eHealth Interventions for Dutch Cancer Care: Systematic Review Using the Triple Aim Lens

Background Globally, the burden of cancer on population health is growing. Recent trends such as increasing survival rates have resulted in a need to adapt cancer care to ensure a good care experience and manageable expenditures. eHealth is a promising way to increase the quality of cancer care and support patients and survivors. Objective The aim of this systematic review was 2-fold. First, we aimed to provide an overview of eHealth interventions and their characteristics for Dutch patients with and survivors of cancer. Second, we aimed to provide an overview of the empirical evidence regarding the impact of eHealth interventions in cancer care on population health, quality of care, and per capita costs (the Triple Aim domains). Methods The electronic databases Web of Science, PubMed, Cochrane, and Ovid PsycINFO were searched using 3 key search themes: eHealth interventions, cancer care, and the Netherlands. The identified interventions were classified according to predetermined criteria describing the intervention characteristics (eg, type, function, and target population). Their impact was subsequently examined using the Triple Aim framework. Results A total of 38 interventions were identified. Most of these were web portals or web applications functioning to inform and self-manage, and target psychosocial factors or problems. Few interventions have been tailored to age, disease severity, or gender. The results of this study indicate that eHealth interventions could positively affect sleep quality, fatigue, and physical activity of patients with and survivors of cancer. Inconclusive results were found regarding daily functioning and quality of life, psychological complaints, and psychological adjustment to the disease. Conclusions eHealth can improve outcomes in the Triple Aim domains, particularly in the population health and quality of care domains. Cancer-related pain and common symptoms of active treatment were not targeted in the included interventions and should receive more attention. Further research is needed to fully understand the impact of eHealth interventions in cancer care on participation, accessibility, and costs. The latter can be examined in economic evaluations by comparing eHealth interventions with care as usual.


Introduction Background
Globally, population health is greatly affected by cancer. An estimated 19.3 million new cancer cases and almost 10 million cancer deaths occurred in 2020 [1]. The related health care expenditure amounted to €103 (US $110) billion in Europe in 2018, corresponding to 6.2% of the total health expenditures [2]. The global cancer incidence is estimated to double by 2035 [3]. Owing to better screening and treatment options, survival rates have increased. Hence, cancer is increasingly becoming a chronic disease. Therefore, it is essential to develop and implement interventions to promote the long-term health and well-being of patients and survivors and to support daily disease coping [4].
Increasing attention is being paid to the use of eHealth to improve cancer care and support patients with cancer and survivors in coping with their illness. The World Health Organization defines eHealth as "the use of information and communication technology in support of health and health-related fields" [5]. There are several definitions of cancer survivors. Here, we use the definition of the National Cancer Institute: "persons with cancer post-treatment until the end of life" [6]. Currently, various eHealth interventions are available for patients with cancer and survivors. These interventions show considerable variations in function, target population, and type of eHealth technology. For instance, interventions can provide patients with and survivors of cancer with information about the disease and its treatment [7,8], support decision-making and self-management [9,10], alleviate physical and emotional problems [11,12], or provide peer social support [13,14]. Furthermore, interventions target different groups of patients with or survivors of cancer using various technologies and can be used as unguided self-help or with the support of health care professionals. Several studies have evaluated specific eHealth interventions in cancer care [15][16][17][18][19][20]. These studies considered a variety of outcomes, such as psychological complaints [15,16], symptom distress [17,19], and insomnia severity [18], and examined the effect of intervention characteristics, such as the amount of support, on intervention efficacy [21].
Currently, a general overview of eHealth interventions in cancer care and their characteristics is lacking. Such an overview would provide insights into the broad range of eHealth interventions available in cancer care, making it easier to compare interventions and their efficacy. In addition, no reviews that investigate the empirical evidence of the impact of eHealth interventions in cancer care are available. The absence of such overviews limits our understanding of the added value of eHealth interventions in cancer care. One way of evaluating interventions is through the Triple Aim framework. This model focuses on (1) improving population health, (2) improving the quality of care and patient experience, and (3) reducing the per capita health care costs [22]. Many areas of health reform can be helped forward and strengthened by Triple Aim framework, including the integration of information technologies such as eHealth. Deploying the Triple Aim lens offers an opportunity for a holistic and versatile evaluation.

