Educational Research Paper proposal

Original Research

Gender and Ethnic Differences in Health-Promoting Behaviors of Rural Adolescents

Lynn Rew, EdD, RN, AHN-BC, FAAN 1 , Kristopher L. Arheart, EdD

2 ,

Sharon D. Horner, PhD, RN, FAAN 1 , Sanna Thompson, PHD, MSW

3 ,

and Karen E. Johnson, PhD, RN 1

Abstract

Although much is known about health-risk behaviors of adolescents, less is known about their health-promoting behaviors. The purpose of this analysis was to compare health-promoting behaviors in adolescents in Grades 9–12 by gender and ethnicity and explore how these behaviors changed over time. Data were collected from 878 rural adolescents (47.5% Hispanic; mean age at baseline 14.7 years). Males from all ethnic groups scored significantly higher than all females on phys- ical activity; non-Hispanic Black males and females scored significantly higher than other ethnic groups on safety behaviors. Hispanic and non-Hispanic White females scored higher than males in these ethnic groups on stress management. Nutrition, physical activity, and safety behaviors decreased significantly for most participants from Grade 9 to 12 whereas stress management remained relatively stable. Findings are similar to those from nationally representative samples that analyzed cross-sectional data and have implications for school nursing interventions to improve health-promoting behaviors in rural adolescents.

Keywords

health/wellness, high school, cultural issues, exercise, nutrition

Despite an expanding literature about the factors that relate

to and predict adolescent health-risk behaviors, less is

reported in the literature about health-promoting behaviors

in adolescents—particularly among those living in rural areas.

Health-promoting behaviors emphasize lifestyle choices that

improve physical health and well-being (Steinberg, 2014).

Health-promoting behaviors such as safety, stress manage-

ment (Groft, Hagen, Miller, Cooper, & Brown, 2005), physi-

cal activity (Kalak et al., 2012), and adequate nutrition

(Williams & Mummery, 2012) contribute to positive rather

than adverse health outcomes. Research that focuses on the

presence of health-promoting behaviors, as opposed to the

risk-focused perspective, is essential to understand how to

help young people make the successful transition from child

to adult. As part of an interdisciplinary education team com-

mitted to student success, school nurses are in ideal settings to

collaborate with others (e.g., health education teachers, food

service staff, and school health advisory councils) to deliver

interventions that enhance adolescent health.

Purpose

This article is a report of findings from a large longitudinal

study of health-risk and health-promoting behaviors among

rural adolescents as they progressed through high school (in

Grades 9–12). The specific aims of this analysis, which was

a component of the larger study, were to (1) compare the

health-promoting behaviors of adolescents by gender and

ethnicity and (2) explore how health-promoting behaviors

of these adolescents changed during the high school years.

Findings about health-risk behaviors have been published

previously (Horner, Rew, & Brown, 2012).

We sought to answer two research questions: (1) what are

the gender and ethnic differences in health-promoting beha-

viors among Hispanic, non-Hispanic Black (NHB), and non-

Hispanic White (NHW) adolescents residing in three rural

communities and (2) do health-promoting behaviors change

1 The University of Texas at Austin School of Nursing, Austin, TX, USA

2 Department of Epidemiology and Public Health, The University of Miami,

Miami, FL, USA 3

The University of Texas at Austin School of Social Work, Austin, TX, USA

Corresponding Author:

Lynn Rew, EdD, RN, AHN-BC, FAAN, The University of Texas at Austin,

School of Nursing, 1710 Red River, Austin, TX 78701, USA.

Email: [email protected]

The Journal of School Nursing 2015, Vol. 31(3) 219-232 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1059840514541855 jsn.sagepub.com

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

as adolescents matriculate through Grades 9–12? Our

hypotheses were as follows

Hypothesis 1: Of all groups examined (gender, grade,

and race/ethnicity), racial and ethnic minority males will

exhibit the fewest health-promoting behaviors.

Hypothesis 2: Adolescents’ health-promoting behaviors

will decrease each year from Grade 9 through Grade 12.

Background

Adolescent Health Behavior

During childhood, patterns of behavior are initiated that con-

tribute to the individual’s health and well-being throughout

the life span. Health behaviors are reflected in a continuum

from those that promote and enhance optimum development

and well-being (i.e., health-promoting behaviors) to those

that threaten development and well-being (i.e., health-risk

behaviors). Engaging in health-promoting behaviors, such

as eating nutritional snacks, engaging in physical activity,

and managing stress effectively, has been shown to protect

adolescents from adverse health outcomes (Iannotti & Wang,

2013; Peterhans, Worth, & Woll, 2013) and has been asso-

ciated with better health outcomes in later adulthood than

engaging in health-risk behaviors during adolescence (Dorn,

Beal, Kalkwarf, Pabst, Noll, & Susman, 2013; Olshansky

et al., 2005). Engaging in health-promoting behaviors contri-

butes to development of a healthy lifestyle (Kelder et al.,

2003). For example, running for 30 min daily was shown to

improve sleep quality and psychological functioning in

healthy adolescents (Kalak et al., 2012).

Previous studies have shown gender and age differences

in health-promoting behaviors. For example, Williams and

Mummery (2012) found that compared with males, females

were more likely to exhibit healthy eating patterns. In con-

trast, cross-sectional findings from the 2011 Youth Risk

Behavior Survey (YRBS) suggest males were more likely

than females to exhibit various health-promoting behaviors,

including eating three or more daily servings of fruit, three

or more servings of vegetables, three or more servings of

milk, and participating in 60 min of physical activity every-

day for a week prior to completing the survey (Centers for

Disease Control and Prevention [CDC], 2012, p. 29). Similar

to Williams and Mummery’s findings (2012), ninth graders

reported a higher prevalence than 12th graders of these same

nutritional and physical activity behaviors (CDC, 2012).

Although we have ample evidence of risk and protective

factors related to health-risk behaviors in adolescents (Gott-

fredson & Hussong, 2011; Leeman, Hoff, Krishnan-Sarin,

Patock-Peckham, & Potenza, 2014; Taliaferro, Muehlen-

kamp, Borowsky, McMorris, & Kugler, 2012; Thompson,

Dewa, & Phare, 2012 ), we have much less evidence of those

factors related to health-promoting behaviors. Previous stud-

ies of health-promoting behaviors in adolescents have been

primarily cross-sectional and involved small samples that

were mostly White (Mahon, Yarcheski, Yarcheski, & Hanks,

2007; Yarcheski, Mahon, & Yarcheski, 1997). The biannual

reports from the YRBS, such as the CDC (2012) report men-

tioned previously, describe a nationally representative sample

of adolescents, but they too are cross-sectional. Moreover,

these studies focus on trends in the population as opposed

to changes over time within a particular population such as

those living in rural areas.

In this study, we began with a public health approach

grounded in the premise that ‘‘health is a product of lifestyle

shaped heavily by social and physical environments’’

(Crosby, Kegler, & DiClemente, 2009, p. 4). Basic social

and physical attributes such as ethnicity, sex, socioeconomic

status (SES), parent’s level of education, and marital status

have a profound effect on learned behaviors, including those

that are health related. Identifying these attributes and their

effects on health-promoting behaviors may influence the

development of interventions that can be tailored to adoles-

cents with diverse personal and cultural characteristics.

Similarly, identifying if and when these health-promoting

behaviors change over time may influence the development

and testing of interventions targeted at specific developmen-

tal stages of adolescence when adult lifestyles are being

shaped. These interventions are needed to ensure a healthy

generation of adults.

Importantly, the health behavior of adolescents living in

rural communities is studied less often than those of youth

in urban or suburban environments—particularly those

from racial/ethnic minority backgrounds (Curtis, Waters,

& Brindis, 2011). For example, a survey of health status

and clinic use among ninth graders in rural Mississippi

yielded a response rate of only 27.6% for a mostly White school and a 2.6% response rate for a school that was pre- dominantly African American (Bradford & O’Sullivan,

2007). Compared to urban and suburban areas, rural areas

rank low in most population health indicators including

health behaviors and maternal and child health (Hartley,

2004). As an example of this, Nanney, Davey, and Kubik

(2013) evaluated policies and practices of secondary

schools in 28 states and found that schools in smaller towns

and rural areas did not have as many healthy eating policies

and practices as schools in urban/suburban areas.

Well over half of rural counties (65%) experience shortages in health care providers and access to health ser-

vices, with this percentage being higher in rural counties

where people of color are the majority (Probst, Moor,

Glover, & Samuels, 2004). These disparities create a social-

environmental context for adolescent health and development

that is distinctly different from that of suburban or urban

contexts. Geographic isolation and lack of community

resources such as confidential health care services may

present barriers to rural adolescents having the supports

they need to engage in health-promoting behaviors (Curtis

et al., 2011). Curtis, Waters, and Brindis (2011) conducted

220 The Journal of School Nursing 31(3)

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

a secondary analysis of rural adolescents ages 12 through

17 from the cross-sectional 2005 California Health Inter-

view Survey and found disparate levels of sexual activity,

substance use, depressive symptoms, and risk factors for

obesity (i.e., poor diet and low levels of physical activity).

