Background: A stroke is a sudden, life-altering event with potentially devastating consequences for survivors and their loved ones. Despite advances in endovascular and neurocritical care approaches to stroke treatment and recovery, there remains a considerable unmet need for interventions targeting the emotional impact of stroke for both patients and their informal caregivers. This is important because untreated emotional distress becomes chronic and negatively impacts quality of life in both patients and caregivers. Our team previously used mixed methods to iteratively develop a six-session modular dyadic intervention to prevent chronic emotional distress in patients with stroke and their informal caregivers called "Recovering Together" (RT) using feedback from dyads and the medical team. The aim of the current study is to test the feasibility of recruitment, acceptability of screening and randomization methods, acceptability of RT, satisfaction with RT, feasibility of the assessment process at all time points, and acceptability of outcome measures. Secondarily, we aimed to explore within-treatment effect sizes and change in clinically significant symptoms of depression, anxiety, and post-traumatic stress (PTS). The larger goal was to strengthen methodological rigor before a subsequent efficacy trial. Methods: We conducted a feasibility randomized controlled trial to evaluate the RT intervention relative to minimally enhanced usual care (MEUC) in stroke patients admitted to a Neurosciences Intensive Care Unit (Neuro-ICU). Dyads were enrolled within 1 week of hospitalization if they met specific eligibility criteria. Assessments were done via paper and pencil at baseline, and electronically via REDCap or over the phone at post-intervention (approximately 6 weeks after baseline), and 3 months later. Assessments included demographics, resiliency intervention targets (mindfulness, coping, self-efficacy, and interpersonal bond), and emotional distress (depression, anxiety, and PTS). Primary outcomes were feasibility and acceptability markers. Secondary outcomes were depression, anxiety, PTS, mindfulness, coping, self-efficacy, and interpersonal bond. Results: We consented 20 dyads, enrolled 17, and retained 16. Although many patients were missed before we could approach them, very few declined to participate or dropped out once study staff made initial contact. Feasibility of enrollment (87% of eligible dyads enrolled), acceptability of screening, and randomization (all RT dyads retained after randomization) were excellent. Program satisfaction (RT post-test M = 11.33/12 for patients M = 12/12 for caregivers), and adherence to treatment sessions (six of seven RT dyads attending 4/6 sessions) were high. There were no technical difficulties that affected the delivery of the intervention. There was minimal missing data. For both patients and caregivers, participation in RT was generally associated with clinically significant improvement in emotional distress symptoms from baseline to post-test. Participation in MEUC was associated with clinically significant worsening in emotional distress. Although some of the improvement in emotional distress symptoms decreased in the RT group between post-test to 3 months, these changes were not clinically significant. RT was also associated with substantial decrease in frequency of individuals who met criteria for clinically significant symptoms, while the opposite was true for MEUC. There were many lessons that informed current and future research. Conclusions: This study provided evidence of feasibility and signal of improvement in RT, as well as necessary methodological changes to increase recruitment efficiency before the future hybrid efficacy-effectiveness trial. Trial registration: NCT02797509. © 2020 The Author(s).