Wardle J, Johnson F. Weight and dieting: Examining levels of
weight concern in British adults. Int J Obes 2002;26:1144-49.
(5.) Strauss RS. Self-reported weight status and dieting in a
cross-sectional sample of young adolescents. Arch Pediatr Adolesc Med
1999;153:741-47.
(6.) Kaltiala-Heino R, Kautiainen S, Virtanen SM, Rimpela A,
Rimpela M. Has the adolescents' weight concern increased over 20
years? Eur J Public Health 2003;13:4-10.
(7.) Blokstra A, Burns CM, Seidell JC. Perception of weight status
and dieting behaviour in Dutch men and women. Int J Obes 1999;23:7-18.
(8.) Villaneva E. The validity of self-reported weight in US
adults: A population based cross-sectional study. BMC Public Health
2002;1:11.
(9.) Paeratakul S, White MA, Williamson DA, Ryan DH, Bray GA. Sex,
race/ethnicity, socioeconomic status, and BMI in relation to
self-perception of overweight. Obes Res 2002;10(5):345-50.
(10.) Whisenhunt BL, Williamson DA. Perceived weight status in
normal weight and overweight women. Eat Behav 2002;3:229-38.
(11.) Mack KA, Anderson L, Galuska D, Zablotsky D, Holtzman D,
Ahluwalia I. Health and sociodemographic factors associated with body
weight and weight objectives for women: Behavioural risk factor
surveillance system. J Wom Health 2004;13(9):1019-32.
(12.) Befort CA, Greiner KA, Hall S, Pulvers KM, Nollen NL,
Charbonneau A, et al. Weight-related perceptions among patients and
physicians: How well do physicians judge patients' motivation to
lose weight? J Gen Intern Med 2006;21:1086-90.
(13.) Lean MEJ, Han TS, Morrison CE. Waist circumference as a
measure for indicating need for weight management. Br Med J
1995;311:158-61. Available at:
http://www.bmj.com/cgi/content/full/311/6998/158 (Accessed March 15,
2009).
(14.) Zhu S, Wang Z, Heshka S, Heo M, Faith MS, Heymsfield SB.
Waist circumference and obesity-associated risk factors among whites in
the third National Health and Nutrition Examination Survey: Clinical
action thresholds. Am J Clin Nutr 2002;76:743-49.
(15.) Janssen I, Katzmarzyk PT, Ross R. Waist circumference and not
body mass index explains obesity-related health risk. Am J Clin Nutr
2004;79:279-84.
(16.) Murray S. Is waist-to-hip ratio a better marker of
cardiovascular risk than body mass index? CMAJ 2006;174:308.
(17.) Pischon T, Boeing H, Hoffmann K, Bergmann M, Schulze MB,
Overvad K, et al. General and abdominal adiposity and risk of death in
Europe. NEJM 2008;359:2105-20.
(18.) Bryant H, Robson PJ, Ullman R, Friedenreich C, Dawe U.
Population-based cohort development in Alberta, Canada: A feasibility
study. CDIC 2006;27(2):55-63.
(19.) Robson PJ, Lo Siou G, Ullman R, Bryant HE. Sociodemographic,
health and lifestyle characteristics reported by discrete groups of
adult dietary supplement users in Alberta, Canada: Findings from The
Tomorrow Project. Public Health Nutr 2008;11(12):1238-47.
(20.) Health Canada. Canadian Guidelines for Body Weight
Classification in Adults, 2003. Available at:
http://www.hc-sc.gc.ca/fn-an/
nutrition/weightspoids/guide-ld-adult/qa-qr-pub_e.html (Accessed January
7, 2008).
(21.) Collett D. Modelling Binary Data. London, UK: Chapman &
Hall, 1991.
(22.) Leemis LM, Trivedi KS. A comparison of approximate interval
estimators for the Bernoulli Parameter. Am Statistician
1996;50(1):63-68.
(23.) Blaak E. Gender differences in fat metabolism. Curr Opin Clin
Nutr Metab Care 2001;4(6);499-502.
(24.) Chang VW, Christakis NA. Self-perception of weight
appropriateness in the United States. Am JPrev Med 2003;24(4):332-39.
(25.) Bigaard J, Tjonneland A, Thomsen BL, Overvad K, Heitmann BL,
Sorensen TI. Waist circumference, BMI, smoking, and mortality in
middle-aged men and women. Obes Res 2003;11(7):895-903.
(26.) Hojgaard B, Gyrd-Hansen D, Olsen KR, Sogaard J, Sorensen T.
Waist circumference and body mass index as predictors of health care
costs. Plos ONE 2008;3(7):e2619.
(27.) Zhang C, Rexrode KM, van Dam RM, Li TY, Hu FB. Abdominal
obesity and the risk of all-cause, cardiovascular, and cancer mortality:
Sixteen years of followup in US women. Circ 2008;117:1658-67.
