In early 1989, the Research Committee of the American Academy of Physical Medicine and Rehabilitation (AAPM&R) established a subcommittee to develop methods to monitor academic progress in physical medicine and rehabilitation (PM&R) units in the US. To develop an indirect baseline of academic productivity in PM&R, the rates and types of publications by PM&R researchers were assessed in eight peer review medical journals. The journals selected consisted of all issues of the following (published in calendar years 1987 to 1989): Archives of Physical Medicine and Rehabilitation, American Journal of Physical Medicine and Rehabilitation, Physical Therapy, Archives of Neurology, Pain, Stroke, Paraplegia, and Arthritis & Rheumatism. The sampling frame consisted of 3,553 journal articles. Affiliation with a PM&R unit or other clinical science unit (other unit), extramural funding sources, and type of manuscript (eg, case report or scientific investigation) were identified and coded. Sixteen percent of all articles were authored by members of PM&R units. The prevalence of scientific reports written by other unit authors (71%) was comparable to that written by PM&R authors (67%) (chi 2[3] = 5.54; p less than .20). There was a greater prevalence of funding by the US Department of Education of studies written by PM&R authors (10%) than of studies written by members of other units (2%) (chi 2[1] = 79.4; p less than .0001). Reports authored by members of other units had a greater prevalence rate of funding from all other sources--federal and private (47% vs 33%; chi 2[1] = 41.2; p less than .0001).(ABSTRACT TRUNCATED AT 250 WORDS)