Validity of the serial seven procedure

Serial subtraction by seven (Serial Seven Test, SST) is frequently used in mental status evaluation for dementia as a measure of concentration. Validation research on the SST has been limited. The literature to date raises significant concern about the nature of the procedure. The purpose of the present study was to evaluate the nature and validity of the SST. The participants were 80 consecutive outpatient referrals to the neuropsychology consultation service of two large general medical hospitals. All subjects were administered a comprehensive neuropsychological assessment battery, including the Mini Mental State Examination version of the SST. Multiple regression analysis indicated that calculation skill is at least as important as concentration, in predicting SST performance. Measures of overall level of cognitive dysfunction, education, and psychopathology were unrelated to SST performance. The results suggest that SST performance is heavily influenced by basic arithmetic skill and that the procedure should be used with caution as a measure of concentration or other nonacademic mental abilities.