Exercise Preferences for Young People At-Risk for and Living with HIV: A Short Report

Abstract

Exercise can improve physical and mental health for young people at risk for and with HIV, but prevalence rates remain low. This study explored exercise preferences and barriers among young people at risk for and with HIV, and potential gender differences. A total of 129 participants (66.7% male, mean age = 23.8 [SD=2.1; range: 19–28], 35.7% Black/African American) at-risk for or with HIV were recruited from a larger study and completed an online survey of exercise preferences and barriers. Overall, participants preferred an exercise program that takes place at a fitness center, occurs alone, has a coach/instructor present, is between 30–60 minutes, lasts longer than 8 weeks, and includes aerobic and resistance exercises. The fatiguing nature of exercise and cost were common barriers among all genders. Few gender differences emerged. These results should be used to design exercise programs for young adults at risk for and with HIV.

INTRODUCTION

Approximately 1.1 million people in the US have HIV (PWH) with 40,000 annual new diagnoses (Centers for Disease Control, 2019). Currently, HIV is a chronic condition (Do et al., 2014) but to prevent transmission, near perfect (>95%) antiretroviral therapy (ART) adherence is required (Robbins et al., 2014). An increase in life expectancy has placed an increased importance on managing comorbidities (e.g., cardiovascular disease) (Triant, 2013), factors that negatively impact adherence (e.g., mental illness) (Blashill et al., 2015; Do et al., 2014), and quality of life (SantaBarbara et al., 2020).

Exercise may be just such an intervention, given its low barriers to entry (e.g., low cost, ease of access) and ability to improve mental and physical health (Nosrat et al., 2017; SantaBarbara et al., 2017; Schuch et al., 2018) without any negative effects on disease progression (Jaggers & Hand, 2016). Still, most studies often experience recruitment challenges and attrition (Heissel et al., 2019; Nosrat et al., 2017; Ozemek et al., 2020). Exercise preferences (e.g., type,, intensity, etc.) and perceived barriers vary widely across populations (Abrantes et al., 2011; Busch et al., 2016; Jones & Courneya, 2002), but exercise programs are often implemented without considering the unique preferences and barriers of this population (Busch et al., 2016). As a result, low program initiation, commitment, and retention result from a mismatch between the exercise program, the needs, and beliefs towards exercise of the target population.

Furthermore, gender is an important factor to consider when building an exercise program. A study exploring exercise preferences of people seeking substance use treatment found that women preferred supervised group exercise compared to men (Abrantes et al., 2011). Another study that explored preferences for exercise as an adjunctive treatment for depression also found that women preferred a graded exercise program and yoga, whereas men preferred moderate intensity program and weight lifting (Busch et al., 2016).

Overall, by tailoring exercise programs to individual’s exercise preferences, the effectiveness of, and adherence to, an exercise program may increase. However, no such data exists with people living with and at risk for HIV. Thus, the aims of the current investigation are to explore: 1) preferences for the content and structure of an exercise program; 2) perceived barriers to exercise; and 3) gender differences in these two aims.

METHODS

Participants and Procedures

We invited PWH and those at increased risk for HIV (n=1360) from larger HIV studies, who had previously consented to being informed about future studies, to complete an online survey. Further details are reported elsewhere (Arnold et al., 2019; Rotheram et al., 2019; Swendeman et al., 2019). Participants were 18–28 years of age, able to read and write in English, and had no orthopedic condition that prevents exercise.

Participants were recruited via emails and text messages. Email notifications with a brief study description and informed consent link were distributed through Qualtrics_®^XM (QualtricsXM) and text notifications through Twilio (Twilio). Participants received a $20 electronic gift card for survey completion. Data were collected between March and May 2021. This project was approved by the University’s Institutional Review Board.

Measures

Participant Characteristics

Demographic data were collected, and participants were classified into one of three categories: (cisgender) men, (cisgender) women, and gender expansive.

Preferences for the Content and Structure of an Exercise Intervention

Participants reported on factors important to the design of an exercise intervention (e.g., preferred location, social environment, frequency, type). This questionnaire was adapted from previous studies exploring exercise preferences in patients in substance use treatment (Abrantes et al., 2011) and individuals experiencing depression (Busch et al., 2016).

