Background:Social media is a common source of cancer information for patients, caregivers, and survivors. Factors influencing engagement with oncology nutrition content on social media are not well-defined, despite the high prevalence of social media use for health information. The impact of bilingual content strategies on that engagement is also poorly understood. Objective:This study aims to describe changes in postlevel engagement on a Facebook page used to disseminate evidence-based cancer nutrition information during the period before and after the implementation of a bilingual, multiformat content strategy and to identify creative and linguistic features associated with audience engagement at the post level. Methods:This cross-sectional study analyzed 306 Facebook posts from Cook for Your Life over 12 months, comprising 9 months before the strategy change (T1) and 3 months after (T2). Postlevel engagement metrics (likes, comments, shares, and link clicks) and aggregated user demographics were obtained from Meta Business Suite. Trained coders annotated visual and creative features, including branding, post shape, and infographics, as well as topical categories. Linguistic Inquiry and Word Count characterized language features. Group differences across time periods were tested using t tests and chi-square tests. Generalized linear models estimated associations between post characteristics and engagement, adjusting for the time period. Results:Users who engaged with oncology nutrition Facebook content (N=10,196) were mostly female (n=9500, 93.2%) participants, over 35 years of age (n=9311, 91.4%), based in the United States (n=6708, 71.4%), and English-language users (n=9419, 93.4%). Engagement tripled from T1 (mean 5.3, SD 4.2) to T2 (mean 17.7, SD 37.9; P<.001). Creative features such as branding and format were positively associated with engagement. Linguistically, there was a shift toward longer sentence structures, simplified vocabulary, increased analytical language, and a decline in emotionally charged and cognitive language in posts during T2. In models controlling for time period, posts that used bilingual content had 1.72-fold higher odds of audience engagement than English-only posts (95% CI 1.05-2.80). Conclusions:Implementing a bilingual, multiformat content strategy may promote engagement with oncology nutrition information on Facebook and enhance the equitable dissemination of health information. The use of creative visual design elements in combination with language that is more analytical and emotionally neutral was associated with higher engagement with educational public health content. Future research should test bilingual language presentation more directly and examine whether exposure to such content influences health beliefs and downstream behaviors across additional platforms.
Background:Digital health solutions that incorporate electronic patient-reported outcomes (ePROs) hold promise for enhancing communication and patient engagement in pediatric oncology and palliative care. While ePROs are increasingly used in adult populations, their use in children, especially when combined with gamification, remains underexplored. Objective:This MyPal4Kids study aimed to assess the feasibility, acceptability, and user engagement of a digital health platform that integrates ePROs and a serious game, designed for pediatric oncology patients aged 6-17 years and their caregivers. Methods:A multicenter observational feasibility study was conducted between December 2020 and September 2022 at 3 clinical sites in Germany and the Czech Republic. A total of 83 children and adolescents with cancer, their parent or legal guardian, as well as 10 health care professionals (HCPs) participated. Primary outcomes included platform acceptability and engagement, assessed through recruitment, participation, and attrition rates, adherence, and usability ratings. Secondary outcomes focused on the feasibility of ePRO-based data capture and the perceived impact on HCP workflows and communication. A mixed methods design was used, combining in-app data and surveys with qualitative insights from focus groups. Results:The recruitment rate was 55%, with an attrition rate of 18%. Usability was rated positively, with most users finding the platform intuitive. Sustained engagement declined over time, particularly among older children who found the serious game insufficiently engaging. Adherence varied by age group and HCP involvement. Participants reported increased motivation when symptom reports were acknowledged by HCPs, although the impact on communication was perceived inconsistently. Conclusions:The developed platform shows potential for use in clinical care and research. However, maintaining long-term engagement remains a challenge. Tailoring content to different age groups and strengthening feedback mechanisms from HCPs are critical for improving the user experience. Successful implementation in routine clinical practice will require integration into existing workflows and digital infrastructure, along with ongoing user-driven development.
