Predicting Mortality in Patients With Rheumatoid Arthritis

[Guest guest]

Predicting Mortality in Patients With Rheumatoid Arthritis

Wolfe F, Michaud K, Gefeller O, Choi HK

Arthritis Rheum., 2003;48:1530-1542.

Background

Most studies have shown that patients with rheumatoid arthritis (RA) have a

significantly shorter life expectancy than the general population. However,

relatively few studies have evaluated predictors of mortality in these

patients. An accurate understanding of mortality predictors in RA patients

is important both in terms of public health and clinical management of RA.

The primary objective of this study was to quantify the predictive strength

of clinical and demographic variables that are commonly used in clinical

trials and easily measurable in clinical practice.

Experimental Design

A total of 1,387 consecutive RA patients were seen in a single clinic over a

20-year period beginning in 1981. At each visit, a wide range of clinical

and demographic assessments, including blood pressure, body mass index,

tender joint count, grip strength, morning stiffness, Health Assessment

Questionnaire (HAQ) disability index scores, visual analog scale (VAS) for

pain and global severity, Arthritis Impact Measurement Scales (AIMS),

anxiety and depression scales, erythrocyte sedimentation rate (ESR),

hemoglobin level, rheumatoid factor (RF) status, and others were recorded

and entered into a computer database. Radiographs of the hands were

generally obtained at 2-year intervals and read using the Larsen method.

Major Results

HAQ disability was found to be by far the most important univariate

predictor of mortality. Among the clinical and laboratory variables, HAQ was

the best predictor of mortality, followed by global disease severity, pain,

depression, anxiety, and grip strength. A 1­standard deviation (SD) change

in HAQ was associated with a 26.2% greater increase in the odds ratio for

mortality than a 1-SD change in global disease severity. The HAQ was

predictive across its full range, compared with variables such as the ESR,

which were predictive only at higher levels, and consequently, in fewer

patients.

Commentary

The HAQ was first designed, utilized, and reported by Dr. F. Fries in

1980. The HAQ assesses how patients are performing their activities of daily

living, such as lifting over their heads or getting out of a chair. The

results of the HAQ are used to assess an individual¹s level of physical

function and physical disabilities. Shortly after the HAQ was validated,

investigators found that it was a strong predictor of how RA patients were

responding to their medications and that it was the best predictor of

response to antirheumatic medications.

This study employed a longitudinal database of RA patients who Dr. Wolfe

followed for several years in his practice. He found that when the HAQ was

completed by patients when they were first evaluated in the clinic and then

annually for a few years, it was the strongest predictor of mortality, even

after adjusting for relevant confounding variables that could influence the

result. Translating this result into clinical practice, it would be

appropriate for clinicians to obtain an HAQ measurement when RA patients are

initially evaluated by a new rheumatologist and then to repeat the HAQ

measurement annually for a few years. The goal of the exercise would then be

to try to prescribe therapies that reduced the HAQ score, which would

presumably improve life expectancy.

Common clinical and laboratory values, such as tender and swollen joint

counts and ESR, are probably more important for determining initial

therapeutic decisions than for predicting long-term functional outcomes.

While the data presented herein are meticulously analyzed and results are

carefully discussed, validation of these findings should be done before

these results are widely applied in clinical practice. Other large

rheumatology clinical practices with more than one clinician should perform

a similar study so the rheumatology community will know whether to accept or

refute these findings.