Background/Purpose:   Limited information exists whether RA
patients are likely to develop more morbid conditions after RA diagnosis
compared to those without RA and the trajectory of multimorbidity over the
course of RA remains unclear. We aimed to examine the accumulation of incident morbidities
after onset of RA compared to patients with hypertension (HTN)- a highly
prevalent chronic non-inflammatory condition.  

Methods: Using claims data from a U.S commercial health plan, we identified
new onset RA patients with ³2 ICD-9 codes of RA, >7 to <365 days apart,
and new initiation of ³1 DMARD after ³12 months of enrollment. HTN Patients
were identified similarly using ICD-9 codes of HTN and new initiation of ³1 HTN
medication. The index date was defined as first prescription of a DMARD or
respective HTN medication. We excluded patients with any prior use of DMARDs, HTN
medication, or other chronic conditions any time prior to the index date. RA
and HTN patients were 1:4 matched on age, sex, obesity status, and index date. Beginning
on the index date, patients were followed for development of any of the 16 chronic
conditions according to the widely used Deyo-adapted Charlson Index (Table). We calculated incidence rates (IR)
per 1,000 patient-years (PY) for each condition. Cumulative percentage of
patients developing at least one additional chronic condition over time was
plotted separately for RA and HTN patients using Kaplan-Meier curves.  

Results: We includeda total of 4,803 RA and 19,212 HTN patients. Mean age was 49 years , 71% was female,
0.5% obesity. Mean time of follow up was 25 months in the RA and 32 months in
the HTN cohort. In the RA cohort the highest IR was found for chronic pulmonary
disease (66.2/1,000 PY) compared to HTN (47.3/1,000 PY; p<0.01). The IR of
cancer was higher in RA patients (RA 32.6/1,000 PY vs. HTN 25.3/1,000 PY p<0.01)
whereas the IR for diabetes was higher in the HTN cohort (RA 27.9/1,000 PY vs.
HTN 40.1/1,000 PY p<0.01; table).
The cumulative incidence of patients with ³1 co-morbidity was higher for RA patients
within the first two years of follow-up (FU) (16.8% vs. 14.6% up to 1 year, 24.6%
vs. 21.4% up to 2 years; p<0.01), after 3 years of FU Kaplan Meier curves
revealed no significant differences between RA and HTN patients (log rank test
p=0.35).

Conclusion: Within the first 2 years
after disease onset, a higher incidence of multimorbidity was observed in RA
patients. The highest IR was found for pulmonary disease, which might have an
important impact on many outcomes including morbidity and mortality.  

Table.  Incidence rate of 16 different
morbid conditions per 1,000 person-years for the rheumatoid arthritis (RA) and
the hypertension (HTN) cohorts.

 ADDIN EN.REFLIST 1. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for
use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-9.