The purpose of this study was to examine the accuracy, validity, and clinical utility of a brief cognitive screening instrument in identifying cognitive impairment in SUD patients in a clinical research setting. Results generally support its appropriate and practical use in this population. Based on its agreement with a reference criterion, the MoCA showed evidence of criterion-related validity and good accuracy in correctly classifying cognitive impairment cases and non-cases.
The Neuropsychological Assessment Battery-Screening Module (NAB-SM) served as the reference criterion, and was used for determining agreement and rates of correct and incorrect decision classifications. The NAB-SM and MoCA similarly sample a broad range of cognitive domains. Results of the present study showed good agreement between the MoCA and the five NAB-SM cognitive sub-domains, including attention, language, memory, spatial, and executive. Thus, among the processes sampled across the NAB-SM cognitive sub-domains, none are disproportionately weighted in the MoCA. However, unlike the NAB-SM, the MoCA does not include performance tasks related to psychomotor speed or visual learning and delayed recognition. Also, the NAB-SM assesses verbal memory through learning and delayed recall of verbally-presented narrative, whereas the MoCA assesses verbal memory through immediate and delayed recall of five unrelated words. How these differences might affect overall test performance is not apparent.
Based on patient acceptability and other practical considerations the MoCA has good clinical utility. Assessment of patient acceptability indicated that patients in general did not find the MoCA particularly unpleasant or demanding. It also provides an accurate and valid screening measure that is easy to use, time-efficient, and resource-conscious. This makes conducting cognitive assessment with SUD patients more practical for treatment settings and providers, such that patients who screen positive may be referred to more comprehensive evaluation. Further, the MoCA, including protocol sheet, instructions for administration and scoring criteria are available at no cost by the test developer (http://www.mocatest.org/
). The MoCA may be used, reproduced, and distributed without permission for clinical and educational non-commercial purposes. For non-commercially funded research, it may be used with prior written permission. If used for commercially funded research, prior written permission and a licensing agreement are required. In contrast, the list price of the NAB-SM is $825, plus the cost of additional screening module record forms ($94) and response booklets ($52) per every 25 administrations. Purchase of the NAB-SM is restricted to professionals who meet competency-based qualification guidelines; and who have completed the registration and qualification process attesting to their eligibility on the basis of training, education, and experience.
In comparison to the 10-minute administration time required for the MoCA, the NAB-SM takes approximately 45 minutes to complete. Further, hand-scoring the NAB-SM can take 30 minutes or longer. Scoring software is available (i.e., NAB Software Portfolio; NAB-SP) that can reduce the time needed to score the NAB-SM, but requires a PC-based computer with CD-ROM drive for installation, and Internet connection or telephone for software activation. In comparison to the MoCA, administration of the NAB-SM also requires significantly more space - i.e., a larger working surface with sufficient space to spread-out testing materials including puzzle pieces and the stimulus book.
Finally, in addition to English, the MoCA has been translated into 22 languages. Multiple language versions of the MoCA have shown high sensitivity for screening patients with mild cognitive impairment, including the Korean (Lee, et al., 2008
), Arabic (Rahman & El Gaafary, 2009
), and Chinese (Wen, Zhang, Niu, & Li, 2008
) language versions.
There are several strengths of this investigation. The natural heterogeneity of the sample is a strength of the present study because it demonstrates the validity of the MoCA for standard clinical practice. In other words, the ecological validity of the study findings is maximized without compromising the internal validity of the neuropsychological measures. Another strength is that the criterion measure has specifically been recommended for use with SUD patients because of its strong sensitivity and specificity in classifying patients with present or absent cognitive impairment in this population. The most common neuropsychological batteries typically require several hours of administration time, scoring, and interpretation (Rabin, Barr, & Burton, 2005
). The 45-minute administration time and computerized scoring of the NAB-SM enabled a more comprehensive evaluation across cognitive domains while conferring two additional benefits: (a) enabled testing in a single day, and (b) avoided test fatigue that almost invariably results from lengthy neuropsychological batteries. Another strength of the present study is the counter-balanced presentation of the MoCA and NAB-SM, which avoids possible test order effects.
A limitation of this study is the unknown influence of abstinence duration on overall prevalence of cognitive impairment. An exclusion criterion for the present study was known substance use within seven days prior to study participation. This is consistent with the recommendations made by some investigators who suggest a duration of at least one week between admission to treatment and testing (Miller, 1985
; Parsons & Farr, 1981
). The rationale for this recommendation is that some cognitive recovery generally follows abatement of intoxication and acute abstinence effects. As a result, rates of detection among newly admitted patients may be artifactually inflated due to the effects of residual intoxication or withdrawal. However, because the goal of the study was not to study prevalence of cognitive impairment, but rather the concordance between the MoCA and NAB-SM, the latter should be relatively insensitive to any potential inflation.
The relationship between duration of abstinence and cognitive impairment, furthermore, is unclear. Paradoxically, cognitive performance may actually deteriorate slightly over the first few weeks of abstinence before gradually improving. For example, a recent study showed a gradual worsening in most neuropsychological categories, such that cocaine dependent persons with a positive urine drug screen for cocaine (typically indicating use within the past 72 hours) perform better on a broad range of neuropsychological measures in comparison to cocaine dependent individuals with a negative screen (Woicik, et al., 2009
). These findings are consistent with a previous study showing that the scope of neuropsychological deficit among currently abstinent cocaine dependent persons actually increased from 72 hours to 14 days (Berry, et al., 1993
). These findings suggest that the influence of abstinence duration on cognitive performance may not be linear. However, it is unknown how exclusive such “non-linear” effects of abstinence duration may be to predominantly heavy psycho-stimulants users. Therefore, there may be little relevance to the present study given that only 17% of patients met criteria for cocaine dependence and 8% met criteria for amphetamine or methamphetamine dependence.
A body of evidence is emerging showing that cognitive impairment in patients with substance use disorders (SUD) has a significant and negative impact on treatment outcomes and therapeutic mechanisms of change. Specialized treatments and enhancements aimed at improving outcomes for SUD patients have shown some success, but this is still an area in its infancy and further research is needed. One of the main challenges associated with developing treatments for cognitively-impaired SUD patients is uncertainty about which patients to target for specialized interventions. To date, the search for a brief cognitive screening instrument sensitive to the mild-to-moderate impairment observed in SUD patients has been unsuccessful. The present findings show the Montreal Cognitive Assessment addresses a critical need in the addiction treatment research community by providing a quick and accurate screening instrument that can expedite the progression of research in this area.