The Effect of Person-First Language on Obesity Stigma

Abstract

Purpose

To test the assumption that person-first language (PFL) reduces obesity stigma, mediated by perceived personal responsibility for obesity.

Design

Cross-sectional, experimental.

Setting

Online, United States.

Participants

299 young adults.

Measures

Participants read a vignette using PFL or identity-first language (IFL) or about someone without obesity. Participants reported perceived personal responsibility for obesity, and 3 operationalizations of obesity stigma: prejudice, stereotypes, and support for punitive policies. Mediation analyses were used to test if the manipulation affected obesity stigma, through perceived personal responsibility.

Results

There was no indirect effect of PFL vs IFL on the 3 outcomes (95% CIs contained zero). However, the indirect effects of PFL vs no-obesity condition were significant (prejudice: β = −0.10, SE = 0.05, 95% CI [−0.22, −0.01]; stereotypes: (β = 0.07, SE = 0.03, 95% CI [0.01, 0.14]); punitive punishment: (β = −0.06, SE = 0.04, 95% CI [−0.15, −0.01]). Also, the indirect effects of IFL vs no-obesity condition on stereotypes (β = 0.07, SE = 0.04, 95% CI [0.0003, 0.15]) and punitive punishment (β = −0.06, SE = 0.04, 95% CI [−0.15, −0.0002]) were significant.

Conclusion

PFL may not affect obesity stigma as it does in the context of other marginalized groups. The effect of PFL and IFL, compared to the no-obesity condition, suggests future routes for intervention.

Keywords

obesity stigma person-first language intervention

Purpose

Obesity stigma, or marginalization of people with higher weight, is pervasive and contributes to negative health outcomes such as depression and anxiety, health care avoidance, and weight gain.¹ In recent years, researchers have attempted to identify interventions to reduce obesity stigma. One such intervention is the use of “person-first” language (PFL; ie, person with obesity) over “identity first” language (IFL; ie, obese person), an approach that has been widely endorsed by major public health organizations.^2,3 PFL is assumed to be effective because mentioning the term “person” before the “condition” should frame the target as an individual, apart from their stigmatized status. Alternatively, IFL presents one’s stigmatized condition before mentioning the target’s humanity, and thus is thought to frame the condition as a central part of their identity. Preventing perceivers from defining the target by their condition (eg, higher weight), is thought to lower stigma.³

There is evidence that PFL reduces stigma towards other stigmatized groups, such as people with schizophrenia⁴ and addiction.^5,6 However, no studies have experimentally tested whether PFL reduces obesity stigma, despite recommendations to adopt such language. Additionally, none of the experiments showing support for the use of PFL (in the context of other marginalized groups) tested the mechanism by which it leads to lower stigma. While it is assumed that PFL reduces stigma by separating the stigmatized condition (eg, obesity) from the individual, this remains untested. A potential mechanism that may explain the assumed effect of PFL on stigma is lower perceived personal responsibility for one’s stigmatized status.⁶ Emphasizing the target’s humanity independent of their condition, may facilitate consideration of other factors that influence a person’s stigmatized condition besides their personality or volitional behaviors (eg, genetics, environment). If reducing perceptions that a target is personally responsible for their “condition” is a mechanism by which PFL reduces stigma, there is reason to believe that PFL would be especially effective in reducing obesity stigma. It is well established that higher personal responsibility attributions (eg, laziness, overeating) are a leading cause of weight stigma.⁷

We use a three-group experimental design to explore the effects of PFL on obesity stigma, and a mechanism by which they take place. All participants are randomly assigned to read about an individual with obesity, using PFL or IFL, or a person without obesity (no-obesity comparison group). The main aim of the study is to test if the use of PFL (vs IFL) when reading about an individual with obesity leads to lower obesity stigma, and if the effect is explained by a lower perceived personal responsibility for obesity. We also assess if encountering a target with obesity, using PFL, results in lower obesity stigma compared to not encountering a target with obesity (no-obesity condition), mediated by lower perceived personal responsibility for obesity. Exposure to individuals that belong to a marginalized group can lead to more positive attitudes towards that group, but only when the target is portrayed in a favorable way.⁸ Thus, if participants in the PFL condition report lower obesity stigma compared to participants in the no-obesity condition, we can infer that PFL is beneficial to obesity stigma in general, and not just compared to using language that integrates obesity into one’s identity (IFL).

