Rational treatment of low back trouble?

The treatment of low back trouble remains, with rare exceptions, largely empirical. The scientific validation of the various available therapeutic options ultimately rests in controlled clinical trials. This paper discusses the problems inherent in the construction of such trials, and suggests that resources should be directed to the identification of clinical syndromes of predictable natural history. It is recommended that future trials of therapeutic efficacy be considered in relation to the natural history of back trouble.