Approximately 18.1 million new cancer cases and 9.6 million cancer-related deaths occurred globally in 2018 [1]. The rising tide of cancer diagnoses in many developed countries has been attributed to both population aging and the increasing prevalence of primary risk factors, including physical inactivity, obesity, and metabolic disease [1,2]. As the cancer population continues to grow, there is an urgent need to improve supportive care services [2].

Supportive care focuses on assisting people with cancer and their families to cope with the disease and its treatment [3]. The management of cancer treatment-related symptoms and side effects and the maintenance of health-related quality of life from early diagnosis to end-of-life are key aims of supportive cancer care [3,4]. Supportive care interventions vary and may involve multidisciplinary team support, including doctors, nurses, pharmacists, and allied health professionals [5]. Recently, a shared follow-up approach between primary and secondary providers has been promoted to successfully meet increasing demands for survivorship care [6,7], as innovative methods for long-term cancer care are constantly needed [2].

There have been ongoing attempts to improve access to supportive care cancer services through the use of digital health technology [8]. Digital health interventions, with telemedicine as its oldest form dating back to the 1920s, have been increasing dramatically in recent years [9]. The terms digital health and eHealth are frequently used interchangeably, with numerous varied definitions. Eysenbach [10] defined eHealth as an emerging field in the intersection of medical informatics, public health, and business, referring to health services and information delivered or enhanced through the internet and related technologies. Elbert [11] and McLean et al [12] assert 3 key elements of digital health: (1) data obtained from the patient, (2) electronic transfer of data over a distance, and (3) patient-tailored feedback from a health care professional. Furthermore, a recent conceptual model proposed by Shaw et al [13] acknowledges the role of telehealth consultations, web-based forums, mobile devices and apps, and social media, in enabling real-time communication between health professionals and consumers.

Recent systematic reviews evaluating the impact of digital health interventions on health and health care costs provide promising evidence of effectiveness and cost-effectiveness [11,14]. Digital health has demonstrated potential in engaging people in their care [15], including as a tool for the treatment and self-management training of chronically ill patients [16,17]. Digital health interventions have been shown to be effective for managing cancer-related fatigue [18], may improve physical activity among cancer survivors [19], and can lead to positive effects addressing the supportive cancer care needs of individuals with different preferences and priorities [20]. Cancer survivors have been found to have a positive attitude toward digital health [21], suggesting that digital health interventions have the potential to overcome common challenges associated with access to supportive care in this population. Health professional-led, digital health–enabled, supportive care interventions may prove particularly useful in increasing accessibility of services to those with limited access because of location, health, time, and public health emergencies [22-30]. Despite the abundance of recent digital health literature, there remains an acknowledged lack of quality evidence regarding the effectiveness of supportive digital health care interventions for people with cancer [19,20,31-33].

Over the past several decades, studies have investigated the implementation and effects of digital health interventions in people with cancer. In previous systematic reviews evaluated in 2013, the design features of supportive digital health interventions for patients with cancer [20]; in 2014, the use of technology in cancer follow-up [31]; in 2015 and 2017, the effect of telehealth interventions in cancer survivors’ general quality of life [32,34], and in 2020, the benefits and limits of digital health for optimal supportive care in oncology [35]. The last review by Aapro et al [35] conducted an article search up to November 2018 and focused on the technical features of digital technologies. This is a rapidly growing area of health care because of advances in information technology and the uptake of digital technologies by both health professionals and patients. Therefore, this systematic review aims to explore the effect of supportive care interventions assisted by digital technologies on the outcomes of patients with cancer.

This systematic review was registered in the PROSPERO (International Prospective Register of Systematic Reviews) and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [36]. A search was performed in August 2020, using the following web-based databases: MEDLINE (OvidSP, Wolters Kluwer), PubMed (National Center for Biotechnology Information, US National Library of Medicine), Embase (OvidSP, Wolters Kluwer), PsycINFO (American Psychological Association), Cochrane Central Register of Controlled Trials (John Wiley & Sons), and Scopus (SciVerse, Elsevier). Three main keywords were searched: supportive care, digital health, and cancer patients. Additional search terms were included based on synonyms of these keywords and medical subject headings. Multimedia Appendix 1 indicates the search strategy used.

The inclusion criteria were as follows: (1) studies in English describing a randomized controlled trial (RCT), published between January 2000 and August 2020; (2) intervention recipients were adults with a diagnosis of cancer; (3) involved clinician-led digital health interventions; and (4) interventions implemented to provide supportive cancer care.

The determination of digital health interventions was on the basis of the conceptual model of Shaw et al [13], which consists of 3 core domains:

To be included in the review, it was essential that the intervention satisfied the 2 domains health in our hands and interacting for health. The third domain, data enabling health was deemed nonessential because of the known inconsistent reporting of these criteria. The essential criterion health in our hands was captured in the Intervention column and outlines the nature of digital health experience. Interacting for health was captured in the Interactions column of review data and outlines the individuals involved in any form of communicative exchange that supports the health and well-being of the patient and caregiver.

Studies with interventions involving automated systems, such as interactive voice response and similar web-based systems, to monitor symptoms were included if the intervention featured an internet or web-based component and triggered health professional or researcher involvement when a threshold, such as a pain score, was reached.

Studies were excluded if they were (1) not RCTs; (2) reported only self-managed interventions, patient-to-patient interventions, prevention tools, or alternative treatments, or (3) focused solely on interventions involving telephone delivery that replicated a clinical service.

