Summaries of Must-Read Clinical Literature, Guidelines, and FDA Actions
Identifying RA Through Use of Administrative Data
Arthritis Care Res; 2018 Oct; Curtis, Xie, et al
Claims‐based algorithms can accurately identify incident rheumatoid arthritis (RA), according to a recent study that evaluated an approach to identify incident RA in health plan claims data. Both Medicare and commercial claims data were linked to a US RA registry. Researchers evaluated the accuracy of year of RA onset in the registry (gold standard) vs different claims algorithms, varying International Classification of Diseases, Ninth Revision codes for RA/arthritis, duration of health plan enrollment preceding diagnosis (minimum of 1 vs 2 years), and use of RA medications. Results were reported as positive predictive values (PPVs) of the claims‐based algorithm for incident RA. They found:
- Depending on the algorithm tested and whether patients were enrolled in Medicare or the commercial health plan, the PPVs for incident RA ranged from 68–81%.
- A 2‐year clean period free of all RA‐related diagnoses and medications was somewhat more optimal although, by comparison, a 1‐year clean period yielded similar PPVs and retained approximately 90% more RA patients for analysis.
Curtis JR, Xie F, Chen L, Greenberg JD, Zhang J. Evaluation of a methodologic approach to define an inception cohort of rheumatoid arthritis patients using administrative data. Arthritis Care Res. 2018;70(10):1541-1545. doi:10.1002/acr.23533.
In the current age of Big Data, many clinical research studies are conducted and published without any direct personal interaction between the investigator and patients who are the subjects of the research. Individual patients need not be seen or examined by the investigator, who uses a computer to select relevant data from data bases that have been collected by large epidemiology studies or information required to take claims for services provided. Investigators can also enter electronic medical records from various providers, sometimes supplemented by self-report questionnaires. This methodology study evaluates methods of identifying patients with incident rheumatoid arthritis (RA). The Corrona RA Registry was used, with 6,509 patients participating in Medicare. ICD-9 codes for RA that were submitted to Medicare for payment of doctor visits were compared with year of RA onset as recorded by Corrona. For patients whose first RA ICD-9 code came after at least 2 years on Medicare, the assumed year of RA onset agreed with that reported by Corrona in 81% of cases.—Harold E. Paulus, M.D.; Emeritus Professor; University of California, Los Angeles; Division of Rheumatology.