Background:Mobile health (mHealth) apps offer new opportunities to support cancer survivors in managing their health, adopting healthier lifestyles, and increasing awareness of recurrence symptoms. However, despite their potential, little is known about survivors' use of such tools in routine follow-up care, or which factors are associated with their engagement. Objective:The aim of the present study was to identify survivor characteristics associated with the use of the Lifestyle and Empowerment Techniques in Survivorship of Gynecologic Oncology (LETSGO) app. In addition, we aimed to describe engagement patterns among gynecologic cancer survivors during the first year of follow-up after treatment. Methods:App data from 378 participants in the intervention group of the LETSGO multicenter clinical trial were included in the analysis. Using multiple logistic regression, we analyzed app data from the first year of LETSGO app use to identify participant characteristics associated with use. In addition, we described the frequency of user interactions with app features and identified the most frequently used features during the first year of the intervention in a subset of 49 participants with complete app log data. Results:Of the 378 participants in the intervention group, 267 (70.6%) used the LETSGO app at least twice and were defined as app users. The median age of app users was 63 (53-70) years vs 70 (60-78) years among app nonusers (P<.001). App use was associated with younger age (odds ratio [OR] 0.96 per year, 95% CI 0.93-0.99; P=.003), higher education (OR 2.2, 95% CI 1.1-4.6; P=.03), and having ovarian cancer (OR 2.9; 95% CI 1.1-7.5; P=.03). The LETSGO app included a monthly reminder for symptom self-registration, and monthly peaks in the symptom monitoring feature corresponded with the timing of this reminder. App log data from the full first year of the intervention were available for 49 participants. Among these, user engagement was highest for the physical activity and activity goal-setting features, whereas the disease information feature showed steady but less frequent use. Conclusions:Our findings indicate that LETSGO app use was associated with survivor characteristics such as age, education level, and cancer type. Considering user characteristics when tailoring mHealth apps may support user engagement in digital follow-up care and inform the development of mHealth tools that better align with survivors' needs and preferences. Engagement with the LETSGO app during the first year of follow-up among a subset of participants with complete app log data was highest for the physical activity and activity goal-setting features. The pattern of symptom registration following reminders indicates that scheduled prompts may support regular app engagement, while steady use of the disease information feature suggests an ongoing need for accessible health information during follow-up.
Background:Accelerating the transition to value-based health care (VBHC) is essential to ensure sustainable care delivery. VBHC maximizes patient value by optimizing outcomes, controlling costs, and leveraging data to improve quality and patient-doctor communication. Integrating real-world data in health care is important to achieve informed decisions, especially in palliative care, where choices are complex. Objective:This study described the development and pilot-testing of an information tool that incorporates real-world outcome data for women with metastatic breast cancer who are initiating CDK4/6 inhibitor treatment. Real-world insights are needed to better inform these patients, in particular because real-world outcomes are known to differ from trial results because of larger heterogeneity in patient characteristics, and differences in frequency of check-ups and handling side effects. Methods:We developed an information tool together with patient representatives and clinicians using a participatory development approach that consisted of five key steps: (1) establishment of a multidisciplinary steering group (n=10 steering group members), (2) mapping of the patient journey and patients' needs through focus groups (n=9) and semistructured interviews (n=8), (3) extraction of real-world outcome data from electronic health records systems of 229 patients, (4) prototyping of the tool (n=10), and (5) pilot evaluation with the targeted patient population using semistructured interviews (n=38). We used qualitative analysis methods to analyze the focus group and interview data. Results:We developed a tool consisting of (1) a communication aid for use during doctor-patient consultations (ie, KIJKgesprek [Stichting Kijksluiter]) and (2) a 2-component companion app with informational videos for use at home, both incorporating real-world outcome data (ie, KIJKbericht and KIJKsluiter [Stichting Kijksluiter]). Participants valued the tool for its clarity and structured design, reporting that the outcome data reinforced their experiences and facilitated the setting of realistic expectations. However, some participants described the outcome data as overwhelming, underscoring the importance of careful framing and delivery. Preferences regarding the type, level of detail, and timing of information presentation varied among participants, highlighting the necessity of individualizing information tools to meet diverse informational needs. Conclusions:Most patients valued the inclusion of real-world outcome data in the information tool, although many found it challenging to process. Preferences for the type and presentation of information varied widely among individuals. Information tools incorporating outcome data have the potential to enhance patient understanding and support informed decision-making about care that they value most. However, these tools must be designed to allow for customization, ensuring they address individual informational needs and preferences effectively.
