There are no studies in the literature on the effect of exercise in reversing osteoporosis of the upper extremities for people with neurological disorders. The purpose of this study was to explore what conditions respond to a home exercise program for the upper extremities. Sixteen patients, divided randomly into experimental and control groups, were recruited for this trial for at least one year. Both upper extremities of the patients, affected and unaffected, were tested pre and post trial with the Hologic QDR for bone area, bone mineral content (BMC), bone mineral density (BMD), fat, muscle mass, muscle mass plus BMD, total weight, and relative fat content (percent). The home exercise program involved the subject squeezing a ball, one hand at a time, 20 times each, three times a day, at least three times a week, while sitting with the elbow flexed and the arm resting comfortably at the side of the trunk or on an armrest. A significant difference (p < 0.02) was found in bone area, BMC, BMD or relative fat content between the affected and sound sides both before and after the study, although all parameters were much different for each patient. We found a statistically significant difference in a change in the bone area for the affected upper extremities of the control group, but not of the experimental group. Subjective reactions of the subjects were positive. They stated that they felt they were improving and contributing to their recovery. Patients with Parkinson's disease, but not those with cerebral infarction, seemed to react to the exercise with an increase in bone mineral content, although the difference was not significant due to the scarce number of subjects. This is a preliminary pilot study to help develop further research on what condition is likely to respond to exercise for protection against osteoporosis.