Specifically, we hypothesize that: participants in the PFL condition will report lower obesity stigma (represented by prejudice, negative stereotypes, and support for policies that penalize individuals with obesity) than participants in the IFL and no-obesity conditions. The effect will be mediated by lower perceived personal responsibility for obesity.

Methods

Sample

The study was administered online to 299 undergraduate students between February and April 2023. An a priori power analysis run on G*power⁹ indicated that a sample of 214 was needed to achieve a power of .90 and to detect an effect size 0.25. We recruited additional participants to account for possible missing data and failed attention checks. The response rate was 100%, as students were self-selected from the university’s psychology subject pool. Participants were 51.2% female and 48.8% male. Most identified as White (63.5%). Other commonly represented racial/ethnic groups were Asian (22.1%), Hispanic/Latino/a/x (10.7%), and Black (8.7%). Additionally, 70.6% had a “normal” body mass index (BMI)¹⁰ and M_bmi = 23.38 (SD = 3.78). To ensure data quality, 4 attention checks and a manipulation check were embedded in the survey.

Design

Participants were randomly assigned to 1 of 3 conditions: PFL (n = 99), IFL (n = 100), and a no-obesity condition (n = 100). All participants were provided with a definition of obesity (BMI >30) before reading a vignette about a student named “Jesse,” adapted from prior literature.^6,11 In the PFL condition, the target needed to be identified as a person before their obesity was mentioned, thus, Jesse was described as “a person with obesity.” In the IFL condition, Jesse was described as “an obese person,” as IFL involves mentioning the stigmatized condition before the individuality of the target. In the no-obesity condition, the target was described as having a BMI of 22, which is considered healthy (ie, does not indicate overweight or obesity, nor underweight) by the Centers of Disease Control and Prevention.¹⁰ Besides the description of Jesse’s weight status, all information was uniform across the 3 conditions and were equal in length. The vignettes did not contain information about the cause of Jesse’s weight and contained neutral information about their health behaviors. Jesse was meant to represent a typical college student with “pretty good health” and varied health behaviors, sometimes being physically active and eating a healthy diet, and sometimes less so, especially with Jesse was busy with school. Jesse’s experience having COVID-19 was also described, given that the coronavirus was a salient health indicator during the time that the study was administered. The name “Jesse” was chosen as it is gender-neutral and would not confound results. Full vignettes are available in supplemental materials. The study was approved by The George Washington University Office of Human Research (IRB #: NCR224692). All participants consented before starting the study.

Measures

All measures were adapted to avoid PFL or IFL language (eg, wording changed from “obese people” to “people with a BMI of at least 30”).

Demographics

Biological sex, thin idealization (the preoccupation with being thin),¹² implicit theories of body weight (the perception that weight is malleable [incremental theory] vs fixed [entity theory]),¹³ and friends/family who may have BMI >30 were covariates in analyses because of their associations with the proposed mediator and/or outcome variables.

Perceived Personal Responsibility for Obesity

To measure personal responsibility, 7 items were adapted from the Causal Attributions for Weight subscale.^14,15 The stem read “How much do you think the following factors cause obesity?” Sample items include “laziness” and “overeating.” Response options ranged from 1 (not at all) to 5 (extremely). A mean was calculated (α = 0.86) with higher scores indicating stronger beliefs that obesity is caused by one’s own character and controllable behaviors.

