AIDS PATIENT CARE and STDs ORIGINAL ARTICLE

Volume XX, Number XX, 2017

ª Mary Ann Liebert, Inc.
DOI: 10.1089/apc.2016.0259

Body Image and Risk Behaviors in Youth with HIV

Ronald H. Dallas, PhD,1 Megan M. Loew, MS,2 Megan L. Wilkins, PhD,1,2 Yilun Sun, MS,3
Li Tang, PhD,3 Jerlym S. Porter, PhD,2 and Aditya H. Gaur, MD1

Abstract

Body image concerns are common among people living with HIV. Among adults with HIV, body image
concerns have been shown to be related to risky sexual behaviors; little research has been conducted among
youth living with HIV (YLWH). The current study examined the predictors, including body image, of sexual
risk behaviors among YLWH. Adolescents from a single clinic (n = 143; age range 16–24 years; 69% male;
95% African American) completed a computerized self-report survey to assess demographic, behavioral, and
body image domains. Demographic and clinical data were abstracted from the medical record. Logistic re-
gression analyses assessed associations between risk factors and risky sexual behaviors. Results indicated that
YLWH who reported less favorable body image perceptions ( p = 0.04) and more sexual partners ( p = 0.05) were
less likely to use condoms during their last sexual encounter. YLWH with six or more sexual partners were
more likely to use drugs or alcohol during their last sexual encounter ( p = 0.03). A belief that their HIV
medications changed their body physically ( p = 0.05), history of HIV-related complications ( p = 0.03), an
undetectable viral load at their most recent clinical laboratory draw ( p = 0.01), and having a high school
diploma or equivalent ( p = 0.001) were independently associated with disclosure of participant’s HIV status to a
romantic/sexual partner. Findings suggest that body image perceptions may influence risky sexual behavior in
YLWH. Further study is warranted to understand and intervene upon this relationship to improve individual and
public health outcomes.

Keywords: HIV, body image, risky sexual behaviors, youth, adolescents

Introduction such as depressive symptoms, which among YLWH have

been associated with reduced medication adherence and in-
creased risk behaviors.6–8 This is of concern given that youth
C oncerns about body image are a common occurrence
among adults living with HIV.1 Body image among
adults living with HIV has also been shown to be related
accounted for *26% of new HIV infections in the United
States in 2010.9 As a result, research is urgently needed ex-
to risky sexual behavior.2,3 Multiple physical, psychosocial, amining novel pathways that predict HIV transmission risk
and cultural factors influence one’s body image. In turn, among YLWH to inform effective intervention development.
body image may impact upon an individual’s engagement in High-risk sexual behaviors among people living with HIV
HIV-related self-care behaviors (e.g., transmission risk re- can have detrimental individual and public health level
duction, and medication adherence). Emerging evidence consequences.10 Research has shown a relationship between
among adults living with HIV has highlighted the role body body image and sexual risk behaviors across multiple age,
image disturbance may exert on HIV self-care behaviors that ethnic, and sexual preference groups.11 Poor body image may
influence both individual and public health related outcomes; impact condom self-efficacy, increasing one’s risk of ac-
however, there is sparse evidence among youth living with quiring or transmitting HIV.12 Evidence suggests that more
HIV (YLWH).2 YLWH engage in risky sexual behaviors compared to adults
YLWH are prone to struggle with health and/or poor body with HIV.13 In another study examining risk behaviors
image, as well as engage in sexual transmission risk behav- among YLWH, almost half of youth reported engaging in
iors.4,5 Body dissatisfaction has been shown to contribute to unprotected sex since learning they had HIV.14 In addition to
negative emotions and psychosocial problems in YLWH adolescent risk factors such as use of social media to connect

Departments of 1Infectious Diseases, 2Psychology, and 3Biostatistics, St. Jude Children’s Research Hospital, Memphis, Tennessee.

