Examining 24-Hour Activity and Sleep Behaviors and Related Determinants in Latino Adolescents and Young Adults With Obesity

Abstract

Background.

Few studies have examined 24-hour activity and sleep behaviors and their contribution to type 2 diabetes (T2D) in Latino adolescents and young adults with obesity.

Aim.

This study included quantitative data on T2D risk and 24-hour activity and sleep behaviors and qualitative data on individual, social, and environmental behavioral determinants.

Method.

A 7 day, 24-hour, wrist-worn accelerometer protocol assessed moderate-to-vigorous physical activity (PA), sedentary behaviors (SB), sleep, and sleep regularity, in adolescents (N = 38; 12–16 years) and young adults (N = 22; 18–22 years). T2D-related outcomes included adiposity (BMI, BF%, waist circumference), fasting, and 2-hour glucose. A subsample of participants (N = 16 adolescents, N = 15 young adults) completed interviews to identify behavioral determinants.

Results.

High levels of PA were observed among adolescents (M = 103.8 ± 67.5 minutes/day) and young adults (M = 96.8 ± 78.8 minutes/day) as well as high levels of SB across both age groups (≥10 hours/day). Sleep regularity was negatively associated with adiposity (all ps < .05) in both age groups as well as fasting and 2-hour glucose in young adults (all ps < .05). Social support was associated with PA in both age groups as well as SB in younger youth. Auditory noises, lights, and safety inhibited sleep in both age groups.

Conclusion.

PA is critical for disease reduction, yet reducing SB and improving sleep are also important targets for reducing T2D risk in Hispanic adolescents and young adults. Future health promotion and disease prevention strategies should leverage qualitative findings regarding behavioral determinants.

Background

Disparities in type 2 diabetes (T2D) emerge in early adolescence and are due in part to unhealthy lifestyle behaviors (Goran et al., 2003; Menke et al., 2016). Latino adolescents with obesity are the most insulin resistant pediatric subgroup in the United States with higher rates of prediabetes (22.9%) than non-Latino White youth (15.1%; Menke et al., 2016). Adolescence is an important life stage for disease risk where youth typically experience a decrease in disease-related behaviors like physical activity (PA; Goran et al., 2003; Troiano et al., 2008). Health behaviors and patterns established during adolescence persist into young adulthood and are predictive of future health outcomes (Goran et al., 2003). To address T2D disparities in this high-risk population, there is a critical need for disease prevention strategies that address lifestyle behaviors, particularly during the critical, transitional stages of adolescence to young adulthood (McCurley et al., 2017).

Current lifestyle interventions in high-risk youth and young adults have been largely ineffective (Mendoza-Vasconez et al., 2016; Reis et al., 2016) due in part to their singular focus on PA (Rosenberger et al., 2019). While it has been established that PA improves insulin resistance, more emergent research has shown that the cumulative time spent in PA, sedentary behaviors (SB), and sleep, across the 24-hour cycle, significantly contribute to glucose metabolism (Rosenberger et al., 2019). For example, it is estimated that youth with obesity spend 60% of their day in SB (Belcher et al., 2010; Carson et al., 2015; Carson, Hunter, et al., 2016), which is associated with increased insulin resistance (Gordon-Larsen et al., 2000; Saunders et al., 2013). Only 30% of adolescents meet current sleep recommendations of 8 to 10 hours of sleep/night (Carson, Tremblay, et al., 2016). Sleep duration, in addition to measures of sleep quality like sleep regularity, are inversely related to T2D risk (Dorenbos et al., 2015; Javaheri et al., 2011; Morselli et al., 2012). This evidence has led to a paradigm shift to look beyond the independent contribution of any one behavior such as PA, to consider the contributions of 24-hour activity and sleep behaviors on health and disease outcomes (Chastin et al., 2015; Rosenberger et al., 2019).

