12 and a better LR− of .63, a PPV of .23 and a NPV of .86. Table 3 also shows that introduction of the D and E criteria results in an expected increase in specificity (.82) at the expense of a decrease in sensitivity (.21). Also restricting BD symptoms to substance free periods (criteria D and E) and removing functional impairment (criterion C: allowing BD I to be included as a case) did not improve

the sensitivity and specificity of the MDQ to detect BD (Table 3, last column). The positive and negative likelihood ratios (LR+, LR−) ranged from 1–1.42 to 1–.63, respectively. Validity indicators based on the MDQ assessment at T1 were very similar and certainly not better than those based on the MDQ assessment at baseline (T0) (data not shown). Of the 170 patients with a SCID at T1, 159 (94%) also completed all the other diagnostic

instruments (DIS, SIDP-IV) at T1. Of the 31 patients selleck (19.5%) with BD, 8 (25%) Lapatinib also had BPD, 2 (6.4%) APD and 10 (32.2%) ADHD. Of the 128 patients without BD, 15 (11.7%) had BPD, 29 (22.7%) APS, and 38 (29.7%) ADHD. The relative risks of the presence of BPD, APD and ADHD in patients with BD compared to patients without BD were 2.2 (95% CI 1.03–4.72) for BPD, 0.28 (95% CI 0.07–1.13) for APD, and 1.09 (95% CI 0.61–1.93) for ADHD, meaning that BD relatively often co-occurred with BPD, the BD tended to co-occur less often with APD and that ADHD was equally present in patient with and without BD. The standard MDQ operating characteristics with BD, BPD, APD, ADHD and any externalizing disorder (BD and/or BPD and/or APD and/or ADHD) as external criterion for this population

are shown in Table 4. In order to compare the performance of the MDQ for these different external criteria, we calculated areas under the curve. The AUCs ranged Fossariinae from .51 (BD) to .63 (ADHD). The 95% CI of the AUCs of BD, BPD and APD all included 0.50, indicating that the standard MDQ performed not better than chance for these three disorders. The performance to detect ADHD and any externalizing disorder was slightly better with AUCs of .63 (95%CI .54–.72) and .60 (95%CI .51–.68) respectively, but 95% CI’s largely overlapped with those of the AUC of the other external criteria (BD, BPD and APD). The primary objective of this study was to evaluate the screening properties of the MDQ to detect BD in a treatment seeking population of patients with SUD. Our first hypothesis that the MDQ would be a valid screen due to an expected relatively high prevalence of BD in this population was not confirmed. With the SCID diagnosis of BD-I, DB-II or BD-NOS as “golden standard” (prevalence 21%), the performance of the MDQ in this population was very disappointing: sensitivity = .43, specificity = .57, PPV = .21, NPV = .80, and AUC = .50.