1
2 DALLAS ET AL.

with partners, low concern about HIV infection, and infrequent the MBSRQ-Appearance Scales (AS).18 The MBSRQ-AS is a
discussion about HIV status,15 body image may play a role in shortened 34-item measure from the MBSRQ that consists of
increasing risks. Despite the recognized relationship between five subscales: Appearance Evaluation, Appearance Orienta-
body image and high-risk sexual behavior, studies examining tion, Overweight Preoccupation, Self-Classified Weight, and
this among YLWH are lacking and urgently needed. the Body Areas Satisfaction Scale (BASS). To estimate overall
To address this identified gap in the literature, the current body image from this measure, a combination score from the
study aimed to examine the predictors, including body image Appearance Evaluation and BASS was derived.18
measures, of sexual risk behaviors among YLWH. It was
hypothesized that negative body image perceptions would be Figure Rating Scale. The Figure Rating Scale used for
significantly related to increased HIV sexual transmission this study is modeled after Stunkard’s19 widely used nine-
risk behaviors. For the purposes of this study, youth will be figure scale. The current version is a culturally relevant
defined as those between the ages of 16 and 24 years old. measure of body size satisfaction; the figures have hair and
facial features resembling people with multi-ethnic back-
Methods grounds. It consists of a series of nine schematic figures of
varying size. Findings indicate good test–retest reliability and
Participants and procedure moderate correlations with other measures of body dissatis-
This study was a secondary analysis from an existing da- faction, eating disturbance, and overall self-esteem.20 Parti-
taset. The dataset used was collected as a cross-sectional cipants were asked to identify which image they find most
exploratory pilot study of body image in a sample of YLWH. similar to their current body type, as well as their ideal or
The purpose of the original study16 was to explore self- preferred body type.
reported body image, health behaviors, clinical, and socio-
demographic factors among YLWH. Youth Risk Behavior Survey. The Youth Risk Behavior
Participants were recruited using a census approach Survey monitors priority health-risk behaviors and the
from an adolescent infectious diseases (ID) clinic in the prevalence of obesity and asthma among youth in the United
Mid-southern United States. Inclusion criteria included the States.21 The Youth Risk Behavior Surveillance System
following: youth (aged 16–23 years old inclusive) with (YRBSS)21 includes a national school-based survey con-
documented diagnosis of HIV infection, awareness of HIV ducted by the Centers for Disease Control and Prevention
diagnosis, primary language was English, and ability to pro- (CDC) and state, territorial, tribal, and district surveys con-
vide informed consent. Youth with significant motor, sen- ducted by education and health agencies/governments. The
sory, or other impairments that precluded participation in YRBSS monitors six categories of priority health-risk be-
the study as determined by the clinic psychologist were ex- haviors among youth and young adults, including behaviors
cluded. Participants completed an audio computer-assisted that contribute to unintentional injuries and violence; tobacco
self-interview (ACASI)17 in a private area of the clinic and use; alcohol and other drug use; sexual behaviors related to
received $10 compensation for their time and effort. The unintentional pregnancy and sexually transmitted infections,
electronic survey was reviewed by the ID clinic’s youth including HIV; unhealthy dietary behaviors; and physical
community advisory board, and recommendations were in- inactivity.21 For the purposes of this study, select questions
corporated into the final script. Relevant demographic and pertaining to sexual behaviors and drug use were included.
clinical data were also abstracted from the youth’s medical
record. This study was approved by the clinic site’s institu- Center for Epidemiologic Studies Depression Scale–Short
tional review board. Form. The Center for Epidemiologic Studies Depression
Scale–Short Form (CES-D-10) is the abbreviated version of
one of the most commonly used self-report depression scales
Measures
designed for use in clinical and research settings.22 The
Demographics questionnaire. Participants were asked CES-D-10 has been found to perform, as well as the full item
several demographic questions, including gender, education questionnaire.23 Cronbach’s alpha for the CES-D-10 scale
status, sexual identity, sexual history [e.g., men who have sex reliability has consistently exceeded 0.8, and item-to-total
with men (MSM)], and race. Social/environmental questions correlations have been low (<0.5).22 High internal consis-
were also collected and included with whom the participant tency also has been reported.22,23 The range of scores is 0–30
lives and whether/not they receive public assistance (e.g., with higher scores representing a more depressed mood. With
public health insurance, food stamps, or other public financial all 10 items, the CES-D-10 can be regarded as a measure
support). of affect (i.e., the presence of negative affect and, to a lesser
extent, the absence of positive affect). A cutoff score of 10
Multi-dimensional Body-Self Relations Questionnaire. has been recommended for the categorization of depres-
The Multi-dimensional Body-Self Relations Questionnaire sion22; therefore, those with a score of 10 or higher were
(MBSRQ) is a well-validated self-report inventory for the categorized as having elevated depressive symptoms for the
assessment of body image.18 Internal consistency of the purpose of these analyses.
subscales on the MBSRQ ranges from a = 0.70–0.91 for male In addition to the measures above, participants currently
respondents and a = 0.73–0.90 for female respondents. Test– taking antiretroviral (ARV) medications were asked if they
retest reliability estimates ranged from 0.71 to 0.89 for male believed their ARVs were causing changes to their body.
respondents and 0.74 to 0.94 across subscales at 1-month Female participants who were currently prescribed hormonal
follow-up for female respondents. The MBSRQ is intended contraceptives were asked if they believed their birth control
for use with individuals 15 years or older. This study used medications caused weight loss or weight gain and if they
BODY IMAGE AND RISK BEHAVIORS IN YOUTH WITH HIV 3