Perceived Barriers to Physical Activity

Barriers to physical activity were assessed through a 12-item measure that uses a 4-point Likert scale (Busch et al., 2016). Frequencies of barriers reported were later dichotomized as disagree (0) and agree (1).

Self-reported Exercise

The Godin Leisure-Time Exercise Questionnaire (GLTEQ) (Amireault & Godin, 2015) was used to measure self-reported leisure-time exercise during a typical week. Participants were asked how many times in a typical week they participate in 15 minutes or more of minimal (e.g., easy walking), moderate (e.g., weight lifting), and strenuous (e.g., vigorous running) exercise. Validated scoring and physical activity cutoffs were used (Amireault & Godin, 2015; Garber et al., 2011), and we added a question about changes in exercise since the COVID-19 pandemic (mid-March 2020).

Data Analysis

Descriptive statistics are presented as means and standard deviations for continuous variables, and percentages for categorical variables. Relationships between gender, exercise preferences, and perceived barriers were examined using Analysis of Variance (ANOVA). We first conducted omnibus F tests followed by post hoc tests for pairwise differences if F tests were significant. All analyses were conducted using IBM SPSS 28, and significance was set a priori at p < 0.05.

RESULTS

Descriptives

The sample (N=129; See Table 1) was mostly male (66.7%) with a mean age of 23.8 (SD = 2.1). Most participants were Black/African American (35.7%) and most were considered at-risk for HIV (82.9%).

Table 1.

Sample descriptives (N=129)

	n, %
Gender
Male	86(66.7)
Female	21(16.3)
Transgender male*	8(6.2)
Transgender female*	5(3.9)
Other gender identity*	9(6.9)
Age(years)
Mean(SD)	23.8(2.1)
Range	19–28
Body Mass Index
Mean(SD)	28.2(7.3)
Range	17.9–54.2
Racial Group
Black/African American	46(35.7)
White	37(28.7)
Other	24(18.6)
Asian	11(8.5)
Do not know	5(3.9)
Do not want to answer	6(4.7)
Hispanic/Latina(o)
Yes	42(32.6)
No	91(65.9)
Do not want to answer	2(1.6)
Sexual Orientation
Homosexual	64(49.6)
Heterosexual	23(17.8)
Bisexual	22(17.1)
Other	12(9.3)
Do not want to answer	8(6.2)
Living situation
Living alone	31(24.0)
Living with a romantic partner	29(22.5)
Living with a spouse	2(1.6)
Living with roommates	43(33.3)
Other	20(15.5)
Do not want to answer	4(3.1)
Are you currently homeless/without a permanent place to stay?
Yes	12(9.3)
No	115(89.1)
Do not want to answer	2(1.6)
In the past, have you ever been homeless/without a permanent place to stay?
Yes	46(35.7)
No	81(62.8)
Do not want to answer	2(1.6)
Education
Less than high school graduate	1(0.8)
High school graduate	22(17.1)
Some vocational school/college	51(39.5)
Completed vocational school	7(5.4)
Completed college	34(26.4)
Some graduate school	5(3.9)
Completed masters/doctoral degree	9(7.0)
Employment Status
Employed full time	56(43.4)
Employed part time	30(23.3)
Unemployed (looking)	31(24.0)
Unemployed (not looking)	13(10.1)
Retired	0(0.0)
Disabled	2(1.6)
Furloughed (due to COVID-19)	4(3.1)
Student Status
Not a student	86(67.2)
Full time student	31(24.2)
Part time student	11(8.5)
Current Annual Income
<$10,000	44(34.1)
$10,001-$25,000	37(28.7)
$25,001-$40,000	23(17.8)
$40,001-$55,000	11(8.5)
>$55,000	6(4.7)
Do not know	6(4.7)
Do not want to answer	2(1.6)
Currently living with HIV
Yes	20(15.5)
No	107(82.9)
Do not want to answer	2(1.6)

^*Considered as Gender Expansive in further analyses

Current Exercise Behaviors

The mean total leisure-time exercise score (i.e., strenuous and moderate intensity) for the sample was 34.1 (SD = 26.1), which indicates sufficiently active (>24 on GLTEQ). A total of 67.5% and 42.6% of participants are classified as active according to their strenuous and moderate intensity, respectively. Additionally, for total leisure-time exercise score, 68.7% males, 76.2% females, and 54.5% gender expansive, for strenuous-intensity 43.0% males, 57.1% females, and 28.5% gender expansive, and for moderate-intensity 23.2% males, 23.8% females, and 28.6% gender expansive are considered active. In terms of changes in exercise behaviors since the COVID-19 pandemic, over half (58.9%) of the sample reported exercising less, 26.4% reported no change, and 14.7% reported an increase in exercise.

