Medical disability assessment of the back pain patient for the Social Security Administration: the weighting of presenting clinical features.
We investigated how physicians use history and physical findings when assessing disability for low back pain. Thirty-six North Carolina physicians, either practitioners experienced in disability determinations (26) or employees of the Social Security disability agency (10), responded to 48 clinical vignettes. They rated each case on a scale of 0 to 1.0, according to their degree of certainty that the patient was disabled. All combinations of five patient variables were presented in the vignettes: pain (mild or severe), physical examination (normal, reflex loss or muscular weakness), mobility (normal or restricted), X-rays (normal or osteoarthritis), and occupational history (normal or light work). The mean certainties for the individual vignettes ranged from 0.08 to 0.43. Mean certainty estimates across physicians ranged from 0 to 0.61, indicating substantial variability in how physicians assess disability. Practicing physicians had higher certainty of patients' disability than did physicians employed by Social Security, 0.37 vs 0.07 (p less than 0.01). Degree of pain was not associated with certainty of disability. All other clinical factors were highly significant predictors of physician assessment of certainty of disability. The emphasis on physical and radiographic findings over history places disability evaluation distinctly apart from other medical assessments.
Duke Scholars
Published In
DOI
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Social Security
- North Carolina
- Middle Aged
- Medical Records
- Male
- Humans
- Epidemiology
- Disability Evaluation
- Back Pain
- Analysis of Variance
Citation
Published In
DOI
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Social Security
- North Carolina
- Middle Aged
- Medical Records
- Male
- Humans
- Epidemiology
- Disability Evaluation
- Back Pain
- Analysis of Variance