Background:Delayed cancer diagnosis leads to poorer outcomes. Unintended weight loss (UWL) is a nonspecific symptom associated with cancer and other serious conditions, which can make it complex to identify the underlying cause. Clinical decision support systems (CDSSs) can provide evidence-based recommendations to facilitate timely investigation and diagnosis. Objective:This study aimed to pilot the implementation of a CDSS for improving early cancer detection in primary care patients with UWL. Methods:Five practices reviewed patients identified by the UWL CDSS. Staff were interviewed on the acceptability and feasibility of the CDSS. Interviews were analyzed thematically using 2 relevant frameworks: the acceptability of health care interventions and the sociotechnical model for evaluation of digital interventions framework. Clinical audits assessed the correct identification of UWL, follow-up rates, and patient characteristics. Results:Of 60 patients identified by the CDSS as potentially having UWL, 36 (60%) had true UWL. Among the misclassified cases, most patients (16/55, 30%) were intentionally trying to lose weight; this intention was documented as free text in the clinical notes for 98% (58/60) of patients and in a structured field for only 20% of patients. Of the 36 patients, 5 (14%) were actively recalled by their practices for further follow-up; the others (31/36, 86%) were not recalled, as most were deemed already under appropriate follow-up. By 6 months, 94% (34/36) of the cohort had received follow-up care regardless of whether they had been formally recalled. One-third (12/36, 33%) of patients had no additional symptoms, while 36% (13/36) had recorded additional abdominal symptoms. Mental health conditions accounted for 19% of diagnoses linked to UWL consultation. Practice staff were generally receptive to the UWL CDSS concept. It was particularly appreciated by practices with strong quality improvement processes and prior CDSS experience. A high proportion of misclassified patients, poor workflow integration, and conflicting patients' agendas were cited as implementation barriers. Conclusions:Our study revealed challenges and potential benefits of implementing a CDSS for identifying patients with UWL at risk of cancer in primary care. While integration issues and misclassification of patients were noted, high rates of follow-up care were observed regardless of the CDSS. These high follow-up rates prompt consideration of whether they reflect Australian primary care and whether a CDSS with these characteristics is the most appropriate approach to support early cancer detection. Future research should focus on improving CDSS integration with existing workflows, enhancing correct identification through access to clinical notes and the use of more sophisticated digital methods (eg, AI). These findings highlight the potential of CDSS in improving patient care and the complexities involved in their successful implementation.
Background:Pediatric cancer presents many challenges for young patients and families that extend beyond active treatment. Families report significant unmet information and supportive care needs at the end of treatment (EOT). Digital health solutions, such as patient apps, offer a potential solution to address current gaps in care. Objective:This study aimed to develop and test a smartphone app called LotusLab, designed to improve access to survivorship information for parents of childhood cancer survivors aged 0 to 18 years, and for young people (aged 12 years and older) at the EOT. In addition, this study aimed to examine parents' and health care providers' (HCPs) perceptions of LotusLab during β-testing to inform further app development and refinement. Methods:Participants were invited to explore the content and key functionality of LotusLab, including an embedded patient-reported outcome measure (PROM). Evaluation of the usability, acceptability, usefulness, and feasibility of the app was conducted through quantitative surveys and qualitative semistructured interviews. Results:Participants comprised 8 parents whose children had completed cancer treatment, and 9 HCPs. All participants (100%) "agreed" or "strongly agreed" that the app was acceptable; however, HCPs reported more neutral acceptability responses compared with parents. Parents scored app usability more highly, with a mean System Usability Scale (SUS) score of 86.88 (SD 8.74), than HCPs' mean SUS score of 66.39 (SD 10.16). A common perceived benefit of LotusLab was the provision of trusted information and psychosocial resources. Suggestions for improvement included adding a search function, enhancing the layout, and adding additional topics. All parents completed the app-embedded PROM. Both cohorts noted the value of the PROM in enhancing EOT consultations, while highlighting considerations for the effective integration of and response to PROM data. Conclusions:Initial data indicate that LotusLab is easy for parents to use and provides value through the provision of EOT information and the opportunity to enhance the use of the PROM. Consistent with the co-design methodology, these findings are guiding the ongoing development and implementation of LotusLab.