Two reviewers (MM and DM) independently reviewed the titles and abstracts, followed by a full-text review of all publications. In cases of disagreement, a consensus was sought through discussion. Disagreement persisted for 4 studies; therefore, a third reviewer (TS) was consulted to adjudicate.

Endnote software (Clarivate Plc) [37] was used to manage references, and Covidence software (Veritas Health Innovation Ltd) was used to import and extract studies [38]. Two reviewers (MM and DM) independently applied the revised Cochrane risk of bias tool (The Cochrane Collaboration) [39] to establish the quality of the included studies. A matrix was developed by the authors and applied in the collection and analysis of structured data. Matrix criteria also included whether a study adhered to the CONSORT (Consolidated Standards of Reporting Trials) eHealth Checklist [40], a tool developed to improve the standard of reporting in digital health trials.

Using the revised Cochrane Collaboration’s tool for assessing risk of bias (RoB) in randomized trials (Risk of Bias 2.0) [39], each study’s methodological quality was assessed in 5 major domains: randomization process, deviation from intentional interventions, missing outcomes, measurement of outcomes, and selection of reported results. The RoB for each domain was rated as some concerns, low, or high. The overall RoB for each study was rated as some concerns or high.

The initial search generated 972 records. After title and abstract review, 135 publications were retained. The full text for each of these 135 publications was reviewed for eligibility, resulting in the identification of 17 publications. Three additional studies were found using a reference search, generating a total of 20 digital health–enabled supportive cancer care interventions for inclusion in the review (Figure 1).

With regard to the overall RoB, no included studies were rated as low overall RoB; instead, 8 had some concerns, and 12 were high risk. A summary of the RoB assessment can be found in Multimedia Appendix 3 [41-60], describing the methodological quality of each domain according to the Cochrane tool for assessing RoB in randomized trials (RoB 2.0). Figures 2 and 3 describe a graphical representation of the RoB assessments. Figure 2 includes studies in which an intention-to-treat analysis was performed, whereas Figure 3 contains studies with a per-protocol analysis.

This review compiles evidence regarding the potential of digital health interventions for supportive cancer care in different settings, including remote areas and emergency situations [33]. However, the challenges facing public health systems worldwide in terms of emergencies, such as the COVID-19 pandemic, have rapidly increased the use of digital health interventions [29]. Health care systems, including cancer care, are adapting in response to the need for social distancing, lockdowns, and other public health initiatives. Cancer clinics have reduced clinical appointments, administration encounters, and postponed elective cancer surgeries [28], which has increased the need for follow-up and management without visiting hospitals [76]. This situation has advanced the use of digital health and telehealth apps and programs worldwide [30]. Digital supportive cancer care has been implemented out of necessity and is becoming a common delivery model [77]. Two of the suggested strategies to enable supportive cancer care during COVID-19 are (1) empowering patients and caregivers through the use of digital communication and (2) increasing the use of existing digital health platforms [28].

There is an urgent need to agree on relevant outcomes, methods of assessment, and there is a need for improved quality of primary studies and RCTs, as shown by our quality assessment. The lack of high-quality randomized trials identified in this review reflects the ongoing problem of low-quality research. Moreover, only 2 of the included publications referenced the CONSORT Digital Health Checklist, which was published in 2011. Only 3 included studies were published before 2011, making it disappointing that most later publications failed to reference this standard. Comments of Dickinson in BMC Cancer [31] remain relevant in 2020:

This insight has been recurrent in different systematic reviews [32,74].

Although the review was a rigorous evaluation of RCTs, there were a small number of included studies that indicated there may be significant literature in the phase I or II feasibility spectrum. Although 20 studies were included, the interventions and outcomes were heterogeneous and did not enable meta-analyses. Owing to the nature of this review, there was also a heterogeneous population, variable outcome measures, variable study quality, and methodological limitations. These characteristics have been found to contribute to a lack of evidence regarding the benefits of digital health [19]. It is difficult to draw conclusions and synthesize studies with inconsistent outcome measures, and a systematic approach to using standardized measures is required. The CONSORT Digital Health Checklist should be used routinely. The review was limited to studies written in English. Therefore, it is possible that research papers published in other parts of the world were missed. As with many studies in oncology, this review found that breast cancer survivorship dominated. Thus, the development of supportive care interventions across other tumor streams is required. As most of the included studies were conducted in the United States and Europe, it is unclear whether the findings from these studies can be generalized to other countries and populations, particularly in developing nations.

Living well with cancer has gained greater relevance as the survival rates of many cancer types are increasing. The future of digital health in oncology supportive care brings a range of new and exciting possibilities. There is a need to evaluate the efficacy and efficiency of digital interventions in real-world conditions and standardize a core set of outcomes included in all studies to facilitate comparisons between interventions and digital technologies.

Digital health–enabled supportive cancer care is capable of improving health-related quality of life, symptom burden including self-report of pain and fatigue, depression, and, to a lesser extent, functional capacity. Supportive digital interventions in the field of cancer are being used and have been reported to be helpful for patients, independent of other factors. However, there is a need for higher quality research and clearer reporting than is evident in the current RCTs. Future research should focus on using valid, standardized outcome measures, increasing the methodological rigor of studies undertaken, and the development and evaluation of strategies to improve both patient and health professional engagement in the design and delivery of supportive digital health interventions.