Aim: To compare the effects of short-term hospitalization individualization therapy and family therapy on childhood autism. Methods: A total of 77 children with autism, including 40 ones received hospitalization therapy and 37 ones received family therapy from October 1998 to May 2001 were selected from Child and Adolescent Psychological Health Center of Chinese Psychological Health Association. Patients in the hospitalization individualization therapy group were treated with training schemes, which were designed by special autistic trainers according to children's symptoms. One trainer took charge of one child for all day long, and the contents referred to speech, social intercourse, behavior, cognition, self-care in life etc. Children in family therapy group were trained by parents at home according to the instructions and advices of trainers. The time of training was 3-4 months. Subjects were scaled by autism behavior checklist, childhood autism rating scale and adaptive behavior scale before and after the short-term training to compare the curative effects of two groups. Results: A total of 77 subjects were involved in the analysis of results. 1 There were no significant differences in scores of independence, cognition and social/self-control factors in autism behavior checklist, childhood autism rating scale and adaptive behavior scale between the two groups before therapy (t=0.839-1.517, P > 0.05). 2 After short-term therapy, the pathogenetic conditions of children in the hospitalization individualization therapy group were significantly ameliorated than those in the family therapy group, including the scores of three factors in three scales mentioned above (t=2.469-6.008, P < 0.05), whereas in children of the family therapy group, all scale scores were remarkably reduced except the independence factor score in child adaptive behavior scale (t=2.427-4.902, P < 0.05). 3After therapy, the total score of childhood autism rating scale scores in the hospitalization individualization therapy group was significantly lower than the family therapy group (t=2.508, P < 0.05). The scores of independence factor and social/self-control factor of adaptive behavior scale in the hospitalization individualization therapy group were significantly higher than family therapy group (t=1.803-2.205, P < 0.05). Conclusion: Both short-term family therapy and hospitalization individualization therapy can positively affect the functional recovery and development of children with autism, while as to the amelioration effect on independent capacity of autistic children, that of hospitalization individualization therapy is superior to family therapy, which indicate that comprehensive strategies should be adopted to ameliorate the social capacity of autistic children, and therapy that isolated with society or congregation is not adoptable.