Revisiting sleep duration and mental distress among US high school students: National YRBS, 2011–2023

Abstract

Background

Adolescent mental distress is a prevalent global public health concern. Aligning with the goals of Healthy People 2030, this study aimed to characterize the sleep duration associated with the lowest prevalence of mental distress among US high school students.

Methods

This study pooled and examined cross-sectional data from 7 waves of the national Youth Risk Behavior Survey (YRBS) in 2011–2023. All data analyses were sex-stratified. We estimated the prevalence of mental symptoms and suicidal behaviors across sleep duration, and examined both linear and quadratic associations. We also calculated adjusted odds ratios (AORs) with 95% CIs and compared the prevalence of mental distress indicators among sleep duration categories.

Results

Overall, 24.39% of adolescents reported 8–9 h of sleep (female: 22.51%; male: 26.21%). Compared with 8–9 h, insufficient sleep (< 8 h) was associated with significantly higher odds of all five mental distress indicators among female students: current bad mental health (AOR, 1.93; 95% CI, 1.61–2.30), feeling sad/hopeless (AOR, 1.59; 95% CI, 1.40–1.80), suicidal thoughts (AOR, 1.68; 95% CI, 1.49–1.91), suicide plan (AOR, 1.41; 95% CI, 1.25–1.59), and attempted suicide (AOR, 1.41; 95% CI, 1.20–1.66). Results were similar among male students. Across all five indicators in both sexes, the lowest prevalence was observed at 8–9 h; however, most comparisons at ≥ 10 h were not statistically significant.

Conclusions

The lowest prevalence of mental distress across all indicators and both sexes was observed at 8–9 h of sleep; however, most ≥ 10-hour comparisons were not significant, likely reflecting the small proportion of adolescents in this category.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12916-026-05074-6.

Background

Improving adolescent well-being is an urgent global priority [1] to achieve the Sustainable Development Goals. One in six people worldwide are adolescents aged 10–19 years [2]. And one in seven adolescents globally experiences mental distress, which contributes to 15% of the disease burden in this age group [2]. Adolescent mental health conditions, if left unaddressed, may compromise adult physical and mental well-being [3]. The benefit-to-cost ratio in adolescent health ranges from 5 to 10 [4], supporting large-scale investment in this neglected age group. Multiple indicators of adolescent mental distress, according to the Youth Risk Behavior Survey (YRBS) explore [5], demonstrated a sustained upward trend in the US. For example, the prevalence of persistent feelings of sadness or hopelessness increased from 29.9% in 2013 to 39.7% in 2023—a 9.8%-point rise over the past decade. The 2023 national YRBS [6] found that 28.5% reported poor mental health. Furthermore, 20.4% seriously considered suicide, and 9.5% had attempted suicide (Healthy People 2030 target is 1.8% [7]).

Humans need sufficient sleep to stay healthy. The relationship between sleep and mental health has garnered considerable attention in recent years. Healthy People 2030 establishes 355 core objectives, including increasing the proportion of high school students who get sufficient sleep (≥ 8 h) [8]. The YRBS introduced a measure of sleep duration in 2007 to monitor adolescent sleep, following the 2015 [9] and 2016 [10] recommendations for adolescents. According to YRBS Explorer, the prevalence of US adolescents sleeping 8 or more hours daily declined significantly from 31.7% in 2013 to 23.2% in 2023—an 8.5%-point reduction over this decade. Category 10 or more hours was not emphasized in most analyses due to its low prevalence (< 3%) [11].

Insufficient sleep has been defined as a strong factor associated with mental distress [6, 12–14], and safety risk behaviors [15]. Also, there are U-shaped associations of sleep duration with various adolescent health consequences, including obesity [16], biological aging [17], depression [18], and suicidality [19]. However, a meta-analysis [20] found that short sleep duration, not long sleep duration, is associated with mental disorders. The contradictory findings suggest the need for further research to understand the underlying reasons. A global study [21] indicates that healthy sleep durations vary across countries; hence, it is crucial to re-examine these associations within a specific country using robust statistical methods.

This study uses data from the national YRBS, which provides consistent indicators and standardized collection protocols across multiple waves within a nationally representative sample, minimizes potential biases in previous meta-analyses, and offers a more rigorous evaluation of the relationship between sleep duration and mental health. Specifically, the objectives of this study were to characterize the sleep duration associated with the lowest prevalence of various mental distress indicators among US high school students.