Examining 24-hour activity and sleep behaviors may provide a more comprehensive picture of behavioral factors that influence chronic diseases (Rosenberger et al., 2019). However, few studies have used objectives measures like accelerometry to rigorously assess PA, SB, and sleep across the 24-hour cycle in Latino youth or young adults (Chaput et al., 2014; Dunstan et al., 2010; Patel et al., 2010; Tremblay et al., 2016). Furthermore, because research to date has focused primarily on determinants of PA, little is known regarding the individual, social, and environmental determinants that drive SB and sleep, particularly among Latino adolescents and young adults (Keadle et al., 2017). Therefore, the primary aim of this study was to use accelerometry to measure 24-hour activity and sleep behaviors, compare these behaviors across age groups and examine associations among behaviors and T2D risk outcomes in Latino adolescents and young adults with obesity. The secondary aim was to use qualitative interviews to identify individual, social, and environmental determinants of 24-hour activity and sleep behaviors. Our focus on Latino adolescents and young adults provides a unique opportunity to gather novel information on T2D-related behaviors and their determinants in two high-risk populations across two important, transitional life periods where adolescents are growing out of childhood and young adults are growing into adulthood. The information on 24-hour behaviors and determinants gained from this study will inform the development of diabetes prevention strategies in two vulnerable populations.

Method

Participants

Participants were recruited through a vast network of community organizations, pediatric clinics, and local Spanish-language media in Phoenix, AZ, using flyers, advertisements, and in-person announcements. Participants were screened using the following inclusion criteria: (1) self-identification as Latino(a), (2) aged 12 to 16 years for adolescents and 18 to 22 years for young adults, and (3) with obesity, defined as body mass index (BMI) ≥ 95th percentile for age and sex or BMI ≥ 30 kg/m². Participants were excluded if they were diagnosed with a condition or sustained an injury that would impair cognition, metabolism, or their ability to be active. Given that our staff were bilingual and able to consent and collect data in Spanish or English, we did not have criteria regarding language. This study was approved by the institutional review board at Arizona State University. Informed consent was obtained from young adults and parental consent and adolescent assent was obtained prior to any study procedures.

Study Design

In this mixed-methods study, data were collected sequentially starting with quantitative data in Phase 1 and qualitative data in Phase 2. Data collected in Phase 2 was used to provide insights on quantitative data collected in Phase 1 (Creswell & Creswell, 2018). To complete Phase 1, we recruited 60 participants (38 adolescents, 22 young adults) and collected quantitative data on 24-hour activity and sleep behaviors using accelerometry. Phase 2, immediately followed Phase 1. In Phase 2, we randomly selected a convenience sample of 31 participants (16 adolescents, 15 young adults) from Phase 1 and collected qualitative data on determinants of 24-hour activity and sleep behaviors to provide a deeper understanding of contextual factors that influence quantitatively assessed behaviors (Creswell & Creswell, 2018). The quantitative and qualitative data were integrated at the interpretation stage of the study (Curry & Nunez-Smith, 2015).

Quantitative Data Collection

Measures of Adiposity.

All participants arrived at the clinical research unit at Arizona State University at approximately 8 a.m. after an overnight fast. Participants completed a demographic survey. Height and weight were assessed using a scale and stadiometer to calculate BMI and participants were categorized using definitions from the Centers for Disease Control and Prevention (2020, 2021). Body fat percentage (BF%) was assessed by Dual-energy X-Ray Absorptiometry (GE Lunar, Madison, WI). Waist circumference was assessed in triplicate at the umbilicus using a flexible measuring tape (Gullick II, Model 67020). These assessments are in line with recommendations for screening in obese youth (Barlow, 2007).

Diabetes Risk.

A standard 2-hour, 75-gram oral glucose tolerance test was used to assess diabetes risk. Blood samples were collected at fasting and 2 hours to determine glucose tolerance. Measures of glucose tolerance derived from glucose tolerance testing are valid, reliable, and highly correlated with the gold-standard hyperinsulinemic-euglycemic clamp (r = .78, p < .0005; Matsuda & DeFronzo, 1999; Yeckel et al., 2004). According to the American Diabetes Association, participants with fasting glucose <100 mg/dl or 2-hour glucose <140 mg/dl were considered to have normal glucose tolerance and participants with glucose ≥ 100 mg/dl or 2-hour glucose ≥ 140 mg/dl were considered to have impaired glucose tolerance or prediabetes (American Diabetes Association, 2021).

24-Hour Activity and Sleep Behaviors.

