A comparative study of self-report, urinalysis and hair analysis in the detection of methamphetamine in Yaba users

Three diagnostic methods have dominated drug-abuse research: self-report, urinalysis and hair analysis. Previous studies have compared detection rates for various drugs, but none has focused a three-pronged concordance study on the use of methamphetamine (MA). To determine and compare the rates of MA detection in urine and hair of subjects who reported consuming MA in the form of Yaba. Self-reports of Yaba use, as well as biological specimens for chemical analyses, were collected from paid volunteers participating in a larger project studying risk-taking behavior of young adults in northern Thailand. All subjects in the present study reported using Yaba within 90 days of enrollment. Hair analysis for MA followed a validated protocol that coupled solid phase microextraction (SPME) with gas chromatography-mass spectrometry (GC-MS). Preliminary urinalysis was by means of REMEDi-HS. Positive urine was confirmed for MA by the SPME/GC-MS protocol. The MA detection rate by hair analysis (34.3%, n = 172) was significantly higher than by urinalysis (19.1%, n = 96) (p < 0.01; McNemar's test). All subjects with MA-positive urine samples reported using Yaba within 30 days of testing, while hair analysis gave positive results for self-reports up to 90 days. Urinalysis showed greater concordance with self-report than hair analysis if testing occurred within seven days of most recent admitted Yaba use. The reverse was true after 14 days. Agreement of laboratory findings with self-reports increased if test results for the two biological matrices were combined. There was no strong agreement between hair analysis and urinalysis for subjects reporting most recent use within 30 days of testing (kappa = 0.131; 95% CI = 0.022-0.240). For the Yaba users in the present study, urinalysis for MA significantly detected more positives than hair analysis if the most recent use reportedly occurred within seven days of testing. Hair analysis yielded better results after an interval of 14 days, with its window of detection extending up to three months. There were no urine positive samples for reported use after 30 days. Combining urinalysis and hair analysis increased the probability of detecting recent MA use. Both urinalysis and hair analysis significantly under-detected MA in the biological samples collected. The combined detection rate was 44.4%. This discrepancy might have resulted from over-reporting of Yaba use due to social/psychological factors and/or insufficient MA consumption