How Often, and How Late, Writing Difficulty Accompanies ADHD

What the study found

By age nineteen, 64.5 percent of boys with ADHD had met criteria for a written-language disorder, against 16.5 percent of boys without it. Among girls the figures were 57.0 percent and 9.4 percent. Both comparisons reached significance at P below .001.

Those are cumulative incidence figures from a population-based birth cohort: every child born between 1976 and 1982 to a mother living in one Minnesota school district, 5,699 of them once nineteen children with severe intellectual disability were excluded, followed until death, emigration or graduation. Kouichi Yoshimasu and colleagues at the Mayo Clinic and Children’s Hospital Boston reported the results in the journal Pediatrics in 2011. Three hundred and seventy-nine of the children had ADHD, three-quarters of them boys.

Nobody in the cohort had to seek out a diagnosis to be counted. School files were searched for any documented concern about learning or behaviour, which flagged 1,961 children, of whom 1,509 had complete testing. Paired IQ and writing achievement scores were then run through three separate psychometric formulas, and a child counted as affected if any one of the three was satisfied. The writing problems named in the diagnostic manual of the period were grammatical and punctuation errors, poor paragraph organisation, multiple spelling errors, and excessively poor handwriting.

Eight hundred and six children in the cohort met at least one writing criterion. Six hundred and three of those also met a criterion for reading disability, leaving 203 whose writing was impaired while their reading was intact. Both groups were substantially more common among the children with ADHD. Where reading was also affected, cumulative incidence by nineteen reached 45.9 percent with ADHD against 9.6 percent without, an adjusted hazard ratio of 5.02. Where reading was unaffected, the figures were 16.6 percent and 3.3 percent, a hazard ratio of 6.80. The models were adjusted for the child’s race and for the mother’s age and years of education.

The sexes diverged in only one of those two groups. Among children who had a reading disability as well, the adjusted hazard ratio associated with ADHD was 4.21 for boys and 9.78 for girls, and the interaction between sex and ADHD was significant. That girls’ estimate rests on 95 girls with ADHD and carries a confidence interval running from 6.76 to 14.14, so the magnitude is loosely fixed even though the elevation itself is clear. Where reading was intact, boys and girls with ADHD were at closely similar risk.

What it means for you and your child

Among the children whose writing was impaired, those whose reading was intact met the criteria at a median of 11.3 years, roughly three years later than those whose reading was affected as well. The study reports that interval without accounting for it.

Those three years have a familiar shape at home. A child reads above grade level, talks well about what they have read, and then produces a few laboured sentences in the time the lesson allows, and the fluent reading is taken as proof that language is intact. The writing gets attributed to effort or motivation, while the volume of written work a school requires increases every year. Spelling and handwriting are remarked on in passing, and a teacher’s comment about neatness is not a measured score.

This cohort was schooled through the 1980s and 1990s under discrepancy-based definitions of learning disability that have since been revised, and the population was 96 percent white and largely middle class, so these percentages should not be read as current national rates. The design also fixes the age at which research criteria were met rather than the age at which either condition began, which leaves open whether ADHD precedes the writing difficulty, follows it, or whether both follow from something shared. The association survived when each of the three psychometric formulas was evaluated on its own.

The authors’ own recommendation is procedural. An evaluation for ADHD, in their view, should include a thorough psychoeducational assessment for written-language disorder, which means writing tested directly, covering spelling, handwriting, punctuation and the organisation of connected text. A strong reading score offers no reassurance about written expression.

Reading is routinely measured in detail while written expression is reduced to a single composite score, and the paper’s recommendation is that it be assessed in its own right. If you have an assessment ahead of you, the sentence to bring into the room is this: I want written expression tested and reported separately from reading, with the standard scores for each.