Background:Large language model (LLM)-based conversational agents are increasingly used in health care, yet their capacity to support genuine multiturn dialogue remains underexplored. In oncology, where patients and caregivers experience complex informational and emotional needs throughout the disease trajectory, conversational agents may support information provision, symptom consultation, and emotional assistance. However, research specifically examining multiturn conversational agents designed for patients with cancer and informal caregivers remains limited. Objective:This scoping review aimed to map the research landscape of LLM-based multiturn conversational chatbots developed for patients with cancer and informal caregivers, focusing on system design, intervention purposes, evaluation approaches, safety considerations, and transparency of LLM-related components. Methods:This scoping review followed the Joanna Briggs Institute methodology and PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. Six databases-PubMed, Embase, Scopus, Web of Science, CINAHL, and PsycINFO-were searched for studies published between January 2022 and January 2026, with supplementary searches conducted in IEEE Xplore Digital Library and ACM Digital Library in May 2026. Studies were included if they described LLM-based chatbots designed for patients with cancer or informal caregivers that supported multiturn conversational interaction. Two reviewers independently conducted the study selection and data extraction. Results:Eight studies met the inclusion criteria. Most studies focused on prototype development, with limited research evaluating clinical outcomes. ChatGPT-based models were the most commonly used LLMs, and retrieval-augmented generation techniques were applied in several studies. Chatbots were primarily designed for emotional support or information provision. Evaluation approaches varied widely, including response quality, psychological outcomes, and user experience. However, no studies evaluated interaction-level characteristics such as conversational continuity or context retention, and only 2 studies reported any conversational memory mechanism. Reporting on safety risks, mitigation strategies, prompt design, model parameters, and adherence to LLM reporting guidelines was often limited or absent. Conclusions:This scoping review identified only 8 studies on LLM-based multiturn conversational chatbots for patients with cancer and informal caregivers. The field remains at an early stage, characterized by prototype-oriented development, heterogeneous design and evaluation approaches, and inconsistent safety and transparency reporting. Future development should prioritize genuine conversational capability, safety management, and transparent reporting.
Background:Cardiovascular disease is a leading noncancer cause of morbidity and mortality among individuals diagnosed with cancer. Although modern cancer therapies have improved survival, many are associated with cardiotoxic effects that increase the risk of cardiovascular complications. Early identification of cardiovascular symptoms and signs may support timely clinical management. However, tracking cardiovascular health during cancer care can be complex for both patients and medical team participants, and important indicators, such as blood pressure changes, heart rate variability, fluid retention, chest pain, or new-onset fatigue, may be missed between clinic visits. Mobile health technologies offer potential tools to facilitate remote monitoring, yet behavioral factors influencing cardiovascular tracking in cardio-oncology remain insufficiently understood. Objective:This study aimed to identify barriers and facilitators influencing cardiovascular symptom tracking among patients with cancer and medical team participants using the capability, opportunity, motivation-behavior (COM-B) framework. Methods:This qualitative descriptive study included adult patients with cancer (n=12) receiving cardiotoxic therapies and members of their health care teams (n=12), including oncologists, nurses, advanced practice providers, and allied health professionals. Participants were recruited from a cardio-oncology clinic. Semistructured interviews explored perceptions of tracking cardiovascular symptoms and signs. Data were analyzed using rapid qualitative analysis and organized according to COM-B domains. The study received ethical approval, and participants provided informed consent. Results:Patients with cancer and medical team participants described cardiovascular symptom tracking as a shared responsibility. Capability-related themes included uncertainty regarding which symptoms to monitor, appropriate thresholds for concern, and variability in communication practices. Opportunity-related themes included time pressures, competing clinical demands, and the influence of patient-medical team participant relationships and social support. Motivation-related themes reflected perceived benefits of tracking, including reassurance for patients and improved clinical insight for medical team participants, alongside concerns about information burden and unclear actionability. Both groups expressed interest in digital tools that provide clear guidance, defined response pathways, and integration within health care workflows. Conclusions:Cardiovascular symptom tracking in cancer care is shaped by interrelated behavioral and contextual factors affecting both patients and medical team participants. Digital systems intended to support tracking should address knowledge clarity, workflow integration, and actionability of reported information. Understanding these determinants may inform the design and implementation of cardiovascular monitoring strategies in cardio-oncology settings.