Although nationally representative samples of adolescents

include those from rural areas, we found no published

longitudinal studies of how behaviors in rural areas may

be different from or the same as behaviors in urban and

suburban youth. This longitudinal analysis examining

health-promoting behaviors among rural adolescents can

extend Curtis et al.’s cross-sectional findings and help to

fill gaps in the literature regarding rural adolescent health.

Method

Design and Setting

A cohort-sequential longitudinal design was used to explore

changes in the development of health behaviors in adoles-

cents residing in rural communities in central Texas. Four

cohorts of students were initially recruited over a period of

2 years by sending letters to parents of children who were

then in Grade 4 through Grade 6 (Rew, Horner, & Brown,

2011). These participants were followed through Grade 8,

and subsequently recruited again for a second longitudinal

study when they were in Grade 9 (Rew, Arheart, Thompson,

& Johnson, 2013). Sixty-seven percent of the original sam-

ple were retained for this study.

Protection of Human Participants

The study was reviewed annually by the institutional review

board at the first author’s university. Both written parental

consent and adolescent assent were collected each year of

the study for all participants. When adolescents reached

18 years of age, they provided their own consent.

Sample

The sample for this analysis was drawn from a total of 1,294

adolescents who were recruited for a longitudinal study

when they entered high school in Grade 9 and consisted of

878 adolescents who were retained at the final data collec-

tion point when they were in Grade 12 (68% retention over 4 years). This retention rate reflects an average loss of

approximately 14% of the sample each succeeding year of the study. Participants were an average of 14.7 years old

in Grade 9 and 17.18 years old in Grade 12. The sample con-

sisted of four Cohorts (i.e., A, B, C, and D) that reflect the

grade the participant was in at the beginning of the previous

longitudinal study to which this was a 4-year follow-up. For

example, Cohort A would have been in Grade 6 during the

first year of the previous study, Cohort B would have been

in Grade 5, and so on.

Measures

Two measures were used for this analysis: a demographic

form and the Adolescent Lifestyle Questionnaire (ALQ). The

demographic form was developed by the principal investiga-

tor of the study and consisted of age, sex, race, and ethnicity.

Health-promoting behaviors were measured using four

subscales from the ALQ: nutrition, physical activity, safety,

and stress management (Gillis, 1997). Three other subscales

of the ALQ, identity awareness, health awareness, and social

support, were not included in the present analysis because

they are conceptually different from health-promoting beha-

viors. The ALQ consists of 43 six-point Likert-type items

(6 ¼ always and 1 ¼ never); high scores mean greater engagement in health-promoting behaviors; we used only

the 23 items that comprised the four subscales used in this

analysis. Examples of items are ‘‘I usually make informed

choices about sexual relationships’’ (safety); ‘‘I participate

in a regular program of sports/exercise at school’’: (physical

activity); ‘‘I read labels on packaged foods I eat’’ (nutrition);

and ‘‘I usually use helpful strategies to help me deal with

stress’’ (stress management; Gillis, 1997, pp. 38–39).

Procedures

Following approval from the institutional review board, par-

ents signed informed consent forms and adolescents under

age 18 signed informed assent forms annually; adolescents

18 years of age and older signed their own consents. In the

first 2 years of the study, data were gathered through home

visits using computer-assisted self-interviewing (CASI) or

via a secure website that the adolescent could access from

home. In the final 2 years, data were gathered by mailed sur-

vey owing to increased difficulty in making appointments

for home visits. Scheduling difficulties arose, as adolescents

became older and involved in more after-school and evening

activities.

Data Analysis

Descriptive statistics (mean + standard error or percentage) were used to describe the demographic data by year and

cohort within year. Descriptive statistics (M + SD) and Cronbach’s a reliability coefficient were computed for each health-promoting behavior for each year and cohort within

year. To address Research Question 1—what are the gender

and ethnic differences in health-promoting behaviors among

Hispanic, non-Hispanic Black, and non-Hispanic White

adolescents residing in three rural communities—and

Hypothesis 1—of all groups examined (gender, grade, and

race/ethnicity), racial and ethnic minority males will exhi-

bit the fewest health-promoting behaviors, we used sepa-

rate general linear mixed models for each year and

health-promoting behavior. The fixed effects of interest

included in the models were gender, race/ethnicity, and the

interaction of gender and race/ethnicity. Fixed covariates

for age, two-parent household (yes/no), subsidized school

Rew et al. 221

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

lunch (yes/no), and mother’s highest education were included

to control for their possible confounding effects. A random

effect was included for cohort. Planned comparisons were

made for gender differences within each race/ethnic category

and between race/ethnic categories for each gender.

To address Research Question 2—do health-promoting

behaviors change as adolescents matriculate through Grade

9 through Grade 12 and Hypothesis 2—adolescents’ health-

promoting behaviors will decrease each year from Grade 9

through Grade 12, we used a general linear mixed model for

each health-promoting behavior to perform a linear growth

curve analysis. The model included random terms for the

intercepts and slopes (trajectories) for each gender-race/

ethnicity combination. A random term was included for per-

son nested within cohort. Statistical Analysis Software (SAS)

9.3 (SAS Institute, Inc., 2013) was used for all analyses. Sta-

tistical tests resulting in a probability level less than .05 were

considered to be statistically significant.

Results

Demographic data by year and cohort within year are sum-

marized in Table 1. Table 2 is a summary of the means,

standard errors, and Cronbach’s a for each of the four health-promoting behaviors (nutrition, physical activity,

safety, and stress management) measured for each year and

Table 1. Demographic Characteristics in Sample of Rural Adolescents in Grade 9 Through Grade 12.

Grade Cohort n

Age, M + SE

Female (%)

Hisp (%)

NHB (%)

NHW (%)

Both Parents (%)

Subsidized Lunch (%)

<HS a

(%) HS

a

(%)

Some College

a

(%) BS

a

(%) >BS

a

(%)

Grade 9 814 14.7 + 0.02 56 47 13 40 59 57 15 34 31 12 8 A 205 15.0 + 0.05 58 47 13 40 54 59 13 38 32 7 10 B 177 14.9 + 0.03 58 44 9 47 66 59 13 36 29 11 11 C 254 14.5 + 0.03 56 48 15 37 56 52 17 35 31 12 5 D 176 14.4 + 0.04 54 48 16 36 63 58 16 28 31 15 10

Grade 10 825 15.5 + 0.02 57 48 13 39 60 57 16 32 31 12 9 A 119 16.2 + 0.04 68 50 12 38 58 59 14 37 30 6 13 B 294 15.5 + 0.03 55 47 11 12 62 57 14 32 32 12 10 C 218 15.3 + 0.03 54 46 16 38 56 55 19 33 31 12 5 D 194 15.2 + 0.04 55 49 14 37 64 58 16 29 30 16 9

Grade 11 818 16.3 + 0.02 58 45 14 41 60 59 15 31 33 12 9 A 213 16.7 + 0.04 62 43 15 42 55 62 13 32 36 7 12 B 210 16.3 + 0.04 61 40 13 47 61 62 12 30 33 13 12 C 214 16.1 + 0.04 55 47 15 38 59 54 17 30 34 13 6 D 181 16.1 + 0.04 55 51 20 19 66 59 18 31 28 15 8

Grade 12 707 17.2 + 0.02 61 48 12 40 62 60 14 32 32 12 10 A 170 17.5 + 0.0 67 47 14 39 58 61 13 35 32 7 13 B 211 17.1 + 0.03 67 43 10 47 62 64 11 30 33 14 12 C 193 17.0 + 0.04 59 49 14 37 61 55 18 33 31 12 6 D 133 17.0 + 0.04 53 53 12 35 69 61 14 29 30 19 8

Note. M ¼ mean; SE ¼ standard error; Hisp ¼ Hispanic; NHB ¼ non-Hispanic Black; NHW ¼ non-Hispanic White; HS ¼ high school; BS ¼ bachelor’s degree. a Mother’s highest level of eduction.

Table 2. Overall Means, Standard Errors, and Cronbach’s a Coefficients for Health-Promoting Behaviors in Adolescents.

