Web and mobile-based symptom management interventions for physical symptoms of people with advanced cancer: A systematic review and meta-analysis

Machine Learning Approaches to Predict Symptoms in People With Cancer: Systematic Review

BACKGROUND

People with cancer frequently experience severe and distressing symptoms associated with cancer and its treatments. Predicting symptoms in patients with cancer continues to be a significant challenge for both clinicians and researchers. The rapid evolution of machine learning (ML) highlights the need for a current systematic review to improve cancer symptom prediction.

OBJECTIVE

This systematic review aims to synthesize the literature that has used ML algorithms to predict the development of cancer symptoms and to identify the predictors of these symptoms. This is essential for integrating new developments and identifying gaps in existing literature.

METHODS

We conducted this systematic review in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist. We conducted a systematic search of CINAHL, Embase, and PubMed for English records published from 1984 to August 11, 2023, using the following search terms: cancer, neoplasm, specific symptoms, neural networks, machine learning, specific algorithm names, and deep learning. All records that met the eligibility criteria were individually reviewed by 2 coauthors, and key findings were extracted and synthesized. We focused on studies using ML algorithms to predict cancer symptoms, excluding nonhuman research, technical reports, reviews, book chapters, conference proceedings, and inaccessible full texts.

RESULTS

A total of 42 studies were included, the majority of which were published after 2017. Most studies were conducted in North America (18/42, 43%) and Asia (16/42, 38%). The sample sizes in most studies (27/42, 64%) typically ranged from 100 to 1000 participants. The most prevalent category of algorithms was supervised ML, accounting for 39 (93%) of the 42 studies. Each of the methods-deep learning, ensemble classifiers, and unsupervised ML-constituted 3 (3%) of the 42 studies. The ML algorithms with the best performance were logistic regression (9/42, 17%), random forest (7/42, 13%), artificial neural networks (5/42, 9%), and decision trees (5/42, 9%). The most commonly included primary cancer sites were the head and neck (9/42, 22%) and breast (8/42, 19%), with 17 (41%) of the 42 studies not specifying the site. The most frequently studied symptoms were xerostomia (9/42, 14%), depression (8/42, 13%), pain (8/42, 13%), and fatigue (6/42, 10%). The significant predictors were age, gender, treatment type, treatment number, cancer site, cancer stage, chemotherapy, radiotherapy, chronic diseases, comorbidities, physical factors, and psychological factors.

CONCLUSIONS

This review outlines the algorithms used for predicting symptoms in individuals with cancer. Given the diversity of symptoms people with cancer experience, analytic approaches that can handle complex and nonlinear relationships are critical. This knowledge can pave the way for crafting algorithms tailored to a specific symptom. In addition, to improve prediction precision, future research should compare cutting-edge ML strategies such as deep learning and ensemble methods with traditional statistical models.

Understanding perceptions of tumor genomic profile testing in Black/African American cancer patients in a qualitative study: the role of medical mistrust, provider communication, and family support

Tumor genomic profiling (TGP) examines genes and somatic mutations specific to a patient's tumor to identify targets for cancer treatments but can also uncover secondary hereditary (germline) mutations. Most patients are unprepared to make complex decisions related to this information. Black/African American (AA) cancer patients are especially at risk because of lower health literacy, higher levels of medical mistrust, and lower awareness and knowledge of genetic testing. But little is known about their TGP attitudes or preferences. Five in-person focus groups were conducted with Black/AA cancer patients (N = 33) from an NCI-designated cancer center and an affiliated oncology unit in an urban safety-net hospital located in Philadelphia. Focus groups explored participants' understanding of TGP, cultural beliefs about genetics, medical mistrust, and how these perceptions informed decision-making. Participants were mostly female (81.8%), and one-third had some college education; mean age was 57 with a SD of 11.35. Of patients, 33.3% reported never having heard of TGP, and 48.5% were not aware of having had TGP as part of their cancer treatment. Qualitative analysis was guided by the principles of applied thematic analysis and yielded five themes: (1) mistrust of medical institutions spurring independent health-information seeking; (2) genetic testing results as both empowering and overwhelming; (3) how provider-patient communication can obviate medical mistrust; (4) how unsupportive patient-family communication undermines interest in secondary-hereditary risk communication; and (5) importance of developing centralized patient support systems outside of treatment decisions. Results improve understanding of how Black/AA patients perceive of TGP and how interventions can be developed to assist with making informed decisions about secondary hereditary results.

