Skip Navigation Archive: U.S. Department of Health and Human Services U.S. Department of Health and Human Services
Archive: Agency for Healthcare Research Quality
Archival print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

  • Publication # 13-RA009

Pharmacy and coded-diagnosis information do not accurately predict tuberculosis cases in patients with rhematoid arthritis

Patient Safety and Quality

Certain drugs used to treat rheumatoid arthritis, which suppress the immune system, can boost a person’s risk for tuberculosis (TB). Data from diagnostic codes or pharmacy prescription files have had mixed results in their ability to identify TB in individuals taking these drugs. A new study revealed that using administrative data alone results in a high false-positive rate of TB. 

Researchers identified 18,094 patients with rheumatoid arthritis who were receiving Medicaid benefits in one State. They used three strategies to identify patients with TB. The first approach was based on ICD-9-CM diagnostic codes for TB. A second strategy used pharmacy claims data to see which patients filled prescriptions for two or more anti-TB medications on the same day.

The final approach combined both strategies to identify patients with TB. Cases of TB identified in these ways were compared to actual confirmed cases obtained from a TB registry used by TB control programs. Within the rheumatoid arthritis study population, 10 persons were confirmed with TB during 61,461 years of followup, resulting in an incidence rate of 16.3 per 100,000 person-years. There was a wide variation in the number of TB cases identified by the 3 strategies, ranging from 6 to 449. 

All three approaches had low positive predictive values. There were high false-positive rates of TB detection for the diagnostic code (98.7 percent) and pharmacy claims (95.8 percent) strategies. However, when used together, six false-positive cases were identified, dropping the false positive rate to 75 percent. Thus, adding pharmacy claims to diagnosis coding data results in only a slight improvement in detecting cases of TB. Given these findings, the researchers suggest using actual confirmed TB case data when conducting drug-epidemiology research. 

The study was supported in part by AHRQ (T32 HS13833). See "Accuracy of pharmacy and coded-diagnosis information in identifying tuberculosis in patients with rheumatoid arthritis," by Christina T. Fiske, M.D., Marie R. Griffin, M.D., M.P.H., Ed Mitchel, M.S., and others in Pharmacoepidemiology and Drug Safety 21, pp. 666-669, 2012.


Page last reviewed June 2013
Internet Citation: Pharmacy and coded-diagnosis information do not accurately predict tuberculosis cases in patients with rhematoid arthritis: Patient Safety and Quality. June 2013. Agency for Healthcare Research and Quality, Rockville, MD.


The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care