Methods

Study design and participants

The YRBS is a school-based cross-sectional survey conducted biennially since 1991 among representative samples of high school students in grades 9–12 at national, state, and large urban school districts in the United States. Questionnaires were administered in the classroom during one class period. Students completed their responses directly on a computer-scannable questionnaire. Detailed information about the YRBS methodology is available elsewhere [22].

The sleep duration was introduced into YRBS in 2007, bullying victimization was added in 2009, and bullying victimization and physical activity were added in 2011. We used 7 waves of national YRBS data [22], including the main sample and supplementary American Indian or Alaska Native sample, from 2011 to 2023 (94,338 students) for our analysis (Additional file 1: Table S1), because bullying, cyberbullying, and physical activity are important covariates for addressing the association between sleep duration and mental distress.

Student participation in the YRBS was anonymous and voluntary, and local procedures were used to obtain parental permission. The Institutional Review Board of the Centers for Disease Control and Prevention (CDC) approved the national YRBS [22]. Ethical approval was not required for this secondary analysis as the study used existing openly available and de-identified YRBS datasets.

Measures

The YRBS measures sleep duration and various indicators of mental distress (Table 1). The mental distress indicators included feeling sad, suicidal thoughts, suicide plans, and suicide attempts, as suggested by a previous study [23], as well as current mental health, which was added in YRBS 2021.

Table 1.

Youth Risk Behavior Survey (YRBS) Questionnaire Items and Analytic Coding

Variables Assessed	YRBS Questionnaire Item	Analytic Coding
Exposure
Sleep duration	On an average school night, how many hours of sleep do you get?	≤ 4, 5, 6, 7, 8, 9,≥10
Outcome
Mental distress
Current bad mental health a	During the past 30 days, how often was your mental health not good? (Poor mental health includes stress, anxiety, and depression.)	Most of the time/Always vs. Never/Rarely/Sometimes
Feeling sad/hopeless	During the past 12 months, did you ever feel so sad or hopeless almost every day for 2 weeks or more in a row that you stopped doing some usual activities?	Yes vs. no
Seriously considered attempting suicide	During the past 12 months, did you ever seriously consider attempting suicide	Yes vs. no
Made a suicide plan	During the past 12 months, did you make a plan about how you would attempt suicide?	Yes vs. no
Attempted suicide	During the past 12 months, how many times did you actually attempt suicide?	≥ 1 vs. 0
Covariates
Aerobic physical activity	During the past 7 days, on how many days were you physically active for a total of at least 60 min per day? (Add up all the time you spent in any kind of physical activity that increased your heart rate and made you breathe hard some of the time.)	7 days vs. 0–6 days
Current cigarette use	During the past 30 days, on how many days did you smoke cigarettes?	≥ 1 days vs. 0 days
Current alcohol use	During the past 30 days, on how many days did you have at least one drink of alcohol?	≥ 1 days vs. 0 days
Current marijuana use	During the past 30 days, on how many days did you use marijuana?	≥ 1 days vs. 0 days
Ever cocaine use	During your life, how many times have you used any form of cocaine, including powder, crack, or freebase?	≥ 1 times vs. 0 times
Bullying victimization	During the past 12 months, have you ever been bullied on school property?	Yes vs. no
	During the past 12 months, have you ever been electronically bullied? (Count being bullied through texting, Instagram, Facebook, or other social media.)
Sexual trauma	Have you ever been physically forced to have sexual intercourse when you did not want to?	Yes vs. no
	During the past 12 months, how many times did someone you were dating or going out with physically hurt you on purpose? (Count such things as being hit, slammed into something, or injured with an object or weapon.
	During the past 12 months, how many times did someone you were dating or going out with force you to do sexual things that you did not want to do? (Count such things as kissing, touching, or being physically forced to have sexual intercourse.)
Sexual identity	Which of the following best describes you?	Heterosexual, Sexual Minority, Other/Questioning

^aData were pooled from the YRBS 2021–2023, because current mental health item was added in YRBS 2021

Adolescent sleep duration in our study was categorized into 3 categories: <8 h, 8 to 9 h, and > = 10 h. Generally, the response rates were about 80% for sleep hours and over 95% for most mental health indicators.