Supportive Digital Health Service During Cancer Chemotherapy: Single-Arm Before-and-After Feasibility Study

BACKGROUND

Digital supportive cancer care is recommended to improve patient outcomes. A portal was designed and embedded within the electronic medical record and public health portal of Iceland, consisting of symptom and needs monitoring, educational material, and messaging.

OBJECTIVE

This study aims to assess (1) portal feasibility (adoption, engagement, usability, and acceptability), (2) potential predictors of usability and acceptability, and (3) the potential impact of the portal on patient-reported outcomes.

METHODS

This was a single-arm, before-and-after feasibility study at a university hospital among patients with cancer who were undergoing chemotherapy. Participation included filling out the Edmonton Symptom Assessment System-Revised (ESASr) weekly and the Distress Thermometer and Problem List (DT&PL) 3 times; reading educational material and messaging; and completing study questionnaires. Clinical and portal engagement data were collected from medical records. Data from patients were collected electronically at baseline and 7 to 10 days after the third chemotherapy round. Usability was assessed using the System Usability Scale (score 0-100), and acceptability was assessed using a 35-item survey (score 1-5). Patient-reported outcome measures included ESASr and DT&PL; a single-item scale for quality of life, family support, and quality of care; and multi-item scales for health literacy (Brief Health Literacy Screener), health engagement (Patient Health Engagement Scale), self-care self-efficacy (Self-Care Self-Efficacy scale), symptom interference (MD Anderson Symptom Inventory), knowledge expectations (Hospital Patients' Knowledge Expectations), and received knowledge (Hospital Patients' Received Knowledge). Health care professionals were interviewed regarding portal feasibility.

RESULTS

The portal adoption rate was 72% (103/143), and the portal use rate was 76.7% (79/103) over a mean 8.6 (SD 2.7) weeks. The study completion rate was 67% (69/103). The combined completion rate of the ESASr and DT&PL was 78.4% (685/874). Patients received a mean 41 (SD 13) information leaflets; 33% (26/79) initiated messaging, 73% (58/79) received messages, and 85% (67/79) received follow-up phone calls. The mean System Usability Scale score was 72.3 (SD 14.7), indicating good usability. Usability was predicted by age (β=-.45), ESASr engagement (β=.5), symptom interference (β=.4), and received knowledge (β=.41). The mean acceptability score, 3.97 (SD 0.5), was above average and predicted by age (β=-.31), ESASr engagement (β=.37), symptom interference (β=.60), self-care self-efficacy (β=.37), and received knowledge (β=.41). ESASr scores improved for total symptom distress (P=.003; Cohen d=0.36), physical symptoms (P=.01; Cohen d=0.31), and emotional symptoms (P=.01; Cohen d=0.31). Daily symptom interference increased (P=.03; Cohen d=0.28), quality of life improved (P=.03; Cohen d=0.27) and health engagement (P=.006; Cohen d=0.35) improved, while knowledge expectations decreased (P≤.001; Cohen d=2.57). Health care professionals were positive toward the portal but called for clearer role delineation and follow-up.

CONCLUSIONS

This study supports the feasibility of a support portal and the results indicate the possibility of improving patient outcomes, but further developments are warranted.

Mobile App for Gynecologic Cancer Support for Patients With Gynecologic Cancer Receiving Chemotherapy in China: Multicenter Randomized Controlled Trial

BACKGROUND

Patients with gynecologic cancer receiving chemotherapy often report unmet supportive care needs. Compared with traditional face-to-face clinical interventions, mobile health can increase access to supportive care and may address patients' needs. Although app-based support programs have been developed to support patients with gynecologic cancer, their efficacy has not been adequately tested.

OBJECTIVE

The aim of this study was to examine the efficacy of a mobile app for gynecologic cancer support (MGCS) for patients with gynecologic cancer receiving chemotherapy in China.

