Developing and validating the Communicating Needs with Care Partners Scale (CONNECT): a person-centered brain injury outcome measure

About This Grant

PROJECT SUMMARY In this multi-site study, we will develop and validate the Communicating Needs with Care Partners (CONNECT) scale, which assess recovery of basic communication, a highly valued, but rarely measured milestone of severe brain injury recovery. Severe brain injury affects more than one million people worldwide each year. The most common cause of in-hospital deaths is withdrawal of life-sustaining treatment due to poor prognosis. There are no treatments proven to accelerate recovery from acute brain injury and no prognostic tools that can accurately predict outcome at the individual patient level. One factor contributing to these gaps is the imprecision of brain injury outcome assessment tools which rely on broad categories of global function to determine whether recovery is “favorable” or “unfavorable”. Multiple studies by our group and others indicate that the ability to communicate is a meaningful and desired outcome for persons with brain injury and their caregivers. Communication restores autonomy and provides a neuroanatomically-based target for intervention. Yet, there are no robust measures of communication that are designed for persons with brain injury. We developed a prototype of CONNECT and propose to finalize its content, assess its validity and reliability, and evaluate its Rasch measurement properties. CONNECT is administered to care partners as a structured interview that can be completed remotely. Assessment ranges from attempt to communicate at the floor to accurate expression of basic needs at the ceiling. First, we will work with clinicians, investigators, persons with brain injury, and care partners to finalize CONNECT items, establish content and face validity, and develop an administration and scoring guide. Then we will assess CONNECT’s concurrent validity, predictive validity, inter- rater reliability, test-retest reliability, and care partner variability in patients with brain injury enrolled at two acute care hospitals. We will use Rasch analyses to determine whether CONNECT is monotonic (i.e., rating scale options occur sequentially), hierarchical (i.e., items are arranged along a gradient of increasing difficulty), and unidimensional (i.e., items reflect the same underlying latent trait, recovery of communication). This study leverages an interdisciplinary team of neuroscientists, rehabilitation measurement experts, occupational therapists, and speech language pathology specialists, with expertise and experience in developing diagnostic and prognostic tools for patients with severe brain injury. CONNECT has the potential to accelerate treatment discovery, detect functional improvement and decline that would otherwise go undetected, and contribute to the development of precise prognostic models that inform decisions about withdrawing life- sustaining treatment after brain injury.