Demographic characteristics assessed in this study included sex (female or male), race/ethnicity (White, Black, Hispanic, other), and grade (9, 10, 11, or 12). Based on previous literature [23, 24], we also included other binary controlled variables (Table 1), including aerobic physical activity (being physically active for ≥ 60 min per day on all 7 days), current cigarette use, currently drank alcohol, current marijuana use, ever cocaine use, bullying victimization(on school property or online), and sexual trauma (sexual assault, physical and sexual intimate partner violence). Bullying victimization and sexual trauma were two computed binary variables. Bullying victimization was recoded as “yes” if either item was endorsed as “yes”, and “no” if both items answered “no”. Similarly, sexual trauma was recoded as “yes” if any item was endorsed as “yes”, and “no” if all three items answered “no”.

Statistical analysis

We adjusted for complex sampling and survey weights for all the analyses using the suite of svy commands in Stata/SE 18.0 (StataCorp LLC). We followed the CDC’s best practices for combining YRBS data across years and sites [25] and didn’t divide the weights by the number of combined datasets when combining the 7 waves of national YRBS data. The statistical significance cutoff threshold in our study was 0.05 (two-tailed). Complete case analyses were used.

Prevalence estimates with 95% CIs were calculated using Taylor series linearization [26]. Differences in the distribution of sleep duration across demographic characteristics were tested using Rao-Scott corrected χ2 statistics [27]. All associations of sleep duration with mental distress were stratified by sex as suggested by previous studies [26, 28]. In initial analyses of association, sleep duration was treated as an ordinal variable ranging from 1 (≤ 4 h) to 7 (≥ 10 h); logistic regression analysis was used to assess both linear and quadratic associations of sleep duration with mental distress by adjusting for survey year, race/ethnicity, and grade. In subsequent categorical analyses, logistic regression was used to estimate sex-stratified adjusted odds ratios (AORs) with 95% CIs, comparing the prevalence of mental distress indicators in insufficient (< 8 h) and long sleep duration (≥ 10 h) groups with a referent group of 8–9 h. To assess whether the association between sleep duration and mental distress differed by sex, we included sex × sleep duration category interaction terms in the multivariable logistic regression models, in addition to the sex-stratified analyses.

Sensitivity analyses

To evaluate whether the COVID-19 pandemic influenced the association between sleep duration and mental distress, we conducted a sensitivity analysis stratifying the sample by pre-COVID-19 (2013–2019) and post-COVID-19 (2021–2023) periods. We repeated the multivariable logistic regression models described above separately for each period. Current bad mental health was excluded from this analysis because it was introduced in YRBS 2021, precluding pre/post comparison.

Results

The study comprised 94,338 students with complete data on sleep duration in the combined YRBS 2011–2023 sample. Half were female (47,125 [49.23%]), most identified as white (44,568 [53.86%]), 23,121 (23.42%) identified as Hispanic or Latino, 13,124 (12.61%) identified as black or African American, and 11,917 (10.11%) identified as other race or ethnicity. Grade distribution was balanced (minimum [12th grade], 21,809 [23.50%], and maximum [9th grade], 24,639 [26.78%]). Data volume was consistent across survey years (minimum [YRBS 2017],11,721 [13.65%]; maximum [YRBS 2023],17,332 [17.77%]) (Additional file 1: Table S1).

Table 2 shows the distributions of sleep duration across three categories (< 8 h, 8–9 h, and ≥ 10 h). The prevalence of self-reported 8–9 h of sleep was 24.39% and varied by sex, race/ethnicity, and grade. It was higher among male students (26.21%) than among female students (22.51%), declined from 31.74% in 9th grade to 18.64% in 12th grade, and was highest among White students (26.11%) and lowest among Black students (19.93%).

Table 2.

Weighted Percentage (%) of Sleep Duration Categories by Demographic Group in YRBS 2011–2023

