Validation of the automated neuropsychological assessment metrics for assessing cognitive impairment in systemic lupus erythematosus

Objective We previously demonstrated the utility of the Automated Neuropsychological Assessment Metrics (ANAM) for screening cognitive impairment (CI) in patients with systemic lupus erythematosus (SLE) and developed composite indices for interpreting ANAM results. Our objectives here were to provide further support for the ANAM’s concurrent criterion validity against the American College of Rheumatology neuropsychological battery (ACR-NB), identify the most discriminatory subtests and scores of the ANAM for predicting CI, and provide a new approach to interpret ANAM results using Classification and Regression Tree (CART) analysis. Methods 300 adult SLE patients completed an adapted ACR-NB and ANAM on the same day. As per objectives, six models were built using combinations of ANAM subtests and scores and submitted to CART analysis. Area under the curve (AUC) was calculated to evaluate the ANAM’s criterion validity compared to the adapted ACR-NB; the most discriminatory ANAM subtests and scores in each model were selected, and performance of models with the highest AUCs were compared to our previous composite indices; decision trees were generated for models with the highest AUCs. Results Two models had excellent AUCs of 86 and 89%. Eight most discriminatory ANAM subtests and scores were identified. Both models demonstrated higher AUCs against our previous composite indices. An adapted decision tree was created to simplify the interpretation of ANAM results. Conclusion We provide further validity evidence for the ANAM as a valid CI screening tool in SLE. The decision tree improves interpretation of ANAM results, enhancing clinical utility.

A-247 Overall Test Battery Means from ANAM4 TBI-MIL and D-KEFS: A Within-Subject Comparison of a Traumatic Brain Injury Cognition Composite Score

Objective Several cognition composite scores have been developed for potential use in traumatic brain injury clinical trials. This analysis examined the equivalence of overall test battery mean (OTBM) from two different test batteries administered consecutively to the same subjects. Methods Soldiers were administered the Automated Neuropsychological Assessment Metrics (version 4) TBI-MIL (ANAM4) computerized battery and D-KEFS as part of a larger study comparing within-subject performance from different neuropsychological test batteries. Data from 121 soldiers with complete and valid data on both ANAM4 and D-KEFS and no recent TBI were used in this analysis. OTBMs were calculated for ANAM4 and the seven D-KEFS achievement scores. The OTBMs from the 121 soldiers were ranked from lowest to highest and the percentile rankings from each battery were compared. For each soldier, the differences between the percentile ranks from each battery were also calculated. Results Only 53.8% of soldiers who scored below the 20th percentile on ANAM4 also scored below the 20th percentile on D-KEFS. Furthermore, only 47.8% of soldiers who scored at or above the 80th percentile on ANAM4 also scored in that range on D-KEFS. Some soldiers’ performance on each battery diverged by large amounts, for example from 59.5 to 82.6 percentage points. Correlation analysis revealed that the OTBMs and percentile rankings from both batteries were modestly correlated (OTBM r = 0.515, p < 0.001, percentile rank r = 0.499, p < 0.001). Conclusion These results suggest that comparing similar cognition composites from different neuropsychological test batteries from different studies in a meta-analytic manner may not be feasible due to psychometric difference between batteries.

A-244 Examining the Effects of Computer Operating Configuration on Automated Neuropsychological Assessment Metrics V 4.0 (ANAM4) Reaction Time: A Repeated Measures Approach with Clinical and Research Implications

Objective Computerized cognitive tests are vulnerable to measurement error due to hardware and software configurations of the administration computer. Though automatic score corrections are often provided to account for this, they may mask subtle individual variations in test performance. We investigated the group and individual impact of correcting for measurement error while accounting for possible order effects using data from a larger study of computer platform effects on reaction time measurement. Methods ANAM4 TBI Military’s Simple Reaction Time (SRT) subtest was administered repeatedly on three different computer platforms. We compare two of those platforms: Platform 1, a Dell D630 configured for ANAM4 TBI Military administration; Platform 3, a Dell E6540 with default settings. A mechanical arm with a preset RT was first used to capture measurement error on each platform. SRT was then administered to healthy US Army soldiers in a randomly selected administration order, such that platform 1 was first and platform 3 was third for n = 58 and platform 3 was first and platform 1 second for n = 53. Results Platform 3 was on average 43.06 ms slower than Platform 1. Group analyses suggested applying a 43.06 ms correction to Platform 3 scores resulted in scores similar to Platform 1. However, it appears there were subtle practice and fatigue effects for some individuals that were possibly masked by the score correction. Conclusion Correcting for error due to computer configurations is important in RT measurement, though there may be subtle individual statistical and clinical implications that are lost at a group analysis level.

Automated Neuropsychological Assessment Metrics (v4) Military Battery: Military Normative Data

Introduction The Automated Neuropsychological Assessment Metrics (ANAM®) is a validated computerized cognitive assessment battery that has been routinely used for cognitive health assessments for active duty service members. The purpose of the current study was to provide updated normative data for the ANAM Military-Expanded battery. Materials and Methods This study was a retrospective review of deidentified predeployment baseline data obtained for service members completing ANAM as part of the mandated Department of Defense clinical testing policy. The effects of age, gender, race/ethnicity, and education were examined for all subtests and descriptive measures of performance were computed. Rates of low scores were also examined. Results Data from 1,067,899 active duty service members (88% male) were analyzed and represented all services and the Coast Guard. The sample ranged in age from 17 to 65 years (mean = 28.1, standard deviation [SD] = 8.4) was 66% White with a majority having education of 12 years or less (72%). Generally, small effect sizes were observed for each of the demographic factors with the largest effects observed for age. Rates of low scores on two or more tests in the battery were 10.6% using a more conservative cutpoint (≥1.3 SD below the mean) or 2.4% using the most stringent cutpoint (≥2 SD below the mean). Conclusions The current norms provide an update to previously published normative data and represent the largest, most representative set of data currently available for military personnel. Enhancements over previous norms include an expanded sample size, stratification by age (5-year increments), gender, race/ethnicity and education, and inclusion of all 10 tests in the ANAM Military-Expanded battery. These data are provided to assist healthcare providers using ANAM to assess and manage service member health.