Obesity Prejudice

Prejudice was measured with an adapted version of the dislike subscale of the Antifat Attitudes Questionnaire (AFA).¹⁶ Participants were instructed to think of people who have a BMI of at least 30 and indicate how much they agree or disagree with 8 statements (eg, “I really don’t like them much”). Response options ranged from 1 (strongly disagree) to 7 (strongly agree). A mean was calculated (α = 0.95) with higher scores indicating more prejudice.

Negative Stereotypes About People with Obesity

The Fat Phobia Scale- Short Form¹⁷ and Negative Obesity Stereotypes Scale¹⁸ were adapted for use. Participants were presented with 14 pairs of adjectives “sometimes used to describe people who have a BMI of at least 30” and asked to select the number closest to the adjective that best describes their feelings and beliefs about them (eg, lazy-industrious). Response options ranged from 1 to 5. A mean was calculated with lower scores indicating more negative stereotypes (α = 0.90).

Punitive Policy for Obesity

Two items from the “price raising policies” subscale of Public Policy Support¹⁹ were adapted to be relevant to college students (eg, “GW health insurance fees should be higher for individuals who have a BMI of 30 or higher”). Response options ranged from 1 (strongly opposed) to 7 (strongly support). A mean was calculated (r = 0.32) with higher scores indicating more policy support.

Analysis

Participants (n = 62) were excluded from analyses if they failed 2 or more of the attention checks and/or manipulation check,²⁰ leaving N = 237 (PFL n = 80, IFL n = 77, no-obesity n = 80).

First, descriptive statistics were run. The hypothesis was then tested with mediation analyses using Hayes’ PROCESS²¹ model 4 with an indicator coding system for a multicategorical independent variable and 5000 bootstrapped samples. The PFL condition was set as the reference group, allowing us to compare it with the IFL and no-obesity conditions. In all models perceived personal responsibility was entered as the mediator. A model was tested for each of the 3 types of stigma (prejudice, stereotypes, punitive policy).

When we found that there was no significant difference between the PFL and IFL conditions, but there was a significant difference between PFL and the no-obesity condition, we decided to run a post-hoc analysis comparing IFL and no-obesity condition. The no-obesity condition was set as the reference group with perceived personal responsibility as the mediator. Again, a model was tested for each of the 3 operationalizations of stigma (prejudice, stereotypes, punitive policy).

Results

Descriptive Statistics

The means of the 3 outcome variables did not significantly differ between the 3 conditions (see Table 1). Table 2 shows descriptive statistics and correlations for the primary variables. Additional tables and figures displaying results are available in supplemental materials.

Table 1.

Outcome Variables by Condition.

	PFL		IFL		No-obesity		F	p
	M	SD	M	SD	M	SD	F	p
Prejudice	1.64	1.01	1.80	0.97	1.75	0.97	0.53	.58
Negative stereotypes	2.54	0.51	2.56	0.53	2.57	0.47	0.01	.91
Punitive policy	1.78	0.85	1.87	0.92	1.77	0.89	0.32	.73

Note. M = mean, SD = standard deviation.

Table 2.

Correlation Matrix.

	1	2	3	4	5	6	7	8
1. Sex	-
2. Thin idealization	−.30^a	-
3. Implicit theories	−.35^a	.13^b	-
4. Close others with obesity	−.07	.09	−.04	-
5. Personal responsibility attribution	.32^a	−.01	−.21^a	−.08	-
6. Prejudice	.20^a	.04	−.10	−.18^a	.39^a	-
7. Negative stereotypes	−.22^a	.18^a	.14^b	.16^b	−.50^a	−.46^a	-
8. Punitive policy	.23^a	−.05	−.10	−.16^b	.28^a	.43^a	−.33^a	-
Mean	1.49	4.30	2.99	2.82	3.27	1.73	2.56	1.80
SD	0.50	1.50	1.07	2.36	0.77	0.98	0.50	0.88

Note. SD = standard deviation.

Sex coded as 1 = female, 2 = male.

Thin Idealization and Implicit Theories coded 1-7.