Demographic Group	Hours of sleep			χ2 P Value *
	< 8(N = 69,784)	8 to 9 (N = 22,903)	≥ 10 (N = 1,651)
Total population	73.97(73.31–74.62)	24.39(23.76–25.03)	1.64(1.54–1.76)	NA
Sex
Female	76.01(75.27–76.74)	22.51(21.81–23.23)	1.48(1.33–1.64)	< 0.001
Male	71.99(71.17–72.80)	26.21(25.43-27.00)	1.80(1.65–1.97)
Race/ethnicity
White	72.63(71.84–73.41)	26.11(25.35–26.89)	1.25(1.13–1.39)	< 0.001
Black	77.59(76.12–78.99)	19.93(18.72–21.20)	2.48(2.10–2.92)
Hispanic	73.68(72.60-74.74)	24.41(23.39–25.47)	1.90(1.70–2.13)
Other	78.28(76.98–79.53)	20.06(18.90-21.27)	1.66(1.35–2.04)
Grade
9	65.93(64.89–66.95)	31.74(30.75–32.76)	2.33(2.09–2.60)	< 0.001
10	72.88(71.89–73.85)	25.29(24.35–26.26)	1.82(1.60–2.08)
11	77.89(76.91–78.85)	20.91(20.00-21.85)	1.20(1.01–1.41)
12	80.35(79.44–81.24)	18.64(17.78–19.53)	1.01(0.84–1.21)

Abbreviation: NA, not applicable; YRBS, Youth Risk Behavior Survey. ^* Rao-Scott Pearson Chi Square was used

In initial analyses, sleep duration was treated as an ordinal variable, and mental distress indicators were strongly associated with it (Table 3). Among female and male students, all the mental distress indicators, including current bad mental health, feeling sad, suicidal thoughts, suicide plans, and suicide attempts, demonstrated both a linear and quadratic association with sleep duration, with the lowest prevalence of all the mental distress indicators occurring at 8–9 h of sleep.

Table 3.

Linear and Nonlinear Associations Between Sleep duration and Mental Distress Among High School Students by Sex in YRBS 2011–2023

Mental Distress	Hours of sleep, weighted %
	≤ 4	5	6	7	8	9	≥ 10	Type of Association b
Female Students	N = 4,197	N = 6,923	N = 11,688	N = 12,949	N = 8,537	N = 2,070	N = 761
Current bad mental health a	63.23	51.59	42.34	31.19	28.25	24.26	41.03	Linear and quadratic
Feeling sad/hopeless	69.92	58.74	47.83	38.13	32.84	33.75	40.94	Linear and quadratic
Seriously considered attempting suicide	46.12	32.93	25.12	19.16	15.73	17.65	22.50	Linear and quadratic
Made a suicide plan	38.51	25.90	20.10	14.21	13.02	14.05	20.26	Linear and quadratic
Attempted suicide	26.44	15.01	10.94	7.44	7.37	7.96	14.57	Linear and quadratic
Male Students	N = 4,142	N = 5,564	N = 10,644	N = 13,677	N = 9,825	N = 2,471	N = 890
Current bad mental health a	36.46	25.79	20.85	13.55	11.05	10.49	21.66	Linear and quadratic
Feeling sad/hopeless	45.55	36.63	27.26	19.68	15.12	14.56	21.54	Linear and quadratic
Seriously considered attempting suicide	30.33	18.84	14.42	9.60	7.61	7.90	14.44	Linear and quadratic
Made a suicide plan	27.47	17.18	11.25	7.96	5.68	7.04	12.67	Linear and quadratic
Attempted suicide	20.00	7.92	5.12	3.61	2.92	4.21	12.59	Linear and quadratic

Abbreviation: YRBS, Youth Risk Behavior Survey

^aData were pooled from the YRBS 2021–2023, because current mental health item was added in YRBS 2021

^bThe linear trend test used the Stata contrast postestimation command and controlled survey year, race/ethnicity and grade, and all P values are less than 0.001

We also calculated the predicted probability of mental distress indicators being 1 for different levels of sleep duration, holding all other variables at their mean values (Fig. 1). The results indicated that among female students, those reporting 9 h of sleep had the lowest probability of experiencing sadness, suicidal ideation, suicide planning, and suicide attempts compared to peers with shorter or longer sleep durations, whereas the lowest probability of current bad mental health occurred at 8 h of sleep. Among male students, the lowest probability of experiencing current bad mental health, suicidal ideation, suicide planning, and suicide attempts occurred at 8 h of sleep, whereas the lowest probability of experiencing sadness occurred at ≥ 10 h of sleep.

Fig. 1.