METHODS

A multicenter randomized controlled trial was conducted in 2 university-affiliated hospitals in China. A total of 168 Chinese patients with gynecologic cancer were recruited and randomized to receive routine care or MGCS program plus routine care for 24 weeks. The Mishel uncertainty in illness theory guided the development of MGCS program, which has 4 modules: weekly topics, emotional care, discussion center, and health consultation. The primary outcome of this program was the assessment of the uncertainty in illness. The secondary outcomes were quality of life, symptom distress, and social support. All health outcomes were evaluated at baseline (T0), 12 weeks (T1), and 24 weeks (T2). Repeated measures analysis of covariance was used to assess the efficacy of the MGCS program.

RESULTS

In this trial, 67 patients in the control group and 69 patients in the intervention group completed 2 follow-up assessments (response rate, 136/168, 81%). At 12 weeks, no significant differences were observed in any of the health outcomes between the 2 groups. At 24 weeks, compared to patients in the control group, those in the intervention group reported significant decreased uncertainty in illness (P<.001; d=-0.60; adjusted mean difference -7.69, 95% CI -11.31 to -4.07) and improved quality of life (P=.04; d=0.30; adjusted mean difference 4.77, 95% CI 0.12-9.41).

CONCLUSIONS

The MGCS program demonstrated efficacy in supporting patients with gynecologic cancer receiving chemotherapy. This trial illustrates that an app-based program can be incorporated into routine care to support patients with cancer and suggests that allocation of more resources (grants, manpower, etc) to mobile health in clinics is warranted.

TRIAL REGISTRATION

Chinese Clinical Trial Registry ChiCTR2000033678; https://www.chictr.org.cn/showproj.html?proj=54807.

Symptom profiles and related factors among patients with advanced cancer: A latent profile analysis

Objective

This study aimed to investigate symptom subgroups and associated influencing factors in patients with advanced cancer.

Methods

A cross-sectional study was conducted, involving 416 patients with advanced cancer. The study examined five symptoms: fatigue, pain, sleep impairment, anxiety, and depression. Latent Profile Analysis (LPA) was utilized to classify symptom subgroups. A multiple logistic regression analysis was conducted to explore factors associated with the identified symptom subgroups.

Results

The analysis revealed three distinct subgroups among the participants: "all low" (58.2%), characterized by normal symptoms except for moderate sleep quality; "all moderate" (35.1%), exhibiting normal symptoms except for poor sleep quality and fatigue; and "all high" (6.7%), experiencing normal pain, moderate depression, moderate anxiety, poor sleep quality, and fatigue. Malnutrition risk, cancer diagnosis, and cancer survivorship duration were found to be associated with a more severe symptom burden.

Conclusions

Patients in the "all high" subgroup faced an increased risk of malnutrition and a longer cancer survivorship duration. Additionally, patients in the "all moderate" subgroup were distinguished by having a breast cancer diagnosis. These findings have significant implications for allocating medical resources and implementing person-centered symptom management strategies.

Digital health and telehealth in cancer care: a scoping review of reviews

The COVID-19 pandemic necessitated remote cancer care delivery via the internet and telephone, rapidly accelerating an already growing care delivery model and associated research. This scoping review of reviews characterised the peer-reviewed literature reviews on digital health and telehealth interventions in cancer published from database inception up to May 1, 2022, from PubMed, Cumulated Index to Nursing and Allied Health Literature, PsycINFO, Cochrane Reviews, and Web of Science. Eligible reviews conducted a systematic literature search. Data were extracted in duplicate via a pre-defined online survey. Following screening, 134 reviews met the eligibility criteria. 77 of those reviews were published since 2020. 128 reviews summarised interventions intended for patients, 18 addressed family caregivers, and five addressed health-care providers. 56 reviews did not target a specific phase of the cancer continuum, whereas 48 reviews tended to address the active treatment phase. 29 reviews included a meta-analysis, with results showing positive effects on quality of life, psychological outcomes, and screening behaviours. 83 reviews did not report intervention implementation outcomes but when reported, 36 reported acceptability, 32 feasibility, and 29 fidelity outcomes. Several notable gaps were identified in these literature reviews on digital health and telehealth in cancer care. No reviews specifically addressed older adults, bereavement, or sustainability of interventions and only two reviews focused on comparing telehealth to in-person interventions. Addressing these gaps with rigorous systematic reviews might help guide continued innovation in remote cancer care, particularly for older adults and bereaved families, and integrate and sustain these interventions within oncology.

