All demographic characteristics were based on parent report as collected in the NHIS Sample Child interview and included the child’s sex, age group, race/ethnicity (dichotomized as non-Hispanic white vs all other races and ethnicities combined due to limited sample size for more detailed subgroups), family income relative to federal poverty level, and highest parental education attainment among residential parents. ADHD diagnosis was also collected in the NHIS Sample Child interview and was based on affirmative report to a set of survey questions about whether the child ever received an ADHD diagnosis from a health care provider, and if yes, a follow-up question asked whether the child currently has ADHD.

We based our examination of health behavior and social-emotional well-being entirely from teen report in NHIS-Teen, and descriptions of measures included in analyses are outlined in Table 1. Briefly, indicators of difficulties making friends, lack of social and emotional support, peer support, parent support, or community support, bullying (as a victim or perpetrator), mood disorder symptoms for depression and anxiety, low life satisfaction, unmet mental health treatment needs, sleep behaviors, weekday screen time use (excluding schoolwork), physical activity, and weight concerns were included in analyses. Our study represents a secondary analysis of a previously collected, de-identified dataset.

We stratified the sample’s demographic characteristics by parent-reported current ADHD diagnoses. Next, we stratified and examined health behavior and well-being factors by current ADHD diagnoses. In addition, we examined associations between health behavior and well-being factors and current ADHD after adjusting for all demographic characteristics. Wald χ² tests with an α of 0.05 were used to define statistically significant differences across sociodemographic factors. Prevalence estimates with Korn-Gruabard 95% confidence intervals (CIs) and prevalence ratios (aPR) with 95% CI adjusting for teen age, sex, and family income were calculated. All estimates shown meet NCHS data presentation standards for proportions (www.cdc.gov/nchs/data/series/sr_02/sr02_175.pdf).

Analyses were conducted using SAS-callable SUDAAN version 11.0.1 (RTI International; Cary, NC). We used NHIS-Teen sampling weights in all analyses. Sample weights account for parent permission, teen non-response, and were post-stratified to population totals obtained from the NHIS allowing us to calculate nationally representative estimates (www.cdc.gov/nchs/data/nhis/teen/nhis-teen-weighting-18m-report.pdf). Item-level missingness was low within the analytic sample; 0.6% of responses were missing for ADHD diagnosis, and missing responses for health behavior and well-being factors ranged from 0% to 4.6%. Less than 1% of responses were missing for sex, and 3.2% were missing for highest parental educational attainment. NHIS imputed income files (ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/2022/NHIS2022-imputation-techdoc-508.pdf) were used, which use multiple imputation methodology to account for when family income was missing (10.8%). The analytic sample size was n = 1169; respondents with missing data for ADHD diagnosis (n = 7) were excluded from analyses, and those with missing data for health behavior and well-being factors were excluded from respective item-level analyses for associations with ADHD.

Among teens ages 12 to 17 years, 10.4% had an ADHD diagnosis. Teens with ADHD significantly differed from teens without ADHD by sex, age group, and race and Hispanic origin; teens with ADHD were more likely to be male (64.4%), aged 15 to 17 years (61.5%), and non-Hispanic White (66.8%; Table 2). Family income and highest parental education attained were not statistically different between those with and without ADHD.

Teens with ADHD more often reported having difficulties making friends compared with those without ADHD (23.6% vs 13.3%, aPR = 1.83, CI = 1.15–2.90; Table 3). By contrast, fewer teens with ADHD reported a lack of peer support (35.7%) compared with those without ADHD (51.5%) (aPR = 0.70, CI = 0.51–0.96). Lack of parental support was higher among teens with ADHD and those without ADHD (43.4% vs 35.1%) but not statistically significant (aPR = 1.29, CI = 0.97–1.71). Lack of social and emotional support did not differ significantly between teens with ADHD and those without ADHD (43.5% vs 41.3%) nor did community support (15.9% vs 19.8%). Compared with those without ADHD, teens with ADHD more often experienced bullying victimization (49.1% vs 31.6%; aPR = 1.64, CI = 1.27–2.11); rather than bullying perpetration, which was similar among teens with and without ADHD. No significant differences were observed between those with ADHD and those without, respectively, for any depression or anxiety symptoms within the past 2 weeks (30.4% vs 26.7%) or unmet mental health needs (12.7% vs 21.1%). The prevalence of low life satisfaction was almost twice as high for teens with ADHD compared with those without (10.4% vs 5.9%), although the difference was only borderline significant (aPR = 1.93, CI = 1.00–3.74).