Grade n

Nutrition Physical Activity Safety Stress Management

M + SE (a) M + SE (a) M + SE (a) M + SE (a)

Grade 9 814 25.1 + 0.3 (0.89) 15.6 + 0.2 (0.89) 36.7 + 0.2 (0.76) 14.3 + 0.2 (0.65) Range by cohort 176–254 24.1 + 0.6 - 25.9 + 0.7 15.4 + 0.4–16.2 + 0.5 36.2 + 0.4–37.1 + 0.4 14.0 + 0.2–14.5 + 0.4 Grade 10 825 25.0 + 0.3 (0.89) 15.2 + 0.2 (0.90) 36.0 + 0.2 (0.80) 14.3 + 0.2 (0.66) Range by cohort 119–294 24.3 + 0.8–25.3 + 0.6 14.2 + 0.6–15.8 + 0.4 35.2 + 0.7–36.3 + 0.4 14.0 + 0.3–14.5 + 0.3 Grade 11 818 25.1 + 0.3 (0.90) 14.5 + 0.2 (0.90) 36.3 + 0.2 (0.79) 14.3 + 0.2 (0.62) Range by cohort 181–214 24.9 + 0.6–25.1 + 0.6 14.0 + 0.5–15.1 + 0.5 35.5 + 0.4–36.7 + 0.4 14.1 + 0.3–14.9 + 0.3 Grade 12 707 24.8 + 0.3 (0.91) 13.6 + 0.2 (0.89) 35.9 + 0.2 (0.77) 14.2 + 0.2 (0.68) Range by cohort 133–211 23.4 + 0.7–25.7 + 0.8 12.6 + 0.5–14.3 + 0.6 34.9 + 0.5–36.7 + 0.4 14.1 + 0.4–14.5 + 0.4

Note. Nutrition: 8 items, score range (8–48); Physical Activity: 4 items, score range (4–24); Safety: 7 items, score range (7–42); Nutrition: 8 items, score range (8–48). M ¼ mean; SE ¼ standard error.

222 The Journal of School Nursing 31(3)

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

T a b

le 3 .

N u tr

it io

n H

e al

th B e h av

io r

b y

R ac

e /E

th n ic

it y

an d

G e n d e r

W it h

C o va

ri at

e s.

M al

e F e m

al e

C o va

ri at

e s

G ra

d e

R ac

e /E

th n ic

it y

M al

e (M

+ S E )

F e m

al e

(M +

S E )

M al

e ve

rs u s

F e m

al e

p H

is p

p N

H B

p H

is p

p N

H B

p F ix

e d

b +

S E

p M

o m

e d

M +

S E

p

9 H

is p

2 5 .2

0 +

0 .7

3 2 5 .3

0 +

0 .6

2 .9

0 6

— —

— —

A ge

� 0 .5

4 +

0 .5

0 .2

8 3

< H

S 2 4 .7

9 +

0 .8

8 .8

4 0

N H

B 2 5 .0

5 +

1 .2

3 2 4 .6

8 +

1 .1

9 .8

1 8

.9 1 6

— .6

3 3

— P ar

e n t

1 .4

6 +

0 .6

4 .0

2 3

H S

2 4 .7

4 +

0 .5

5 N

H W

2 4 .2

0 +

0 .7

5 2 6 .5

5 +

0 .6

7 .0

1 4

.3 4 0

.5 5 5

.1 7 3

.1 6 3

L u n ch

� 0 .9

4 +

0 .7

1 .1

8 5

S o m

e C

2 4 .8

2 +

0 .5

7 —

— —

— B S

2 5 .7

9 +

0 .9

5 —

— —

— >

B S

2 5 .6

9 +

1 .0

9 1 0

H is

p 2 5 .0

7 +

0 .7

1 2 5 .2

2 +

0 .6

2 .8

6 6

— —

— —

A ge

� 0 .5

4 +

0 .4

8 .2

6 4

< H

S 2 4 .6

2 +

0 .8

6 .2

7 2

N H

B 2 4 .2

3 +

1 .2

9 2 4 .5

4 +

1 .1

5 .8

5 2

.5 5 6

— .5

9 6

— P ar

e n t

1 .4

1 +

0 .6

5 .0

2 9

H S

2 4 .5

6 +

0 .5

6 N

H W

2 3 .3

0 +

0 .7

5 2 7 .5

7 +

0 .6

7 <

.0 0 1

.0 8 7

.5 3 7

.0 1 0

.0 2 1

L u n ch

0 .2

6 +

0 .7

1 .7

1 8

S o m

e C

2 4 .2

8 +

0 .5

7 —

— —

— B S

2 6 .5

8 +

0 .9

2 —

— —

— >

B S

2 4 .8

9 +

1 .0

6 1 1

H is

p 2 4 .9

5 +

0 .7

3 2 5 .2

9 +

0 .6

1 .7

0 5

— —

— —

A ge

� 0 .2

6 +

0 .5

1 .6

0 9

< H

S 2 3 .2

7 +

0 .8

5 .4

3 0

N H

B 2 3 .8

4 +

1 .2

4 2 3 .8

5 +

1 .1

0 .9

9 3

.4 2 3

— .2

4 4

— P ar

e n t

1 .4

9 +

0 .6

3 .0

1 7

H S

2 4 .9

1 +

0 .5

7 N

H W

2 3 .1

7 +

0 .7

3 2 6 .8

3 +

0 .6

4 <

.0 0 1

.0 8 6

.6 4 6

.0 8 5

.0 1 8

L u n ch

0 .6

3 +

0 .7

0 .3

7 4

S o m

e C

2 4 .7

3 +

0 .5

4 —

— —

— B S

2 4 .8

0 +

0 .8

9 —

— —

— >

B S

2 5 .5

7 +

1 .0

2 1 2

H is

p 2 3 .8

6 +

0 .8

8 2 5 .5

1 +

0 .7

4 .0

8 1

— —

— —

A ge

� 0 .1

2 +

0 .6

1 .8

4 2

< H

S 2 3 .1

9 +

1 .0

2 .2

7 9

N H

B 2 1 .9

2 +

1 .5

0 2 2 .8

8 +

1 .2

9 .6

0 5

.2 3 0

— .0

5 5

— P ar

e n t

0 .9

6 +

0 .6

8 .1

6 0

H S

2 3 .2

0 +

0 .7

2 N

H W

2 3 .8

7 +

0 .9

0 2 7 .4

1 +

0 .7

8 .0

0 1

.9 8 7

.2 4 4

.0 4 4

.0 0 1

L u n ch

� 0 .7

5 +

0 .7

5 .3

2 1

S o m

e C

2 4 .4

6 +

0 .7

0 —

— —

— B S

2 5 .2

6 +

1 .0

4 —

— —

— >

B S

2 5 .1

0 +

1 .1

5

N o te

.G e n e ra

ll in

e ar

m ix

e d

m o d e lf

o r

fi x e d

ra ce

/e th

n ic

it y

an d

ge n d e r

e ff e ct

s ad

ju st

e d

fo r

fi x e d

co va

ri at

e e ff e ct

s an

d a

ra n d o m

co h o rt

e ff e ct

.B o ld

e d

va lu

e s

ar e

st at

is ti ca

lly si

gn if ic

an t.

H is

p ¼

H is

p an

ic ;M

o m

e d ¼

m o th

e r’

s h ig

h e st

le ve

l o f e d u ca

ti o n ; M ¼

m e an

; N

H B ¼

n o n -H

is p an

ic B la

ck ; N

H W ¼

n o n -H

is p an

ic W

h it e ; S E ¼

st an

d ar

d e rr

o r;

H S ¼

h ig

h sc

h o o l;

C ¼

co lle

ge ; B S ¼

b ac

h e lo

r’ s

d e gr

e e .

223

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

T a b

le 4 .

P h ys

ic al

A ct

iv it y

H e al

th B e h av

io r

b y

R ac

e /E

th n ic

it y

an d

G e n d e r

W it h

C o va

ri at

e s.