Mobile applications in radiation oncology-current choices and future potentials

PURPOSE

To review existing scientific literature on mobile applications (apps) in the field of radiation oncology and to evaluate characteristics of commercially available apps across different platforms.

METHODS

A systematic review of the literature for publications presenting apps in the field of radiation oncology was carried out using the PubMed database, Cochrane library, Google Scholar, and annual meetings of major radiation oncology societies. Additionally, the two major marketplaces for apps, App Store and Play Store, were searched for available radiation oncology apps for patients and health care professionals (HCP).

RESULTS

A total of 38 original publications which met the inclusion criteria were identified. Within those publications, 32 apps were developed for patients and 6 for HCP. The vast majority of patient apps focused on documenting electronic patient-reported outcomes (ePROs). In the two major marketplaces, 26 apps were found, mainly supporting HCP with dose calculations.

CONCLUSION

Apps used in (and for) scientific research in radiation oncology are rarely available for patients and HCP in common marketplaces.

mHealth program for patients with advanced cancer receiving treatment in a public health hospital in Brazil

OBJECTIVE

Previous studies have suggested the benefit of routine screening for biopsychosocial symptoms among patients with cancer. In recognition of the lack of data from low- and middle-income countries, this study sought to test and determine the effect of a mHealth program to screen biopsychological symptoms among patients with advanced breast or gynecological cancer.

METHODS

This was a quasi-experimental pre-post study conducted in a public hospital located in central western Brazil. Patients diagnosed with advanced breast of gynecological cancer who were about to initiate chemotherapy treatment at this institution and had access to Internet by smartphone, computer or tablet were invited to participate. Patients received training on using the app Comfort, a program developed to rate their physical and emotional symptoms during the 6-month of the proposed study. Patients were also asked to complete the EuroQOL 5D (EQ-5D-3L) every month. Mann-Whitney U test was used to determine differences among groups of patients (engaged and non-engaged). RM-ANOVA was used to determine the effect of time on mean visual analog scale (VAS) score.

RESULTS

A total of 125 patients were recruited (median age = 46.6 years old, 41.6% married). Mostly, patients possessed lower levels of education and had relatively low monthly incomes. Notably, 67.2% of patients engaged with the Comfort program, and few patients (4%) withdrew due to lack of engagement with the program or issues with internet connection. In general, patients who engaged with the program reported improvement in physical and emotional symptoms (p < 0.01), as well as in their overall quality of life (VAS; p = 0.009), compared with patients who did not engage with the program.

CONCLUSIONS

This is the first mHealth program developed in Brazil for patients in a low resource setting. Our findings suggest that Comfort could be an effective resource to assist patients and health care providers track symptoms and improve patients' quality of life.

Mobile health-based remote interaction management intervention for patients with low anterior resection syndrome: study protocol for a randomised controlled trial

INTRODUCTION

Low anterior resection syndrome (LARS) involves bowel dysfunction after sphincter-preserving surgery for rectal resection that significantly impacts patients' quality of life (QoL). The improvement of LARS largely depends on patient self-management behaviour; however, insufficient information about supportive care and weak awareness of self-management lead to poor self-management behaviour. Motivational interviewing (MIs) explore and change patients' ambivalence during the conversation, thereby changing and maintaining healthy behaviours to enhance effective participation. In recent years, mobile health has been widely used in clinical practice, providing continuous information support and remote interaction. However, current online information on LARS is suboptimal, websites are highly variable, important content is often lacking and the material is too complex for patients. Therefore, this study will evaluate the impacts of a remote LARS interaction management intervention based on a WeChat applet ('e-bowel safety') and MIs on patients with LARS.

METHODS AND ANALYSIS

This study will be a single-blind, two-arm randomised controlled trial involving patients with LARS in three tertiary grade A general hospitals who will be randomised into two groups. The intervention group will use the 'e-bowel safety' applet and the intervention team will conduct a monthly MI about syndrome management. The control group will receive an information booklet that contains the same information as that provided in the 'e-bowel safety' informational module. The intervention will last for 3 months, followed by 3 months of follow-up. The primary outcome will be global QoL; the secondary outcomes will include bowel function, social support, self-management measured at the baseline, 3 months and 6 months for three times and patients' thinkings at the end of the intervention (at 3 months).

