Perceptions of weight discrimination: prevalence and comparison to race and gender discrimination in America.

Abstract

Limited data are available on the prevalence and patterns of body weight discrimination from representative samples. This study examined experiences of weight/height discrimination in a nationally representative sample of US adults and compared their prevalence and patterns with discrimination experiences based on race and gender.
Data were from the National Survey of Midlife Development in the United States, a 1995-1996 community-based survey of English-speaking adults aged 25-74 (N=2290). Reported experiences of weight/height discrimination included a variety of institutional settings and interpersonal relationships. Multivariate regression analyses were used to predict weight/height discrimination controlling for sociodemographic characteristics and body weight status.
The prevalence of weight/height discrimination ranged from 5% among men to 10% among women, but these average percentages obscure the much higher risk of weight discrimination among heavier individuals (40% for adults with body mass index (BMI) of 35 and above). Younger individuals with a higher BMI had a particularly high risk of weight/height discrimination regardless of their race, education and weight status. Women were at greater risk for weight/height discrimination than men, especially women with a BMI of 30-35 who were three times more likely to report weight/height discrimination compared to male peers of a similar weight.
Weight/height discrimination is prevalent in American society and is relatively close to reported rates of racial discrimination, particularly among women. Both institutional forms of weight/height discrimination (for example, in employment settings) and interpersonal mistreatment due to weight/height (for example, being called names) were common, and in some cases were even more prevalent than discrimination due to gender and race.

Full text

Page 1

ORIGINAL ARTICLE
Perceptions of weight discrimination: prevalence
and comparison to race and gender discrimination
in America
RM Puhl, T Andreyeva and KD Brownell
Rudd Center for Food Policy and Obesity, Yale University, New Haven, CT, USA
Objective: Limited data are available on the prevalence and patterns of body weight discrimination from representative
samples. This study examined experiences of weight/height discrimination in a nationally representative sample of US adults and
compared their prevalence and patterns with discrimination experiences based on race and gender.
Method and procedures: Data were from the National Survey of Midlife Development in the United States, a 1995–1996
community-based survey of English-speaking adults aged 25–74 (N¼2290). Reported experiences of weight/height
discrimination included a variety of institutional settings and interpersonal relationships. Multivariate regression analyses were
used to predict weight/height discrimination controlling for sociodemographic characteristics and body weight status.
Results: The prevalence of weight/height discrimination ranged from 5% among men to 10% among women, but these
average percentages obscure the much higher risk of weight discrimination among heavier individuals (40% for adults with
body mass index (BMI) of 35 and above). Younger individuals with a higher BMI had a particularly high risk of weight/height
discrimination regardless of their race, education and weight status. Women were at greater risk for weight/
height discrimination than men, especially women with a BMI of 30–35 who were three times more likely to report
weight/height discrimination compared to male peers of a similar weight.
Discussion: Weight/height discrimination is prevalent in American society and is relatively close to reported rates of racial
discrimination, particularly among women. Both institutional forms of weight/height discrimination (for example, in
employment settings) and interpersonal mistreatment due to weight/height (for example, being called names) were common,
and in some cases were even more prevalent than discrimination due to gender and race.
International Journal of Obesity advance online publication, 4 March 2008; doi:10.1038/ijo.2008.22
Keywords: discrimination; bias; gender; race
Introduction
Overweight individuals are frequent targets of weight
stigmatization and prejudice.1,2Weight bias occurs from
employers,3,4
health-care professionals,5–9
and even family members.13
include impairments in psychological well-being14–15and
physical health, such as avoiding preventive health-care
services,16engaging in unhealthy eating patterns13,17–19and
avoiding physical activity.18,20–22
Despite evidence of weight bias documented in case
studies, experimental work, treatment and convenience
educators10–12
Its negative implications
samples, limited information is available on the prevalence
and patterns of weight discrimination in a nationally
representative sample. Discrimination is distinct from pre-
judice and beliefs in that it refers to negative, unequal
treatment of people because of their membership in a
particular group.23Prejudice reflects attitudes, while discri-
mination depicts behavior, thus prejudiced attitudes do not
necessarily translate into discriminatory behavior. There is
abundant evidence illustrating prejudiced attitudes toward
obese persons, but very little work has documented dis-
crimination perceived by obese individuals or the prevalence
of these experiences in the US population. Carr and
Friedman24
examined the frequency of institutional and
interpersonal discrimination among a nationally representative
sample of 3437 adults and found that compared to normal
weight individuals, obese persons reported significantly
more frequent daily discrimination, work-related discrimi-
nation and health-related discrimination. Individuals in the
Received 13 September 2007; revised 25 November 2007; accepted 19
January 2008
Correspondence: Dr RM Puhl, Rudd Center for Food Policy and Obesity, Yale
University, 309 Edwards Street, New Haven, CT 06520-8369, USA.
E-mail: rebecca.puhl@yale.edu
International Journal of Obesity (2008) 1–9
& 2008 Nature Publishing Group All rights reserved 0307-0565/08 $30.00
www.nature.com/ijo