M al

e F e m

al e

C o va

ri at

e s

M al

e F e m

al e

M al

e ve

rs u s

H is

p N

H B

H is

p N

H B

G ra

d e

R ac

e /E

th n ic

it y

M +

S E

M +

S E

F e m

al e

p p

p p

p F ix

e d

b +

S E

p M

o m

e d

M +

S E

p

9 H

is p

1 6 .9

1 +

0 .5

4 1 4 .0

1 +

0 .4

6 <

.0 0 1

— —

— —

A ge

� 0 .6

7 +

0 .3

7 .0

7 2

< H

S 1 5 .1

9 +

0 .6

5 .4

2 2

N H

B 1 9 .2

2 +

0 .9

1 1 4 .5

4 +

0 .8

8 <

.0 0 1

.0 2 4

— .5

8 3

— P ar

e n t

0 .8

6 +

0 .4

7 .0

6 9

H S

1 6 .0

7 +

0 .4

1 N

H W

1 6 .6

4 +

0 .5

6 1 5 .3

3 +

0 .5

0 .0

6 5

.7 2 8

.0 1 6

.0 5 0

.4 2 7

L u n ch

0 .3

0 +

0 .5

3 .5

6 7

S o m

e C

1 6 .0

3 +

0 .4

2 —

— —

— B S

1 6 .0

2 +

0 .7

0 —

— —

— >

B S

1 7 .2

3 +

0 .8

1 1 0

H is

p 1 6 .3

3 +

0 .5

6 1 4 .2

8 +

0 .4

9 .0

0 2

— —

— —

A ge

� 0 .3

0 +

0 .3

8 .4

3 0

< H

S 1 3 .7

1 +

0 .6

7 .0

2 4

N H

B 1 7 .4

1 +

0 .9

9 1 4 .2

5 +

0 .8

9 .0

1 2

.3 1 9

— .9

7 5

— P ar

e n t

0 .4

3 +

0 .4

9 .3

8 1

H S

1 5 .1

4 +

0 .4

5 N

H W

1 6 .3

1 +

0 .5

9 1 4 .6

0 +

0 .5

3 .0

1 9

.9 7 9

.3 3 3

.6 5 1

.7 3 0

L u n ch

0 .8

5 +

0 .5

3 .1

1 1

S o m

e C

1 5 .7

3 +

0 .4

6 —

— —

— B S

1 6 .2

6 +

0 .7

2 —

— —

— >

B S

1 6 .8

1 +

0 .8

2 1 1

H is

p 1 5 .8

9 +

0 .5

7 1 3 .6

4 +

0 .4

8 .0

0 1

— —

— —

A ge

� 0 .1

8 +

0 .4

0 .6

4 4

< H

S 1 2 .3

5 +

0 .6

6 .0

0 1

N H

B 1 7 .8

9 +

0 .9

6 1 3 .5

5 +

0 .8

5 <

.0 0 1

.0 6 2

— .9

2 4

— P ar

e n t

1 .2

6 +

0 .4

8 .0

0 9

H S

1 4 .9

1 +

0 .4

5 N

H W

1 5 .2

1 +

0 .5

7 1 3 .0

3 +

0 .5

0 .0

0 2

.4 0 0

.0 1 6

.3 6 7

.5 8 5

L u n ch

0 .6

3 +

0 .5

4 .2

4 3

S o m

e C

1 5 .5

8 +

0 .4

3 —

— —

— B S

1 5 .4

9 +

0 .6

9 —

— —

— >

B S

1 6 .0

1 +

0 .7

9 1 2

H is

p 1 5 .0

7 +

0 .6

0 1 3 .0

0 +

0 .4

9 .0

0 4

— —

— —

A ge

0 .0

1 +

0 .4

4 .9

7 9

< H

S 1 2 .5

0 +

0 .7

2 .0

7 5

N H

B 1 7 .2

8 +

1 .0

9 1 3 .2

0 +

0 .9

3 .0

0 3

.0 6 5

— .8

4 4

— P ar

e n t

0 .9

6 +

0 .5

1 .0

6 0

H S

1 4 .1

9 +

0 .4

7 N

H W

1 4 .8

9 +

0 .6

2 1 1 .9

9 +

0 .5

2 <

.0 0 1

.8 4 1

.0 5 6

.1 4 8

.2 4 2

L u n ch

0 .3

4 +

0 .5

6 .5

4 9

S o m

e C

1 4 .2

9 +

0 .4

6 —

— —

— B S

1 5 .2

7 +

0 .7

3 —

— —

— >

B S

1 4 .9

6 +

0 .8

1

N o te

.G e n e ra

ll in

e ar

m ix

e d

m o d e lf

o r

fi x e d

ra ce

/e th

n ic

it y

an d

ge n d e r

e ff e ct

s ad

ju st

e d

fo r

fi x e d

T ab

le 5 .B

o ld

e d

va lu

e s

ar e

st at

is ti ca

lly si

gn if ic

an t.

H is

p ¼

H is

p an

ic ;M

o m

e d ¼

m o th

e r’

s h ig

h e st

le ve

lo f e d u ca

ti o n ;M ¼

m e an

; N

H B ¼

n o n -H

is p an

ic B la

ck ; N

H W ¼

n o n -H

is p an

ic W

h it e ; S E ¼

st an

d ar

d e rr

o r;

H S ¼

h ig

h sc

h o o l;

C ¼

co lle

ge ; B S ¼

b ac

h e lo

r’ s

d e gr

e e .

224

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

T a b

le 5 .

S af

e ty

H e al

th B e h av

io r

b y

R ac

e /E

th n ic

it y

an d

G e n d e r

W it h

C o va

ri at

e s.

M al

e F e m

al e

C o va

ri at

e s

M al

e F e m

al e

M al

e ve

rs u s

H is

p N

H B

H is

p N

H B

G ra

d e

R ac

e /E

th n ic

it y

M +

S E

M +

S E

F e m

al e

p p

p p

p F ix

e d

b +

S E

p M

o m

e d

M +

S E

p

9 H

is p

3 5 .5

1 +

0 .5

1 3 7 .1

9 +

0 .4

4 .0

0 7

— —

— —

A ge

� 0 .2

0 +

0 .3

5 0 .5

6 9

< H

S 3 5 .1

5 +

.6 1

.0 0 2

N H

B 3 8 .0

4 +

0 .8

6 3 8 .3

6 +

0 .8

3 .7

8 1

.0 0 9

— .2

0 3

— P ar

e n t

1 .3

7 +

0 .4

4 0 .0

0 2

H S

3 7 .2

4 +

.3 8

N H

W 3 6 .1

5 +

0 .5

2 3 7 .3

0 +

0 .4

7 .0

8 4

.3 8 5

.0 6 1

.8 5 4

.2 6 1

L u n ch

0 .1

1 +

0 .5

0 0 .8

1 7

S o m

e C

3 6 .5

9 +

.4 0

— —

— —

B S

3 8 .4

1 +

.6 6

— —

— —

> B S

3 8 .0

7 +

.7 6

1 0

H is

p 3 4 .9

3 +

0 .5

5 3 6 .5

5 +

0 .4

7 .0

1 7

— —

— —

A ge

� 0 .4

3 +

0 .3

7 0 .2

4 3

< H

S 3 4 .8

4 +

.6 6

.0 6 7

N H

B 3 5 .6

7 +

1 .0

0 3 8 .0

3 +

0 .8

9 .0

6 4

.5 0 6

— .1

3 7

— P ar

e n t

0 .8

5 +

0 .5

0 0 .0

8 7

H S

3 5 .7

5 +

0 .4

3 N

H W

3 5 .9

3 +

0 .5

7 3 6 .6

3 +

0 .5

1 .3

4 4

.2 0 7

.8 1 6

.9 0 7

.1 6 9

L u n ch

0 .2

1 +

0 .5

4 0 .7

0 3

S o m

e C

3 6 .2

6 +

.4 4

— —

— —

B S

3 7 .5

5 +

.7 1

— —

— —

> B S

3 7 .0

5 +

.8 1

1 1

H is

p 3 5 .0

2 +

0 .5

3 3 7 .0

3 +

0 .4

5 .0

0 2

— —

— —

A ge

� 0 .7

4 +

0 .3

7 0 .0

4 6

< H

S 3 5 .3

6 +

.6 2

.0 5 6

N H

B 3 9 .1

7 +

0 .9

1 3 7 .6

0 +

0 .8

1 .1

8 1

< .0

0 1

— .5

3 4

— P ar

e n t

0 .8

2 +

0 .4

6 0 .0

7 3

H S

3 6 .5

7 +

.4 2

N H

W 3 5 .2

3 +

0 .5

3 3 6 .6

9 +

0 .4

7 .0

3 1

.7 7 9

< .0

0 1

.6 0 4

.3 2 7

L u n ch

� 0 .1

2 +

0 .5

2 0 .8

2 2

S o m

e C

3 6 .5

9 +

.4 0

— —

— —

B S

3 7 .1

3 +

.6 5

— —

— —

> B S

3 8 .3

0 +

.7 5

1 2

H is

p 3 4 .1

0 +

0 .6

4 3 6 .4

7 +

0 .5

4 .0

0 1

— —

— —

A ge

� 0 .4

5 +

0 .4

4 0 .3

0 6

< H

S 3 4 .2

0 +

.7 5

.0 2 2

N H

B 3 6 .8

1 +

1 .0

9 3 6 .8

0 +

0 .9

4 .9

9 2

.0 2 0

— .7

4 2

— P ar

e n t

0 .9

2 +

0 .4

9 0 .0

6 4

H S

3 6 .0

0 +

.5 3

N H

W 3 4 .9

1 +

0 .6

6 3 6 .8

0 +

0 .5

8 .0

1 1

.3 3 1

.1 1 6

.6 2 8

.9 9 8

L u n ch

� 0 .6

1 +

0 .5

4 0 .2

6 1

S o m

e C

3 5 .7

9 +

.5 2

— —

— —

B S

3 7 .4

2 +

.7 6

— —

— —

> B S

3 6 .5

0 +

.8 3

N o te

.G e n e ra

ll in

e ar

m ix

e d

m o d e lf

o r

fi x e d

ra ce

/e th

n ic

it y

an d

ge n d e r

e ff e ct

s ad

ju st

e d

fo r

fi x e d

co va

ri at

e e ff e ct

s an

d a

ra n d o m

co h o rt

e ff e ct

.B o ld

e d

va lu

e s

ar e

st at

is ti ca

lly si

gn if ic

an t.

