Predictors of disability in a longitudinal sample of patients with rheumatoid arthritis

J Paul Leigh, James F. Fries

Research output: Contribution to journalArticlepeer-review

87 Scopus citations


Information from the Health Assessment Questionnaire (HAQ) is used to identify which variables measured in 1981 successfully predict the severity of disease in 1989 and the eight year change in severity of disease in a sample of 330 residents of Santa Clara County, California, USA. This study is exploratory and no previous hypotheses are made. Using univariate correlations and stepwise linear regressions, initial values of a number of variables are found to be useful predictors including, in order: the HAQ disability index, pain scale, global health status, tender joints, fewworkhours, age, female sex, nevermarried, widowhood, and occupation as operative (e.g. dry wail installers, assemblers). Data suggest that deterioration over eight years is least rapid for those with severe disease in 1981. A mathematical identity suggests that analyses of the changes in severity versus analyses of 1989 severity are identical, provided that the 1981 disability index is entered as a covariate. The initial level of the disability index of the HAQ is by far the strongest predictive variable and provides a clinically important gauge for the likelihood of future impairment.

Original languageEnglish (US)
Pages (from-to)581-587
Number of pages7
JournalAnnals of the Rheumatic Diseases
Issue number5
StatePublished - 1992
Externally publishedYes

ASJC Scopus subject areas

  • Ophthalmology
  • Sensory Systems
  • Cellular and Molecular Neuroscience
  • Immunology
  • Rheumatology


Dive into the research topics of 'Predictors of disability in a longitudinal sample of patients with rheumatoid arthritis'. Together they form a unique fingerprint.

Cite this