Page 2

highest obese categories were 40–50% more likely to report
any discrimination than normal weight persons. The authors
also estimated that the effect of obesity on perceived
discrimination was similar across race, gender and age
categories.
Important questions remain. First, specific forms of
weight/height discrimination are important to document
so that prevention programs and policies can be targeted to
appropriate settings. The Carr and Friedman study was
complicated in this respect by relying on a combination of
weight and appearance (other than weight or height)
discrimination in their analyses, despite no differences in
body mass index (BMI) among people who reported
appearance-based discrimination and the rest of the sample.
Combining these variables into one measure assumes that
discrimination based on appearance and weight is the same,
even though appearance encompasses numerous physical
attributes unrelated to weight.
Second, it is important to determine how weight discri-
mination compares in its strength and prevalence to
discrimination based on other attributes such as gender
and race. No work to our knowledge has compared
prevalence rates of perceived weight/height discrimination
to racial or gender discrimination.
It is also important to examine gender differences in
perceived weight/height discrimination to determine unique
vulnerabilities of men or women. Carr and Friedman
reported that the effects of obesity on perceptions of general
discrimination were similar for men and women, but gender
was not addressed in the context of weight discrimination.24
In contrast, previous work on prejudiced attitudes toward
obese individuals has reported mixed findings on gender,
with some studies detecting gender differences25–27and
other studies not.13,15Thus, research examining differences
among men and women in their perceptions of weight/
height discrimination would be informative.
To address these issues, and to better understand the
demographic profiles of individuals who report discrimination
due to weight vs other attributes like race or gender, the
present study documents rates and patterns of weight/height
discrimination in comparison to other forms of discrimination
among adults in the United States. We examine the prevalence
and types of self-reported weight/height discrimination in
comparison to racial and gender discrimination, and compare
patterns of sociodemographic predictors for these different
types of discrimination. Data were drawn from the National
Survey of Midlife Development in the United States (MIDUS),
a nationally representative random sample of more than 3000
Americans aged 25–74 years in 1995–1996.
Methods and procedures
Sample
The National Survey of MIDUS is an interdisciplinary study
of patterns, predictors and consequences of midlife health,
well-being and social interactions. Respondents were drawn
from a nationally representative sample of community-based
English-speaking adults aged 25–74 years in the United States.
In the first stage, participating households were selected from
working telephone banks via random digit dialing. In the
second stage, disproportionate stratified sampling was applied
to select individual respondents. Elderly people (ages 65–74)
and men were oversampled along with oversampling in five
metropolitan areas (Atlanta, Boston, Chicago, Phoenix and
San Francisco). Those queried in the survey completed an
initial telephone interview and a self-administered mail
questionnaire, with data collected in 1995. The response rate
was 70% for the telephone interview, 86.8% for completing
the mail questionnaire among the telephone respondents,
with an overall survey response of 60.8%.25
Our study uses data from the MIDUS random core sample
(the Main Data) (our 1995–1996 MIDUS data is from the 2006
MIDUS Release (30 March 2006). The most recent version of
the MIDUS 1995–1996 data was released on 16 April 2007).
The 1995–1996 MIDUS study included data from three
samples: the Main Data (N¼4242 with 3485 national
random sample participants and 757 respondents from
metropolitan oversamples), the Twins Data (N¼1996 twins)
and the Siblings Data (N¼1614 pairs with 951 participants
drawn from the Main Data). The survey assessed physical and
psychological health throughout the respondent’s adult life,
substance abuse, well-being and personal beliefs, socioeco-
nomic status, social support and various forms of perceived
discrimination. Follow-up data were collected 9–10 years
from baseline in 2004–2005. Further details on the MIDUS
data and methodology are available elsewhere.25
We applied several restrictions on the analytic sample.
We excluded 980 people who had incomplete or inconsistent
data on discrimination-related questions. Another exclusion
(215 observations) applied to data with missing responses
pertaining to covariates like BMI and sociodemographics.
As a result of all exclusions, 2290 individuals (1104 men and
1186 women) remained eligible for the study. There were no
significant differences between our sample and the excluded
respondents for the key discrimination measures and
individual characteristics such as BMI and some socio-
demographic characteristics (for example, gender). The
excluded participants, however, were more likely to be from
minority population groups, younger and with lower educa-
tional attainment than an average respondent in our sample.
Yet, given similar discrimination and body weight character-
istics of the excluded and included respondents from the
national random sample, it is unlikely that our data
exclusions introduced bias into our estimates. We also
conducted sensitivity analyses to check robustness of our
results including respondents with inconsistent data on
discrimination questions.
Measures
Weight/height discrimination.
ceived discrimination by requesting participants to report
The MIDUS evaluates per-
Perceptions of weight discrimination
RM Puhl et al
2
International Journal of Obesity