ETHICS AND DISSEMINATION

Ethics approval was granted by the Clinical Medical Research Ethics Committee of the First Affiliated Hospital of Anhui Medical University (PJ2022-07-53).

TRIAL REGISTRATION NUMBER

Chinese Clinical Trial Registry (ChiCTR2200061317).

Enhanced supportive care for advanced cancer patients: study protocol for a randomized controlled trial

BACKGROUND

Early palliative care along with standard cancer treatments is recommended in current clinical guidelines to improve the quality of life and survival of cancer patients. This study protocol aims to evaluate the effect of "Enhanced Supportive Care", an early primary palliative care provided by nurses.

METHODS

A randomized controlled trial (RCT) will be conducted including advanced cancer patients scheduled for first-line palliative chemotherapy (N=360) and their caregivers in South Korea. Participants will be randomly assigned to the intervention or control group in a 1:1 ratio. Participants in the intervention group will receive the "Enhanced Supportive Care", which provides five sessions of symptom management and coping enhancement counseling by nurses. The control group will receive symptom monitoring five times. The primary endpoints are symptoms, coping, and quality of life (QoL) at 3 months. Secondary endpoints are symptoms, coping, and QoL at 6 months, depression and self-efficacy for coping with cancer at 3 and 6 months, symptom and depression change from baseline to 3 months, survival at 6 and 12 months among patients, and depression among caregivers at 3 and 6 months.

DISCUSSION

This RCT will evaluate the effects of "Enhanced Supportive Care" on symptoms, depression, coping, self-efficacy for coping with cancer, QoL and survival of patients, as well as depression of caregivers. It will provide evidence of a strategy to implement early primary palliative care provided by nurses, which may consequently improve cancer care for newly diagnosed patients with advanced stage cancer.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT04407013. Registered on May 29, 2020, https://www.

CLINICALTRIALS

gov/ct2/show/study/NCT04407013 . The protocol version is ESC 1.0.

Telerehabilitation physical exercise for patients with lung cancer through the course of their disease: A systematic review

BACKGROUND AND OBJECTIVE

Evidence is shown for the benefits of physical activity, for patients with lung cancer, at different times through the course of the disease. Telerehabilitation can overcome some of barriers often met by patients to practice physical activity. The objective of this systematic review is to assess feasibility and safety of telerehabilitation for patients with lung cancer, its effects on physical capacity, quality of life, symptoms severity, depression and anxiety, survival, lung function, post-operative outcomes, dyspnoea and body composition. Secondary aim was to distinguish the telerehabilitation efficacy between the different phases of the disease.

DATA SOURCE AND SELECTION CRITERIA

Pubmed, PEDro, Scopus, ScienceDirect, randomized controlled trials and non-randomized controlled trials, written in French or English, of telerehabilitation among patients with lung cancer.

RESULTS

Eight studies were included. Telerehabilitation is safe but was characterized by a low recruitment and attendance rate (<70%). It enhances quality of life, muscle mass, depression and anxiety but it does not improve physical capacity (except in preoperative period), symptoms severity, survival, lung function or dyspnoea. After surgery, it ameliorates quality of life, depression and anxiety. During systemic treatments of lung cancer, it improves quality of life, symptoms severity and muscle mass.

CONCLUSION

Telerehabilitation could be proposed in patients with lung cancer as a complementary intervention of hospital-based programme to increase physical activity volume, compliance and self-efficacy. In case the classic programmes are not possible, it could also be an alternative approach for patients unable to participate to a hospital or community-based training programme.

Development and evaluation of the efficacy of a web-based education program among cancer patients undergoing treatment with systemic chemotherapy: a randomized controlled trial

Purpose

The study aimed to develop a web-based education program among cancer patients undergoing treatment with systemic chemotherapy and to evaluate the efficacy of the program on symptom control, quality of life, self-efficacy, and depression.

Methods

A web-based education program was prepared in line with patient needs, evidence-based guidelines, and expert opinions and tested with 10 cancer patients. The single-blind, randomized controlled study was conducted at a medical oncology unit of a university hospital. Pretests were applied to 60 cancer patients undergoing treatment with systemic chemotherapy, and the patients (intervention: 30, control: 30) were randomized. The intervention group used a web-based education program for 3 months, and they were allowed to communicate with researchers 24/7 via the website. The efficacy of a web-based education program at baseline and after 12 weeks was evaluated. The CONSORT 2010 guideline was performed.