A wrist-worn GT3X-BT accelerometer (Actigraph, Pensacola, FL) was used to assess SB, defined as <1 metabolic equivalents (METs), light physical activity (LPA; < 3 METs), and moderate-to-vigorous physical activity (MVPA; > 6 METs; Crouter et al., 2013). While placement of the accelerometer on the hip provides more accurate measures of PA, use of wrist-worn actigraphy has been shown to provide valid and acceptable measures of activity as compared with device placement on the hip, in both youth (Scott et al., 2017; Trost et al., 2014) and young adults (Dieu et al., 2017). Use of wrist-worn actigraphy to measure sleep has been validated against polysomnography, the gold standard, and provides more accurate measures of sleep parameters compared to placement of the device at the hip in both youth (Martin & Hakim, 2011; Smith et al., 2019) and young adults (Acebo et al., 1999; Hjorth et al., 2012). Furthermore, placement of the device on the wrist can increase wear compliance and was selected to align with procedures used in National Health and Nutrition Examination Survey (Fairclough et al., 2016; Hildebrand et al., 2014). Participants were asked to wear the accelerometer for 24 hours a day for 7 days (Migueles et al., 2017). A valid day of wear was defined as ≥ 18 hours of wear on ≥ 4 days, one of which had to be a weekend day (Migueles et al., 2017). A 7-day monitoring protocol provides reliable estimates of usual PA behavior and accounts for weekday and weekend activity patterns (Migueles et al., 2017). Previous research has demonstrated that different cut points yield different PA estimates across youth and adults (Evenson et al., 2012; Loprinzi et al., 2012). Thus, we used cut points that were specific to youth and specific to adults. All accelerometer data was analyzed in the ActiLife software. To conduct data processing within the software, we indicated that device was worn on the wrist. Nonwear time was assessed using the Choi algorithm, which has been shown to lead to improvements in the estimation of PA and SB (Choi et al., 2011). For adolescents, behavioral data was categorized using Evenson cut points: SB (<100 counts per minute), LPA (101–2295 counts per minute), MVPA (≥2296 counts per minute; Evenson et al., 2008). For young adults, behavioral data was categorized using Troiano cut points: SB (<100 counts per minute), LPA (101–2019 counts per minute), MVPA (≥2020 counts per minute; Troiano et al., 2008). The Sadeh algorithm was used to assess sleep duration in both age groups and sleep times were validated using participant wear logs (Sadeh et al., 1994). Data are reported as average minutes/day for activity average hours/day for sedentary and sleep behaviors.

We used sleep times to calculate the sleep regularity index (SRI) to determine the regularity or irregularity of day-to-day sleep patterns (Phillips et al., 2017). The SRI is calculated using an algorithm as the likelihood that any two time-points (minute-by-minute) 24 hours apart were the same sleep/wake state, across all days (Phillips et al., 2017). The SRI was originally validated in college attending young adults and has a possible range of 0 to 100 with a higher SRI score indicating greater sleep regularity (Phillips et al., 2017) with a lower score indicating greater sleep irregularity, which is associated with greater cardiometabolic risk youth and young adults (Phillips et al., 2017; Spruyt et al., 2011).

Qualitative Data Collection

Interview Procedures.

Youth and young adults were interviewed between August 2018 and August 2019 by two trained interviewers. Interviews took place in a private room at Arizona State University or the Phoenix Public Library, were audio-recorded, and lasted 40 to 60 minutes. All participants elected to conduct the interview in English. A semistructured interview guide was informed by the Expanded Ecodevelopmental Model, which maps the direct and indirect determinants of health and disease outcomes originating from multiple-ecological levels. This model provides a framework for investigating the complex interactions of individual-, social-, and community-level factors across critical life periods such as adolescence and young adulthood (Castro et al., 2009). Guided by this framework, interview questions were developed to elicit information about individual-level, family and friend social-level, and environmental-level determinants of activity and sleep behaviors. Interview questions are presented in Table 1.

Table 1.

Interview Questions.

1.	Tell me about the last time that you felt like you were being active? Describe where you were, who you were with, and what you were doing.
2.	Tell me about the last time you felt like you were sitting for too long? Describe where you were, who you were with, and what you were doing.
3.	Which of these statements about how you sleep are true for you:
	(a) Sleep is important to me, so I try to make sure I get enough sleep.
	(b) I often stay up late and I feel like I don’t get enough sleep.
	(c) Sleep doesn’t really affect me, sometimes I get enough sleep and sometimes I don’t get enough sleep.
4.	Tell me more about why this statement is true for you?
5.	What are some ways that your friends influence your health behaviors?
6.	What are some ways that your family influences your health behaviors?

Statistical Analyses

Standard descriptive statistics were used to assess and compare participant characteristics and behavioral data. Spearman correlations were used to examine relationships among MVPA, LPA, SB, sleep, and SRI, with measures of adiposity (BMI, BF%, waist circumference) and T2D risk (fasting and 2-hour glucose). All analyses were conducted in the IBM SPSS statistical software (IBM Corp, 2019).

