Abstract

Physical therapy and rehabilitative care are highly necessary for the maintenance and recovery of
health and well-being. Aside from healing light muscle sprains, physical therapy prevents the
development of more post-operative complications down the line. Furthermore, physical therapy
allows patients to improve their quality of life and prevents other conditions from occurring.
However, in the United States, multiple barriers exist to attaining the level of rehabilitative care
due to various social determinants of health (SDH). These determinants play a very big role,
especially in physical therapy, which requires patients to visit the clinics consistently, increasing
the cost and transportation. Socioeconomic disparities are a very large barrier, due to the low
number of private clinics accepting Medicaid or government insurance for low-income patients.
Racial barriers also limit access to care, as African-Americans tend to be under-referred by
physicians, not being able to receive the necessary care. Further examination of other factors
including gender, educational attainment, and geographic location reveal more trends of high
barriers in access to physical therapy. This paper seeks to address the question: “How do
differences between various social factors and demographic variables affect access to physical
therapy care?” By examining the most relevant disparities with machine learning techniques,
insight can be gathered for more targeted public health policies and actions to make physical
therapy more accessible.