Risk Model for Severe Rheumatoid Arthritis Within 12 Years

 

Patient selection: rheumatoid arthritis with < 5 years disease duration

 

Endpoint: severe rheumatoid arthritis within 12 years (based on the highest tertile Sharp score modification of the van der Heijde score)

 

Parameters:

(1) rheumatoid factor

(2) erosive changes on joint radiographs

(3) swollen joint count (SJC)

Rheumatoid Factor	Erosive Changes	Swollen Joint Count	Chance Severe RA
negative	absent	< 6	< 35%
negative	absent	> 10	> 82%
negative	present	<4	< 65%
negative	present	> 8	> 93%
positive	absent	< 4	< 38%
positive	absent	> 9	> 89%
positive	present	< 3	< 60%
positive	present	> 6	> 90%

 

where:

• Figure 1 on page 388 does not make sense as published. The data above is based on what I think is the correct interpretation.

• Note that there is a gap in the swollen joint count in each pair. For example, in the first pair there is no coverage for a SJC of 6 to 10. In the implementation I will use the range between the two chances given in Figure 1.

• The chances of developing severe disease in the "low risk" group seem relatively large. Almost a third of the people with negative findings will develop severe disease.

 

Findings in Table 2 seen in patients with a mild course:

(1) 0 or very low swollen joint count

(2) very low Ritchie score

(3) usually rheumatoid factor negative

(4) no erosions on radiographs

(5) no disability

(6) 0 or very low Sharp score

 

Findings in Table 2 seen in patients with a severe course:

(1) multiple swollen joints

(2) positive Ritchie score

(3) often rheumatoid factor positive

(4) sometimes erosions seen on radiographs

(5) some disability

(6) positive Sharp score