Significantly more teens with ADHD reported having sleep problems than those without ADHD (79.9% vs 69.1%, aPR = 1.17, CI = 1.05–1.31). There were also significant differences for specific sleep problem indicators; difficulty getting up was more common among teens with ADHD (50.2%) compared with those without (40.2%; aPR = 1.29, CI = 1.02–1.64). Not having a regular waketime was twice as prevalent among teens with ADHD than those without (29.0% vs 14.0%; aPR = 2.17, CI = 1.45–3.25). Having 4 or more hours of screen time was more prevalent among teens with ADHD compared with teens without ADHD (63.6% vs 50.0%, aPR = 1.26, CI = 1.05–1.52). Teens with ADHD did not differ from those without ADHD for several measures including having less than 60 minutes of daily physical activity (75.5% vs 74.3%) and having a concern about their weight (29.4% vs 33.2%). As there were no meaningful differences seen between crude and adjusted prevalence ratios, only adjusted prevalence ratios are presented in Table 3. For crude prevalence ratios, see Supplemental Table 1, http://links.lww.com/JDBP/A488.

This study has several limitations. First, the study design was cross-sectional; therefore, the associations between ADHD, social-emotional well-being, and health behaviors examined in the study cannot be used to draw causal inferences. Second, ADHD diagnoses were based on parent report of a health care provider diagnosis and have not been validated against medical records, which may have resulted in misclassification. Third, health behavior and well-being measures were reported by the teenagers; they may not reflect the same measures that would be reported by parents or directly observed. Fourth, while the NHIS-Teen includes questions on a variety of factors related to health behaviors and social-emotional well-being, it does not a include questions on items such as sleep duration and size of friend group that may influence the association between having an ADHD diagnosis and the health behaviors and social-emotional well-being outcomes of interest. Fifth, due to limitations in sample size and statistical power, we were unable to detect significant differences for some findings. Sixth, there is a small yet significant bias after weighting adjustments, including an underrepresentation of teens with ADHD in the NHIS-Teen sample.²⁹

In addition, we were not able to examine the prevalence and differences for co-occurring disorders (e.g., mental, emotional, or behavioral conditions) that may play a role in the social-emotional well-being and health-related factors we considered. It was not possible to determine whether the reported associations with ADHD overall were consistent for individuals with ADHD alone and those with co-occurring disorders. Related, we were unable to examine differences in prevalence and examine the use of ADHD medications or behavioral therapies to assess their role in impacting the overall pattern of findings due to lack of ADHD treatment questions on the surveys.

Furthermore, our sample size did not provide enough power to examine demographic differences, such as age, sex, or race and ethnicity by individual health behavior and well-being factors. Both ADHD prevalence and experience and self-report of social and emotional factors can be influenced by sex, for example.^2,17 Finally, as expected from other studies, our sample of teens with ADHD diagnosis was significantly more often male, older, and more often White non-Hispanic than teens without ADHD.²⁰ Demographic differences including gender, age, and race or ethnicity can influence whether a child receives an ADHD diagnosis; therefore, our findings represent teens with diagnosed ADHD but may not be generalizable to teens with undiagnosed ADHD.

This study’s findings have several implications for public health. In our study, many teens with and without ADHD reported sleep problems, high levels of screen time use, and mental health difficulties. As discussed, teens with ADHD were more likely to have difficulties across several health behavior and well-being domains compared with teens without ADHD, including having difficulty making friends, experiencing bullying victimization, reporting sleep problems, and engaging in high daily screen time. Future public health research could explore the interrelation among these factors. Parent-reported healthy lifestyle factors of teens with ADHD (e.g., screen time limits, adequate sleep, and daily physical activity) have been found to be protective against peer difficulties and bullying among teenagers,⁵ but research is limited on teen self-report of healthy lifestyle factors regarding their peer relationships and social-emotional well-being. By using data drawn from a teen perspective, this study sheds light on important concerns for teens with ADHD, including different measures of social and emotional well-being and health behaviors. Future work could continue looking at these factors and how teens’ access to care, medications, and therapies affect these outcomes.

It has been recommended that health care providers such as pediatricians and schools can support teens with and without ADHD to have healthy and active lifestyles, including media and screen use.²⁴ CDC’s Whole School, Whole Community, Whole Child approach³⁰ describes the importance of integrating a range of health factors including social and emotional climate, physical health, mental and physical health services, among others, as well as having families and communities work together with schools to support teen development and health.³⁰ Health care providers have an opportunity to assess teens for engagement in health-related behaviors and address those if warranted.³¹ Behavioral interventions to support healthy sleep, for example, have been shown to be effectively delivered to children with ADHD by health care providers.³² Our results represent a step toward informing opportunities for public health interventions to decrease the negative health effects of ADHD and encourage healthy habits among teenagers with ADHD.