H is

p ¼

H is

p an

ic ;M

o m

e d ¼

m o th

e r’

s h ig

h e st

le ve

l o f e d u ca

ti o n ; M ¼

m e an

; N

H B ¼

n o n -H

is p an

ic B la

ck ; N

H W ¼

n o n -H

is p an

ic W

h it e ; S E ¼

st an

d ar

d e rr

o r;

H S ¼

h ig

h sc

h o o l;

C ¼

co lle

ge ; B S ¼

b ac

h e lo

r’ s

d e gr

e e .

225

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

T a b

le 6 .

S tr

e ss

M an

ag e m

e n t

H e al

th B e h av

io r

b y

R ac

e /E

th n ic

it y

an d

G e n d e r

W it h

C o va

ri at

e s.

M al

e F e m

al e

C o va

ri at

e s

M al

e F e m

al e

M al

e ve

rs u s

H is

p N

H B

H is

p N

H B

G ra

d e

R ac

e /E

th n ic

it y

M +

S E

M +

S E

F e m

al e

p p

p p

p F ix

e d

b +

S E

p M

o m

e d

M +

S E

p

9 H

is p

1 3 .6

0 +

0 .4

0 1 4 .5

6 +

0 .3

4 .0

4 8

— —

— —

A ge

0 .0

3 +

0 .2

7 .9

0 0

< H

S 1 4 .3

4 +

0 .4

8 .6

6 9

N H

B 1 4 .8

2 +

0 .6

8 1 4 .9

5 +

0 .6

5 .8

8 .1

0 7

— .5

9 0

— P ar

e n t

0 .8

0 +

0 .3

5 .0

2 3

H S

1 4 .6

6 +

0 .3

0 N

H W

1 2 .9

5 +

0 .4

1 1 5 .0

4 +

0 .3

7 <

.0 0 1

.2 5 7

.0 1 8

.3 3 6

.9 0 0

L u n ch

0 .1

8 +

0 .3

9 .6

3 7

S o m

e C

1 4 .2

4 +

0 .3

1 —

— —

— B S

1 3 .8

3 +

0 .5

2 —

— —

— >

B S

1 4 .5

3 +

0 .6

0 1 0

H is

p 1 3 .9

5 +

0 .3

9 1 4 .9

9 +

0 .3

4 .0

2 9

— —

— —

A ge

0 .0

4 +

0 .2

6 .8

7 9

< H

S 1 3 .8

2 +

0 .4

7 .4

6 1

N H

B 1 4 .4

1 +

0 .7

0 1 5 .4

5 +

0 .6

3 .2

4 8

.5 5 4

— .5

1 1

— P ar

e n t

0 .1

7 +

0 .3

5 .6

2 7

H S

1 4 .1

9 +

0 .3

0 N

H W

1 2 .8

0 +

0 .4

1 1 4 .9

1 +

0 .3

6 <

.0 0 1

.0 4 1

.0 4 8

.8 7 9

.4 5 4

L u n ch

0 .2

5 +

0 .3

8 .5

2 4

S o m

e C

1 4 .3

6 +

0 .3

1 —

— —

— B S

1 4 .5

2 +

0 .5

0 —

— —

— >

B S

1 5 .2

0 +

0 .5

7 1 1

H is

p 1 3 .8

9 +

0 .4

0 1 5 .0

1 +

0 .3

4 .0

1 6

— —

— —

A ge

0 .0

8 +

0 .2

7 .7

8 1

< H

S 1 3 .0

6 +

0 .4

6 .0

1 1

N H

B 1 5 .5

8 +

0 .6

6 1 3 .9

7 +

0 .5

8 .0

5 3

.0 2 0

— .1

0 7

— P ar

e n t

0 .3

6 +

0 .3

2 .2

6 5

H S

1 4 .7

7 +

0 .3

2 N

H W

1 2 .7

4 +

0 .4

0 1 4 .4

5 +

0 .3

6 <

.0 0 1

.0 3 3

< .0

0 1

.2 3 1

.4 5 8

L u n ch

0 .5

5 +

0 .3

6 .1

3 3

S o m

e C

1 4 .5

6 +

0 .3

1 —

— —

— B S

1 4 .2

3 +

0 .4

8 —

— —

— >

B S

1 4 .7

3 +

0 .5

5 1 2

H is

p 1 3 .1

2 +

0 .4

3 1 4 .6

7 +

0 .3

4 .0

0 2

— —

— —

A ge

0 .1

6 +

0 .3

1 .6

2 1

< H

S 1 3 .1

6 +

0 .5

1 .0

4 1

N H

B 1 5 .0

7 +

0 .7

8 1 5 .5

2 +

0 .6

6 .6

4 5

.0 2 4

. .2

4 6

— P ar

e n t

0 .6

4 +

0 .3

7 .0

8 0

H S

1 4 .8

3 +

0 .3

3 N

H W

1 2 .1

2 +

0 .4

4 1 4 .8

0 +

0 .3

7 <

.0 0 1

.1 0 3

.0 0 1

.8 0 0

.3 3 2

L u n ch

0 .0

7 +

0 .4

0 .8

6 8

S o m

e C

1 4 .5

0 +

0 .3

2 —

— —

— B S

1 4 .0

1 +

0 .5

2 —

— —

— >

B S

1 4 .5

7 +

0 .5

8

N o te

.G e n e ra

ll in

e ar

m ix

e d

m o d e lf

o r

fi x e d

ra ce

/e th

n ic

it y

an d

ge n d e r

e ff e ct

s ad

ju st

e d

fo r

fi x e d

co va

ri at

e e ff e ct

s an

d a

ra n d o m

co h o rt

e ff e ct

.B o ld

e d

va lu

e s

ar e

st at

is ti ca

lly si

gn if ic

an t.

H is

p ¼

H is

p an

ic ;M

o m

e d ¼

m o th

e r’

s h ig

h e st

le ve

l o f e d u ca

ti o n ; M ¼

m e an

; N

H B ¼

n o n -H

is p an

ic B la

ck ; N

H W ¼

n o n -H

is p an

ic W

h it e ; S E ¼

st an

d ar

d e rr

o r;

H S ¼

h ig

h sc

h o o l;

C ¼

co lle

ge ; B S ¼

b ac

h e lo

r’ s

d e gr

e e .

226

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

cohort within each year. Results to answer the first research

question and test the first hypothesis are presented in

Tables 3–6. Results to answer the second research question

and test the second hypothesis are in Table 7. The number

of participants in each table varies because those who did

not provide complete data on a particular variable were not

included in that analysis.

Demographic Attributes

In addition to age, sex, and ethnicity, Table 1 also shows the

percentage of participants who lived with both parents, the

percentage who received subsidized lunch (low SES), and

the highest grade level of the participants’ mothers.

Annual Means of Health-Promoting Behaviors

Table 2 shows the means and standard errors for each type of

health-promoting behavior measured in each of the 4 years

of the study and the ranges for each of the four cohorts. See

our previous report on that study for a fuller description of

the cohorts (Rew et al., 2011). Table 2 also shows the relia-

bility coefficients (a) for each of the health-promoting

behavior subscales. These ranges for cohorts were .88–.91

for nutrition, .88–.92 for physical activity, .73–.86 for safety,

and .59–.71 for stress management.

Differences in Nutrition-Related Behaviors

There were no statistically significant gender differences in

nutrition-related health-promoting behaviors for Hispanic and

NHB participants in any of the four grades; however, NHW

females engaged in significantly more healthy eating beha-

viors such as avoiding foods high in fat and salt than NHW

males in each of the four grades. Although NHW females

exhibited a higher frequency of these nutrition-related beha-

viors than Hispanic and NHB females in all four grades, the

differences were statistically significantly greater than Hispa-

nics in Grade 10 only and greater than NHB females in

Grades 10, 11, and 12.

Differences in Physical Activity Behaviors

There were statistically significant sex differences in physi-

cal activity behaviors such as participating in sports or exer-

cising regularly for all ethnic groups in each of the four

Table 7. Trajectories of Health Behaviors Over Time by Race/Ethnicity and Gender.