Page 3

occurrences of discrimination over their lifetime and on a
day-to-day basis in interpersonal relationships. While parti-
cipants are asked about the primary reason for discrimina-
tion,they can reportmultiple
Specifically, the survey asks ‘What was the main reason for
the discrimination you experienced? If more than one
reason, circle all that apply’, with response choices including
‘Your Age, Gender, Race, Height or weight, Ethnicity or
nationality, Physical disability, Some aspect of appearance
other than weight or height, Sexual orientation, Religion,
and Other reason’.
The key outcome of this study is perceived discrimination
due to the respondent’s weight or height. As the MIDUS
survey used one category for weight discrimination combi-
ning height and weight, we refer to this variable as weight/
height discrimination throughout the article. We compared
average body weight, height and BMI (defined as weight in
kg relative to height in m2) between the survey participants
reporting weight/height discrimination and the rest of the
sample to test if weight was more likely than height to be a
source of discrimination. On average, body weight and BMI
were significantly higher in the group reporting weight/
height discrimination relative to other participants (BMI of
32 vs 27 for men and BMI of 33 vs 25 for women, Po0.01).
There was, however, no difference in height so that a short
body stature was unlikely to be a source of weight/height
discrimination. Among men, body height was even slightly
higher in the group reporting weight/height discrimination
(Po0.10), so a short stature does not appear to drive weight/
height discrimination responses among men. Finally, we
verified that rates of weight/height discrimination did not
change in any meaningful way if people with extreme height
were excluded (bottom 1% or top 1% height). Therefore,
we have reasonable evidence to believe that reports of
discrimination due to weight or height primarily reflect
higher body weight and obesity rather than height.
We evaluated lifetime experiences of institutionally based
discrimination (employment, medical care, education) and
interpersonal discrimination on a day-to-day basis. Lifetime
experiences were reported in the question: ‘How many times
in your life have you been discriminated against in each of
the following ways because of such things as your race,
ethnicity, gender, age, religion, physical appearance, sexual
orientation, or other characteristics?’ Eleven forms of life-
time discrimination were evaluated, including: ‘discouraged
by a teacher or advisor from seeking higher education’,
‘denied a scholarship’, ‘not hired for a job’, ‘not given a job
promotion’, ‘fired’, ‘prevented from renting or buying a
home in the neighborhood you wanted’, ‘prevented from
remaining in a neighborhood because neighbors made life
uncomfortable’, ‘hassled by the police’, ‘denied a bank loan’,
‘denied or provided inferior medical care’, and ‘denied
or provided inferior service by a plumber, car mechanic,
or another service provider’.
Interpersonal discrimination experiences were evaluated
with the question: ‘How often on a day-to-day basis do you
reasons if applicable.
experience each of the following types of discrimination?’
with nine response items including: ‘you are treated with less
courtesy than other people’, ‘you are treated with less respect
than other people’, ‘you receive poorer service than other
people at restaurants or stores’, ‘people act as if they are
afraid of you’, ‘people act as if they think you are dishonest’,
‘people act as if they think you are not as good as they
are’, ‘you are called names or insulted’, and ‘you are
threatened or harassed’. Participants said how frequently
they had experienced these situations using the categories
‘Often’, ‘Sometimes’, ‘Rarely’ or ‘Never’.
We constructed a measure of perceived discrimination
indicating whether an individual reported at least a single
occurrence of discrimination in any setting. For inter-
personal discrimination, only responses that were endorsed
Often or Sometimes counted as indicators of discrimination.
We weighted multiple and single occurrences of discrimina-
tion equally so that a person reporting a single occurrence of
discrimination would be treated like someone reporting
several experiences. We also constructed a separate measure
of lifetime discrimination occurrences and one of mistreat-
ment in personal relationships.
We compared patterns of weight/height discrimination
among US adults ages 25–74 with the profile of those
discriminated by race and gender, the two most prevalent
types of discrimination reported in the MIDUS data. Using
the same measurement approach as for weight/height
discrimination, we constructed indicators of race and gender
discrimination in either daily interpersonal relationships or
major lifetime settings. We also considered occurrences
of multiple forms of discrimination (for example, self-
reported discrimination due to race and weight/height).
Independent variables.
height discrimination included a set of sociodemographic
variables and measures of relative body weight. We measured
obesity based on the NIH clinical guidelines for the
classification of overweight and obesity in adults,26distin-
guishing between underweight (BMIo18.5), normal weight
(BMI 18.5–24.9), overweight (BMI 25.0–29.9), obesity class I
(BMI 30.0–34.9), obesity class II (BMI 35.0–40.0) and obesity
class III (BMI 40.0þ). We separated the obesity group
(BMIX30) into moderate (BMI 30–35) and severe obesity
(BMIX35) to account for potentially nonlinear effects of
obesity on weight discrimination by degree of obesity. The
sample size of people with BMIX40 was too small to
enable meaningful estimation. Sociodemographic covariates
included gender, age, race, educational achievement, marital
status and current occupation. On the basis of statistical tests
of interactions between gender and sociodemographic
covariates, we did not include any interacting terms and
conducted regression analyses for men and women jointly.
Table 1 provides rates of weight/height discrimination across
sociodemographic and body weight covariates for the total
sample and by gender.
The predictors of perceived weight/
Perceptions of weight discrimination
RM Puhl et al
3
International Journal of Obesity