Background:Emerging evidence indicates that inflammation plays a crucial role in cancer prognosis. Inflammatory response biomarkers are recognized as promising prognostic factors for mortality in patients with cancer. Objective:This study aims to evaluate the prognostic significance of the systemic inflammatory response index (SIRI), systemic immune-inflammation index (SII), platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), inflammatory prognostic index (IPI), and C-reactive protein-albumin-lymphocyte (CALLY) index. Methods:Weighted Cox regression analyses, restricted cubic spline models, Kaplan-Meier survival curves, and receiver operating characteristic analyses were performed to assess the predictive value of the 6 inflammatory markers for mortality. Subgroup analyses and sensitivity analyses were conducted to examine associations within specific subpopulations. Results:Cox regression models demonstrated that SIRI, NLR, IPI, and CALLY were significant predictors of all-cause mortality (tertile 3 vs tertile 1; hazard ratio [HR]: SIRI: 1.72, 95% CI 1.29-2.27; NLR: 1.33, 95% CI 1.02-1.74; IPI: 1.48, 95% CI 1.14-1.92; CALLY: 0.66, 95% CI 0.51-0.85). IPI (HR 1.91, 95% CI 1.11-3.27) and CALLY (HR 0.53, 95% CI 0.31-0.90) were significantly associated with cancer-specific mortality, whereas only SIRI was able to predict cardiovascular mortality (P value for trend=.04). Dose-response relationships were observed between the 6 inflammatory markers and mortality outcomes. Kaplan-Meier survival curves further illustrated significant differences between tertile groups (log-rank test, P<.001). The 6 inflammatory indices exhibited moderate predictive ability for all-cause mortality. IPI yielded the highest area under the curve (AUC) for cancer-specific mortality (AUC=0.6338), and SIRI was the most efficient predictor of cardiovascular mortality (AUC=0.687). No significant interactions were observed between the 6 inflammatory markers and most subgroup variables. Conclusions:SIRI, NLR, IPI, and CALLY represent convenient and cost-effective prognostic tools for predicting mortality in patients with cancer. In contrast, SII and PLR may not be reliable prognostic biomarkers.
Abstract Background Chemotherapy-induced alopecia is among the most psychologically distressing adverse effects of systemic cancer therapy. Although scalp cooling is increasingly used to mitigate hair loss, it is still largely perceived as a cosmetic intervention. Its broader psychological relevance and the biological basis of treatment success, particularly the preservation of follicular integrity under ongoing cytotoxic exposure, remain insufficiently explored. Objective This study aimed to reconceptualize scalp cooling beyond visible hair preservation by examining its psychological impact on patients receiving highly alopecia-inducing chemotherapy, while integrating quantitative objective hair preservation metrics with structural and ultrastructural analyses of hair follicle damage to identify avenues for improving follicular integrity and scalp-cooling efficiency. Methods A total of 82 patients undergoing highly alopecia-inducing chemotherapy consisting of a sequential anthracycline-taxane regimen (4 cycles of epirubicin and cyclophosphamide followed by 12 weekly paclitaxel applications) received standardized scalp cooling. Objective hair preservation was quantified using the hair mass index (HMI) as a standardized and reproducible measure of hair retention. Structural and ultrastructural follicular integrity was assessed using light microscopy as well as scanning and transmission electron microscopy. Objective hair preservation metrics were analyzed in relation to patient-reported quality-of-life outcomes (EORTC [European Organisation for Research and Treatment of Cancer]–based measures), subjective treatment burden, and cognitive appraisal of the scalp-cooling experience. Multivariable regression models were applied to identify determinants of posttherapeutic quality of life. Results Visible chemotherapy-induced alopecia was successfully prevented in more than half of the treated patients. Scalp cooling resulted in substantial objective hair preservation as quantified by the HMI. However, HMI values showed only a limited association with posttherapeutic quality-of-life outcomes. In contrast, the cognitive appraisal of scalp cooling emerged as a central determinant of posttherapeutic quality of life, independent of the degree of objective hair retention. Structural and ultrastructural analyses demonstrated that the preservation of follicular integrity was closely associated with successful macroscopic hair retention under ongoing cytotoxic exposure, supporting a biological basis for the clinical effectiveness of scalp cooling. Conclusions The clinical relevance of scalp cooling extends beyond objective and visible hair preservation and appears to reside predominantly in its psychological impact on patients undergoing highly alopecia-inducing chemotherapy. Importantly, the identification of structural and ultrastructural markers of follicular vulnerability provides a mechanistic foundation for the future optimization of scalp-cooling approaches and for the development of adjunct follicle-directed protective strategies to enhance follicular integrity and support patient well-being during cytotoxic therapy.