Intercept Trajectory (Slope)

Race/Ethnicity–Gender b + SE p b + SE p

Nutrition Hispanic males 25.04 + 0.98 <.001 �0.31 + 0.17 .066 Non-Hispanic Black males 24.18 + 1.18 <.001 �0.57 + 0.28 .046 Non-Hispanic White males 23.67 + 1.09 <.001 �0.09 + 0.19 .627 Hispanic females 25.16 + 0.95 <.001 �0.06 + 0.13 .631 Non-Hispanic Black females 23.93 + 1.10 <.001 �0.55 + 0.29 .055 Non-Hispanic White females 27.49 + 1.05 <.001 0.26 + 0.13 .048

Physical activity Hispanic males 15.08 + 0.74 <.001 �0.37 + 0.12 .002 Non-Hispanic Black males 17.01 + 0.89 <.001 �0.40 + 0.18 .033 Non-Hispanic White males 14.44 + 0.82 <.001 �0.39 + 0.12 .002 Hispanic females 12.34 + 0.72 <.001 �0.33 + 0.10 .001 Non-Hispanic Black females 12.35 + 0.83 <.001 �0.24 + 0.19 .203 Non-Hispanic White females 12.55 + 0.80 <.001 �0.74 + 0.11 <.001

Safety Hispanic males 34.66 + 0.72 <.001 �0.17 + 0.14 .206 Non-Hispanic Black males 37.27 + 0.88 <.001 0.20 + 0.26 .430 Non-Hispanic White males 35.07 + 0.80 <.001 �0.35 + 0.12 .003 Hispanic females 36.31 + 0.70 <.001 �0.06 + 0.10 .512 Non-Hispanic Black females 37.19 + 0.81 <.001 �0.39 + 0.18 .029 Non-Hispanic White females 36.36 + 0.77 <.001 �0.21 + 0.10 .027

Stress management Hispanic males 13.04 + 0.51 <.001 �0.06 + 0.10 .538 Non-Hispanic Black males 14.48 + 0.62 <.001 0.18 + 0.19 .352 Non-Hispanic White males 12.02 + 0.57 <.001 �0.08 + 0.10 .431 Hispanic females 14.22 + 0.50 <.001 0.01 + 0.07 .912 Non-Hispanic Black females 14.15 + 0.58 <.001 �0.06 + 0.14 .696 Non-Hispanic White females 14.12 + 0.55 <.001 �0.01 + 0.08 .870

Note. General linear mixed model for fixed race/ethnicity–gender and year nested within race/ethnicity–gender effects adjusted for fixed covariates for subsidized lunches, two-parent households, and mother’s education. Random effects were intercept, trajectory, and child nested within cohort. SE ¼ standard error.

Rew et al. 227

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

grades (i.e., Grades 9–12). Males in all racial/ethnic groups

reported significantly more of these physical activity beha-

viors than their respective racial/ethnic females in each of

the four grades with the exception of NHW in Grade 9

(p ¼ .065). NHB males reported statistically significantly more of these physical activity behaviors than Hispanic

males in Grade 9 and more than NHW in Grades 9 and 11.

The only statistically significant difference among females

by race/ethnicity was between Hispanic and NHW females

in Grade 9: NHW females reported engaging in significantly

more of these physical activity behaviors than Hispanic

females. Although not statistically significant, NHW females

engaged in more of these physical activity behaviors than eth-

nic minority females in Grades 9 and 10, but Hispanic females

engaged in these behaviors more than NHB or NHW in Grade

11 and NHB engaged in the greatest number of physical activ-

ity behaviors in Grade 12.

Differences in Safety Behaviors

Overall, NHBs reported the highest frequency of safety

health-promoting behaviors such as wearing a seatbelt or

refusing to ride with a driver who is drinking alcohol. At all

time points, Hispanic and NHW females reported higher

levels of these safety behaviors than Hispanic and NHW

males, respectively. Hispanic males scored lower than all

other males in all grades, but these differences were not

statistically significant. The NHB males engaged in signifi-

cantly more of these safety behaviors than Hispanic males

at all time points except Grade 10, and significantly more than

NHW males in Grade 11. There were no statistically signifi-

cant race/ethnicity differences in frequency of these safety

behaviors among females in any of the four grades.

Differences in Stress Management

There were statistically significant sex differences in stress

management health-promoting behaviors such as having

friends to talk to between Hispanic males and females and

between NHW males and females in each of the four grades.

Females engaged in significantly more of these stress man-

agement behaviors than males in each of the 4 years. There

were no statistically significant sex differences for NHB

participants.

The NHB males engaged in significantly more of these

stress management behaviors than NHW males in all grades;

they also engaged in statistically significantly more of these

behaviors than Hispanic males in Grades 11 and 12. There

were no statistically significant differences among females

in Grade 9 through Grade 12.

Hypothesis 1

The first hypothesis that racial and ethnic minority males

would exhibit fewer health-promoting behaviors than NHW

males and all females during high school (Grades 9–12) was

only partially supported. Nutrition behaviors of Hispanic

and NHB males were greater than those of NHW males in

all grades, except Grade 12 when NHW exhibited slightly

more of these behaviors than NHB and Hispanic males, but

these differences were not statistically significant. NHB

males exhibited the greatest frequency of physical activity

in all grades and, although not statistically significant, His-

panic males exhibited a greater frequency of physical activ-

ity than NHW males in all four grades. Similarly, NHB

males exhibited a greater frequency of safety behaviors and

stress management behaviors than NHW or Hispanic males

in all four grades.

Females consistently engaged in fewer physical activity

behaviors than males, which does not support the hypoth-

esis; however, NHW females engaged in more nutrition-

related behaviors than all ethnic minority males and females

across all four grades, which provides partial support for the

hypothesis.

Hypothesis 2

Hypothesis 2, that adolescents’ health-promoting behaviors

would decrease each year from Grade 9 through Grade 12,

was partially supported. Table 7 shows the growth curve or

trajectory for each health-promoting behavior over time.

All statistically significant trajectories are negative, which

means that the behaviors decreased over time. None of the

trajectories for stress management changed significantly

over time. The only statistically significant positive trajec-

tory was for non-Hispanic white females’ nutrition. This

change indicates that their eating behaviors were better

over time.

The trajectories for each health-promoting behavior

indicate particular changes by race and sex. For example,

the nutrition behaviors of Hispanic, NHB, and NHW males

all decreased over time. The change was statistically signif-

icant for the NHB males (p ¼ .046), but not for the Hispa- nic males (p ¼ .066), nor for the NHW males (p ¼ .627). Physical activity behaviors declined for all racial/ethnic

groups over time and all were statistically significant

except for the NHB females. Similarly, safety behaviors

declined for all racial/ethnic groups over time except for

NHB males whose frequency of engaging in safety beha-

viors increased over time, but the trajectory was not statis-

tically significant (p ¼ .430). Stress management behaviors decreased over time in all groups except NHB males and

Hispanic females, but these changes were not statistically

significant. This means that participants continued to talk

with friends, family, teachers, and coaches about the stres-

sors in their lives with similar frequency in all four grades.

Discussion

The specific aims of this analysis were to compare the

health-promoting behaviors of adolescents by gender and

ethnicity, and explore how health-promoting behaviors of

these adolescents changed during the high school years.

228 The Journal of School Nursing 31(3)

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

Overall, the findings from this rural sample are similar to

other recent cross-sectional findings in national samples of

adolescents. For all race/ethnic groups, females engaged in

more safety behaviors than males 3 of the 4 years (i.e.,

Grades 9, 10, and 12). NHB females engaged in fewer of

these safety behaviors than males only in Grade 11. These

disparate findings may be related to timing of learning to

drive and having one’s own car or access to a family vehicle.

It could be that parents are more protective of female chil-

dren and give male and female children different messages

about safety. This interpretation warrants further study.

As other studies have shown, we found that males

engaged in more sports and exercise activities than females

over all 4 years of the study. This finding supports those of

the 2011 YRBS (CDC, 2012) and the National Health and

Nutrition Examination Survey (NHANES; Liu, Sun, Beets,

& Probst, 2013). The finding that NHW females reported

significantly more of these physical activity behaviors than

Hispanic females in Grade 9 supports previous findings that

suggest NHW females participate in higher levels of physi-

cal activity, including sports teams, than racial/ethnic minor-

ity females (Biddle, Whitehead, O’Donovan, & Nevill,

2005; CDC, 2012; Liu et al., 2013). These disparities may

be related to barriers to participation in sports commonly

faced by racial/ethnic minority females. These barriers

include lower SES (Biddle et al., 2005; Glennie & Stearns,

2012), the higher prevalence of overweight/obesity among

racial/ethnic minority females (Kimm et al., 2002), lack of

resources in the home/yard or neighborhood (Graham, Wall,

Larson, & Neumark-Sztainer, 2014), and cultural differ-

ences in perceptions of various physical activities. In Grade

11 however, unexpectedly, NHW females reported signifi-

cantly lower levels of physical activity behaviors than NHB

females and nearly significantly lower levels in Grade 12. It

could be that these females are driving or riding in cars more

than walking or biking. By Grade 11, many students are being

advised to take on more service-learning projects and other

volunteer activities to improve their chances for admission

to the college or university of their choice. These other extra-

curricular activities can reduce the available time for enga-

ging in physical activity (Spring, Grimm, & Dietz, 2008).

