Abstract

We investigated the classification accuracy of the Inventory of Problems − 29 (IOP-29), its newly developed memory module (IOP-M) and the Fifteen Item Test (FIT) in an Australian community sample (N = 275). One third of the participants (n = 93) were asked to respond honestly, two thirds were instructed to feign mild TBI. Half of the feigners (n = 90) were coached to avoid detection by not exaggerating, half were not (n = 92). All measures successfully discriminated between honest responders and feigners, with large effect sizes (d ≥ 1.96). The effect size for the IOP-29 (d ≥ 4.90), however, was about two-to-three times larger than those produced by the IOP-M and FIT. Also noteworthy, the IOP-29 and IOP-M showed excellent sensitivity (>90% the former, > 80% the latter), in both the coached and uncoached feigning conditions, at perfect specificity. Instead, the sensitivity of the FIT was 71.7% within the uncoached simulator group and 53.3% within the coached simulator group, at a nearly perfect specificity of 98.9%. These findings suggest that the validity of the IOP-29 and IOP-M should generalize to Australian examinees and that the IOP-29 and IOP-M likely outperform the FIT in the detection of feigned mTBI.

摘要

我们在澳大利亚社区样本（N  = 275）中调查了问题清单-29（IOP-29），其新开发的内存模块（IOP-M）和十五项测试（FIT）的分类准确性。三分之一的参与者（n  = 93）被要求诚实回答，三分之二被指示假装轻度TBI。半数伪装者（n  = 90）接受过指导，以免被夸大而避免被发现，而另一半则没有（n  = 92）。所有的措施成功地歧视诚实反应和feigners之间，具有大的影响大小（d  ≥1.96）。IOP-29的效果大小（d ≥4.90），比IOP-M和FIT产生的值大2到3倍。同样值得注意的是，IOP-29和IOP-M在完美和无条件的假装条件下均表现出出色的敏感性（前者> 90％，后者> 80％）。相反，未训练的模拟器组中FIT的灵敏度为71.7％，而训练的模拟器组中FIT的灵敏度为53.3％，几乎完美的特异性为98.9％。这些发现表明，IOP-29和IOP-M的有效性应推广到澳大利亚的考生，并且IOP-29和IOP-M在假装mTBI的检测中可能胜过FIT。