ever stopped taking their birth control medication due to Table 2. Factors Associated with Using
concerns that it was causing body weight changes. a Condom During Last Sexual Encounter
Univariate Multivariate
Medical record/chart abstraction. Demographic and health analyses analyses
status variables, such as disease and treatment status, laboratory
values, and vital signs, were abstracted from the participant’s Odds ratio Odds ratio
medical record by study staff. Specific values collected for these (95% CI) p (95% CI) p
analyses included: participant’s most recent height, weight, and
No of sexual partners
body mass index (BMI). Calculated BMI scores also were cat- One to five people Ref
egorized based on the CDC standard weight status categories Six or more people 0.39 0.03 0.39 0.04
(underweight, normal, overweight, and obese) for BMI.24 (0.16–0.93) (0.16–0.96)
Appearance evaluation 2.02 0.04 1.93 0.05
Analyses and body areas (1.05–3.90) (0.99–3.77)
satisfaction
Descriptive statistics was used to describe and characterize combination score
the sample. Associations with risky sexual behaviors were as-
sessed using univariate and multiple logistic regression models. CI, confidence Interval.
Risk factors with a p-value below 0.1 in univariate logistic
regressions were included in multiple logistic regressions, and a American. Approximately 83% of the sample had acquired
backward selection was performed until all remaining factors HIV through behavioral contact and 82.8% of males in the
were significant at the 0.1 level. Data analyses were conducted study identified as MSM. The mean BMI for the total sample
using SAS version 9.3 and IBM SPSS version 23. was 25.3 (SD = 6.13), CD4 count was 653.4 (SD = 312.1), and
viral load was 12,782.1 copies/mL (SD = 32,538.5). See
Results Table 1 for further sample characteristics.
Using a condom at last sexual encounter was associated
A total of 148 potential participants were approached for with the combined appearance evaluation and body areas
study participation. Of these 148, 5 declined to participate satisfaction scale (AE BASS) score [odds ratio (OR) 1.93;
due to lack of time or being uninterested in the study. Parti- 95% confidence interval (CI) 0.99–3.77; p = 0.05] and num-
cipants were 143 HIV-infected YLWH with an average age ber of sexual partners (OR 0.39; 95% CI 0.16–0.96; p = 0.04),
of 20.7 years (SD = 1.98; range = 16.02–23.98 years). The with participants having higher AE BASS combination
majority was male (69.2%), and 95.1% were black/African scores and fewer number of sexual partners being more likely
to use a condom during their last sexual encounter (Table 2).
Compared to participants with one to five total sexual part-
Table 1. Sample Characteristics ners, participants with six or more sexual partners were more
N = 143 likely to use drugs or alcohol during their last sexual en-
counter (OR 2.49; 95% CI 1.11–5.57; p = 0.03; Table 3).
Age, median (range) 20.98 (16.02–23.98) Participants who received their high school diploma or
Gender, n (%) General Education Development (GED) were more likely to
Male 99 (69.2) have disclosed their diagnosis to a romantic or sexual partner
Female 44 (30.8) (OR 8.33; 95% CI 2.30–30.23; p = 0.001). Those who en-
Race/Ethnicity, n (%) dorsed believing that their ARVs had changed their body
Black 136 (95.1) were more likely to have disclosed their diagnosis to a ro-
White 6 (4.2) mantic or sexual partner (OR 2.98; 95% CI 1.0–8.91;
Other 1 (0.7)
Mode of HIV acquisition, n (%) Table 3. Factors Associated with Using Drugs
Perinatal 24 (16.8) or Alcohol at Last Sexual Encounter
Behavioral 119 (83.2)
MSM, n (%) Univariate Multivariate
Not MSM 17 (17.2) analyses analyses
MSM 82 (82.8) Odds ratio Odds ratio
Prescribed ARVs, n (%) (95% CI) p (95% CI) p
Yes 108 (75.5)
No 35 (24.5) No of sexual partners
Body mass index, mean (SD) 25.3 (6.13) One to five people Ref
CD4 absolute, mean (SD) 653.4 (312.1) Six or more people 2.49 0.03 2.49 0.03
CD4 percentage, mean (SD) 30.8 (10.7) (1.11–5.57) (1.11–5.57)
HIV viral load (copies/mL), 12,782.1 (32,538.5) Depressive symptoms
mean (SD) No Ref
Average weight (kg), 69.4 (44.7–159.1) Yes 1.94 0.09
median (range) (0.89–4.24)
ARVs, antiretrovirals; MSM, men who have sex with men; SD, Backward selection analysis eliminated depressive symptoms at
standard deviation. p < 0.10 level from the final model shown.
4 DALLAS ET AL.

