John P. K. Bernstein

Characteristics of a Mild Traumatic Brain Injury Sample Recruited Using Amazon’s Mechanical Turk

Several barriers exist in the recruitment of individuals with a history of mild traumatic brain injury (mTBI). The current study is the first to examine whether crowdsourcing, which has been used in other clinical populations, is useful in the recruitment of those with mTBI.

I'd Do Anything for Research, But I Won't Do That: Interest in Pharmacological Interventions in Older Adults Enrolled in a Longitudinal Aging Study

Alzheimer’s disease (AD) ranks as the 6th leading cause of death in the United States, yet unlike other diseases in this category, there are no disease-modifying medications for AD. Currently there is significant interest in exploring the benefits of pharmacological treatment before the onset of dementia (e.g., in those with mild cognitive impairment); however, recruitment for such studies is challenging. The current study examined interest in pharmacological intervention trials relative to other types of clinical interventions. A total of 67 non-demented older adults enrolled in a longitudinal cognitive aging study completed a questionnaire assessing interest in participating in a variety of hypothetical research study designs. Consistent with past research, results showed that the opportunities for participants to advance science, receive feedback about their current health, and help themselves or others, were associated with increased interest in clinical trial participation. Some factors were not associated with change in interest (e.g., a doctor not recommending participation) while others were associated with decreased interest (e.g., having to come in for multiple visits each week). Relative to other types of interventions, pharmacological intervention trials were associated with the least interest in participation, despite pharmacological interventions being rated as more likely to result in AD treatment. Decreased interest was not predicted by subjective memory concerns, number of current medications, cardiovascular risk, or beliefs about the likely success of pharmacological treatments. These results highlight the challenges faced by researchers investigating pharmacological treatments in non-demented older individuals, and suggest future research could contribute to more effective ways of recruiting participants in AD-related clinical trials.

Comparison of self-reported and informant-reported depressive symptoms in an outpatient neuropsychology clinic sample

ABSTRACT Introduction: Depressive symptoms are common in patients assessed by neuropsychologists, and these symptoms can have an effect on cognitive functioning. Discrepancies between patient- and informant-reported depressive symptoms have been found in some diagnostic groups, raising concerns as to the value of assessing depressive symptoms via self-report alone. This study aimed to examine the relationship between these discrepancies and cognitive functioning in a heterogeneous outpatient neuropsychological sample. Method: Sixty-two patients with a variety of diagnoses (e.g., stroke, mild cognitive impairment) completed a measure of depressive symptoms and a battery of neuropsychological tests. Informants also rated the patients’ depressive symptoms and completed a measure of the patients’ functional abilities. Results: Patient and informant reports of depressive symptoms were moderately correlated; larger differences between patient- and informant-reported symptoms were associated with poorer verbal learning and memory and greater functional impairments. Conclusions: The moderate level of agreement found for depressive symptoms is consistent with past studies using specific neuropsychological patient groups and other studies in both clinical and non-clinical populations. These results highlight the utility of collecting informant reports of depressive symptoms in neuropsychological evaluations.

What types of recommendations are we giving patients? A survey of clinical neuropsychologists

Abstract Objective: To identify types of recommendations that neuropsychologists most frequently give to patients, and determine which specific recommendations are most and least consistently given to patients across and within different diagnostic populations. Method: A total of 309 clinical neuropsychologists completed a survey evaluating the frequency with which they made particular types of recommendations (e.g. driving, employment and education, health and rehabilitation referrals) to patients with various disorders (e.g. traumatic brain injury, dementia). Results: Recommendations in some categories were common across diagnoses (e.g. recommendations related to health). Neuropsychologists reported being more likely to give patients with dementia (relative to other clinical populations) recommendations related to educational resources, supervision and independence, and driving. Patients with a TBI diagnosis were more likely to be given recommendations related to employment and education, health and rehabilitation referrals, and mental health and substance use. Patients with a psychiatric diagnosis were more likely to be given recommendations related to mental health and substance use. Irrespective of diagnosis, neuropsychologists reported being most likely to give recommendations that could be implemented by the patient or caregiver without assistance from outside sources. Neuropsychologists varied in their use of recommendations to seek evaluations and treatment from health and rehabilitation referrals and in providing recommendations related to driving. Conclusions: Recommendations provided within diagnostic groups are consistent with the etiology and prognosis of those conditions. Recommendations for the use of compensatory strategies to address cognitive deficits and recommendations to improve health were endorsed by the vast majority of neuropsychologists across diagnoses.

