Table 1. Trends in Mental Health Care Among 230 070 Adolescent Participants of the National Survey on Drug Use and Health, 2005-2018, by Sociodemographic Characteristics.
Characteristic | No. of participants | Mental health care, weighted % of participants | Change in odds of receiving carea | Test of departure from linearityb | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
2005-2006 | 2007-2008 | 2009-2010 | 2011-2012 | 2013-2014 | 2015-2016 | 2017-2018 | aOR (95% CI) | P value | F test | P value | ||
Overall | 230 070 | 20.3 | 19.7 | 18.7 | 18.9 | 19.9 | 19.7 | 20.8 | 1.01 (0.96-1.06) | .60 | 26.11 | <.001 |
Sex | ||||||||||||
Female | 112 848 | 22.8 | 22.8 | 21.8 | 22.1 | 23.5 | 23.5 | 25.4 | 1.11 (1.04-1.19) | .001 | 16.06 | <.001 |
Male | 117 222 | 17.8 | 16.8 | 15.8 | 15.8 | 16.4 | 16.0 | 16.4 | 0.90 (0.84-0.96) | .002 | 8.83 | .004 |
Race/ethnicity | ||||||||||||
Non-Hispanic white | 131 882 | 20.4 | 20.4 | 19.3 | 19.6 | 21.2 | 21.3 | 22.7 | 1.08 (1.03-1.14) | .004 | 21.43 | <.001 |
Non-Hispanic black | 31 139 | 22.6 | 20.8 | 19.9 | 19.4 | 20.0 | 17.9 | 18.5 | 0.76 (0.67-0.85) | <.001 | 1.35 | .25 |
Hispanic | 43 927 | 18.8 | 17.6 | 16.9 | 17.3 | 17.7 | 18.0 | 19.4 | 1.02 (0.91-1.13) | .75 | 6.64 | .01 |
Other | 23 122 | 17.6 | 17.2 | 16.7 | 17.4 | 17.7 | 17.6 | 17.6 | 1.03 (0.88-1.20) | .74 | 0.07 | .80 |
Age, y | ||||||||||||
12-13 | 72 913 | 19.6 | 19.9 | 18.5 | 19.3 | 20.2 | 18.6 | 20.1 | 1.00 (0.92-1.08) | .96 | 1.19 | .28 |
14-15 | 77 939 | 21.6 | 20.5 | 19.9 | 19.9 | 20.9 | 20.7 | 21.3 | 0.97 (0.90-1.04) | .41 | 6.95 | .01 |
16-17 | 79 218 | 19.5 | 18.7 | 17.8 | 17.4 | 18.6 | 19.6 | 20.9 | 1.07 (1.00-1.15) | .06 | 30.31 | <.001 |
Family income compared with federal poverty level, % | ||||||||||||
≤100 | 46 394 | 22.0 | 22.3 | 20.0 | 21.1 | 20.7 | 20.6 | 19.9 | 0.88 (0.80-0.98) | .02 | 0.10 | .76 |
>100 to 200 | 51 754 | 20.8 | 20.0 | 19.1 | 18.5 | 20.5 | 19.2 | 19.2 | 0.90 (0.81-1.00) | .051 | 1.58 | .21 |
>200 | 131 922 | 19.5 | 18.8 | 18.1 | 18.2 | 19.3 | 19.5 | 21.7 | 1.11 (1.04-1.18) | .001 | 37.03 | <.001 |
County of residence | ||||||||||||
Large metropolitan | 102 410 | 20.6 | 19.3 | 19.0 | 18.8 | 20.2 | 20.0 | 20.8 | 1.04 (0.97-1.11) | .31 | 11.75 | .001 |
Small metropolitan | 78 279 | 20.0 | 20.6 | 18.8 | 19.1 | 20.3 | 19.6 | 21.3 | 1.01 (0.94-1.09) | .75 | 6.92 | .01 |
Nonmetropolitan | 49 381 | 19.5 | 19.3 | 17.8 | 18.6 | 18.1 | 18.6 | 20.1 | 0.93 (0.83-1.03) | .15 | 5.73 | .02 |
Health insurance | ||||||||||||
Private | 139 864 | 19.4 | 18.6 | 17.7 | 18.1 | 19.3 | 19.0 | 21.2 | 1.11 (1.04-1.18) | .002 | 31.58 | <.001 |
Medicaid/CHIP | 66 577 | 24.3 | 24.0 | 22.2 | 21.3 | 21.9 | 21.4 | 21.6 | 0.87 (0.81-0.94) | .001 | 4.90 | .03 |
Other | 7936 | 21.0 | 19.3 | 19.5 | 17.9 | 21.5 | 18.3 | 17.8 | 0.88 (0.69-1.12) | .30 | 0.01 | .90 |
No health insurance | 12 928 | 15.8 | 16.4 | 14.7 | 16.0 | 15.9 | 16.2 | 12.7 | 0.92 (0.75-1.14) | .45 | 1.29 | .26 |
Abbreviations: aOR, adjusted odds ratio computed in logistic regression; CHIP, Children’s Health Insurance Program.
Based on binary logistic regressions of each dummy-coded variable on the variable of survey year. The variable of survey year was scaled to range from 0 (representing 2005-2006) to 1 (representing 2017-2018). Thus, regression coefficients represent change from the 2005-2006 to 2017-2018 period. The logistic regression models were adjusted for sex, age, race/ethnicity, family income compared with the federal poverty level, health insurance, and metropolitan status of county of residence.
Indicates adjusted Wald F test, with 1 and 110 df associated with the squared survey year variable.