Table 4. Factors Associated with Disclosing HIV Diagnosis to Romantic and/or Sexual Partners
Univariate analyses Multivariate analyses
Odds ratio (95% CI) p Odds ratio (95% CI) p
Belief that ARVs have caused physical changes to body
No Ref
Yes 2.22 (0.88–5.61) 0.09 2.98 (1.00–8.91) 0.05
Received high school diploma or GED
No Ref
Yes 3.48 (1.39–8.71) 0.01 8.33 (2.30–30.23) 0.001
Undetectable viral load
No (‡ 400 copies/mL) Ref
Yes (< 400 copies/mL) 2.18 (1.02–4.66) 0.05 4.53 (1.40–14.69) 0.01
History of HIV related diagnosis
No Ref
Yes 2.40 (0.90–6.46) 0.08 6.18 (1.24–30.94) 0.03
No of sexual partners
One to five people Ref
Six or more people 2.83 (1.23–6.50) 0.01
Backward selection analysis eliminated number of sexual partners at p < 0.10 level from the final model shown.
GED, General Education Development.

p = 0.05). Participants with a history of a HIV related com- more sexual partners in the final adjusted multivariate model
plication (OR 6.18; 95% CI 1.24–30.94; p = 0.03) and an (Table 5). No factors were significantly associated with those who
undetectable viral load (OR 4.53; 95% CI 1.40–14.69; reported engaging in sexual intercourse before the age of 17 years.
p = 0.02) also were more likely to have disclosed their HIV
diagnosis to a romantic and/or sexual partner (Table 4).
Discussion
Participants who reported disclosing their diagnosis to their
most recent sexual partner (OR 3.37; 95% CI 1.36–8.35; Consistent with previous literature,2 these findings suggest
p = 0.009) were not overweight or obese based on the BMI CDC that body image perceptions among YLWH may influence
range ( p = 0.007) and reported using drugs or alcohol at their HIV transmission risk behaviors. Specifically, youth with
most recent sexual encounter (OR 2.65; 95% CI 1.04–6.75; positive feelings about their overall appearance and con-
p = 0.04), but were significantly associated with having six or tentment with most areas of their body were more likely to

Table 5. Factors Associated with Six or More Sexual Partners

Univariate analyses Multivariate analyses
Odds ratio (95% CI) p Odds ratio (95% CI) p
Body mass index (CDC Categorization)
Normal/underweight Ref
Overweight 0.27 (0.10–0.73) 0.01 0.18 (0.06–0.55) 0.003
Obese 0.51 (0.20–1.33) 0.17 0.40 (0.14–1.14) 0.09
Disclosed HIV status to romantic and/or sexual partner
No Ref
Yes 2.83 (1.23–6.50) 0.01 3.37 (1.36–8.35) 0.009
Used drugs or alcohol at most recent sexual encounter
No Ref
Yes 2.49 (1.11–5.57) 0.03 2.65 (1.04–6.75) 0.04
Used condoms at most recent sexual encounter
No Ref
Yes 0.39 (0.16–0.93) 0.03
Gender
Male Ref
Female 0.42 (0.19–0.97) 0.04
Belief that ARV medications caused physical changes in body
No Ref 0.058
Yes 2.29 (0.97–5.39)
Backward selection analysis eliminated condom use, gender, and ARV body changes at p < 0.10 level from the final model shown.
BODY IMAGE AND RISK BEHAVIORS IN YOUTH WITH HIV 5