Sex-stratified predicted probabilities of five mental distress indicators by sleep duration, adjusted for covariates. The controlled covariates included race/ethnicity, grade, aerobic physical activity, current cigarette use, current alcohol use, current marijuana use, ever used cocaine, bullying victimization, and sexual trauma

In categorical analyses, sleep duration was categorized into a three-level variable(< 8 h, 8–9 h, and ≥ 10 h). Compared with the 8–9 h sleep reference group, individuals of both sexes who slept < 8 h had significantly higher odds of experiencing each mental distress indicator(Table 4). However, sleep duration ≥ 10 h was not significantly associated with elevated odds of most mental distress indicators in either female (N = 761) or male (N = 890) students, although both sexes exhibited elevated prevalence across all mental distress indicators compared to the 8–9-hour sleep group. We also examined the interaction between sex and sleep duration categories in relation to mental distress indicators and found that the interaction term was significant only for “feeling sad/hopeless” and “made a suicide plan” for the 8–9 h sleep duration category (Additional file 1: Table S2).

Table 4.

Multivariable Logistic Regression of the Association of Sleep Duration Categories with Mental Distress by Sex in YRBS 2011–2023

Mental Distress	Hours of sleep
	< 8		8–9 (Reference)	≥ 10
	weighted %	AOR (95%CI, P value)	weighted %	weighted %	AOR (95%CI, P value)
Female Students	N = 35,757		N = 10,607	N = 761
Current bad mental health a	43.03	1.93(1.61–2.30, < 0.001)	27.46	41.03	1.78(0.91–3.47, 0.092)
Feeling sad/hopeless	48.89	1.59(1.40–1.80, < 0.001)	33.01	40.94	1.22(0.85–1.74, 0.279)
Seriously considered attempting suicide	26.82	1.68(1.49–1.91, < 0.001)	16.09	22.50	1.01(0.69–1.49, 0.958)
Made a suicide plan	21.14	1.41(1.25–1.59, < 0.001)	13.21	20.26	1.25(0.84–1.87, 0.262)
Attempted suicide	12.15	1.41(1.20–1.66, < 0.001)	7.48	14.57	1.26(0.79-2.00, 0.328)
Male Students	N = 34,027		N = 12,296	N = 890
Current bad mental health a	20.80	2.00(1.55–2.59, < 0.001)	10.94	21.66	2.46(1.20–5.03, 0.014)
Feeling sad/hopeless	27.75	1.90(1.66–2.17, < 0.001)	15.01	21.54	0.83(0.56–1.24, 0.366)
Seriously considered attempting suicide	14.96	1.73(1.49–2.02, < 0.001)	7.67	14.44	0.94(0.60–1.49, 0.805)
Made a suicide plan	12.67	2.02(1.66–2.46, < 0.001)	5.95	12.67	1.12(0.68–1.85, 0.646)
Attempted suicide	6.52	1.69(1.29–2.22, < 0.001)	3.17	12.59	1.21(0.58–2.52, 0.618)

Abbreviation: YRBS, Youth Risk Behavior Survey; AOR, adjusted odds ratio (adjusted for survey year, race/ethnicity, grade, aerobic physical activity, current cigarette use, current alcohol use, current marijuana use, ever cocaine use, bullying victimization, and sexual trauma)

^aData were pooled from the YRBS 2021–2023, because current mental health item was added in YRBS 2021

We stratified the sample into pre-COVID-19 (2013–2019) and post-COVID-19 (2021–2023) periods to assess whether the pandemic altered the sleep duration–mental distress association. Insufficient sleep was significantly associated with elevated odds of every mental distress outcome in both sexes and both periods, with one exception: the association between insufficient sleep and attempted suicide among male students was significant pre-COVID-19 (AOR, 1.87; 95% CI, 1.35–2.60; P < 0.001) but nonsignificant post-COVID-19 (AOR, 1.52; 95% CI, 0.95–2.44; P = 0.081). Comparisons in the ≥ 10-hour category were mostly non-significant across both periods and sexes (Additional file 1: Table S3).

Discussion

Our study supports that insufficient sleep duration (< 8 h) was consistently associated with elevated odds of mental distress across all five indicators in both sexes. The pattern for long sleep duration (≥ 10 h) was directionally similar but did not reach statistical significance, likely reflecting the small proportion of adolescents in this category (1.64%). Taken together, our findings suggest that within the existing 8-10-hour recommendation [7, 8], the 8–9 h range was associated with the lowest prevalence of mental distress.

Most studies indicate that insufficient sleep duration is negatively associated with adolescent mental distress [3, 13, 29], whereas a minority report a positive association [12]. Our findings strengthen the evidence-based consensus that insufficient sleep duration is associated with elevated risks for depression [30], psychological distress [30], and suicide [31]. Insufficient sleep duration may compromise mental health by increasing susceptibility [32], altering emotional response processing [33], and impairing neurocognitive function [34].