An Investigation of the Accuracy of Reaction Time Measurements on ANAM4 TBI-MIL Across Three Computer Platforms

Study Design Using two observational methods and a within-subjects, counterbalanced design, this study aimed to determine if a computer’s hardware and software settings significantly affected reaction time (RT) on the Automated Neuropsychological Assessment Metrics (Version 4) Traumatic Brain Injury Military (ANAM4 TBI-MIL). Methods Three computer platforms were investigated: Platform 1—older computers recommended for ANAM4 TBI-MIL administration, Platform 2—newer computers with settings downgraded to run like the older computers, and Platform 3—newer computers with default settings. Two observational methods were used to compare measured RT to observed RT on all three platforms: 1, a high-speed video analysis to compare the timing of stimulus onset and response to the measured RT and 2, comparing a preset RT delivered by a robotic key actuator activated by optic detector to the measured RT. Additionally, healthy active duty service members (n = 169) were administered a brief version of the ANAM4 TBI-MIL battery on each of the three platforms. Results RT differences were observed with both the high-speed video and robotic arm analyses across all three computer platforms, with the smallest discrepancies between observed and measured RT on Platform 1, followed by Platform 2, then Platform 3. When simple reaction time (SRT) raw and standardized scores obtained from the participants were compared across platforms, statistically significant and clinically meaningful differences were seen, especially between Platforms 1 and 3. Conclusions A computer’s configurations have a meaningful impact on ANAM SRT scores. The difference in an individual’s performance across platforms could be misinterpreted as clinically meaningful change.

Clinical Use of the Automated Neuropsychological Assessment Metrics TBI-Mil Expanded Battery in Evaluating Concussion Recovery: A Retrospective Study

Introduction This retrospective study evaluated the use of the Automated Neuropsychological Assessment Metrics (ANAM4) Expanded battery in a clinical setting to determine if the resolution of physiological symptoms, in the absence of neurocognitive assessment, was sufficient data in the return-to-duty (RTD) determination. Materials and Methods In this study, 508 U.S. Naval Academy midshipmen were diagnosed with concussion and prescribed a standard treatment protocol. As directed by the protocol, they were evaluated and tracked by medical providers until determined to have normal balance and to be asymptomatic at rest and with exertion. Upon the resolution of these physiological symptoms, the midshipmen were referred for neurocognitive (ANAM4) testing. When results indicated a return to neurocognitive baseline, a RTD determination was made. Results The analysis of data in this study indicated that RTD determinations based solely on the resolution of physiological symptoms would have resulted in 25.1% of the sample being returned to duty before neurocognitive recovery. Additional analysis of the ANAM4 reliable change index (RCI) data for each of the concussed patients revealed a pattern of scores that correlated with an expected length of recovery. Individuals with at least one RCI greater than or equal to −1.64 returned to neurocognitive baseline in 8.92 days, whereas individuals with 2 or more RCIs greater than or equal to −1.25 (but less than −1.64) returned to baseline in 5.78 days, and those with a difference in measured reaction times that were greater or equal to −1.25 returned to baseline in 3.20 days. Furthermore, findings indicated that female service members required additional time for the resolution of physical symptoms as compared to their male counterparts. The mean number of days from injury to being deemed symptoms free in males was approximately 14, as compared to females who were deemed symptom free in 21 days. This difference is statistically significant. Conclusions Findings of this investigation revealed three specific outcomes. First, a computerized neurocognitive assessment instrument should be used as an adjunct measure in evaluating the resolution of physiological symptoms following a concussive injury. Second, results revealed that based on the RCIs of postinjury ANAM4 assessments, it is possible to estimate the remaining recovery time needed for a return to neurocognitive baseline. Third, results of this analysis revealed that gender appears to be a factor in time between concussive injury and resolution of symptoms.

Chronic decompression illness cognitive dysfunction improved with hyperbaric oxygen: a case report

Introduction: Altitude chamber exposures are used for training to allow aircrew to experience their hypoxia and pressure effect symptoms. Decompression illness (DCI) can occur subsequent to altitude chamber training or in operational aircraft when the cabin altitude is at least 18,000 feet. Definitive emergent treatment is hyperbaric oxygen (HBO2) to decrease bubble size, dissipate excess nitrogen hyperoxygenate tissue and reduce inflammation. Case report: A 27-year-old female underwent altitude chamber training to 25,000 feet. She developed tingling in both legs and left arm, headache, dizziness, malaise, then difficulty talking. She underwent two HBO2 treatments. Over the next 12 months she had paresthesia, decreased memory and cognitive function similar to symptoms seen following traumatic brain injury. She was referred 14 months after the event for evaluation. Using pre-deployment Automated Neuropsychological Assessment Metrics (ANAM) and serial tests over 58 HBO2 treatments, the patient demonstrated near-return to her pre-deployment test scores. Discussion: The reason for HBO2 treatment was based on previous experience with chronic traumatic brain injury subjects where HBO2 improved outcome. The patient’s chronic neurological symptoms mimicked chronic TBI. The patient was unique in that baseline cognitive tests existed that could be used to monitor her changes during the treatment series.