Bower published a comprehensive review that traced the history of rehabilitation and cerebral palsy from the early 1900s to the present. She noted that Phelps was one of the first clinicians to treat patients with cerebral palsy in the 1930s, followed by Peto and Bobath, who became interested in cerebral palsy a few years later. The major developments in the use of physical therapy in the treatment of cerebral palsy, however, didn't occur until the 1950s. This was largely due to the decline in the incidence of poliomyelitis after the Salk vaccine was discovered, and the shifting of research dollars and therapeutic techniques to persons with cerebral palsy.

Bower's description of the major therapeutic approaches to the treatment of cerebral palsy over the past 50 years is briefly summarized below.

Phelps, an orthopedic surgeon, recommended 5 years of institutionalized therapy from a team approach (physical, occupational, speech and other therapies). He described five different categories of cerebral palsy. His work included the use of motion pictures to describe children and evaluate the effectiveness of treatment. He also endorsed tenotomies (surgery on individual muscles). Paine researched the work of Phelps in 1962 and reported that children with mild spastic hemiplegia improved with and without treatment, and that Phelps' treatment strategies were not effective for children with athetosis. The treatment proposed by Phelps did not change whether or not a child would need surgery. The work of Rood was based on afferent sensory stimulation, but no empirical research has ever been carried out on the efficacy of this therapy.

Vojta, a German neurologist, combined the techniques of Fay, Rood and Kabat. In 1981, research was carried out on one subject with cerebral palsy whose subluxed hip was reduced after 3 years of treatment. In 1984, Kanda looked at Vojta's work with 29 subjects with cerebral palsy. Eight were children with spastic diplegia who had gotten treatment early in their lives, and 21 children with spastic diplegia were treated several months later. All walked by the age of 3. The children who were treated earlier seemed to have an improved gait pattern compared to the children treated later. Vojta also reported on his own work but there are disputes concerning how many of the 207 infants he wrote about (who supposedly became "normal" in mental and motor performance) were children with cerebral palsy.

In 1991, Mayo studied one of the most widespread therapeutic techniques used to treat children with cerebral palsy, the Bobath or neurodevelopmental approach (NDT). Started in the 1940's by Karel and Berta Bobath, this treatment was based on the view that cerebral palsy results from interference with development of normal posture against gravity. The brain lesion leads to the loss of inhibition of abnormal primitive reflexes. Mayo examined the effectiveness of NDT and did not find this technique to be any better than other forms of physical therapy. Wright and Nicholson researched Bobath's work and concluded that there were no differences between children treated with the Bobath method compared to children who received no treatment. The one exception was the ability of children with quadriplegia to roll.

In summarizing her findings, Bower noted that although each of the above methods have been used for decades in treating children with cerebral palsy, there is very little scientific evidence to prove that any one therapy is more or less effective in improving the long-term function of children with cerebral palsy, and that often therapy continues when it may no longer be needed. She noted that "research needs to be undertaken by clinical scientists to assess the relative merits of the various approaches of therapy. Research methodologies used in psychology and the social sciences may well prove to be more useful for this purpose than those used in traditional medical research."

In 1988, a classic study by Palmer and coworkers entitled, "Effects of Physical Therapy on Cerebral Palsy," and published in the highly respected New England Journal of Medicine, investigated two early intervention programs in 48 infants with mild to severe spastic diplegia (12 to 19 months of age). One intervention involved neurodevelopmental therapy and the other intervention was a published infant stimulation program called Learningames. The investigators found that there was no motor, cognitive, or social advantage for infants receiving physical therapy after six and 12 months of treatment, and that trends favored the infant stimulation program. More frequent contact between therapist and patient may have been necessary to make physical therapy more beneficial to infants with cerebral palsy, but if so, this would require higher costs. The investigators concluded with the following statement:

"This clinical trial offers no support for the idea that neurodevelopmental physical therapy is a preferred intervention in infants with mild to severe spastic diplegia. Although it is possible that there are longer-term benefits of physical therapy or benefits in domains not reported in this study, the goal of improved motor development was not achieved in infants receiving physical therapy as compared with infants receiving infant stimulation. Furthermore, physical therapy applied earlier offered no advantage over physical therapy applied six months later. The findings underscore a fundamental issue in developmental pediatrics and public policy affecting developmentally disabled children: the immediate and long-term effectiveness of traditional interventions must be examined critically. Alternative, less costly outcomes may improve function (p. 807)."

In 1995, Graves reviewed the literature on therapeutic methods for cerebral palsy and came to a similar conclusion:

"In reviewing these methods it becomes clear that while all have their zealous proponents, the methodology of evaluation studies is often flawed and the results are inconclusive. Again the call is often made for more and better studies, but the reality is that these methods have had several decades to provide proof of efficacy and none has been forthcoming. Significantly, there has been little or no attention to the negative effects of therapy (p. 26)."

Graves went on to say that although "...therapy does not lead to dramatic improvements in the neurological status of children with cerebral palsy, therapists have an important role in helping families understand how to work with their children in a physical and social setting."

In an extensive review of therapeutic techniques for persons with cerebral palsy, Bleck wrote: "after half a century of sincere and intense effort by professionals to 'treat' cerebral palsy, most now acknowledge that these remedial efforts have been unsuccessful in achieving function. Perhaps it is time to give up trying to 'cure' the neurological deficits by remedial methods, stop looking for positive studies, and get on with the task of helping children and their families."

Despite periodic calls for scientific inquiry into the effects of physical therapy in the treatment of children with cerebral palsy, there are only a handful of clinical trials in the literature that have examined the efficacy of physical therapy. As a result of small, heterogeneous samples, substantial attrition among subjects, and nonrandom assignment of treatment, it is difficult to interpret what are the most effective treatment strategies for improving the motor performance in persons with cerebral palsy. Fetters and Kluzik noted that "this lack of scientific evidence of treatment effectiveness is true for many types of physical therapy for children with cerebral palsy."

This cursory review of the literature on cerebral palsy and therapeutic techniques provides a strong incentive for the task force to develop new intervention strategies that will prove useful in the overall physical, psychological and social development of persons with cerebral palsy. Clearly, we must establish a more ecological approach to treatment and must develop safe and effective exercise guidelines for persons with cerebral palsy that can be implemented in community-based fitness settings.