Nurse-Led AI-Supported Psychosocial Support for Palliative Caregivers

This randomized controlled trial will evaluate the effect of a nurse-led, generative artificial intelligence (AI)-supported individualized psychosocial support program on the care readiness and psychological resilience of primary caregivers of patients receiving palliative care. Palliative care is a holistic approach that aims to improve the quality of life of patients and their families facing problems associated with life-threatening illnesses through the early identification, accurate assessment, and treatment of pain and other physical, psychosocial, and spiritual problems (WHO, 2020). Palliative care also includes a team-based approach that provides support to patients, their families, and caregivers. Primary caregivers undertake a substantial proportion of patients' daily living and care needs, including assistance with activities of daily living and self-care, medication management, communication with healthcare professionals, and emotional, social, and spiritual support (Pinto et al., 2026).

As the disease progresses, increasing caregiving responsibilities, uncertainty about the future, and the possibility of death may result in considerable physical, emotional, and psychosocial challenges for caregivers. Caregivers may not always feel adequately prepared for the caregiving role and may experience difficulties accessing the support they need (Hudson, 2003; McAndrew et al., 2023). These challenges may become particularly evident during the transition from hospital to home. Previous research involving palliative care patients and their caregivers has shown that positive transition experiences are associated with adequate support, the ability to manage medications, and the fulfillment of healthcare needs, whereas unclear caregiving responsibilities, medication-related confusion, uncertainty, and emotional distress may complicate the transition process (Saunders et al., 2021). Caregiver needs also continue after discharge, including adaptation to changing caregiving responsibilities, emotional regulation, coping with difficulties, use of available support resources, and maintenance of self-care (Hudson, 2003; McAndrew et al., 2023).

Care readiness is an important factor influencing caregivers' ability to participate effectively in the caregiving process. Care readiness refers to the extent to which caregivers perceive themselves as prepared to fulfill the physical, emotional, and practical requirements of the caregiving role and to cope with the stress associated with caregiving (Karaman & Karadakovan, 2015). Higher levels of care readiness have been associated with lower levels of depression and caregiver burden among caregivers of individuals with Alzheimer's disease and related dementias (Hancock et al., 2022). Among family caregivers of patients with advanced cancer, support from healthcare professionals and communication about the illness have also been associated with higher levels of care readiness (Häger Tibell et al., 2024). A systematic review and meta-analysis indicated that psychoeducational, educational, supportive, and self-care-oriented interventions may have beneficial effects on caregivers' readiness for caregiving (Bilgin & Özdemir, 2022). Similarly, an intervention based on the identification of individual caregiver support needs was associated with a significant increase in care readiness scores following the intervention (Norinder et al., 2024). However, a web-based psychoeducational intervention for family caregivers of patients receiving palliative home care did not demonstrate a significant effect on care readiness compared with the control group, potentially due to limited user engagement; the authors recommended greater integration of such programs into clinical care processes (Bauman et al., 2025).

Psychological resilience represents another important psychosocial resource associated with adaptation to caregiving. Psychological resilience refers to an individual's capacity to adapt to adversity, maintain functioning, and recover following stressful experiences. It is considered a multidimensional construct involving self-perception, future orientation, behavioral and structural characteristics, social skills, family functioning, and social support (Basım & Çetin, 2011). Among family caregivers of patients with advanced cancer, higher psychological resilience has been associated with greater care readiness and lower levels of anxiety and depressive symptoms (Dionne-Odom et al., 2021). Among family caregivers of patients in the terminal stage, adaptive coping strategies such as active coping, acceptance, positive reframing, planning, and humor have been associated with higher psychological resilience (Shimizu et al., 2025). Psychoeducational interventions may support caregivers' ability to cope with caregiving-related stress and promote psychological adaptation. In a quasi-experimental study of caregivers of patients with amyotrophic lateral sclerosis, a six-week online psychoeducational program resulted in significantly higher psychological resilience and self-compassion scores in the intervention group compared with the control group (Duran & Aydoğdu, 2024). Similarly, a randomized controlled trial among caregivers of adolescents and young adults with cancer found that a psychoeducational intervention delivered through mobile instant messaging reduced unmet caregiver needs, anxiety, and depressive symptoms (Cheng et al., 2024).

Generative AI may provide opportunities to deliver individualized explanations, case examples, and psychoeducational feedback according to caregivers' needs. However, because generative AI systems have important limitations regarding accuracy and reliability, they should not replace clinical assessment by nurses or other healthcare professionals and should be used under appropriate professional supervision (Tischendorf et al., 2025). In a study evaluating a domain-specific generative AI agent developed to support families of children and adolescents with anorexia nervosa, 91.6% of 477 responses were considered clinically appropriate by experts. However, the study was limited to system development and expert evaluation, and its effects on caregivers' psychosocial outcomes were not examined (Hanzawa et al., 2026). AI-supported systems in palliative care may have potential to provide patients and caregivers with information, emotional support, and practical assistance (Jafari Dehnayebi et al., 2026). A scoping review of AI applications for individuals with serious illnesses and their family caregivers in home settings indicated that these technologies may facilitate individualized care, continuity of care, emotional support, and caregiver support. Nevertheless, most existing studies have focused on feasibility, usability, and user experience, while research evaluating clinical and psychosocial outcomes remains limited (Xiao et al., 2026). Furthermore, a systematic review of AI applications in palliative care reported that 86% of the included studies were retrospective or proof-of-concept studies, with only seven randomized controlled trials and six prospective studies identified. These limitations highlight the need for controlled studies evaluating the effects of AI-supported interventions on psychosocial outcomes among caregivers in palliative care (Bozkurt et al., 2025).

In this trial, generative AI will be used as a nurse-supervised supportive tool rather than as an autonomous clinical decision-maker. The intervention content will be developed by the research team and validated through expert review. Predefined safety rules and standardized core content will be used throughout the intervention. The six-week program will consist of six modules: (1) adaptation to the caregiving role and care readiness, (2) understanding caregiving needs and coping with uncertainty, (3) stress management and emotional regulation, (4) adaptive coping and problem solving, (5) self-care and social support, and (6) meaning-making and strengthening personal resources related to the caregiving experience. The standardized core content will be maintained for all participants, while explanations, information cards, case examples, brief exercises, mini-questions, and psychoeducational feedback will be individualized according to the caregiver's psychosocial needs, caregiving experience, responses during the program, and basic characteristics of the patient's care needs.

The generative AI system will not independently diagnose psychological or clinical conditions, make treatment decisions, modify medical treatment, or replace professional nursing assessment. The nurse will maintain an active supervisory role throughout the intervention and will evaluate participant responses according to predefined monitoring procedures. When responses indicate a need for clinical assessment, additional psychological support, or referral to another healthcare professional, the nurse will initiate the appropriate referral process.

The primary outcomes of the trial will be care readiness and psychological resilience among primary caregivers of palliative care patients. Changes in these outcomes will be evaluated between the intervention and control groups at predefined assessment time points. The study aims to determine whether a nurse-led, generative AI-supported individualized psychosocial support program can improve caregivers' preparedness for the caregiving role and psychological resilience.

By evaluating a professionally supervised generative AI-supported psychosocial intervention in a randomized controlled design, this study is expected to contribute empirical evidence to the limited literature on the clinical and psychosocial effectiveness of generative AI interventions for caregivers in palliative care.