Comment on ‘Comfort Needs of Kidney Transplant Recipients: A Qualitative Analysis Based on Kolcaba's Comfort Theory’

We read with great interest the article by Calazans et al. (2026), titled ‘Comfort Needs of Kidney Transplant Recipients: A Qualitative Analysis Based on Kolcaba's Comfort Theory’. This study provides a valuable holistic understanding of comfort across physiological, environmental, sociocultural and psychospiritual dimensions, highlighting the importance of a comprehensive care approach in post-transplant management. Although the qualitative findings are highly insightful, we would like to explore how digital health technologies could extend this work to enable predictive and personalised comfort management. This study relies on retrospective self-reports obtained through interviews. Although such methods offer rich narrative depth, they are susceptible to recall bias and lack temporal granularity. Comfort is a dynamic state that fluctuates daily due to medication side effects, emotional stress, social interactions and environmental triggers (Habighorst 2025), factors that are difficult to capture through periodic interviews. Although the study identifies key determinants of comfort, including pain, family support, religious beliefs and medication adherence, these factors are not translated into measurable, real-time indicators that could support early intervention. The absence of a continuous monitoring framework limits the ability to anticipate comfort crises before they escalate, potentially delaying supportive care. We propose integrating Kolcaba's Comfort Theory with digital health technologies to establish a clinically operable ‘Digital Phenotyping System for Comfort in Kidney Transplantation’. This system would combine continuous, active and passive data collection to capture multidimensional signals reflecting patients' comfort in daily environments. Leveraging computational models, it could enable dynamic assessment and risk prediction of comfort, shifting care from reactive responses to proactive, anticipatory interventions. The overall implementation workflow of this system is illustrated in Figure 1. A smartphone application tailored for transplant rehabilitation could integrate multi-level data acquisition. At the vocal level, patients could record short voice diaries, allowing automatic detection of emotional trends and psychological stress using acoustic features (e.g., pitch, speech rate and amplitude) and natural language processing. At the behavioural level, wearable devices (Wang et al. 2023) or built-in smartphone sensors could continuously monitor gait stability, daily activity, sleep patterns and mobility, indirectly reflecting physiological comfort and social participation. At the self-report level, micro-surveys could capture real-time perceptions of pain, environmental conditions, family support and psychological well-being. The system could integrate multidimensional time-series data using machine learning to identify patterns of comfort deterioration. Behavioural data might be preprocessed using time-series feature extraction methods, while voice data could be analysed via acoustic and text feature fusion models. Self-report data could be encoded as multimodal input vectors. Sequential predictive models, such as LSTM (Kong et al. 2025), could then generate individualised comfort risk predictions, identifying patients at risk of pain or depressive symptoms in the near term. For instance, a sustained decline in daytime activity combined with low night-time voice sentiment could signal elevated risk, enabling a shift from post-symptom intervention to pre-emptive risk alerts. To achieve closed-loop management, the system could include clinical decision support functionality. High-risk patterns would trigger graded alerts to the nursing team, along with concise data summaries for remote follow-up or outpatient intervention. Simultaneously, personalised recommendations could be delivered to patients, aligned with Kolcaba's comfort contexts, such as environmental adjustments for physical discomfort or mindfulness exercises and psychosocial resources for psychological distress. Translating this system into clinical practice requires addressing key challenges: (1) validating digital phenotyping indicators and their correlation with conventional assessments; (2) conducting prospective trials to evaluate impact on patient-reported comfort and readmission rates; and (3) ensuring privacy, ethics and accessibility in system design. Integrating digital phenotyping into post-transplant care aligns with the holistic perspective of Kolcaba's theory while enhancing predictive precision. By shifting from retrospective assessments to real-time predictive modelling, clinicians can move from reactive responses to proactive comfort management. Future research should explore the feasibility, acceptability and effectiveness of such digital tools in transplant populations. This approach has the potential to transform comfort from a subjective experience into a measurable and actionable dimension of rehabilitation. We commend Calazans et al. for their foundational work and encourage interdisciplinary collaboration among nursing science, data engineering and behavioural informatics to build upon it. Declaration of Generative AI and AI-Assisted Technologies in the Writing Process: Generative AI tools (Paperpal and ChatGPT-4o) were used solely for language polishing and stylistic refinement. They did not contribute to study design, data analysis, interpretation or scientific content. All intellectual contributions remain the responsibility of the authors, who reviewed and approved the final manuscript for accuracy and originality. The authors have nothing to report. The authors have nothing to report. The authors have nothing to report. The authors declare no conflicts of interest. The authors have nothing to report.