(28.) Orpana HM, Berthelot JM, Kaplan MS, Feeny DH, McFarland B,
Ross NA. BMI and mortality: Results from a national longitudinal study
of Canadian adults. Obes 2009;1:214-18.
(29.) Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths
associated with underweight, overweight, and obesity. JAMA
2005;293:1861-67.
(30.) Sharma AM, Kushner RF. A proposed clinical staging system for
obesity. Int J Obes 2009; advance online publication:1-7.
Jordana Linder, MPH, [1] Lindsay McLaren, PhD, [1] Geraldine Lo
Siou, MSc, [2] Ilona Csizmadi, PhD, [2] Paula J. Robson, PhD [2]
Author Affiliations
[1.] Department of Community Health Sciences, University of
Calgary, Calgary, AB
[2.] Alberta Health Services--Cancer Care, Population Health
Research, Calgary, AB
Correspondence: Ms. Jordana Linder, E-mail: jdlinder@ucalgary.ca
Table 1. Descriptive Statistics for BMI, Waist
Circumference, and Combined Risk Profile for Full
Sample and for Women and Men Separately
Full sample
(N=7,436)
n %
BMI
<18.5: Underweight 51 0.7
18.5-24.9: Normal weight 2481 33.4
25.0-29.9: Overweight 2923 39.3
[greater than or equal to]30.0: Obese 1981 26.6
BMI 30.0-34.9: Obese Class I 1329 17.9
BMI 35.0-40.0: Obese Class II 443 6.0
BMI [greater than or equal to]40.0: 209 2.8
Obese Class III
Waist Circumference (WC) from
Canadian guidelines *
Men: <102 cm; Women: <88 cm 4212 62.6
Men: [greater than or equal to]102 cm; 2521 37.4
Women: [greater than or equal to]88 cm
Combined risk profile--
WC from Canadian guidelines and BMI *
BMI: 18.5-24.9; WC: Men <102 cm, 2381 35.4
Women <88 cm
BMI: 18.5-24.9; WC: Men [greater than 100 1.5
or equal to]102 cm, Women
[greater than or equal to]88 cm
BMI: 25.0-34.9; WC: Men <102 cm, 1831 27.2
Women <88 cm
BMI: 25.0-34.9; WC: Men [greater than 2421 36.0
or equal to]102 cm, Women
[greater than or equal to]88 cm
Women
(N=4,456)
n %
BMI
<18.5: Underweight 47 1.1
18.5-24.9: Normal weight 1834 41.2
25.0-29.9: Overweight 1442 32.4
[greater than or equal to]30.0: Obese 1133 25.4
BMI 30.0-34.9: Obese Class I 685 15.4
BMI 35.0-40.0: Obese Class II 290 6.5
BMI [greater than or equal to]40.0: 158 3.6
Obese Class III
Waist Circumference (WC) from
Canadian guidelines *
Men: <102 cm; Women: <88 cm 2475 62.5
Men: [greater than or equal to]102 cm; 1486 37.5
Women: [greater than or equal to]88 cm
Combined risk profile--
WC from Canadian guidelines and BMI *
BMI: 18.5-24.9; WC: Men <102 cm, 1742 44.0
Women <88 cm
BMI: 18.5-24.9; WC: Men [greater than 92 2.3
or equal to]102 cm, Women
[greater than or equal to]88 cm
BMI: 25.0-34.9; WC: Men <102 cm, 733 18.5
Women <88 cm
BMI: 25.0-34.9; WC: Men [greater than 1394 35.2
or equal to]102 cm, Women
[greater than or equal to]88 cm
Men
(N=2,980)
n %
BMI
<18.5: Underweight 4 0.1
18.5-24.9: Normal weight 647 21.7
25.0-29.9: Overweight 1481 49.7
[greater than or equal to]30.0: Obese 848 28.5
BMI 30.0-34.9: Obese Class I 644 21.6
BMI 35.0-40.0: Obese Class II 153 5.1
BMI [greater than or equal to]40.0: 51 1.7
Obese Class III
Waist Circumference (WC) from
Canadian guidelines *
Men: <102 cm; Women: <88 cm 1737 62.7
Men: [greater than or equal to]102 cm; 1035 37.3
Women: [greater than or equal to]88 cm
Combined risk profile--
WC from Canadian guidelines and BMI *
BMI: 18.5-24.9; WC: Men <102 cm, 639 23.1
Women <88 cm
BMI: 18.5-24.9; WC: Men [greater than 8 0.3
or equal to]102 cm, Women
[greater than or equal to]88 cm
BMI: 25.0-34.9; WC: Men <102 cm, 1098 39.6
Women <88 cm
BMI: 25.0-34.9; WC: Men [greater than 1027 37.0