The effects of long sleep duration on adolescent mental distress lack consensus due to conflicting but equally robust bodies of evidence. Our findings indicated that adolescents with long sleep duration reported a higher prevalence of every type of mental distress in both sexes. Long sleep duration may reduce protective factors against depression, including physical activity [35], social contact [36], sunlight exposure [37, 38], and greenness exposure [39]. In turn, this reduction elevates depression risk. Most of the existing studies with conflicting results on long sleep duration and mental distress were conducted among young adults [20, 30, 40], and one study based on YRBS 2007–2019 [41] indicates the U-shaped associations between sleep duration and adolescent suicidality. Given that only 1.64% of adolescents in our study slept 10 h or more daily, the resulting small sample size made the association between long sleep duration and adolescent mental distress statistically non-significant, necessitating larger, dedicated studies to validate these results.

Despite the higher prevalence of insufficient sleep and all mental health outcomes among female students, the interaction between sex and sleep duration was significant for only two of the examined mental health indicators. This pattern indicates that the association between sleep duration and mental health operates in a largely similar manner across sexes [11], even though females experience a higher overall burden of both sleep problems and psychological distress. Our findings support universal sleep health interventions for adolescents regardless of sex.

The sensitivity analysis stratifying by pre- and post-COVID-19 periods showed broadly consistent associations, suggesting that the pandemic did not fundamentally alter the sleep duration–mental distress relationship. The one exception — attempted suicide among male students — likely reflects reduced statistical power in the smaller post-COVID-19 sample rather than a meaningful change, as the point estimates remained similar.

Our study has important implications for promoting adolescent mental health. Substantial gaps persist in achieving the Healthy People 2030 objectives. Healthy People 2030 aims to increase the proportion of sufficient sleep from 25.4% in 2017 to 27.4% (SH-04) [8], and to reduce the proportion of suicide attempts from 7.4% in 2017 to 1.8% (MHMD-02) [7] among adolescents. The most recent data in 2021 indicates it’s getting worse: the proportion was 22.7% for insufficient sleep and 10.2% for suicide attempt. Our study underscores the importance of sufficient sleep for adolescent mental health and suggests that, within the existing 8–10 h recommendation, the 8–9 h range may warrant particular attention, as it is associated with the lowest prevalence of mental distress.

Limitations

Our study has some limitations. First, the data from the YRBS are cross-sectional, which prevents us from testing causal associations of sleep duration with adolescent mental distress. Second, sleep duration was measured in the YRBS via self-report, which may not necessarily mirror objective sleep duration [41]. Third, the YRBS only measures sleep duration, other sleep indicators (e.g., napping duration, chronotype, and social jetlag) [12] are all closely associated with mental distress. Fourth, the extent of underreporting of self-reported stigmatized health behaviors and mental distress indicators cannot be determined [26]; however, the YRBS questionnaire items demonstrate good test-retest reliability on health risk behaviors and mental distress indicators in grades 9–12 [42, 43]. Fifth, residual confounding from unmeasured variables cannot be ruled out because the YRBS does not collect data on several potentially important confounders, including socioeconomic status, school start times, mental health history, parental support, and comorbid medical conditions.

Conclusions

Our data suggest a pattern consistent with a U-shaped association between sleep duration and adolescent mental distress among female and male students, with 8–9 h per day associated with the lowest prevalence of mental distress across all five indicators in both sexes; however, most comparisons for ≥ 10 h were not significant. Further research will be needed to establish whether the associations are causal, whether they hold across different countries, and whether they remain consistent when measured objectively.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Q.L. developed the conception and design of the work and had full access to all the data in the study. J.Z. and F.X. drafted the first version of the manuscript. W.Z. conducted the data analysis. M.B. and Z.Z. prepared the tables and figures and revised the manuscript. All authors read and approved the final manuscript.

Funding

None.

Data availability

The YRBS data used and/or analyzed during the current study is available from the US Centers for Disease Control and Prevention at https://www.cdc.gov/healthyyouth/data/yrbs/index.htm.

Declarations

Competing interests

The authors declare no competing interests.

Consent for publication

Not applicable.

Ethics Approval and Consent to Participate

Not applicable. Secondary data analyses of the YRBS data do not require Institutional Review Board approval because the data are publicly available and de-identified.