The Pennsylvania State University, along with its subcontractors, is submitting this proposal in response to ?RFA-AG-18-012 Mobile Monitoring of Cognitive Change? (denoted as M2C2). The Data and Statistical Analysis Core for the proposed Project will ensure data interoperability, secure data access, and maintain a uniform data structure, permit data sharing, and oversee data privacy concerns. The significance and innovation of the Data Core for this Project lies in providing (1) a secure data repository for managing data from the M2C2 infrastructure and (2) a rigorous evaluation of reliability and validity of cognitive tests for both between-person and within-person study designs. The Data Core will serve the M2C2 Project by pursuing two primary aims, each of which has multiple sub-aims.
Aim 1 : Operate a Central Data Repository. Develop and operate a central data repository for the M2C2 data, including cognitive tests collected by mobile devices and in-person, self-report, device meta-data (e.g., device model, OS version), and data passively collected by smartphone sensors (e.g., ambient noise, GPS location).
Aim 2 : Conduct analyses to evaluate the psychometric properties and validity of mobile cognitive tests. Provide oversight of all analytics for the M2C2 project and evaluate results against established benchmarks, ensure that evaluations of psychometric test properties correct for bias caused by generalized age effects through stratified analysis and statistical control, and evaluate invariance across age, gender and racial categories to establish validity for population subgroups. The focus of Aim 2 is the conduct of statistical analyses to evaluate the psychometric properties for the test instruments developed in the Project. This includes the evaluation of test reliability by examining the short-term stability of test performance, assess construct validity of candidate mobile tests against gold-standard measures (e.g., NIH Toolbox), and evaluate longitudinal validity of mobile cognitive tests by examining sensitivity for detecting both short-term and long-term cognitive change. In later phases of the Project, the Data Core will replicate reliability and validity in an independent, nationally representative probability based sample, and create nationally representative norms for baseline test performance, and for change in test performance (longitudinal norms). 1
Identification of modifiable risk factors prior to the development of neurological conditions in old age represents a critical challenge for developing effective treatment and preventive measures, controlling health care costs, and ultimately improving the quality of life for seniors and their families. This requires methods to detect subtle cognitive changes years prior to the onset of discernable symptoms. The proposed research will significantly advance these efforts by developing innovative ambulatory methods that rely on mobile and sensor technology to measure the cognitive and behavioral function of people in their everyday life. These methods could provide more cost-effective, accurate, sensitive and `ecologically valid' measurements of early signs of cognitive impairment than are currently available through standard laboratory and clinical practices. By improving our ability to measure cognitive function in daily life, this work will set the stage for the next generation of early intervention and prevention studies to slow or prevent cognitive decline.
