Objective: to evaluate the reliability of diagnostic assessment based on clinical observations by physical therapists.
Design: interobserver studies between two pairs of physical therapists.
Setting: two primary-care physical therapy practices.
Patients: all applying to the practices for treatment.
Exclusions: physical therapy in previous 3 months, or need for acute treatment.
Assessments: schedule derived from the ICIDH.
Results: agreement on disabilities better than on impairments, the latter revealing problems particularly with pain and restricted range of motion.
Conclusions: reliability of assessments of most of the categories considered was reliable; in two categories the reliability was poor, and the categories were modified.