Preferences for an Exercise Program

The preferred location for exercise was a fitness center (41%–57% across gender categories) while a community setting was the least preferred (< 9.5% overall). Most males and gender expansive participants prefer to exercise alone (52.3 and 59.1%, respectively) compared to females who equally prefer exercising alone or as part of a group (38.1% and 23.8%, respectively). Roughly half of each gender group prefer supervised exercise, and more than 80.0% of each gender prefer that a coach/instructor is present. The frequency of 3–4 times per week lasting longer than 8 weeks is the most preferred frequency and program length for all genders. Further details are presented in Table 2.

Table 2.

Exercise preferences by gender (n = 129).

	Male (n=86)	Female (n=21)	Gender Expansive (n=22)	F-value	p-value
Where would your ideal exercise program take place?				0.3	.72
Home	23.3%	23.8%	36.4%
Fitness center (gym)	51.2%	57.1%	40.9%
No preference	10.5%	9.5%	9.1%
Community setting	8.1%	9.5%	4.5%
Other	7.0%	0.0%	9.1%
Would you prefer to exercise alone or as part of a group?				1.9	.15
Alone	52.3%	38.1%	59.1%
In a group	27.9%	23.8%	27.3%
No preference	19.8%	38.1%	13.6%
Would you prefer that this exercise program be supervised by a coach or instructor?				.37	.69
Yes	47.7%	57.1%	50.0%
No	32.6%	28.6%	36.4%
No preference	19.8%	14.3%	13.6%
If yes, what type of supervision would you prefer?				0.3	.75
Coach/instructor physically present	90.2%	83.3%	81.8%
Coach/instructor physically present through video conferencing	4.9%	16.7%	9.1%
Coaching/instructing delivered through email, mobile phone application, chat	4.9%	0.0%	9.1%
How frequently would your ideal exercise program take place?				0.9	.39
Multiples times per day	5.8%	0.0%	13.6%
Once per day	23.3%	9.5%	22.7%
2–3 times per week	22.1%	42.9%	31.8%
3–4 times per week	39.5%	47.6%	27.3%
Once per week	8.1%	0.0%	4.5%
Other	1.2%	0.0%	0.0%
How long would you prefer an exercise program to last?				1.7	.18
1–4 weeks	15.1%	19.0%	18.2%
4–8 weeks	25.6%	38.1%	45.5%
Longer than 8 weeks	55.8%	42.9%	36.4%
Other	3.5%	0.0%	0.0%
What intensity do you prefer?				0.2	.82
Strenuous	17.4%	4.8%	27.3%
Moderate	34.9%	38.1%	18.2%
Mild	7.0%	19.0%	4.5%
Graded (Gets more difficult as you become more fit)	39.5%	38.1%	50.0%
Other	1.2%	0.0%	0.0%
Would you prefer to exercise:				1.0	.37
In one long bout (60–90 minutes)	43.0%	19.0%	36.4%
Shorter bouts (30–60 minutes)	46.5%	71.4%	45.5%
Less than 30 minutes	4.7%	4.8%	9.1%
No preference	5.8%	4.8%	4.5%
Other	0.0%	0.0%	4.5%
What is your most preferred form of exercise?				1.2	.40
Aerobic training (walking, running, cycling)	17.4%	14.3%	4.5%
Weight lifting (strength training)	15.1%	9.5%	9.1%
Yoga	5.8%	4.8%	0.0%
Dance	3.5%	0.0%	4.5%
Group exercise classes	1.2%	4.8%	9.1%
A program that combines aerobic training and weight lifting	2.3%	14.3%	4.5%
Combination of all options	54.7%	52.4%	68.2%

Perceived Barriers to Exercise

Notable barriers included distance from fitness centers which was significantly different between the males and the gender expansive group F(2,126) = 6.5, p<.01, the cost of exercising at a fitness center was significantly different between the males and the gender expansive group F(2,126) = 3.2, p=.04, and the scarcity of places to exercise was significantly different between the males and the gender expansive group F(2,126) = 6.9, p<.01. A more detailed description of perceived barriers to exercise by gender are presented in Table 3.