or equal to]102 cm, Women
[greater than or equal to]88 cm
* Excludes 703 people (495 women and 208 men) with
BMI <18.5 and >35 kg/[m.sup.2] as per Canadian guidelines
for body weight classification in adults which state
that "the WC measurement should be used among those
with BMIs between 18.5 and 34.9 to identify
additional risk"
Table 2. Actual versus Perceived Weight Status
Based on BMI and Stratified by Sex
Perceived Weight Status within
BMI Categories
Actual Weight N
Status (Based on BMI) Underweight (%) About the
Right Weight
(%, 95% CI)
* ([dagger])
Women
BMI <18.5 47 48.9 (34.6-63.2) 51.1 (36.8-65.4)
BMI 18.5-24.9 1834 2.1 (1.4-2.7) 73.0 (71.0-75.0)
BMI [greater than 2575 0 7.0 (6.0-8.0)
or equal to]25
Men
BMI <18.5 4 75.0 (19.4-99.4) 25.0 (0.6-80.6)
([dagger]) ([dagger])
BMI 18.5-24.9 647 8.7 (6.5-10.8) 82.4 (79.4-85.3)
BMI [greater than 2329 0.6 (0.3-0.1)t 28.0 (26.2-29.8)
or equal to]25
Perceived Weight Status within
BMI Categories
Actual Weight N
Status (Based on BMI) Overweight
(%, 95% CI)
Women
BMI <18.5 47 0
BMI 18.5-24.9 1834 24.9 (22.9-26.9)
BMI [greater than 2575 93.0 (92.0-94.0)
or equal to]25
Men
BMI <18.5 4 0
BMI 18.5-24.9 647 9.0 (6.8-11.2)
BMI [greater than 2329 71.4 (69.6-73.3)
or equal to]25
* Percentages apply to each row, and the denominator
is the total n (men, women) within each BMI category
([dagger]) Exact 95% confidence intervals (21,22)
Table 3. Actual versus Perceived Weight Status
According to Waist Circumference and Sex
Perceived Weight Status within
WC Categories
Actual Weight Status N * About the Overweight
(Based on Waist Right Weight (%, 95% CI)
Circumference) (%, 95% CI)
Women
<88 cm 2438 58.2 (56.2-60.1) 41.8 (39.9-43.8)
[greater than or 1485 6.9 (5.6-8.2) 93.1 (91.8-94.4)
equal to]88 cm
Men
<102 cm 1672 62.5 (60.2-64.8) 37.5 (35.2-39.8)
[greater than or 1031 13.3 (11.2-15.4) 86.7 (84.6-88.8)
equal to]102 cm
* Excludes 703 people (495 women and 208 men) with BMI <18.5
and >35 kg/m2 as per Canadian guidelines for body weight
classification in adults which states that "the WC
measurement should be used among those with BMIs between
18.5 and 34.9 to identify additional risk" (20)
([dagger]) Excludes 133 people (60 women and 73 men)
who perceived themselves as underweight
Table 4. Actual versus Perceived Weight Status
Based on a Combined Risk Profile According to
BMI, Waist Circumference and Sex
Perceived Weight Status within
WC Categories
Actual Weight Status N * About the
(Combined Risk Profile) Right Weight Overweight
(%, 95% CI) (%, 95% CI)
Women
BMI 18.5-24.9, WC <88 cm 1705 76.1 (74.0-78.1) 23.9 (21.9-26.0)
BMI 18.5-24.9, WC [greater 91 46.2 (35.9-56.4) 53.8 (43.6-64.1)
than or equal to]88 cm
BMI 25-34.9, WC <88 cm 733 16.5 (13.8-19.2) 83.5 (80.8-86.2)
BMI 25-34.9, WC [greater 1394 4.3 (3.2-5.4) 95.7 (94.6-96.8)
than or equal to]88 cm
Men
BMI 18.5-24.9, WC <102 cm 583 90.1 (87.6-92.5) 9.9 (7.5-12.4)
BMI 18.5-24.9, WC [greater 8 100% 0
than or equal to]102 cm
BMI 25-34.9, WC <102 cm 1089 47.8 (44.8-50.7) 52.2 (49.3-55.2)
BMI 25-34.9, WC [greater 1023 12.6 (10.6-14.6) 87.4 (85.4-89.4)
than or equal to]102 cm
* Excludes 703 people (495 women and 208 men) with
BMI <18.5 and >35 as per Canadian guidelines for body
weight classification in adults which states that "the
WC measurement should be used among those with BMIs
between 18.5 and 34.9 to identify additional risk" (20)
([dagger]) Excludes 133 people (60 women and 73 men)
who perceived themselves as underweight
