Background. Deterioration in school achievement may pre-date adult mental disorders. We studied the association between compulsory school performance and later onset hospital-treated psychiatric morbidity experienced by the Northern Finland 1966 Birth Cohort (N=11017) in adult life.

Methods. School performance was operationalized in two ways: school class level (in normal, i.e. age-appropriate class v. not in normal class, i.e. class below age level or in special school) at the age of 14, and marks for individual school subjects at the age of 16. School class level was ascertained by postal questionnaire and school marks from national application register. These were linked to data on psychiatric morbidity from the National Finnish Hospital Discharge Register. By the end of 1994 (between ages 16 and 28 years), a total of 383 subjects had psychiatric illness. DSM-III-R diagnoses were grouped into three categories: schizophrenia; other psychoses; and non-psychotic disorders. The remaining population with no psychiatric hospitalization served as a single comparison group. School class level and values of school marks in the three diagnostic categories were each compared with this comparison group, stratified by sex.

Results. In the comparison group 6·8% of boys and 3·4% of girls were not in their normal class. In all the diagnosis groups the proportions of those not in normal class were from 2 to 8 times higher than in the comparison group. A majority of those not in normal class and having psychiatric diagnosis were intellectually subnormal (IQ<85). Among adolescents who later developed non-psychotic disorders, means of school marks were lower (P<0·05, adjusted for social class and place of residence) than in the comparison group. Lower marks were not found in categories schizophrenia or other psychoses. Logistic regression analysis confirmed these findings after adjustment for confounding factors.

Conclusions. Not being in the normal class at age 14 predicted future hospital-treated disorders, but low school marks at age 16 predicted only non-psychotic disorders. These findings may be an early manifestation of the disorders themselves, or a marker of vulnerability or other risk factors. The mechanisms may differ between diagnoses.