Table 3.

Perceived Barriers to Exercise by Gender

	Male (n=86)	Female (n=22)	Gender expansive (n=22)	F-value	p-value
Takes too much time	40.7%	14.3%	50.0%	2.8	.06
Exercise tires me	69.8%	66.6%	81.8%	2.4	.09
Places to exercise are too far away**	26.7%	42.9%	50.0%	6.5	<.01
Too embarrassed to exercise	32.6%	38.1%	63.6%	3.2	.05
It costs too much money*	45.4%	28.6%	63.7%	3.2	.04
Scheduling conflicts	22.1%	14.3%	36.3%	1.9	.14
I am fatigued by exercise	41.8%	38.1%	27.2%	0.1	.87
Lack of partner encouragement	19.8%	14.3%	18.2%	0.6	.56
Takes too much time away from family relationships	4.7%	14.3%	9.0%	2.6	.08
People in exercise cloths look funny	9.3%	9.5%	9.0%	0.7	.93
Lack of family encouragement	29.1%	33.3%	13.6%	3.0	.05
Takes too much time away from family responsibilities	9.3%	9.6%	9.1%	0.5	.63
Exercise is hard work for me	48.9%	47.6%	59.1%	1.1	.34
Too few places to exercise*	22.1%	38.1%	50.0%	6.2	<.01

^*Significant at the p<.05

^**Significant at the p<.01

DISCUSSION

Exercise adherence is generally poor among PWH (Nosrat et al., 2017) with a likely factor being a mismatch between the one-size-fits-all exercise program design and participants’ exercise preferences. Exercise preference plays an important role in exercise adherence (Abrantes et al., 2011), and the variability in exercise preferences within and between gender categories underscore poor adherence among PWH.

Overall, the most selected preferences were exercising: 1) in fitness centers, 2) alone, 3) with a coach/instructor, 3) for 30–60 minutes, and 4) for more than 8 weeks. There were no clear overall preferences regarding intensity, frequency, or a specific type of aerobic or resistance training. Some gender differences emerged but were not significant. Specifically, twice as many males and gender expansive participants prefer to exercise daily compared to females (23.3% and 22.7% vs. 9.5%), whereas almost double the number of females preferred to exercise 3–4 days per week. These preferences are consistent with previous work regarding exercise preferences of patients in substance use treatment (Abrantes et al., 2011) and individuals experiencing depression (Busch et al., 2016). However, while these earlier studies have demonstrated gender differences in exercise preferences, the mean age of participants in our study was 23.8, while the mean ages of participants in two earlier studies were 41.6 (Abrantes et al., 2011) and 39 (Busch et al., 2016). Additionally, those early studies did not include a gender expansive group. The lack of significant differences between genders in preferred type and structure, suggests that similar programs could be developed to increase exercise acceptability and retention for all genders with similar demographics to the participants in this study. Previously, individualizing exercise prescriptions and programs have been effective when implemented in other populations with chronic diseases (Thompson et al., 2013).

Most participants reported barriers such as fatigue and hard work associated with exercise. Fatigue is also a barrier in those with depression (Busch et al., 2016) and among people seeking substance use treatment (Abrantes et al., 2011). Several gender differences also emerged, including the distance to and scarcity of exercise facilities and membership cost. In all cases, the gender expansive group reported these as significant barriers compared to other genders.

Limitations include sample size, few women and few PWH, self-report exercise data, and selection bias as participants were recruited through larger ongoing intervention studies. Yet, a major strength of this study is that it is the first to assess the exercise preferences of a diverse sample of young-adult PWH or at-risk for HIV.

CONCLUSIONS

Overall, exercise programs for PWH and those at-risk for HIV, have difficulty with recruitment and retention (Vancampfort et al., 2017). These study findings provide preliminary guidance towards the development of tailored exercise programs to improve health outcomes in this population. Targeting specific sub-groups (e.g., different genders), may allow programs to be more successful by meeting the unique needs of the population they are trying to serve.