Results

In the first phase results of the study, it was found that most of the patients with cancer wanted to receive education about symptom management and the side effects of the treatment. Expert opinions on the developed website were found to be compatible with each other (Kendall’s Wa = 0.233, p = 0.008). According to the randomized controlled study results, patients who received web-based education reported significantly fewer symptoms (p = 0.026) and better quality of life (p = 0.001), but there was no statistically significant difference in the self-efficacy and depression levels during the 3-month follow-up period (p˃0.05). The most frequently visited links in the web-based education program by the patients with cancer were the management of chemotherapy-related symptoms (62.6%).

Conclusion

A web-based education program was found to be efficacy in remote symptom management and improving the quality of life of cancer patients.

Trial registration

www.clinicaltrials.gov, NCT05076916 (October 12, 2021, retrospectively registered).

Supplementary Information

The online version contains supplementary material available at 10.1007/s00520-022-07039-w.

Patient-Centered Palliative Care for Patients With Advanced Lung Cancer

The evidence base demonstrating the benefits of an early focus on palliative care for patients with serious cancers, including advanced lung cancer, is substantial. Early involvement of specialty-trained palliative care clinicians in the care of patients with advanced lung cancer improves patient-reported outcomes, such as quality of life, and health care delivery, including hospice utilization. Since the time that many of these palliative care trials were conducted, the paradigm of cancer care for many cancers, including lung cancer, has changed dramatically. The majority of patients with advanced lung cancer are now treated with immune checkpoint inhibitors or targeted therapies, both of which have had a significant impact on patient's experience and outcomes. With this changing landscape of lung cancer therapeutics, patients are facing new and different challenges, including dealing with novel side effect profiles and coping with greater uncertainty regarding their prognosis. Patients who are living longer with their advanced cancer also struggle with how to address survivorship issues, such as sexual health and exercise, and decision making about end-of-life care. Although palliative care clinicians remain well-suited to address these care needs, they may need to learn new skills to support patients treated with novel therapies. Additionally, as the experience of patients with advanced lung cancer is becoming more varied and individualized, palliative care research interventions and clinical programs should also be delivered in a patient-centered manner to best meet patient's needs and improve their outcomes. Tailored and technology-based palliative care interventions are promising strategies for delivering patient-centered palliative care.

The efficacy of web or mobile-based interventions to alleviate emotional symptoms in people with advanced cancer: a systematic review and meta-analysis

PURPOSE

This review aims to critically evaluate the efficacy of web or mobile-based (WMB) interventions impacting emotional symptoms in patients with advanced cancer.

METHOD

Articles published from 1991 to 2019 were identified using PubMed, PsycINFO, CINAHL, and Scopus. Only interventions involving adults with advanced cancer using a WMB intervention to manage emotional symptoms were included. Risk of bias was assessed using ROBINS-I and ROB2 tools. Studies that reported mean symptom scores were pooled using a random-effects model, and standardized mean difference (SMD) and 95% CIs were calculated.

RESULTS

Twenty-three of the 1177 screened studies met the inclusion criteria, and a total sample of 2558 patients were included. The sample was 57% female, and 33% had advanced cancer with mean age of 57.15 years. Thirteen studies evaluated anxiety, nineteen evaluated depression, and eleven evaluated distress. Intervention components included general information, tracking, communication, multimedia choice, interactive online activities, tailoring/feedback, symptom management support content, and self-monitoring. Overall pooled results showed that WMB interventions' effects on decreasing anxiety (SMD - 0.20, - 0.45 to 0.05, I² = 72%), depression (SMD - 0.10, - 0.30 to 0.11, I² = 73%), and distress (SMD - 0.20, - 0.47 to 0.06, I² = 60%) were not significant for randomized controlled trials (RCTs). In contrast, WMB interventions significantly decreased symptoms of anxiety (p = .002) in a sub-group analysis of non-RCTs.

CONCLUSION

This meta-analysis demonstrated that WMB interventions were not efficacious in alleviating emotional symptoms in adults with advanced cancer. Considering the diversity of interventions, the efficacy of WMB interventions and its impacts on emotional symptoms should be further explored.