Interviews were professionally transcribed and coded by trained coders using a thematic content analysis approach in NVivo (Version 12.5) (Ryan & Bernard, 2003). Coders read each transcript independently and emergent concepts, repeated ideas, and important quotes were identified to generate a list of codes. After the initial reading, coders met to discuss and compare codes before updating their code-books. This process was repeated three times until no new codes were identified. After analysis, codes were organized by levels of influence (individual, social, and environmental) for each behavior (PA, SB, and sleep) and related ideas and responses were grouped to form emerging themes (Bernard & Ryan, 2009). The percentage of participants that discussed each emerging theme were calculated to provide a quantitative representation of each theme within the sample.

Results

24-Hour Activity and Sleep Behaviors and T2D-Related Outcomes

A total of 60 adolescents (n = 38) and young adults (n = 22) completed the study. Activity data was available for 38 participants (n = 25 adolescents, n = 13 young adults) indicating 63% compliance with the accelerometer protocol. Four participants declined to participate in the accelerometer protocol, one device malfunctioned, and 17 participants did not meet criteria for wear time validation. Descriptives on metabolic and behavioral variables are presented in Table 2.

Table 2.

Participant Descriptives.

Variable	Adolescent N = 38 (SD)	Young adult N = 22 (SD)
Age (years)	13.9 (1.5)	20.8 (1.1)
Sex (%)
Male	63.2	36.4
Female	36.8	63.6
Living status (%)
Two-parent household	62.5	73.3
Female-headed household	18.8	6.7
Separated parents	12.5	6.7
Blended family	6.3	13.3
Full-time/part-time school (%)	100	20
Full-time/part-time employed (%)	—	73
Height (cm)	165.1 (8.0)	166.0 (7.5)
Weight (kg)	95.5 (23.0)	109.6 (18.6)
BMI (kg/m2)	33.8 (5.8)	39.9 (6.9)
Waist circumference (cm)	91.2 (21.1)	114.9 (15.8)
Body fat (%)	43.0 (9.9)	46.8 (7.4)
Fasting glucose (mg/dL)	91.6 (6.0)	89.8 (5.7)
2-Hour glucose (mg/dL)	116.6 (26.7)	120.3 (43.1)
Sleep (hours/night)	7.6 (1.1)	8.3 (2.1)
Sedentary behaviors (hours/day)	11.0 (1.3)	11.3 (2.2)
Light physical activity (minutes/day)	225.6 (58.0)	194.0 (60.2)
Moderate-to-vigorous physical activity (minutes/day)	103.8 (67.5)	96.8 (78.8)
Sleep regularity index	57.4 (4.3)	46.8 (11.4)

Average 24-hour activity and sleep cycles are presented in Figure 1. Most adolescents (83%) exceeded U.S. guidelines of 60 minutes of MVPA/day with an average of 103.8 ± 67.5 minutes/day and most young adults (80%) met U.S. guidelines of 150 minutes of MVPA/week with an average of 96.8 ± 78.8 minutes/day (U.S. Department of Health and Human Services, 2018). The majority of adolescents (83%) and young adults (60%) engaged in over 10 hours of sedentary time per day. Less than half of adolescents (43%) achieved the recommended 9 to 11 hours of sleep with an average of 7.6 hours/night (Paruthi et al., 2016). In contrast, 77% of young adults achieved the recommended 7 to 9 hours of sleep with an average of 8.3 hours/night (Watson et al., 2015).

Figure 1.

Average 24-hour activity and sleep profiles in adolescents and young adults.

Note. MVPA = moderate-to-vigorous physical activity; LPA = light physical activity; SB = sedentary behaviors.

In adolescents, SRI was negatively associated with waist circumference (r = −.52, p < .05). In young adults, SRI was negatively associated with weight (r = −.65, p < .05), waist circumference (r = −0.79, p < .01), BF% (r = −.83, p < .01), as well as fasting (r = −0.73, p < .05) and 2-hour glucose (r = −.73, p < .05). No other significant relationships were observed among behaviors and T2D-related outcomes (ps > .05).

Determinants of 24-Hour Activity and Sleep Behaviors

Themes and the percentage of participants from both age groups that discussed each theme are presented in Table 3.

Table 3.

Emergent Themes From Qualitative Interviews and the Percentage of Participants That Discussed Each Theme.