report using a condom at their last sexual encounter. Previous health.28 Given the well-documented relationship between
literature in this area has primarily focused on adults with HIV body image and health behavior among adults infected with
limiting the examination of the relationship between body HIV and the high rate of new adolescent HIV infections in the
image and risk behaviors among YLWH. Thus, the current United States, further exploration among YLWH is needed.
study adds to the literature by providing data for clinicians and Further studies are also needed to evaluate body image and
researchers about the potential relatedness of body image HIV related health behaviors in a larger representative sam-
perceptions and health behaviors in this population. ple, for the development of multi-behavior integrated inter-
Results from a recent meta-analysis found an association ventions. In addition, results from this study highlight the
between high scores of body dissatisfaction and low scores of importance of assessing and intervening on body image
condom use self-efficacy.12 While the present study did not perceptions as opposed to body size or BMI focused inter-
examine condom use self-efficacy, our results did find an ventions as a potential mechanism to impact HIV self-care
association between using a condom with the most recent behaviors. To our knowledge there are no known interven-
sexual partner and increased satisfaction with overall and tions that address body image perceptions among YLWH;
specific areas of the body. It is possible that youth with in- thus, further exploration is warranted and urgently needed.
creased body satisfaction are more likely to engage in be-
haviors to protect their body, such as using a condom.25 As Acknowledgments
such, it will be important to further investigate the relation-
ship between condom use and body image perception, as a The authors acknowledge the contribution of clinical re-
potential component to designing effective interventions to search assistant, Melissa Shenep, and psychology graduate
prevent secondary transmission of HIV. assistant Courtney Peasant, during the data collection phase of
It is interesting to note that participants who indicated that this project. The authors also thank Kirk Knapp for his assis-
they believed their ARV medications had physically changed tance with data abstraction from electronic patient records.
their body were more likely to disclose their HIV diagnosis to
a romantic or sexual partner. This result is in contrast to Author Disclosure Statement
previous findings among adults that did not find a relationship
between diagnosis disclosure and perceived body changes as No competing financial interests exist.
a result of ARV therapy.26 One potential reason for our
findings could be that perceived physical changes in con- References
junction with other HIV clinical progression (e.g., low CD4
1. Sharma A, Howard AA, Klein RS, Schoenbaum EE, Buono
count) indicators have been found to predict HIV status
D, Webber MP. Body image in older men with or at-risk for
disclosure.27 Still, the specific relationship between disclo-
HIV infection. AIDS Care 2007;19:235–241.
sure of HIV status to one’s partner and the belief that ARV 2. Blashill AJ, Goshe BM, Robbins GK, Mayer KH, Safren
medication has caused physical changes in the body is un- SA. Body image disturbance and health behaviors among
clear. Future research is needed to further investigate and sexual minority men living with HIV. Health Psychol 2014;
clarify this relationship. 33:677–680.
Limitations of this study include its cross-sectional design, 3. Wilton L. A preliminary study of body image and HIV sexual
which constricted our ability to make causal inferences. risk behavior in black gay and bisexual men: Implications for
However, these results provide data for the development of HIV prevention. J Gay Lesbian Soc 2009;21:309–325.
future evaluations to further explore body image among 4. DeLaMora P, Aledort N, Stavola J. Caring for adolescents
YLWH. In addition, results from this study have the potential with HIV. Curr HIV/AIDS Rep 2006;3:74–78.
to inform clinical practice by highlighting possible avenues 5. Murphy DA, Wilson CM, Durako SJ, Muenz LR, Belzer M,
of discussion and intervention with patients when address- Adolescent Medicine HIVARN. Antiretroviral medication
ing health behaviors related to HIV transmission risk re- adherence among the REACH HIV-infected adolescent
duction. This study was also conducted at a single clinic site, cohort in the USA. AIDS Care 2001;13:27–40.
which may limit its ability to generalize to other geographic 6. Hughes EK, Gullone E. Emotion regulation moderates re-
regions of the United States. In addition, the demographic lationships between body image concerns and psychologi-
composition of our sample also limits the generalizability cal symptomatology. Body Image 2011;8:224–231.
of the findings. Future research is needed to longitudinally 7. Lehrer JA, Shrier LA, Gortmaker S, Buka S. Depressive
explore patterns of body image perception among a di- symptoms as a longitudinal predictor of sexual risk be-
verse sample of YLWH in relation to health behavior in a haviors among US middle and high school students. Pe-
diatrics 2006;118:189–200.
broader sample.
8. Hosek SG, Harper GW, Domanico R. Predictors of medi-
The results from the present study have the potential to
cation adherence among HIV-infected youth. Psychol Health
inform clinical practice and warrant further study. By ex- Med 2005;10:166–179.
ploring the relationship between body image perception and 9. Hall HI, Walker F, Shah D, Belle E. Trends in HIV diag-
HIV transmission risk behaviors, results from this study noses and testing among U.S. adolescents and young adults.
highlight the importance of addressing psychosocial factors AIDS Behav 2012;16:36–43.
such as body image with the intent of reducing risky sexual 10. Fisher JD, Smith LR, Lenz EM. Secondary prevention of
behaviors among YLWH. Expanding traditional behavioral HIV in the United States: Past, current, and future per-
risk reduction interventions to include aspects such as body spectives. J Acquir Immune Defic Syndr 2010;55 Suppl
image, and using novel techniques such as providing edu- 2:S106–S115.
cation and feedback about health status markers using mobile 11. Gillen MM, Markey CN. Body image and HIV risk among
apps, may strengthen their effectiveness at promoting sexual college students. Am J Health Behav 2014;38:816–822.
6 DALLAS ET AL.