Analysis of data from the National Longitudinal Study of

Adolescent Health (Add Health) showed that extracurricular

activities were related to friendships, particularly among ado-

lescents in high school, when it was more difficult to be part

of a sports team (Schaefer, Simpkins, Vest, & Price, 2011).

The finding that engaging in physical activity declined

for both males and females from Grade 9 to Grade 12 is sim-

ilar to the findings in the national YRBS study of 2011

(CDC, 2012). The national data showed a higher prevalence

of physical activity among adolescents in Grade 9 than in

Grades 10, 11, and 12. These similar findings underscore the

importance of developing activities that keep adolescents,

particularly females, physically active throughout high

school. Further study of how the rural contexts influence

physical activity levels is needed to understand how to trans-

late promising physical activity interventions successfully to

rural adolescents.

The significant gender differences in nutrition health-

promotion behaviors for NHW only, with females scoring

higher than males all 4 years, was a new and somewhat sur-

prising finding, given that males reported higher levels of

fruit, vegetable, and milk intake on the 2011 YRBS (CDC,

2012). These findings, however, are similar to those of

Williams and Mummery (2012) who found that Australian

adolescents’ reports of healthy nutrition behaviors were

greater in girls than in boys. These findings also support

those of a systematic review by Rasmussen et al. (2006),

who found that female gender was a consistent predictor of

higher fruit and vegetable intake among children and adoles-

cents. Although the ‘‘thin ideal’’ that dominates U.S. youth

culture may contribute to eating disorders among females, it

may also lead to the adoption of healthier practices such as

eating more fruits and vegetables than junk foods that are high

in salt and sugar.

Hispanic and NHW females scored significantly higher

than males in their respective ethnic groups on the measure

of stress management. This finding is similar to that of G. S.

Wilson, Pritchard, and Revalee (2005) who found that

females used more coping strategies than males when deal-

ing with stressful experiences. There were no significant

gender differences among NHB participants. This is an

interesting finding that suggests cultural differences in cop-

ing as well as sex differences in coping within cultures. This

finding also has implications for developing gender- and

ethnic-specific interventions to assist adolescents in learning

adaptive coping mechanisms or stress management strate-

gies that contribute to health.

Limitations

This study has several limitations. Data were drawn from a

single geographic area in central Texas and, therefore, do not

represent all high school-age rural adolescents in the United

States. All of the data analyzed in this study were self-report

thus open to self-report bias. Owing to the mobile nature of

the populations of these rural communities, the declining

participation rates are also a limitation. Strengths, however,

include the power of the sample size, large ethnic minority

participation, and the longitudinal design. Despite the long-

itudinal design, this analysis does not reflect individual dif-

ferences over time, but cohort changes only. Nevertheless,

this study yields some important new findings about

health-promoting behaviors with implications for nursing

in general and school nursing in particular.

Implications for School Nurses

School nurses act to prevent adolescents from engaging in

health-risk behaviors such as using alcohol and drugs, using

tobacco products, and having unprotected sex, but we should

Rew et al. 229

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

also strive to enhance the health-promoting behaviors of

adolescents. The role of the school nurse in rural settings

is particularly crucial for promoting health behaviors among

adolescents, given the disparate access to health services

experienced by rural communities (Hartley, 2004), and

school nurse-led programs for rural adolescents in other

countries have shown promise (Barnes, Walsh, Courtney,

& Dowd, 2004). The evidence presented here suggests that

adolescents decrease their health-promoting behaviors dur-

ing the same developmental period that the literature shows

their health-risk behaviors simultaneously increase. School

nurses can play a critical role in the development of health

policies and practices in their school districts. School nurse

participation on the School Wellness Committee is an

important venue for helping shape health-related school pol-

icies that can have long-term benefits for students (National

Association of School Nurses, 2014).

The National Association of School Nurses (2013) holds the

position that professional school nurses are leaders within the

school environment and can be instrumental in setting health

policies for the school in addition to developing and providing

informational and educational programs. School nurses, there-

fore, can have a strong influence on setting policies about avail-

able foods and beverages within the school. They can also

influence policies about physical activity. School nurses are

encouraged to lead interdisciplinary teams within schools and

school districts who will advocate for policies and practices

that promote, rather than risk, the health of adolescents. Such

teams might include other interested professionals such as

school counselors, athletic directors and coaches, social work-

ers, and science teachers. School nurses who work in states that

have State School Nurse Consultants should also partner with

these consultants to advocate and influence the development of

health policies (Broussard & Howat, 2011).

As these findings show, racial/ethnic differences among

girls underscore the need to help racial/ethnic minority girls

find more opportunities to engage in sports and dance. These

findings suggest particular activities that could be increased

in the areas of safety, physical activity, nutrition, and stress

management. For example, interventions to promote physi-

cal activity could be organized for hours immediately after

school (Atkin, Gorely, Biddle, Cavill, & Foster, 2011).

School nurses are already knowledgeable about the signs

of distress in adolescents and may be in positions of leader-

ship where they can refer those who show maladaptive cop-

ing responses to the school counselor or other community

resources (Fitzsimons & Krause-Parello, 2009). Moreover,

many school nurses have opportunities to reach out to the

broader community by making presentations to parent–

teacher organization meetings, organizing health fairs for

parent nights, or creating short community/parent newsletter

items that promote positive strategies for stress management

and other health-promoting behaviors. These strategies have

been used to decrease health-risk behaviors such as smoking

and could also be implemented to enhance health-promoting

behaviors (Hamilton, O’Connell, & Cross, 2004). Previous

research shows that early adolescents, in particular, are

enthusiastic about learning more about how to live a healthy

lifestyle (L. F. Wilson, 2007). The challenge is to embrace

this enthusiasm throughout adolescence.

Conclusion

There are significant gender and ethnic differences in health-

promoting behaviors that may underlie the future health out-

comes of rural adolescents. Findings of this study could

influence the development of school-based interventions for

adolescents. Health-promoting behaviors were found to

decrease over time, which suggests that school nurses, teach-

ers, and parents should pay greater attention to sending mes-

sages to adolescents throughout their high school years

about the health benefits of nutrition, physical activity,

safety, and stress management.

Authors’ Note

The content is solely the responsibility of the authors and does not

necessarily represent the official views of the National Institutes of

Health.

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) disclosed receipt of the following financial support

for the research, authorship and/or publication of this article: This

work was supported by grants from the National institutes of Health

(National Institute of Child Health and Human Development [R01

HD39554] and National Institute of Nursing Research [R01

NR0009856] to the first author.

References

Atkin, A. J., Gorely, T., Biddle, S. J. H., Cavill, N., & Foster, C. (2011).

Interventions to promote physical activity in young people con-

ducted in the hours immediately after school: A systematic review.

International Journal of Behavioral Medicine, 18, 176–187.

Barnes, M., Walsh, A., Courtney, M., & Dows, T. (2004). School

based youth health nurses’ role in assisting young people access

health services in provincial, rural, and remote areas of Queens-

land, Australia. Rural and Remote Health, 4, 279.

Biddle, S. J., Whitehead, S. H., O’Donovan, T. M., & Nevill, M. E.

(2005). Correlates of participation in physical activity for ado-

lescent girls: A systematic review of recent literature. Journal

of Activity and Health, 2, 423–434.

Bradford, J. Y., & O’Sullivan, P. S. (2007). The relationship

between the use of health clinics in rural Mississippi schools

and the Chip-AE Adolescent Health Profile. The Journal of

School Nursing, 23, 293–298.

Broussard, L., & Howat, H. (2011). Influencing school health

policy: The state school nurse consultants. Issues in Compre-

hensive Pediatric Nursing, 34, 198–118.

230 The Journal of School Nursing 31(3)

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

Centers for Disease Control and Prevention. (2012). Youth risk

behavior surveillance, United States, 2011. Morbidity and Mor-

tality Weekly Report, 61, 1–162.

Crosby, R. A., Kegler, M. C., & DiClemente, R. J. (2009). The-

ory in health promotion practice and research. In R. J.

DiClemente R. A., Crosby R. A., & M. C. Kegler (Eds.),

Emerging theories in health promotion practice and research

(2nd ed., pp. 3–17). San Francisco, CA: Jossey-Bass.

Curtis, A. C., Waters, C. M., & Brindis, C. (2011). Rural adolescent

health: The importance of prevention services in the rural com-

munity. The Journal of Rural Health, 27, 60–71.

