ABSTRACT
Objective To explore the methods for estimating premorbid intelligence of patients with cerebral infarction.Methods Ten regression equations were employed to estimate intelligence quotients (IQs) of 74 patients and 74 demographically matched,healthy adults.Those valid estimated IQs were transformed into standard scores and adopted to evaluate the difference with don' t hold' subtests.Results In the healthy group,there were trivial difference between the estimated IQs and obtained IQs with mean discrepancy less than one and the correlation coefficients between them ranged from 0.755 to 0.956.However in the patients group,the estimated IQs were significantly higher than obtained IQs with mean discrepancy of 3.20-10.67.Mean estimated IQs of the patient group were lower than those of healthy group to varing degrees with mean discrepancy of 0.83-8.28,in which the mean IQs estimated just with demographic variables showed small between group effect size of 0.26,and so did IQs estimated with combination of demographic variables and performance on the Information or Figural Matrix (effect size:0.24 and 0.38 respectively).The latter two kinds of estimated IQs were adopted and transformed into standard scores to estimate premorbid intelligence.The two kinds of estimated normal scores of intelligence were higher than normal scores of the Digit Symbol (0.70 and 0.63 respectively) and the Block Design(0.67 and 0.61 respectively)in the patients group.As regarding the healthy group,the counterparts of discrepancies between estimates of intelligence and obtained scores were 0.21,0.21,0.12 and 0.12 respectively,which were significantly smaller than those in the patients group.Conclusion IQ estimated with combination of demographic variables and performance on the Information or Figural Matrix are suggested to estimate premorbid intelligence of patients with cerebral infarction.