Level of influence	Moderate-to-vigorous physical activity		Sedentary behaviors		Sleep
	Adolescents	Young adults	Adolescents	Young adults	Adolescents	Young adults
Individual	Physical appearance (47%)	Exercising with someone (47%)	Video streaming (63%)	Video streaming (67%)	Unaffected by sleep (75%)	Unaffected by sleep (47%)
		Improve health (33%)	Video games (50%)
Social	Family and friend support (75%)	Family and friend support (73%)	Socializing with friends (67%)	None reported	None reported	None reported
	Family T2D status (31%)	Family T2D status (33%)
Environment	School (69%)	Gyms (67%)	School (56%)	Home (67%)	Auditory noises (56%)	Environment disturbances (53%)
	Parks (44%)	Workplace (67%)	Home (38%)	Workplace (27%)	Screen time (44%)	Screen time (40%)

Individual Determinants.

Among adolescents, the most common types of PA included PA during physical education class (69%) or walking (43%). When prompted to describe factors that motivate them to be active, half (47%) of adolescents desired to improve their physical appearance. Eight youth shared that they experienced bullying or struggled with body image.

When my friends told me, they’re like, “Oh you should join cheer.” I’m like, “Do you really think this body would fit in a cheer outfit. And I feel uncomfortable when I’m wearing something short … I don’t like showing my body that much.”

(Female adolescent)

… but some friends at school. They always treat me badly, “Oh you’re fat, you’re fat.” … There’s no day that he hasn’t called me fat.

(Male adolescent)

Walking was also one of the most common types of PA (67%) among young adults, followed by working out at a gym (47%), or being active at work (40%). Young adults were more likely to be motivated by having someone to exercise with (47%) and a desire to improve their health (33%).

I think what kind of keeps me on track too is that I have friends that are also really active. We’ll go to the gym a lot, they’re always inviting me to do things.

(Female young adult)

Just for myself, you know? Like self-care, it’s like my number one thing. I’m just trying to make positive changes in my life or for my body at least.

(Male young adult)

Social Determinants.

Most adolescents (75%) perceived social support from friends and family, particularly from one’s mother, as a social determinant of PA.

I feel that now, because they [participant’s friends] know that I’m trying to have a change in my lifestyle, and the way that I eat, they motivate me. … Like, they encourage me.

(Female adolescent)

… cause I tell her [participant’s mom] I don’t feel good, ‘cause you can feel the weight difference. … And she was like, “You know, you should do it. I know you can do it.”

(Male adolescent)

Most young adults (73%) also perceived social support from friends and family, particularly from one’s mother, as a social determinant of PA.

For sure, [my friends] going to the gym with me. If it fits with our schedules, we can go. If it doesn’t fit with our schedules, they always tell me to go anyways.

(Female young adult)

It does make me feel good because she’s [participant’s mom] my support system. She gives me a lot of advice, she’s just my support.

(Male young adult)

Within the family context, the disease status of family members was discussed by a third of adolescents (31%) and young adults (33%) as a factor that influenced their PA.

Because my dad, my grandma, my mom all have diabetes. … They really care about that [being active] because they don’t want me to go through every day poking my fingers to check my blood sugar.

(Female adolescent)

What does motivate me is my mom because she does have diabetes. Both my mom and my grandma have diabetes … It’s really scary.

(Female young adult)

Environmental Determinants.

The majority of adolescents (69%) perceived school as the primary environment where they are active, followed by parks (44%), and their neighborhood (31%).

My friend has a bike and we been going on the bikes, and we play basketball. … His neighbor has a basketball court.

(Male adolescent)

In comparison, young adults reported being most active at pay for use gyms (67%), their workplace (67%), or local park (40%).

I like to go to the gym a lot, honestly. It helps me relax and I get to listen to my music.

(Male young adult)

Well, we have to walk at least a mile from the parking lot to work … then you got 60 pounds on you of tools. … Then you have to go up six floors of stairs … I mean just work is where I’m the most active.

(Male young adult)

When discussing PA environments, participants in both age groups expressed safety concerns regarding neighborhood PA resources.

To feel more safe walking around. … Because everybody there that lives close to me drinks a lot, and they’re usually outside always.

(Female adolescent)

Since there’s nothing close enough, there’s nothing you can do. … Then where I live isn’t as safe as I would really hope it could be. There’s been shootings. … That’s another downside of why I’m not going out.

(Female young adult)

Determinants of Sedentary Behaviors

Individual Determinants.

