A summary of economic evaluations published in the field of rheumatology and related disciplines.

INTRODUCTION: The relationship between the effectiveness of health care interventions and their costs is of increasing interest. The goal of the present study was to summarize the economic evaluations (EEs) published in the field of rheumatology and related disciplines, and to focus in particular on methodologic issues. METHODS: MEDLINE was searched in order to identify all relevant EEs published between 1966 and February 1995. Titles and abstracts of 1,435 articles were retrieved and independently reviewed by 2 assessors. Overall, 63 articles were identified by either assessor as being definite or possible full EEs (defined as an analysis comparing 2 or more strategies involving the assessment of both costs and consequences), and were assessed independently. RESULTS: Thirty-six articles were determined to be full EEs (33 cost-effectiveness and 3 cost-utility analyses). Most were published in the periods 1984-1990 (31%) and 1991-1995 (61%). Main areas covered were methods of prevention (44%), treatment (31%), and treatment-prevention (22%). Disorders most frequently studied were osteoarthritis (36%), osteoporosis (22%), and rheumatoid arthritis (14%). Direct and indirect costs were measured or estimated in 100% and 28% of the EEs, respectively. The viewpoint of the analysis was stated explicitly in 12 studies (33%). Incremental and sensitivity analyses were presented in 17 (47%) and 23 (64%), respectively. Inadequate use of economic terms was also documented. CONCLUSION: The EE articles reviewed adhered partially to basic analytic methods. Economic evaluations in the field of rheumatology need to be improved to comply with current standards for the evaluation of health care interventions.

Prognostic indicators of joint destruction in systemic-onset juvenile rheumatoid arthritis.

We retrospectively reviewed the charts and radiographs of 38 patients with systemic-onset juvenile rheumatoid arthritis, attempting to identify early in the disease course the clinical and laboratory observations most predictive of the later development of destructive arthritis. In 12 of the patients, destructive arthritis developed within 2 years of disease onset. When first examined, these patients could not readily be differentiated from those in whom joint destruction did not develop, but they more commonly had hepatosplenomegaly (p less than 0.04), serositis (p less than 0.01), and a lower mean serum albumin concentration (26.7 vs 31.3 gm/L; p less than 0.02). However, by 6 months after onset, patients with destructive arthritis more frequently had persistent systemic symptoms (92% vs 12%; p less than 0.0001), polyarthritis (67% vs 19%; p less than 0.0005), a lower mean hemoglobin level (95 vs 114 gm/L; p less than 0.001), a higher mean leukocyte count (21.2 vs 10 x 10(9)/L; p less than 0.0003), a higher mean platelet count (794 vs 400 x 10(9)/L; p less than 0.0001), and a higher mean erythrocyte sedimentation rate (43 vs 24 mm/hr; p less than 0.05). Multivariate analysis of the results at 6 months revealed that persistent systemic symptoms and a platelet count greater than or equal to 600 x 10(9)/L were the variables most highly predictive of the later development of joint destruction. We conclude that patients at high risk for the development of destructive arthritis may be identified within 6 months of disease onset, thereby indicating the need for more aggressive early therapy.