Clinical Utility of Reliable Digit Span in Assessing Effort in Children and Adolescents with Epilepsy

Abstract

The assessment of effort is an important aspect of a comprehensive neuropsychological evaluation, as this can significantly impact data interpretation. While recent work has validated the appropriateness of adult-derived cutoffs for standalone effort measures in younger populations, little research has focused on embedded effort measures in children. The present study includes 54 clinically referred children and adolescents (32 males/22 females; aged 6-17) with a confirmed diagnosis of epilepsy. Reliable Digit Spans (RDSs) were calculated and the Test of Memory Malingering (TOMM) was administered in the context of a comprehensive neuropsychological evaluation. Using a previously published RDS cutoff of ≤6, a pass rate of only 65% was obtained, well below the recommended 90% pass rate for an effective effort index. In contrast, when adult criteria were used on TOMM Trial 2, a 90% pass rate was observed. RDS scores were significantly correlated with IQ estimates (r = .59, p < .001) and age (r = .61, p < .001). The difference between RDS and the TOMM on the participant outcome was statistically significant (χ(2) = 9.05, p = .003). These results suggest that RDS appears to yield a large number of false positives and, therefore, may be of limited utility in detecting poor effort in a pediatric epilepsy population. These findings likely extend to other pediatric populations that are known to have significant cognitive loss.

Assessment of effort in children: A systematic review

“…Embedded indicators of effort are useful because they do not increase assessment time, are relatively resistant to coaching and allow for monitoring of effort in tests that were not originally designed to assess symptom validity (Kirkwood, Connery, Kirk, & Baker, 2013). For example, RDS is one of the few embedded measures investigated in both children and adults because the Digit Span subtest is used on the child and adult versions of the Wechsler Intelligence Scales Welsh et al, 2012;Whitney, Shepard, & Davis, 2013).…”

“…Lowering the cutoffs from embedded measures designed for adults can improve specificity in some pediatric samples, as Kirkwood et al (2011) showed when using an RDS cutoff of 6 in mTBI. However, with more severely impaired samples, that same cutoff yielded a high false positive rate in a more significantly neurologically compromised pediatric sample (Welsh et al, 2012). With severely impaired neurological samples, cutoffs on embedded measures may need to be set so low to yield high specificity rates that the resulting sensitivity rates will be significantly compromised when applied to groups where mild-to-no-significant cognitive impairment would be expected.…”

“…Kirkwood, Hargrave, and Kirk (2011) showed that a lower RDS cutoff score of ≤6 yielded a sensitivity of 61% and a specificity of 92% in children with mTBI. However, Welsh, Bender, Whitman, Vasserman, and Macallister (2012) concluded that RDS yielded a large number (35%) of false positives with a cutoff of ≤6, which limits its utility in detecting poor effort as an embedded effort measure in a pediatric epilepsy population and likely other children with a significant cognitive loss, which would not include mTBI. However, Loughan, Perna, and Hertza (2012) reported that a WISC-IV Digit Span scaled score of ≤4 yielded a specificity of 91% and sensitivity of 43% for detecting poor effort in children with heterogeneous dual diagnoses.…”

Utility of the Test of Memory Malingering (TOMM) in Children Ages 4–7 Years with and without ADHD

“…The impact of intellectual skills on TOMM performance in children has also been the focus of investigation, with most studies showing a positive association between IQ and TOMM performance (Kirk et al, 2011; MacAllister et al, 2009; Welsh, Bender, Whitman, Vasserman, & MacAllister, 2012). Some of these studies also showed that children with low IQ are able to achieve acceptable performance eon the TOMM.…”

“…For example, in MacAllister et al (2009), 7 of 11 participants with an IQ ≤ 70 achieved acceptable performance (including all four children with IQ < 50). Nevertheless, while the proportion of children who do not achieve acceptable performance on the TOMM is relatively low, those whose performance is below threshold are likely to manifest lower IQ or intellectual disability (Kirk et al, 2011; Loughan & Perna, 2014; Welsh et al, 2012). …”