The most common SB discussed among adolescents was screen time, which often included streaming (i.e., Netflix, YouTube; 63%) or video games (50%).

I’m pretty lazy, so when I have nothing else to do, I usually just play my video games in front of my TV for a while, until the afternoon …

(Male adolescent)

I was watching a show on Netflix for too long, and I watched the whole series. It was 14 episodes, and I watched all 14 episodes in one thing.

(Female adolescent)

Similarly, streaming (67%) was the most commonly discussed SB among young adults.

Once you get the catch of the first episode, you’re just hooked on it for the rest of the episodes. It’s like 20 episodes so you just spend your days binging on the shows.

(Female young adult)

Social Determinants.

When prompted to discuss their motivation for screen time, about half of youth (56%) saw it as an opportunity to socialize with friends.

I like being on social media ‘cause that’s the only way I can talk to my friends …

(Female adolescent)

The same friends that I see at school, I also play with them online and stuff and we usually play together on the Xbox and stuff.

(Male adolescent)

In contrast, almost all young adults (93%) reported being alone when streaming and no social determinants of screen time were identified among young adults.

Environmental Determinants.

About half of adolescents (56%) felt most sedentary at school, followed by the home environment (38%).

I just feel like I sit too much at school …’Cause I only have one period, its advanced PE, that’s pretty much it. That’s all I do and the rest are just sitting.

(Male adolescent)

Similarly, most young adults (67%) felt most sedentary at home; followed by the workplace (27%).

Too much sitting, work, for sure. It’s the number one. No way around it. … I’m sitting down mostly at work.

(Male young adult)

Determinants of Sleep

Individual Determinants.

Most adolescents (75%) perceived themselves as unaffected by sleep. However, short sleep durations negatively affected their mood and led to feelings of tiredness.

I’m very tired, and I’m in a bad mood. And when I go to school, I would always get a ton of negatives.

(Female adolescent)

Some young adults (33%) acknowledged that they do not get enough sleep, while about half (47%) also perceived themselves as unaffected by sleep. Similar to adolescents, a short night of sleep also impaired mood and led to feelings of tiredness.

I feel like I’m more annoyed with people … I just don’t wanna like care … in my head I’m like “just be quiet.”

(Female young adult)

If I don’t get a good night’s sleep, my body feels like very drained. I feel like I have to sleep again.

(Male young adult)

Environmental Determinants.

About half of adolescents (56%) discussed auditory noises in their home environments as determinants that inhibit sleep.

My neighbor, they have a German Shepherd called Max, he barks nonstop.

(Male adolescent)

When people stay up late and they’re blasting their music, that’s what I hate the most.

(Female adolescent)

Likewise, 53% of young adults reported environmental disturbances including auditory noises, lights, and feelings of safety.

Our neighbor has his backyard lights on and then it reflects into my. … Well it goes in through my window and it reflects on my mirror.

(Female young adult)

I mean, just where I live in general is a bad area. … Nothing’s happened to my family, so I sleep a little bit good at night. But just knowing that I am in the place that I am, keeps me kind of on my edge.

(Male young adult)

Almost half of youth (44%) and young adults (40%) reported that they engaged in screen time before bed, which led to shorter sleep durations.

Well I go to my room and stay there for an hour and be on my phone til it runs out of battery, and then I go to sleep.

(Female adolescent)

The same thing with Netflix. I stay up late watching the shows and I go to sleep late. Then I wake up early in the morning to go to work.

(Female young adult)

Discussion

The 24-hour paradigm holds that to reduce disparities like T2D among Latino adolescents and young adults, there is a need for prevention strategies that focus comprehensively on 24-hour activity and sleep behaviors (Rosenberger et al., 2019). However, in order to develop 24-hour activity and sleep prevention strategies, there is a critical need to increase our understanding of 24-hour activity and sleep behaviors and determinants, particularly in high-risk populations like Latino adolescents and young adults. Thus, the primary aim of this study was to use accelerometry to rigorously assess 24-hour activity and sleep behaviors and their impact on T2D risk outcomes in Latino adolescents and young adults with obesity. The secondary aim was to use qualitative interviews to contextualize the individual, social, and environmental determinants of 24-hour activity and sleep behaviors.

