Children are diagnosed and treated for attention-deficit/hyperactivity disorder at disproportionate rates based on their race or ethnicity, a study has shown.
Hispanic children were about half as likely as white children to have received a diagnosis by junior high, and African American children were about two-thirds less likely than white children to have received a diagnosis by junior high. In addition, white children with ADHD were about two to three times more likely to be taking medication for their disorder than children of all other racial/ethnic backgrounds, the results showed.
Starting in kindergarten and continuing through eighth grade, racial/ethnic disparities in ADHD diagnosis and medication use were identified in a cohort of 17,100 children, reported Paul L. Morgan, Ph.D., of Pennsylvania State University, University Park, and his associates (Pediatrics 2013 June 24 [doi:10.1542/peds.2012-2390]). Hispanic children had 50% lower odds and African American children had 69% lower odds of being diagnosed compared with white children, after adjustment for confounders. Children of other races/ethnicities had 46% lower odds of being diagnosed with ADHD.
Dr. Morgan’s team identified 6.6% of the initial cohort as being diagnosed with ADHD by eighth grade. The diagnosis was based on parental report that the child had been formally diagnosed by a professional with ADHD, attention-deficit disorder (ADD), or hyperactivity by their kindergarten, first-, third-, fifth-, or eighth-grade years.
The cohort was made up of about 19% Hispanics, 16% non-Hispanic African Americans, 57% non-Hispanic whites, and 8% children of other races/ethnicities, including Asian, Native Hawaiian, Pacific Islander, Native American, and Alaskan Native.
The researchers conducted two analyses: The first included results with only race/ethnicity and time used as predictors from the full cohort of 17,100 kindergarteners. The second, using the 15,100 children for whom the data were available, "included additional child- and family-level predictors measured in kindergarten, as well as time-varying measures of children’s behavioral and academic functioning." The child- and family-level predictors used were low birth weight; mother’s age; health insurance status; English-speaking parents and socioeconomic status, based on family income; and the mother’s and father’s education levels and occupations.
Children’s externalizing and learning-related behaviors were assessed by their kindergarten, first-, third-, and fifth-grade teachers, using the Externalizing Problem Behaviors and the Approaches to Learning subscales of the Social Rating Scale. Averages of the children’s reading and mathematics standardized test scores were used to estimate their academic achievement.
In the first analysis, unadjusted for child or family factors and compared with white children, the odds of being diagnosed with ADHD were 57% lower for Hispanic children, 36% lower for African American children, and 47% lower for children of other races/ethnicities (P less than .001).
In the second model adjusted for other predictors, Hispanic children were 50% less likely (odds ratio, 0.38-0.66), African American children were 69% less likely (OR, 0.24-0.40), and children of other races/ethnicities were 46% less likely (OR, 0.39-0.74) to be diagnosed with ADHD than white children were (P less than .001).
The researchers also used both models in calculating the odds ratios for prescription medication use for ADHD among the children. Parents reported whether their children were taking prescription medication, including methylphenidate, amphetamine, or atomoxetine, related to ADD, ADHD, or hyperactivity while the children were in fifth and eighth grades.
Compared with white children, in the first unadjusted model, Hispanic children were 64% less likely and African American children were 65% less likely to be taking medication for ADHD (P less than .001). Children of other races/ethnicities were 58% less likely to be taking medication for ADHD than were white children (P less than.01).
In the second model, the odds of taking prescription medication for ADHD were 47% lower for Hispanic children (OR, 0.29-0.98; P less than .05), 65% lower for African American children (OR, 0.19-0.62; P less than .001), and 51% lower for children of other races/ethnicities (OR, 0.26-0.95; P less than .05).
The researchers identified other statistically significant predictors of ADHD diagnosis. Children without health insurance had 33% lower odds of being diagnosed with ADHD than children with health insurance (P less than .01). Children were more likely to be diagnosed with ADHD if they had a mother older than 38 when the child was born, compared with mothers aged 18-38 (OR, 1.65; P less than .001) or if the parents were English speaking (OR, 1.86; P less than .05). As has been found in past research, boys were twice as likely to be diagnosed (OR, 1.98; P less than .001).
Children with higher achievement or who engaged in learning-related behaviors were 30% and 41% less likely, respectively, of being diagnosed with ADHD (P less than .001). Children showing externalizing problem behaviors were 46% more likely to be diagnosed (P less than .001).