For Implicit Theory, higher numbers indicate entity (vs. incremental) theory.

^aP < .01.

^bP < .05.

PFL as Reference Group

There was no significant difference between the PFL and IFL conditions on personal responsibility attributions (β = 0.002, SE = 0.12, 95% CI [-0.22, 0.23]). As such, there was no indirect effect of PFL vs IFL on any of the outcomes (prejudice: β = −0.001, SE = 0.05, 95% CI [-0.10, 0.11]; negative stereotypes: (β = 0.0007, SE = 0.04, 95% CI [-0.12, 0.14]); punitive punishment: (β = 0.0006, SE = 0.03, 95% CI [-0.07, 0.07]).

PFL participants reported lower perceived personal responsibility attributions compared to the no-obesity condition (β = −0.24, SE = 0.11, 95% CI [−0.46, −0.01]). Higher personal responsibility led to higher prejudice (β = 0.44, SE = 0.08, 95% CI [0.28, 0.60]), negative stereotyping (β = −0.30, SE = 0.04, 95% CI [−0.38, −0.22]), and support for punitive punishment (β = 0.27, SE = 0.08, 95% CI [0.11, 0.42]). The indirect effects of PFL vs the no-obesity condition on all 3 outcomes via personal responsibility were significant (prejudice: β = −0.10, SE = 0.05, 95% CI [-0.22, −0.01]; negative stereotypes: (β = 0.07, SE = 0.03, 95% CI [0.01, 0.14]); punitive punishment: (β = −0.06, SE = 0.04, 95% CI [-0.15, −0.01]).

No-Obesity as Reference Group

IFL participants also reported lower perceived personal responsibility attributions compared to the no-obesity condition (β = −0.23, SE = 0.11, 95% CI [−0.46, −0.01]). The indirect effects of IFL vs no-obesity condition on prejudice were marginal (β = −0.10, SE = 0.05, 95% CI [−0.21, 0.002]). The indirect effects of IFL vs no-obesity condition on negative stereotypes (β = 0.07, SE = 0.04, 95% CI [0.0003, 0.15]) and punitive punishment (β = −0.06, SE = 0.04, 95% CI [−0.15, −0.0002]) were significant.

Discussion

The current experiment tested the assumption that using PFL (vs IFL) can result in less obesity stigma, and whether the effect is mediated by lower personal responsibility attributions. There were no significant differences between PFL and IFL on any of the 3 tested operationalizations of obesity stigma, which fails to support the assumption that PFL leads to less stigma than IFL. However, compared to the no-obesity condition, both PFL and IFL led to less stigma, mediated by lower personal responsibility attributions.

At first glance, one might consider if the manipulation was too brief to cause an effect of PFL (vs IFL). However, studies showing a positive effect of PFL on other stigmatized groups used similar, brief manipulations,^4-6,22 suggesting that the brevity of the manipulation does not account for the null effect. An alternative explanation is that PFL may not be as beneficial in the context of obesity compared to other stigmatized groups. In contrast to addiction or mental illness, obesity can become more stigmatized when framed as a “condition.” Although framing obesity as a condition can decrease personal responsibility attributions and lead to lower stigma, research shows it can simultaneously increase stigma by implying that weight is an unchangeable essence of a person.^13,23 Labeling obesity as a condition may also lead people to internalize that that fatness is inherently negative.^24,25 Finding that PFL did not reduce obesity stigma more than IFL highlights a need to further explore the effect of language, especially in the context of widespread advocacy and adoption of PFL.

Finding that both the PFL and IFL, compared to the no-obesity, groups reported lower obesity stigma suggests a potential route for future interventions. The manipulation was designed to depict Jesse as a “typical” college student, with varied dietary and physical behaviors. While all participants were exposed to a fellow student that they could connect with, only among PFL and IFL participants did this person have a BMI that indicates obesity. The student that participants in the no-obesity condition read about, and could affiliate with, did not have a high BMI. PFL and IFL participants’ exposure to a student with a larger body who they related to may have reduced blame toward the student for their weight, thus reducing stigma. In contrast, no-obesity participants did not have the opportunity to foster a sense of connection towards a peer with obesity or to reduce weight-related blame due to their “healthy” weight status. Future interventions could attempt to enhance similarity toward people with higher weights and test its efficacy in mitigating obesity stigma.