12. Blashill AJ, Safren SA. Body dissatisfaction and condom of a short form of the CES-D (Center for Epidemiologic
use self-efficacy: A meta-analysis. Body Image 2015;12: Studies Depression Scale). Am J Prev Med 1994;10:
73–77. 77–84.
13. Diamond C, Buskin S. Continued risky behavior in HIV- 23. Carpenter JS, Andrykowski MA, Wilson J, et al. Psycho-
infected youth. Am J Public Health 2000;90:115–118. metrics for two short forms of the Center for Epidemiologic
14. Koenig LJ, Pals SL, Chandwani S, et al. Sexual transmis- Studies-Depression Scale. Issues Ment Health Nurs 1998;
sion risk behavior of adolescents With HIV acquired peri- 19:481–494.
natally or through risky behaviors. J Acquir Immune Defic 24. Centers for Disease Control and Prevention. About BMI
Syndr 2010;55:380–390. for Adults and Teens. 2012. www.cdc.gov/healthyweight/
15. Camacho-Gonzalez AF, Wallins A, Toledo L, et al. Risk assessing/bmi/adult_bmi (Last accessed June 1, 2012).
factors for HIV transmission and barriers to HIV disclo- 25. Wood-Barcalow NL, Tylka TL, Augustus-Horvath CL.
sure: Metropolitan Atlanta Youth Perspectives. AIDS Pa- ‘‘But I Like My Body’’: Positive body image characteris-
tient Care STDS 2016;30:18–24. tics and a holistic model for young-adult women. Body
16. Wilkins ML, Dallas RH, Porter JS, et al. Characterizing Image 2010;7:106–116.
body image in youth with HIV. AIDS Behav 2016;20: 26. Santos CP, Felipe YX, Braga PE, Ramos D, Lima RO,
1585–1590. Segurado AC. Self-perception of body changes in persons
17. Pluhar E, McDonnell Holstad M, Yeager KA, et al. Im- living with HIV/AIDS: Prevalence and associated factors.
plementation of audio computer-assisted interviewing soft- AIDS (London, England) 2005;19 Suppl 4:S14–S21.
ware in HIV/AIDS research. J Assoc Nurses AIDS Care: 27. Chaudoir SR, Fisher JD, Simoni JM. Understanding HIV
JANAC 2007;18:51–63. disclosure: A review and application of the disclosure
18. Cash TF, Pruzinsky T. Body Images: Development, De- processes model. Soc Sci Med 2011;72:1618–1629.
viance, and Change. New York: The Guilford Press; 1990. 28. Saberi P, Siedle-Khan R, Sheon N, Lightfoot M. The use of
19. Stunkard AJ, Sorensen T, Schulsinger F. Use of the Danish mobile health applications among youth and young adults
adoption register for the study of obesity and thinness. In: living with HIV: focus group findings. AIDS Patient Care
Kety S, ed. The Genetics of Neurological and Psychiatric STDS 2016;30:254–260.
Disorders. New York: Raven; 1983:115–120.
20. Pulvers KM, Lee RE, Kaur H, et al. Development of a Address correspondence to:
culturally relevant body image instrument among urban Ronald H. Dallas, PhD
African Americans. Obes Res 2004;12:1641–1651. Department of Infectious Diseases
21. Centers for Disease Control and Prevention. Youth Risk St. Jude Children’s Research Hospital
Behavior Survey. Questionnaire. 2013. Available at: www 262 Danny Thomas Place
.cdc.gov/YRBS (Last accessed June 1, 2012). Memphis, TN 38105
22. Andresen EM, Malmgren JA, Carter WB, Patrick DL.
Screening for depression in well older adults: Evaluation E-mail: ronald.dallas@stjude.org