Background: Individuals with dementia have communication limitations resulting from cognitive impairments that define the syndrome. Whereas there are numerous cognitive assessments for individuals with dementia, there are far fewer communication assessments. The Profiling Communication Ability in Dementia (P-CAD) was developed to address this gap. Aims: The purpose of this study was to examine the concurrent validity, longitudinal validity and inter- and intra-rater reliability of the P-CAD in a population of people with dementia and their communication partners. Method: The P-CAD was administered to 122 people with dementia and their communication partners in Ireland (n = 100) and Canada (n = 22), over a 12-month period. To establish concurrent validity of the P-CAD, scores were compared to scores obtained from existing standardized instruments including the Functional Linguistic Communication Inventory (FLCI), the Mini-Mental State Examination (MMSE-2) and Global Deterioration Scale (GDS). Inter-rater reliability and responsiveness (longitudinal validity) were analysed using data from a subgroup of participants. Outcomes & Results: Overall P-CAD test scores were significantly correlated with FLCI (n = 122; r = 0.875; p < 0.001) and MMSE-2 total scores (n = 122; r = 0.857; p < 0.001). Levels of communication support categories on the P-CAD correlated with GDS rankings (n = 122; rho = -0.539; p < 0.001) and MMSE-2 total scores (n = 122; rho = 0.680, p < 0.001). Inter-rater reliability tested for 20 participants in the Irish sample revealed high levels of agreement between raters in scoring the GDS (n = 20; ICC = 0.969, p < 0.001), MMSE-2 (n = 20; ICC = 0.997, p < 0.001), FLCI (n = 20; ICC = 0.999, p < 0.001) and P-CAD (n = 20; ICC = 0.981, p < 0.001). To establish longitudinal validity to examine if the P-CAD was responsive to changes in cognitive-communication function over time, 12 participants in the Irish sample repeated all tests 3 months after the initial testing. No statistically significant differences in test scores were found for the 12 participants who completed follow-up measures at this time point in any of the three scales. It was not possible to determine sufficient responsiveness as correlations between the change in P-CAD scores over 3 months were insignificant for both the change in MMSE-2 scores (rho = -0.130, p = 0.704) and the FLCI scores (rho = 0.221, p = 0.513). Conclusions & Implications: In this study, P-CAD has demonstrated good concurrent validity and inter-rater reliability in samples collected in two countries with English-speaking participants. The P-CAD is appropriate for use to evaluate communication abilities of people with dementia, including during conversational interactions with caregivers.