Although the current study offers insight into the effects of PFL on obesity stigma, a note-worthy limitation is that the study was conducted among college students who were predominantly White and of a “normal” weight. The effect of PFL on obesity stigma should be tested with a more diverse sample. Additionally, the effects of repeated exposure to PFL should be tested, especially because advocates argue for its continued use.

Conclusions

The current study did not find support for the assumption that the use of PFL (eg, person with obesity) vs IFL (eg, obese person) leads to less obesity stigma. Although further research is necessary, results may indicate that PFL is not as beneficial for obesity stigma as it is for other marginalized groups, perhaps because framing obesity as a “condition” has different implications. However, the finding that both PFL and IFL, compared to the no-obesity condition, led to lower obesity stigma through lower personal responsibility for obesity suggests potential utility of interventions that aim to create a sense of affiliation with people with larger bodies.

So what?

What is Already Known on This Topic?

Experimental studies show that person-first language (PFL) vs. identity-first language (IFL) can result in lower stigma towards some marginalized groups (e.g., people with addiction, schizophrenia), and influential public health organizations promote the use of PFL in the context of obesity.

What Does This Article Add?

This is the first study to experimentally test the assumption that PFL can reduce obesity stigma, compared to IFL, as well as the first the tests a mechanism behind the effect. Results show that there were no significant differences between PFL and IFL on obesity stigma, which fails to support the assumption that PFL leads to less stigma than IFL.

What are the Implications for Health Promotion Practice or Research?

More research is needed to better understand the effects of PFL on obesity stigma, especially because of the widespread push to use that language. Additionally, other intervention routes should be pursued, such as aiming to create a sense of affiliation with people with larger bodies.

Supplemental Material

Supplemental Material - The Effect of Person-First Language on Obesity Stigma

Supplemental Material for The Effect of Person-First Language on Obesity Stigma by Rebecca K. Hoffman, Stacy M. Post, Tonya Dodge, and Michelle L. Stock in American Journal of Health Promotion

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Open Research Statement

This study was pre-registered on Open Science Framework. The analytic plan, code, and materials used are visible at . The dataset and codebook used for analysis are also available at https://osf.io/n5hdq/.

Ethical Statement

ORCID iD

Rebecca K. Hoffman

Supplemental Material

Supplemental material for this article is available online.

References

Tomiyama

Carr

Granberg

, et al. How and why weight stigma drives the obesity “epidemic” and harms health. BMC Med. 2018;16(1):123-126. doi:10.1186/s12916-018-1116-5

Centers for Disease Control and Prevention . Overweight & obesity. 2023. Retrieved from. https://www.cdc.gov/obesity/index.html#:∼:text=We_encourage_the_use_of,%25_of_children_are_obese

National Institutes of Health . Person-first and destigmatizing language. 2023. Retrieved from. https://www.nih.gov/nih-style-guide/person-first-destigmatizing-language

Granello

Gorby

. It’s time for counselors to modify our language: it matters when we call our clients schizophrenics versus people with schizophrenia. Jour of Counseling Develop. 2021;99(4):452-461. doi:10.1002/jcad.12397

Baker

Hamilton

Culpepper

McCune

Silone

. The effect of person-first language on attitudes toward people with addiction. J Addict Offender Couns. 2022;43:38-49. doi:10.1002/jaoc.12102

Kelly

Westerhoff

. Does it matter how we refer to individuals with substance-related conditions? A randomized study of two commonly used terms. Int J Drug Policy. 2010;21(3):202-207. doi:10.1016/j.drugpo.2009.10.010