BACKGROUND:Lung cancer remains a major contributor to cancer mortality in sub-Saharan Africa (SSA), where late diagnosis, driven by low awareness, sociocultural barriers, and health system constraints, limits effective treatment. Despite the growing burden, evidence on patients' quality of life (QoL) and symptom experience in SSA is limited. OBJECTIVE:This study aimed to describe the common symptoms and QoL of patients with lung cancer treated at 2 hospitals in SSA, and to investigate the association of demographics, clinical characteristics, and symptom burden with QoL. METHODS:This was a cross-sectional study that consecutively recruited patients with lung cancer from 2 teaching hospitals in SSA: Bugando Medical Centre (BMC) in Tanzania and the University of the Witwatersrand Centre of Respiratory Excellence (WITS-CORE) in South Africa. Data collected included demographics, clinical information, and performance status using the Eastern Cooperative Oncological Group Performance Scale (ECOG-PS). Health-related QoL was assessed using the 30-item European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30). The study followed all ethical procedures, and data were analyzed using both descriptive and inferential statistics in Stata 18. A P value of <.05 was considered statistically significant. RESULTS:A total of 174 patients with lung cancer were enrolled across the 2 sites. The score on the EORTC QLQ-C30 global health status/QoL subscale was low, with a median of 41.67 (IQR 33.33-41.67), and it varied by site. Patients from WITS-CORE demonstrated higher social functioning scores, while those from BMC reported greater financial difficulties. A low global health status/QoL score was independently associated with the BMC site (adjusted odds ratio [aOR] 3.5, 95% CI 1.3-9.3) and poor performance status (ECOG-PS 3-4; aOR 3.4, 95% CI 1.2-6.6). Furthermore, symptoms such as nausea and vomiting, pain, dyspnea, insomnia, appetite loss, diarrhea, and financial struggles were all associated with a low global health status/QoL score. CONCLUSIONS:QoL among patients with lung cancer in SSA is poor. Low QoL is strongly associated with the Multinational Lung Cancer Control Program study site, poor performance status, and a range of symptoms and financial difficulties. Addressing these factors may help to improve patient outcomes and well-being in SSA.
Background:Tele-oncology addresses geographic barriers to cancer care, but implementation challenges persist in rural settings. AI-enhanced predictive analytics offer opportunities for optimizing deployment through personalized, data-driven strategies; however, evidence in rural tele-oncology contexts remains limited, and critical equity considerations remain underexamined. Objective:This scoping review aimed to map evidence on AI-enhanced predictive analytics in tele-oncology implementation, with particular attention to rural and underserved populations, to identify research gaps and inform implementation science priorities. Methods:We searched 5 databases (PubMed, Embase, CINAHL, Web of Science, and IEEE Xplore) using 4 concept domains (tele-oncology, rural implementation barriers, AI or predictive analytics, implementation science) from January 2015 through November 2025. Two independent reviewers screened 330 unique records (title or abstract; Cohen κ=0.78), with the principal investigator resolving conflicts. Of 138 full-text reviews (κ=0.82), 4 studies met inclusion criteria. Data extraction captured study characteristics, AI applications, implementation factors, and outcomes. We used narrative thematic analysis to map findings into three themes: (1) the current tele-oncology implementation landscape in rural and underserved settings, (2) potential AI applications addressing implementation challenges, and (3) implementation considerations for AI systems themselves. Results:Four included studies (1 pilot feasibility study, 1 proof-of-concept validation study, 1 cross-sectional predictive study, and 1 platform development study; published 2019-2025) demonstrated limited evidence at the intersection of AI, tele-oncology, and rural health equity. Patient characteristics predicted telehealth modality preferences with 86.2% accuracy, revealing that male patients exhibited 66% increased odds of video selection versus female patients (P=.004), and urban residents showed 101% increased odds compared to rural counterparts (P<.001). Liu et al demonstrated that disadvantaged populations engaged with AI-generated health literacy content 2.52-fold more frequently than nondisadvantaged counterparts. However, all 4 studies documented substantial implementation barriers (patient, provider, organizational, and system levels) persisting despite technological sophistication. Organizational threshold effects, where remote monitoring interventions succeeded with adequate provider capacity but failed under resource constraints-suggest that algorithmic innovations cannot overcome structural limitations in rural facilities. No studies explicitly examined algorithmic bias, cross-population validation, or potential harms in rural contexts. Geographic concentration in high-resource countries (United States n=2, Greece n=1, and Singapore n=1) and limited oncology-specific focus underscore structural gaps in knowledge generation for underserved populations. Conclusions:Current evidence remains insufficient to support definitive practice recommendations. The observed evidence gap may reflect broader structural inequities in knowledge generation: populations with the greatest implementation challenges appear to remain substantially underrepresented in AI and digital health literature. Future research should prioritize comparative effectiveness studies in authentic rural contexts with implementation science outcomes, equity-centered cross-population validation, specification of translation mechanisms linking AI predictions to implementation strategies, health economic analyses, and mechanistic research on sociotechnical integration factors, ensuring technological innovation reduces rather than perpetuates disparities in cancer care.