Page 4

Statistical analysis
Our objective was to determine how weight/height discri-
mination was related to sociodemographic and body weight
characteristics among US adults aged 25–74, and to compare
patterns of these relationships to estimates for race and
gender discrimination. To identify predictors of discrimina-
tory experiences for an average adult American, we estimated
a set of models on the MIDUS sample, with all analyses
replicated separately for men and women as a sensitivity
check. To account for the complex sampling design and
obtain nationally representative estimates, we used indivi-
dual sample weights in presenting sample statistics.
Multivariate logistic regression models were used to
generate the odds ratio (ORs) for discrimination across body
weight and sociodemographic characteristics. We checked
robustness of results to the inclusion of annual personal
income and spousal income in the sensitivity analysis. An
additional sensitivity check was the inclusion of indicators
for race and gender discrimination to account for multiple
forms of discrimination. Since obesity may have a differ-
ential effect on the likelihood of weight/height discrimina-
tion by sociodemographic strata due to different perceptions
of ideal body weight across social groups,27–29we tested for
interactions between obesity and sociodemographic charac-
teristics. There was no difference in how obesity was related
to the likelihood of weight/height discrimination across
gender, race and education. Severe obesity, however, had a
different effect with age and occupation, and moderate
obesity did so with occupation (included in the model).
Results
Patterns in weight/height discrimination
The prevalence of weight/height discrimination in US adults
was relatively high, particularly in certain population groups
(Table 1). On average, 10.3% of women reported daily or
lifetime discrimination due to weight/height, while men
were half as likely to report such experiences (4.9%, Po0.01).
Younger cohorts among men and women had the highest
rates of perceived weight/height discrimination (7.3% for
men aged 25–34 years and 14.1% for women aged 35–44
years). Gender-related differences in the prevalence of
weight/height discrimination varied greatly by age from a
5% difference among older adults (aged 65–74 years) to more
than fivefold increase for 45- to 54-year-old women. There
were also large gender differences in weight/height discrimi-
nation across education groups. For example, men with low
educational achievement had few occurrences of weight/
height discrimination, while women in this group showed
the highest prevalence of this discrimination (3.5 and 12.6%,
respectively, Po0.05).
There were substantial differences across weight groups in
the rates of weight/height discrimination among men and
women. Whereas 2% of normal weight men and women
reported weight/height discrimination (potentially due to a
previously higher weight or measurement error), moderately
obese women with a BMI of 30–35 were more than three
times more likely than men in the same weight group to
report weight/height discrimination (20.6 vs 6.1%, Po0.01).
The difference in higher rates of weight/height discrimina-
tion for women vs men narrowed but remained significant in
people with BMI of 35þ (45.4 vs 28.1%, Po0.05).
Among racial groups, weight/height discrimination was
most prevalent among minorities, particularly African-
American women (23.9%) and men (12.7%). Gender-related
differences in discriminatory experiences were similar bet-
Table 1
aged 25–74 by sociodemographic and body weight characteristics
Rates of perceived weight/height discrimination among US adults
Characteristic Sample sizeWeight/height discrimination (%)
Total sampleMenWomen
Gender
Male
Female
1104
1186
4.9
10.3
Age
25–34
35–44
45–54
55–64
65–74
464
533
572
449
272
9.4
9.9
6.9
5.3
4.0
7.3
5.6
2.1
3.9
3.9
10.9
14.1*
10.9*
6.1
4.1
Race
White
Black
Other race/multiracial
2025
150
115
6.9
19.4
9.0
4.4
12.7
5.0
9.0*
23.9
12.9
Highest education
Less than high school
High school
Some college
College and above
211
671
710
698
8.4
6.4
8.9
8.1
3.5
3.9
6.1
5.1
12.6**
8.1**
10.9**
11.3*
Marital status
Married
Single/divorced/widowed
1486
804
7.2
9.5
4.8
5.2
9.4*
11.9*
Weight status
Normal weight
Overweight
Moderate obesity
Severe obesity
905
836
339
165
2.2
5.7
13.5
39.8
2.2
3.5
6.1
28.1
2.2
8.6**
20.6*
45.4**
Occupation
Managerial/professional
Sales/clerk
Service
Worker/mechanic
Laborer/operator
No occupation
677
512
196
171
177
557
7.9
10.3
7.7
5.4
7.3
6.5
5.4
5.0
6.4
4.3
3.3
4.5
10.4**
12.5*
8.5
NA
NA
7.2
Total22907.9 4.910.3*
The reported estimates are weighted. NA, estimates not provided due to a
very small sample in the group.*Significantly different rates of weight/height
discrimination between men and women at 1% level.**Significantly different
rates of weight/height discrimination between men and women at 5% level.
Perceptions of weight discrimination
RM Puhl et al
4
International Journal of Obesity

