Prediction of permanent work disability in a follow-up study of early rheumatoid arthritis: results of a tree structured analysis using RECPAM

Br J Rheumatol. 1996 Jul;35(7):652-9. doi: 10.1093/rheumatology/35.7.652.

Abstract

The objectives of this study are: (a) to determine the occurrence of permanent work disability (PWD) in early rheumatoid arthritis (RA); (b) to identify prognostic groups of patients; (c) to assess the employment rates for these groups over time. Seventy-three gainfully employed consecutive out-patients with early RA (> or = 5 ARA 1958 criteria, disease duration < or = 12 months) at time one (T1) were re-examined at time two (T2) after a mean follow-up of 6 yr (S.D. +/- 2 yr). Potential risk factors, identified at T1, for PWD at T2 were entered in a tree structured survival analysis using RECPAM (RECursive Partition and AMalgamation). Cumulative 3 yr employment rates (3-yrER +/- S.E.M.) were computed from the resulting Kaplan-Meier curves. At T2, PWD occurred in 27 of the 73 patients (37%). The fastest decline in the employment rate was found within the first 3 yr of the disease onset, with a 3-yrER reduced to 73 +/- 5%. The group with the poorest prognosis (n = 14; 3-yrER 14 +/- 9%) was defined by age > or = 50 yr with either ESR > or = 60 mm/h or the combination of modified functional class (1-7) > or = 4 with a disease duration > or = 7 months. An intermediate group (n = 38; 3-yrER 79 +/- 6%) was defined by (a) age > or = 50 yr and low or moderate disease activity, (b) age < 50 yr and more strenuous job-related physical requirements, (c) age < 50 yr and less strenuous work, but joint count > or = 15. No case of PWD occurred in 21 individuals aged < 50 yr with a joint count < 15 and less physically demanding jobs. PWD occurs early in a substantial number of patients with RA. RECPAM defines risk profiles that can readily be applied in actual clinical situations and allow an estimation of the risk of PWD at different time points using the resulting Kaplan-Meier curves.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Arthritis, Rheumatoid / physiopathology*
  • Decision Trees*
  • Disabled Persons*
  • Employment*
  • Female
  • Follow-Up Studies
  • Humans
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Prognosis
  • Time Factors
  • Work Capacity Evaluation