Unlabelled:The Internet of Things (IoT) is transforming various industries, including health care. IoT-based systems are increasingly prevalent in consumer health applications, while intelligent or smart devices equipped with sophisticated sensors are gaining recognition for their potential to improve clinical care practice and decision-making. Cancer care is a particularly promising area for IoT applications, enabling real-time and personalized interventions. However, empirical research on the effects of IoT in this field is limited due to the complexities inherent in cancer as a dynamic disease and the paucity of IoT-generated data available for research. This presents an opportunity to apply mathematical modeling to understand the effects of IoT under various scenarios. These analytical and "in silico" mathematical approaches are instrumental with limited data. Such models support the analysis of treatment uncertainty and patient response while balancing patient preferences, clinical outcomes, and system-level constraints. Grounded in mathematical oncology and health informatics, this paper proposes a conceptual framework that integrates real-time IoT data as dynamic inputs into adaptive mathematical models to simulate cancer dynamics. By exploring applications across multiple levels of analysis, the study demonstrates how IoT-enhanced mathematical models could inform implementation and optimize oncology services, addressing a critical gap in current research.
Unlabelled:Our study describes the development and evaluation of a retrieval-augmented generation-based large language model to improve the quality of responses to provider questions about herbs and dietary supplements.
Informal caregivers play a fundamental role in the diagnosis, treatment, and care received for their family members. Caring for someone with dementia is particularly burdensome, leading to known negative effects on caregivers' health. Despite caregivers being widely recognised as vulnerable, the psychosocial needs of carers who are managing cancer themselves remain unknown. To investigate the experiences and existing gaps in psychosocial support of informal UK caregivers who have received a cancer diagnosis while caring for someone with dementia or memory problems. 25 UK informal caregivers who have received a cancer diagnosis were recruited through convenience sampling to take part in an in-depth semi-structured interview. Reflexive Thematic Analysis was conducted. Study findings were developed into three key themes (1) caring complicates and can take precedence above cancer care, (2) caregivers may experience additional cancer-related distress and anxieties, and (3) increased need for support and coping resources when caregivers face cancer. For dementia carers having cancer can surface/heighten feelings of predeath grief, and worries around future care planning, for their care recipient. Dementia carers viewed multifaceted social support as vital to enable access to and coping with cancer treatment. Caregivers living with cancer themselves continue to shoulder the burden of illness work on behalf of their care recipients. For dementia carers, accessibility and experiences of cancer care may be optimised by person-centred adjustments to care planning and scheduling and dementia-friendly hospital environments. Dementia carers may benefit from tailored psycho-oncology support encompassing carer-role concerns and needs (e.g., predeath grief for their care recipient). N/A
Progression-free survival (PFS) is a critical endpoint in oncology, yet real-world applications of individualised, explainable machine-learning (ML) predictions remain limited. This study aims to develop and validate explainable ML models to predict PFS using retrospective data from a national prostate cancer cohort in Brunei Darussalam. We analysed a retrospective cohort of 212 patients (478 longitudinal observations) treated at the Brunei Cancer Centre (January 2018 to December 2024). Clinical, laboratory, and treatment data were harmonised, with missing values imputed via Extremely Randomised Trees. Longitudinal patterns were captured using a recurrent autoencoder to generate latent representations. We compared four modelling approaches: Cox Proportional Hazards (CPH), Random Survival Forests (RSF), Gradient Boosting Survival (GBS), and Deep Neural Network Survival models. Performance was evaluated using time-dependent AUC, Harrell’s C-index, and Integrated Brier Score (IBS), with SHAP (Shapley Additive exPlanations) used for interpretability. RSF demonstrated improved discriminative performance and balanced calibration, achieving a C-index of 0.906 and AUCs of 0.941 at both 4 and 5 years (IBS = 0.0698). In contrast, the traditional CPH model performed poorly (C-index 0.531; AUC 0.706 at 4 years). Deep survival (AUCs of 0.941 at 4 years and 0.941 at 5 years, C-index 0.719, IBS=0.0590) and GBS (AUCs of 0.765 at 4 years and 0.833 at 5 years, C-index 0.844, IBS=0.0887) models showed moderate performance. SHAP analysis identified sodium (Na), alanine aminotransferase (ALT), MCH, platelet count, and specific treatment categories as key drivers of increased progression risk. Tree-based ensemble approaches, particularly RSF integrated with SHAP, offer high accuracy for personalised risk stratification in prostate cancer. These findings highlight the potential of explainable ML to enhance clinical decision-making. However, external validation in larger multi-institutional, multi-omics dataset is required before routine clinical implementation.