Page 5

ween black and white groups with a somewhat higher
prevalence of weight/height discrimination for Asian/multi-
racial women as compared to men.
Other types of discrimination
Compared to other forms of discrimination among US
adults, weight/height discrimination ranked as the third
most prevalent cause of perceived discrimination among
women (after gender and age discrimination), and the fourth
most prevalent form of discrimination among all adults
(after gender, age and race discrimination) (Figure 1).
In general, rates of self-reported discrimination were high
in this sample with almost every second person reporting at
least one occurrence of any type of discrimination (46%),
particularly among women. Gender discrimination was the
most prevalent type of discrimination reported due to
particularly high rates reported by women (27%), exceeding
by a margin other common causes of discrimination like
race, age and weight/height.
Race-attributed discrimination was reported by 17% of
men and 9% of women, whereas the prevalence of age
discrimination was the same for men and women (10–11%).
Other common causes of discrimination reported by men
were some aspect of appearance other than weight/height
(8.3%),andethnicity/nationality
reported these causes half-as-often. Causes of discrimination
like sexual orientation, religion and physical disability were
relatively uncommon (1–3%).
(6.2%), butwomen
Lifetime experiences of discrimination
Lifetime experiences of discrimination occurred primarily
in employment settings. Almost 60% of participants reporting
weight/height
based discrimination (for example, not hired for a job) on
average four times during their lifetime. This is similar to
experiences of people reporting race discrimination (53%),
and higher compared to individuals reporting gender
discrimination (40%). Other forms of employment discrimi-
nation were similarly prevalent among participants report-
ing weight/height, gender or race discrimination, such as not
given a job promotion reported by approximately a third of
respondents in each category and those who were wrongly
fired by approximately 11–14% of the survey respondents.
Discriminatory experiences outside of employment set-
tings were prevalent across race, gender and weight/height
discrimination, particularly from service providers through
the denial or provision of poorer service, and in educational
institutions through discouragement by a teacher or advisor
from seeking higher education and denial of scholarship.
In general, the distribution of lifetime experiences of discrimi-
nation was similar across types and settings of discrimina-
tion. The number of lifetime experiences among participants
reporting weight/height discrimination was almost identi-
cally distributed, with 20% of the group falling into one of
the five frequencies: 0 time, 1 time, 2–3 times, 4–9 times or
10þ experiences.
discrimination experiencedemployment-
Discrimination in interpersonal relationships
Discriminationdue
reported in daily interpersonal relationships and was compa-
rable to rates of race discrimination and somewhat more
prevalent than gender discrimination. Being treated with less
respect and courtesy than other people and being perceived
to weight/heightwascommonly
Rates of Perceived Discrimination Among Americans Aged 25-74
0%
4%
8%
12%
16%
20%
24%
28%
32%
GenderRaceAge Weight or height Appearance Ethnicity/nationality
Men
Women
Error bars indicate 95% confidence intervals
Figure 1
Rates of perceived discrimination among Americans aged 25–74 years.
Perceptions of weight discrimination
RM Puhl et al
5
International Journal of Obesity