Response to Shapiro and Chandler Letter to the Editor

pay close attention to discriminant validity more generally. In sum, we agree that identifying and characterizing a sample of individuals with mTBI on MTurk is an important objective because access to this group has the potential to greatly accelerate the pace of discovery in nonclinical research. Bernstein and Calamia do many things right, but it is easy for readers unfamiliar with conducting online studies to overlook details that may impact data quality. Furthermore, the rarity of mTBI leaves a narrow margin for error. It is important for researchers using MTurk to understand its limitations and to ensure that they are taking all necessary steps to address them.

Assessing the Longer-Term Effects of Mild Traumatic Brain Injury on Self-Reported Driving Ability

Mild traumatic brain injury (mTBI) can have long‐term cognitive and functional consequences, and recent mTBI has been associated with impaired performance on measures related to driving ability. However, it remains unclear whether mTBI history is associated with poorer driving performance.

Pedometer-assessed steps per day as a predictor of cognitive performance in older adults.

Objective: To examine associations between pedometer-assessed daily steps and several domains of cognitive functioning in cognitive healthy older adults. Method: A total of 582 cognitively healthy older adults enrolled in a longitudinal aging study completed the Uniform Data Set (UDS) neuropsychological battery (n = 374) or the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS; n = 208) and were asked to wear a pedometer for 1 week. Results: Bivariate correlations revealed associations between average daily steps and attention, executive functioning, language, and memory on one or both cognitive batteries. Multiple regression analyses controlling for demographic, health, and mobility variables demonstrated a significant relationship between average daily steps and the executive functioning composite of the UDS battery and average daily steps and the attention index of the RBANS battery. Both of these composites include measures of processing speed and, therefore, the most robust link between daily steps and cognition may be its association with processing speed. Conclusions: Step counts as measured by a pedometer, a simple and inexpensive measure of daily physical activity, are robustly associated with aspects of cognition. As pedometer-based interventions have shown success in increasing daily physical activity in older adults, a greater number of future studies should consider measuring cognitive outcomes of these interventions.

Examining differences in neuropsychiatric symptom factor trajectories in empirically derived mild cognitive impairment subtypes

The aim of this study was to examine neuropsychiatric symptom (NPS) factor severity progression over time in empirically derived (ED) mild cognitive impairment (MCI) subtypes.

Recommendations for Driving After Neuropsychological Assessment: A Survey of Neuropsychologists

Abstract Objective: To determine which patient, provider and practice-related characteristics are associated with increased likelihood of driving-related recommendations following a neuropsychological evaluation. Method: A total of 309 clinical neuropsychologists completed a survey evaluating the frequency with which they made various driving-related recommendations (e.g., take an on-road driving test) to a variety of clinical populations. Information regarding patient characteristics (e.g., perceived likelihood of adhering to recommendations), provider-related characteristics (e.g., years in practice) and practice-related characteristics (e.g., region of North America) were also collected. Correlation coefficients and multiple linear regressions controlling for patient diagnoses were used to examine which characteristics predicted greater likelihood of driving recommendations broadly. Significant characteristics were then entered into regressions to examine independent contributions of each characteristic to the likelihood of each individual driving recommendation. Results: Neuropsychologists reported giving a range of driving recommendations to most clinical populations, with the exception of patients with psychiatric diagnoses. Several characteristics (e.g., patient impairment) were associated with greater likelihood of driving recommendations broadly. After controlling for diagnoses, three significant predictors emerged: higher frequency of individualized recommendations, greater caregiver presence, and greater number of recommendations given. All three characteristics generally predicted frequency of all individual driving recommendations. Conclusions: While patients’ diagnoses may be indicative of whether they receive driving-related recommendations, certain patient and provider-related factors also uniquely contribute to the likelihood of driving-related recommendations after neuropsychological assessment.

Assessing the effects of concussion using the C3Logix Test Battery: an exploratory study

ABSTRACT The C3Logix is a computerized concussion assessment tool that measures a wider array of symptoms (i.e., balance and oculomotor functioning) than other computerized batteries. Although the C3Logix has been used increasingly by athletic organizations at all levels of play, its utility within the concussion population has not been extensively examined. The current study aimed to determine whether the C3Logix is sensitive to the effects of concussion. A total of 54 student-athletes enrolled at a large southern university completed the C3Logix at baseline and within days following a suspected concussion (mean = 2.93, SD = 3.14). Dependent-samples t-tests revealed that relative to their baselines, following concussion, athletes both reported significantly greater postconcussive symptoms and performed more poorly on measures of reaction time and computer-measured balance. Decrements in processing speed, visual acuity, and clinician-observed errors on tests of balance also trended toward significance. Results suggest that inclusion of measures of balance and oculomotor functioning in the assessment of concussion may provide additional clinical utility above and beyond domains typically measured by computerized concussion assessments.