Adolescents in this population met PA recommendations of 60 minutes per day (National Physical Activity Plan Alliance, 2018). The level of PA observed in adolescents in this sample was higher compared with adolescents in the Community Health Study/Study of Latinos (SOL), the largest epidemiologic study among Latinos (Evenson et al., 2019). Yet, our findings are consistent with other accelerometry-based studies reporting that adolescents with obesity achieve 60 minutes of daily PA (Adams et al., 2009; Butte et al., 2007). Of note, these studies as well as behaviors assessed in the SOL study were measured using hip-worn accelerometers and we should use caution in comparing data from hip-worn devices to wrist-worn devices (Kerr et al., 2017).

Adolescents engaged in more PA than young adults, which was expected given that PA declines during adolescence through young adulthood (Troiano et al., 2008). However, young adults in this sample also met recommendations of 150 minutes of MVPA per week (U.S. Department of Health and Human Services, 2018). Levels of PA observed in young adults in this study were comparable to young adults in the SOL study (Arredondo et al., 2016; Evenson et al., 2019). According to our qualitative findings, future PA promotion strategies among Latino adolescents and young adults should target social support to increase PA. A large body evidence has demonstrated that support through modeling, information, and engagement, can increase PA, which is supported in our findings (Donnelly & Springer, 2015). Future studies should also address the presence of T2D in family members. Parental stress from living with chronic conditions is associated with increased adiposity and poor health behaviors in youth (Allport et al., 2019). This also affected young adults in this study as most (73%) still lived with their parents. Given the need for social support and the impact of T2D within a family, these data support the use of peer- and family-based programming that equips friends and family members with the skills and knowledge needed to model and support each other through health behavior changes (Davison et al., 2013; Teufel-Shone et al., 2005).

PA strategies focused on Latino adolescents, should also address emotional health and well-being given that youth experienced weight-based bullying and struggled with body image. Weight-based stigmatization can impair psychological wellbeing, lead to social isolation, and unhealthy coping strategies (Puhl & King, 2013). Our previous work has shown that lifestyle interventions that address emotional health and well-being can lead to long-term improvements in weight-specific quality of life in Latino adolescents with obesity (Soltero et al., 2018). Last, our study showed that lack of safe, neighborhood PA resources limited opportunities for PA in both age groups. Future lifestyle interventions must consider these barriers to ensure that participants are able to safely and effectively engage in behavior change strategies.

Adolescents and young adults in this sample averaged ≥10 hours of sedentary time per day, which is consistent with levels of SB reported in the SOL study (Arredondo et al., 2016). Screen time, the most prevalent SB discussed in our sample, has been identified as particularly detrimental for glycemic control as it promotes mechanical unloading and is often accompanied by snacking (Dempsey et al., 2016; Henson et al., 2016). It has been hypothesized that the protective effect of PA is diminished in the presence of excessive time spent in SB and studies have shown that even when PA recommendations are met, chronic disease risk is not reduced in the presence of excess time in SB (Dunstan et al., 2012; Strizich et al., 2018). These findings may explain, in part, why participants in this study achieved PA recommendations, but were still at high-risk for T2D as demonstrated by the BMI and BMI% which were classified as obese and with 2 hour and fasting glucose values that fell just below the American Diabetes Association criteria for prediabetes (fasting glucose ≥ 100 mg/dl or 2-hour glucose ≥ 140 mg/dl; American Diabetes Association, 2021).

Intervention strategies for reducing SB have been challenging due to the lack of intervention targets and guidelines for meaningful SB reductions (Salmon et al., 2011). Studies using isotemporal substitution techniques have shown that replacing as little as 30 minutes of SB with MVPA is a feasible behavioral target associated with significant improvements in BMI, fasting glucose, and fasting insulin in youth (Li et al., 2018; Rosenberger et al., 2019). While screen time significantly contributes to SB, screen devices can be leveraged to promote PA and reduce SB while still fostering social support, which was a determinant of screen time among adolescents (Williams & Ayres, 2020). For example, exergaming uses video games to increase PA while still allowing players to socialize and connect online (Williams & Ayres, 2020). More studies are needed to examine the use of technology to satisfy the need for social support while improving healthy PA behaviors and reducing SB.

