The MANOVA performed for UPSA-2 data indicated a significant main effect of group [F(7,22) = 5.8, p < 0.01]. Subsequent univariate ANOVAs revealed a main effect of group on the total UPSA score [F(1,28) = 32.8, p < 0.001], comprehension [F(1,28) = 21.58, p < 0.001], finance [F(1,28) = 14.2, p < 0.01], communication [F(1,28) = 10.1, p < 0.01], transportation [F(1,28) = 8.6, p < 0.01], and medication management [F(1,28) = 11.8, p < 0.01]. Bonferroni posthoc tests indicated that the drug-free comparison group exhibited significantly improved performance on the total UPSA-2 score (p < 0.001), comprehension domain ( p < 0.001), financial domain ( p < 0.01), communication domain ( p < 0.01), transportation domain ( p < 0.01) and the medication management domain (p < 0.01) compared to the METH group, but no difference was observed for household skills (). There was a large effect size of group (r values) on the total UPSA-2 score (0.73), comprehension domain (0.66), financial domain (0.58), communication domain (0.52), transportation domain (0.48), and medication management domain (0.54).
UPSA-2 measures for drug-free comparison (n = 15) and METH-dependent (n = 15) subjects. UPSA-2 total score range is from 0 to 120; the score range for each individual domain is from 0 to 20. Higher scores indicate improved functional performance.
We did not observe any significant correlation between UPSA-2 performance and the length of METH use, the amount of total drug consumed, duration of abstinence, or the number of days METH was used in the 12 months prior to testing. There was a trend towards a positive correlation between the age at which drug use was initiated and the total UPSA-2 score (r = 0.51, p = 0.054); this result suggests that participants who started using the drug at an earlier age performed more poorly on the UPSA-2. A significant negative correlation between the frequency of METH use and performance on the UPSA-2 financial score ( r = −0.63, p = 0.016) indicated that more frequent METH use was associated with more impaired performance in this domain.
METH-dependent participants showed significant impairment on the WCST, including more total errors [t (26) = 3.1, p < 0.01], greater perseverative errors [t (26) = 2.8, p < 0.01], and fewer completed categories [t (26) = 3.0, p < 0.01] compared to the drug-free comparison group. Total error T-scores (with higher scores indicating fewer WCST errors) were positively correlated with the total UPSA-2 score and improved performance in the medication, financial, and transportation domains (). Fewer perseverative errors were significantly correlated with better transportation scores and associated with a trend towards a higher total UPSA-2 score and improved performance in the financial domain. Finally, the number of WCST categories completed was positively correlated with the UPSA-2 total, financial and transportation scores. These data suggest that poor performance on the WCST was related to impaired functional ability as assessed by a variety of UPSA-2 measures.
Table 3 Pearson r correlations between WCST measures and UPSA-2 domain measures for all participants (n = 28). WCST total error and perseverative error data are calculated as T scores normalized for age and gender. Higher T scores indicate improved performance (more ...)
The METH-dependent group exhibited significantly higher total (t(28) = −5.28, p < 0.001) and positive symptom (t(28) = −4.12, p < 0.001) PANSS scores compared to drug-free comparison subjects (), but did not differ on the negative symptom scale (t(28) = 0.638, p = 0.53). There were no significant correlations between PANSS scores and UPSA-2 measures in the comparison group. For METH-dependent subjects, the PANSS total score was negatively correlated with the total UPSA-2 score and the household skills score, while the positive symptom PANSS subscale was negatively correlated with the total UPSA-2 score, the household skills score, and the comprehension score ().
Table 4 Pearson r correlations between PANSS Total and Positive Symptoms scores and UPSA-2 domain measures for METH-dependent participants (n = 15). Individuals with higher PANSS scores showed more impaired performance on the total UPSA-2 score and in the comprehension/planning (more ...)
Although some evidence suggests that ADHD may contribute to METH-induced cognitive deficits (Sim et al., 2002
), the three METH-dependent subjects who also met criteria for lifetime ADHD exhibited equivalent performance on the UPSA (average score of 86.7) compared to the non-ADHD METH sample (average score of 83.2). Performance on the PPVT was also unaffected, with average scores of 97.5 and 97.7 for the ADHD and non-ADHD METH-dependent participants, respectively. In addition, ADHD METH-dependent participants showed a tendency towards fewer errors on the WCST (average total error score of 50.7 vs. 40.0 for the non-ADHD METH-dependent sample). These data suggest that the presence of comorbid ADHD in several METH-dependent subjects (at least in this very small sample) does not account for the differences in functional ability observed between METH-dependent and drug-free comparison participants.