Proficiency of drug susceptibility testing for Mycobacterium tuberculosis in Taiwan

Ruwen Jou, C. Y. Chiang, C. Y. Yu, M. H. Wu

Research output: Contribution to journalArticlepeer-review

16 Citations (Scopus)


OBJECTIVE: To evaluate the impact of external quality assessment on the quality of drug susceptibility testing (DST) in clinical mycobacteriology laboratories. DESIGN: A pilot evaluation of DST proficiency was conducted in 2006 and scaled up in 2007. A panel consisting of 20 Mycobacterium tuberculosis isolates was used. Accuracy of 95% in detecting resistance to both isoniazid (INH) and rifampicin (RMP), and 90% to both ethambutol (EMB) and streptomycin (SM), was used to define a competent laboratory. RESULTS: Nine laboratories participated in 2006 and 30 in 2007. In 2006, the mean accuracy in detecting resistance to INH was 91.6%, for RMP it was 96.1%, for EMB it was 90.5% and for SM it was 93.9%. In 2007, the mean accuracy in detecting resistance to INH increased to 95.7% and that for RMP to 97.2%, while the accuracy of EMB resistance detection decreased to 82.0% and that for SM resistance to 86.8%. Quality improvement was observed in those laboratories that had adopted standardised methods. Overall, only five (17%) laboratories fulfilled the competency criteria for all four drugs in 2007. CONCLUSION: The majority of the laboratories that participated in 2006 demonstrated an improvement in DST performance in 2007. It is essential to continue external quality assessment to strengthen the quality of DST.

Original languageEnglish
Pages (from-to)1142-1147
Number of pages6
JournalInternational Journal of Tuberculosis and Lung Disease
Issue number9
Publication statusPublished - Sept 2009
Externally publishedYes


  • Drug susceptibility testing
  • Proficiency
  • Tuberculosis

ASJC Scopus subject areas

  • Pulmonary and Respiratory Medicine
  • Infectious Diseases


Dive into the research topics of 'Proficiency of drug susceptibility testing for Mycobacterium tuberculosis in Taiwan'. Together they form a unique fingerprint.

Cite this