On average, adolescents were less likely to meet sleep recommendations of 8 to 10 hours of sleep (Hirshkowitz et al., 2015). In contrast, young adults were more likely to achieve sleep recommendations of 7 or more hours of sleep (Hirshkowitz et al., 2015). While short sleep durations are associated with insulin resistance, the underlying mechanisms linking insufficient sleep with T2D are largely unknown (Dutil & Chaput, 2017). Sleep quality is more indicative of the restorative aspects of sleep such as time spent in deep sleep or Stage 3 wave activity and is associated with glucose metabolism (Ukraintseva et al., 2020; Zhu et al., 2015). In this study, we found that irregular wake and sleep cycles, which is reflective of poor sleep quality, were correlated with increased measures of adiposity in both age groups and increased T2D risk in young adults. Wake and sleep cycles are a behavioral output of the circadian clock and is hypothesized that one of the mechanism by which irregular wake and sleep cycles affect health outcomes is through circadian misalignment (Johnson et al., 2018). Our findings among young adults, suggest that over time, the lack of quality, deep sleep may lead to chronic circadian misalignment and impaired glucose tolerance.

Our findings suggest a need for strategies to improve sleep duration and quality; however, there is a dearth of research on sleep interventions, particularly among high-risk minority populations (Johnson et al., 2018). Many youth and young adults perceived themselves as unaffected by sleep. Given our finding that sleep irregularity over time may lead to impaired glucose tolerance, there is a need for sleep interventions early in childhood. These interventions should stride to educate youth and parents on the importance of sleep and establishing regular sleep patterns to improve mood and provide energy for other health behaviors like PA (Phillips et al., 2017). Sleep interventions should also provide strategies for developing sleep habits and routines, such as a sleep schedule and sleep management plan (Hiscock et al., 2019). Strategies to improve sleep should also focus on setting a sleep environment to help youth overcome sleep disturbances (i.e., lights, auditory noises) that were addressed in our qualitative work such as use of shades, temperature control, and restricting lighting, especially LED lighting from smartphones prior to bedtime (Rossen et al., 2020). Additionally, cognitive–behavioral approaches that incorporate cognitive restructuring through mindfulness practices along with sleep education and sleep hygiene can be used to increase sleep among high-risk youth (Blake et al., 2018).

The paradigm shift toward 24-hour activity and sleep behaviors requires the use of objective measures like accelerometry to comprehensively assess behaviors across the 24-hour cycle. This is one of the first studies to assess and report 24-hour activity and sleep behaviors in Hispanic youth and young adults. This study also provided qualitative insights confirming existing determinants of behaviors such as social support. Findings also highlighted novel determinants including the association among social support and SB as well as environmental determinants of sleep. The next step in research regarding the 24-hour activity and sleep paradigm will require the use of more sophisticated statistical techniques to examine the combined effect of 24-hour behaviors on health and disease outcomes (Chastin et al., 2015). Informed by contextual determinants of behaviors, intervention strategies should simultaneously target multiple behaviors across the 24-hour cycle. These advancements will expand our knowledge of behavioral factors that contribute to the development of T2D and will lead to more effective prevention strategies.

Strengths of this study include the use of accelerometers to objectively assess behaviors and the use of qualitative methodology to identify novel behavioral determinants. Limitations include the small sample size, which may have limited our ability to find significant relationships between health behaviors and disease outcomes. This study is also limited by placement of the accelerometer on the wrist, which may have overestimated PA levels compared with hip-worn devices (McCurley et al., 2017). Furthermore, there is limited data on validated algorithms for assessing activity from high-frequency, triaxial accelerometers, like the GT3X-BT, worn at the wrist (Cleland et al., 2013). Last, this study is limited by the cross-sectional design, which limits our ability to infer causality (Narayan et al., 2003).

Conclusion

This mixed-method study examined 24-hour activity and sleep behaviors in Latino adolescents and young adults and their impact on diabetes related outcomes as well as the individual, social, and environmental determinants that drive these behaviors. While most youth and young adults achieved PA recommendations, they spent excessive time (≥10 hours/day) in SB, which may contribute to high risk of T2D as was observed in this population of Hispanic youth and young adults. We also found that irregular wake-sleep patterns were associated with higher levels of adiposity in both age groups and greater T2D risk in young adults suggesting a need for sleep focused interventions, particularly among youth. Our data confirm the importance of social support for PA while highlighting the influence of T2D within the family as a determinant that should be addressed in future interventions. Our study also revealed that social support, emotional health, safety, and access to PA resources, are common determinants of SB, particularly social support received through screen time. Our study also revealed new insights on environmental determinants of sleep that should be addressed at sleep interventions aimed at improving sleep behaviors in this population. Moving forward, investigators should continue to comprehensively assess 24-hour activity and sleep behaviors to increase our understanding of how these behaviors contribute to the development of T2D.