Background:With the rapid development of medical technology and the emphasis on early lung cancer screening, the detection rate of multiple primary lung cancer (MPLC) has increased in recent years. However, the prognostic determinants and clinical characteristics of patients with MPLC remain poorly characterized. Objective:This study aimed to develop and validate a nomogram for predicting overall survival (OS) in patients with MPLC using data from the Surveillance, Epidemiology, and End Results database. Methods:This study was reported in accordance with the TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) guidelines. A cohort of 4177 patients with MPLC (2007-2015) was obtained from the Surveillance, Epidemiology, and End Results database. The patients were randomly divided into training (n=2923) and validation (n=1254) cohorts at a 7:3 ratio. Backward stepwise Cox regression identified 11 independent risk factors, which were integrated into a nomogram predicting 3-, 5-, and 8-year OS rates. Results:The nomogram demonstrated superior discriminative ability compared to the American Joint Committee on Cancer staging system, with higher area under the receiver operating characteristic curve values for 3-, 5-, and 8-year OS predictions in both cohorts (training cohort: 0.743, 0.751, and 0.759, respectively; validation cohort: 0.737, 0.734, and 0.695, respectively). Calibration curves and decision curve analysis confirmed its clinical utility. Conclusions:This study establishes a validated nomogram incorporating clinical and socioeconomic variables to optimize prognostic assessment and personalized treatment planning for patients with MPLC.
Background:Effective communication about breast and cervical cancers remains a public health challenge, with widespread misinformation and barriers to cancer-related language understanding. Large language models (LLMs) offer potential for scalable health communication, yet trade-offs between quality, safety, and accessibility of general-purpose and medical-domain LLMs remain underexplored. Objective:This study aimed to propose a comprehensive evaluation framework and systematically assess the performance of LLMs in generating breast and cervical cancer information, with a focus on linguistic quality, safety and trustworthiness, and communication accessibility and affectiveness. Methods:This mixed methods evaluation study assessed outputs from 5 general-purpose and 3 medical LLMs using real-world breast and cervical cancer-related questions curated from publicly available medical datasets. LLM-generated responses were evaluated in a controlled offline setting. Primary outcomes included linguistic quality (fluency, coherence, and accuracy), safety and trustworthiness (toxicity, bias, and harm potential), and communication accessibility and affectiveness (readability, empathy, and clarity). Qualitative ratings were performed by domain experts, while quantitative metrics were compared across models. Statistical analyses included Welch ANOVA to detect differences in metric scores, Games-Howell tests for pairwise comparisons, and Hedges g to assess effect sizes. Results:General-purpose LLMs, particularly Llama 3 and Gemma, demonstrated superior linguistic quality and affectiveness but often produced complex outputs that may limit accessibility. In contrast, medical LLMs (eg, MedAlpaca and BioMistral) generated simpler content suitable for broader audiences but scored lower in safety and empathy due to higher levels of hallucination, bias, and toxicity. Conclusions:While LLMs show promise for improving digital cancer communication, our findings reveal a trade-off between domain specialization and overall communication quality and safety. Future development of health-focused LLMs should prioritize hybrid modeling strategies to enhance trust, clarity, and clinical relevance in patient-facing tools.