Dorn, L. D., Beal, S. J., Kalkwarf, H. J., Pabst, S., Noll, J. G., &

Susman, E. J. (2013). Longitudinal impact of substance use and

depressive symptoms on bone accrual among girls aged 11-19

years. Journal of Adolescent Health, 52, 393–399.

Fitzsimons, V. M., & Krause-Parello, C. A. (2009). Military chil-

dren: When parents are deployed overseas. The Journal of

School Nursing, 25, 40–47. doi:10.1177/1059840508326733

Gillis, A. J. (1997). The adolescent lifestyle questionnaire: Devel-

opment and psychometric testing. Canadian Journal of Nursing

Research, 29, 29–46.

Glennie, E. J., & Stearns, E. (2012). Opportunities to play the

game: The effect of individual and school attributes on partici-

pation in sports. Sociological Spectrum: Mid-south Sociological

Association, 32, 532–557.

Gottfredson, N. C., & Hussong, A. M. (2011). Parental involve-

ment protects against self-medication behaviors during the high

school transition. Addictive Behaviors, 36, 1246–1252.

Graham, D. J., Wall, M. M., Larson, N., & Neumark-Sztainer, D.

(2014). Multicontextual correlates of adolescent leisure-time

physical activity. American Journal of Preventive Medicine,

46, 605–616.

Groft, J. N., Hagen, B., Miller, N. K., Cooper, N., & Brown, S.

(2005). Adolescent health: A rural community’s approach. The

International Electronic Journal of Rural and Remote Health

Research, Education, Practice, and Policy, 5, 366.

Hamilton, G., O’Connell, M., & Cross, D. (2004). Adolescent

smoking cessation: Development of a school nurse intervention.

The Journal of School Nursing, 20, 169–174.

Hartley, D. (2004). Rural health disparities, population health, and

rural culture. American Journal of Public Health, 94, 1675–1678.

Horner, S. D., Rew, L., & Brown, A. (2012). Risk-taking behaviors

engaged in by early adolescents while on school property.

Issues in Comprehensive Pediatric Nursing, 35,90–110. doi:

10.3109701460862.2012.678261

Iannotti, R. J., & Wang, J. (2013). Patterns of physical activity,

sedentary behavior, and diet in U.S. adolescents. Journal of

Adolescent Health, 53, 280–286.

Kalak, N., Gerber, M., Kirov, R., Mikoteit, T., Yordanova, J.,

Pühse, U., . . . Brand, S. (2012). Daily morning running for 3

weeks improved sleep and psychological functioning in health

adolescents compared with controls. Journal of Adolescent

Health, 51, 615–622.

Kelder, S. H., Mitchell, P. D., McKenzie, T. L., Derby, C.,

Strikmiller, P. K., Luepker, R. V., & Stone, E. J. (2003).

Long-term implementation of the CATCH physical education

program. Health Education & Behavior, 30, 463–475.

Kimm, S. Y. S., Glynn, N. W., Kriska, A. M., Barton, B. A., Krons-

berg, S. S., Daniels, S. R., . . . Liu, K. (2002). Decline in physical

activity in black and white girls during adolescence. The New

England Journal of Medicine, 347, 709–715.

Leeman, R. F., Hoff, R. A., Krishnan-Sarin, S., Patock-Peckham,

J. A., & Potenza, M. N. (2014). Impulsivity, sensation-seeking,

and part-time job status in relation to substance use and gambling

in adolescents. Journal of Adolescent Health, 54, 460–466.

Liu, J., Sun, H., Beets, M. W., & Probst, J. C. (2013). Assessing

natural groupings of common leisure-time physical activities

and its correlates among US adolescents. Journal of Physical

Activity & Health, 10, 470–479.

Mahon, N. E., Yarcheski, A., Yarcheski, T. J., & Hanks, M. M.

(2007). Mediational models of health practices in early adoles-

cents. Clinical Nursing Research, 16, 302–316. doi:10.1177/

1054773807307314

Nanney, M. S., Davey, C. S., & Kubik, M. Y. (2013). Rural disparities

in the distribution of policies that support healthy eating in US sec-

ondary schools. Journal of Academic Nutrition and Diet, 113,

1062–1068.

National Association of School Nurses. (2013). Policy and advocacy.

Retrieved from http://www.nasn.org/Policy on November 10, 2013

National Association of School Nurses. (2014). School wellness

policies. Retrieved from www.nasn.org/ToolsResources/

SchoolWellnessPolicies

Olshansky, S. J., Passaro, D. J., Hershow, R. C., Layden, J., Carnes,

B. A., Brody, J., . . . Ludwig, D. S. (2005). A potential decline in

life expectancy in the United States in the 21st century. New

England Journal of Medicine, 352, 1138–1145.

Peterhans, E., Worth, A., & Woll, A. (2013). Association between

health behaviors and cardiorespiratory fitness in adolescents:

Results from the cross-sectional Mo Mo-Study. Journal of Ado-

lescent Health, 53, 272–279.

Probst, J. C., Moor, C. G., Glover, S. H., & Samuels, M. E. (2004).

Person and place: The compounding effects of race/ethnicity and

rurality on health. American Journal of Public Health, 94,

1695–1703.

Rasmussen, M., KrØlner, R., Klepp, K.-I., Lytle, L., Brug, J.,

Bere, E., & Due, P. (2006). Determinants of fruit and vegeta-

ble consumption among children and adolescents: A review of

the literature. Part I: Quantitative studies. International Jour-

nal of Nutrition and Physical Activity, 3, 22.

Rew, L., Arheart, K. L., Thompson, S., & Johnson, K. (2013). Pre-

dictors of adolescents’ health-promoting behaviors guided by

primary socialization theory. Journal for Specialists in Pedia-

tric Nursing, 18, 277–288.

Rew, L., Horner, S. D., & Brown, A. (2011). Health-risk behaviors

in early adolescence. Issues in Comprehensive Pediatric Nur-

sing, 34, 79–96.

SAS Institute, Inc. (2013). SAS Institute, Inc. Cary, NC. Retrieved

from http://www.sas.com/software/sas9/

Schaefer, D. R., Simpkins, S. D., Vest, A. E., & Price, C. D. (2011).

The contribution of extracurricular activities to adolescent

Rew et al. 231

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from

friendships: New insights through social network analysis.

Developmental Psychology, 47, 1141–1152.

Spring, K., Grimm, R., & Dietz, N. (2008). Community service

and service-learning in America’s schools. Washington, DC:

Corporation for National and Community Service.

Statistical Analysis Software [Computer software]. Cary, NC: SAS

Institute.

Steinberg, L. (2014). Adolescence, 10th ed. New York, NY:

McGraw-Hill.

Taliaferro, L. A., Muehlenkamp, J. J., Borowski, I. W., McMorris,

B. J., & Kugler, K. C. (2012). Factors distinguishing youth who

report self-injurious behavior: A population-based sample. Aca-

demic Pediatrics, 12, 205–213.

Thompson, A. H., Dewa, C. S., & Phare, S. (2012). The suicidal

process: Age of onset and severity of suicide behaviour. Social

Psychiatry and Psychiatric Epidemiology, 47, 1263–1269.

Williams, S. L., & Mummery, W. K. (2012). Associations between

adolescent nutrition behaviours and adolescent and parent char-

acteristics. Nutrition & Dietetics, 69, 95–101.

Wilson, G. S., Pritchard, M. E., & Revalee, B. (2005). Individual

differences in adolescent health symptoms: The effects of gen-

der and coping. Journal of Adolescence, 28, 369–379.

Wilson, L. F. (2007). Adolescents’ attitudes about obesity, and

what they want in obesity prevention programs. The Journal

of School Nursing, 23, 229–238.

Yarcheski, A., Mahon, N. E., & Yarcheski, T. J. (1997). Alternate

models of positive health practices in adolescents. Nursing

Research, 46, 85–92.

Author Biographies

Lynn Rew, EdD, RN, AHN-BC, FAAN, is the Denton & Louise

Cooley and Family Centennial Professor in Nursing at The Univer-

sity of Texas at Austin, Austin, TX, USA.

Kristopher L. Arheart, EdD, is Associate Professor at the Depart-

ment of Epidemiology and Public Health in the University of

Miami, Miami, FL, USA.

Sharon D. Horner, PhD, RN, FAAN, is the Dolores V. Sands

Chair in Nursing Research and Associate Dean of Research at

The University of Texas at Austin School of Nursing, Austin,

TX, USA.

Sanna Thompson, PHD, MSW, is Associate Professor at the Uni-

versity of Texas at Austin School of Social Work, Austin, TX,

USA.

Karen E. Johnson, PhD, RN, is Assistant Professor at the University

of Texas at Austin School of Nursing, Austin, TX, USA.

232 The Journal of School Nursing 31(3)

at LOYOLA MARYMOUNT UNIV on October 8, 2015jsn.sagepub.comDownloaded from