Puhl

Heuer

. Obesity stigma: important considerations for public health. Am J Public Health. 2010;100(6):1019-1028. doi:10.2105/AJPH.2009.159491

Schutz

Smout

. Effects of positive versus negative framing on stigma toward individuals recovering from methamphetamine use disorder during randomized brief video exposure. J Psychoactive Drugs. 2024;56(1):88-96. doi:10.1080/02791072.2022.2149436

Erdfelder

Faul

Buchner

. GPOWER: a general power analysis program. Beha Res Meth, Inst Comp. 1996;28(1):1-11. doi:10.3758/BF03203630

10.

Centers for Disease Control and Prevention . About adult BMI. 2022. Retrieved from. https://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html

11.

Martinelli

Meerkerk

Nagelhout

, et al. Language and stigmatization of individuals with mental health problems or substance addiction in The Netherlands: an experimental vignette study. Health Soc Care Community. 2020;28(5):1504-1513. doi:10.1111/hsc.12973

12.

Schaefer

Harriger

Heinberg

Soderberg

Kevin Thompson

. Development and validation of the sociocultural attitudes towards appearance questionnaire-4-revised (SATAQ-4R). Int J Eat Disord. 2017;50(2):104-117. doi:10.1002/eat.22590

13.

Burnette

. Implicit theories of body weight: entity beliefs can weigh you down. Pers Soc Psychol Bull. 2010;36(3):410-422. doi:10.1177/0146167209359768

14.

Foster

Wadden

Makris

, et al. Primary care physicians’ attitudes about obesity and its treatment. Obes Res. 2003;11(10):1168-1177. doi:10.1038/oby.2003.161

15.

Pearl

Lebowitz

. Beyond personal responsibility: effects of causal attributions for overweight and obesity on weight-related beliefs, stigma, and policy support. Psychol Health. 2014;29(10):1176-1191. doi:10.1080/08870446.2014.916807

16.

Crandall

. Prejudice against fat people: ideology and self-interest. J Pers Soc Psychol. 1994;66:882-894.

17.

Bacon

Scheltema

Robinson

. Fat Phobia Scale revisited: the short form. Int J Obes Relat Metab Disord. 2001;25:252-257.

18.

Black

Sokol

Vartanian

. The effect of effort and weight controllability on perceptions of obese individuals. J Soc Psychol. 2014;154(6):515-526. doi:10.1080/00224545.2014.953025

19.

Barry

Brescoll

Brownell

Schlesinger

. Obesity metaphors: how beliefs about the causes of obesity affect support for public policy. Milbank Q. 2009;87(1):7-47. doi:10.1111/j.1468-0009.2009.00546.x

20.

Cobanoglu

Cavusoglu

Turktarhan

. A beginner’s guide and best practices for using crowdsourcing platforms for survey research: the case of Amazon Mechanical Turk (MTurk). Journal of Global Business Insights. 2021;6(1):92-97. doi:10.5038/2640-6489.6.1.1177

21.

Hayes

. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. 2nd ed. New York: Guilford Press; 2018.

22.

Denver

Pickett

Bushway

. The language of stigmatization and the mark of violence: experimental Evidence on the social construction and use of criminal record stigma. Criminology. 2017;55(3):664-690. doi:10.1111/1745-9125.12145

23.

Hoyt

Burnette

Auster-Gussman

Blodorn

Major

. The obesity stigma asymmetry model: the indirect and divergent effects of blame and changeability beliefs on antifat prejudice. Stigma and Health. 2017;2(1):53-65. doi:10.1037/sah0000026

24.

Meadows

Daníelsdóttir

. What’s in a word? On weight stigma and terminology. Front Psychol. 2016;7(OCT):1527-1528. doi:10.3389/fpsyg.2016.01527

25.

Steele

Finucane

. Philosophically, is obesity really a disease? Obes Rev. 2023;24:e13590. doi:10.1111/obr.13590

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.45 MB