Objective
The aim of this systematic review is 2-fold: (1) to provide an overview of available eHealth interventions in cancer care and their characteristics as described in the scientific literature and (2) to provide an overview of the empirical evidence regarding the impact of eHealth interventions in cancer care on population health, quality of care, and per capita costs-the Triple Aim domains [23]. As eHealth interventions are likely to be context-specific or even context-dependent, we will examine eHealth interventions applied in the Dutch context [24]. The Dutch context has been chosen as a case study and serves as an example for other Western countries.

Search Strategy
The following 4 databases were searched electronically from the earliest available date to June 14, 2021, to identify relevant literature: Web of Science, PubMed, Cochrane, and Ovid PsycINFO. Three key search components were used: eHealth interventions, cancer, and the Netherlands. An overview of the search strategies for each database can be found in Multimedia Appendix 1. Other potentially relevant publications were identified by tracking the reference lists of included articles.

Eligibility Criteria
Studies were eligible if the following criteria were met: • Population: the eHealth intervention was offered in the Netherlands and targeted adults (>18 years) diagnosed with cancer who were about to start, are currently undergoing, or have finished treatment (ie, cancer survivors) within the Dutch health care system.

•
Intervention: the study focused on eHealth interventions according to the definition of eHealth by the World Health Organization [5]: "the use of information and communication technology in support of health and health-related fields." Both fully web-based and blended eHealth interventions (ie, interventions combining web-based components with face-to-face contact) were included [25]. The eHealth intervention did not consist of business intelligence and big data solutions, such as analyzing structured and unstructured data to gather information to support decision-making [26].
• Comparison: studies were included independently of the presence and type of control group.
• Outcome: there was no focus on specific research outcomes for the first aim-to provide an overview of available eHealth interventions. The goal was to obtain a broad picture of available eHealth interventions. For the second aim-to provide an overview of empirical evidence regarding the impact of eHealth interventions-only studies that measured one or more of the Triple Aim domains were included.
• Setting: using any study designs except for incomplete trials, editorials, letters, and reviews. Nonetheless, the latter method was used to identify additional relevant studies from the reference lists. We excluded these 3 study designs as they were non-peer-reviewed or did not discuss a specific intervention.
• Time: all years were included as long as the study was published in the Dutch or English language.

Selection Procedure
The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 Statement was used to ensure the validity and reliability of the selection procedure [27]. The PRISMA 2020 checklist can be found in Multimedia Appendix 2 [28]. One investigator (LvD) searched for eligible studies. Subsequently, the reference software program Endnote (Endnote X7; Thomson Reuters) was used to remove duplicates. Two investigators (LvD and LS) independently screened the titles and abstracts of the articles to identify relevant studies. Next, full texts of the potentially relevant articles were assessed. Discrepancies between investigators were mutually resolved through discussion until a consensus was reached. Web-based software Covidence (Veritas Health Innovation) [29] was used for the screening process.
• Functional category: the functional category classification of the interventions was based on CEN (Comité Européen de Normalisation)-ISO (International Organization for Standardization) DTS (Draft Technical Specification) 82304-2:2020 [107]-a document providing quality requirements for health applications. The following categories were distinguished: (1) inform; (2) simple monitoring, to allow users to record health parameters to create health diaries; (3) communicate, to allow 2-way communication; (4) preventive behavior change, to change intended user behavior, such as related to smoking or sexual health; (5) self-management, to help persons with specific health issues to manage their health; (6) treat, to provide treatment for specific health issues or to guide treatment decisions; (7) active monitoring, to automatically record information for remote monitoring; and (8) diagnose, to use data to diagnose health issues.
• Type of eHealth: the classification of the type of eHealth of the intervention was based on the categorization of Nictiz [26], a Dutch knowledge center for national applications of information and communications technology in health care [108]: (1) web application or web portal (offered via a web browser, place, and time-independent), (2) mobile app (available on a smartphone), (3) health sensor (to measure vital bodily functions) or health gateway (to collect and transmit data from health sensors to medical professionals) or wearable devices (health sensors carried on the body), (4) electronic health records or personal health records, and (5) video communication tools.
• Intended setting to use the intervention: primary care, secondary care, or community • Target population: type of cancer, demographics (gender, age, and nationality), and specific characteristics (eg, smokers) • Support of health care professional: yes or no, with an explanation • Use of theory in the development of the intervention: yes or no, with an explanation • Stakeholder involvement in the development of the intervention: yes or no, with an explanation Information on research methods and outcomes was extracted at the study level for each empirical evaluation study. More specifically, we extracted information on the study design and objective, the number of participants included at baseline, description of the control group (if applicable), data collection period, study measures, and outcomes. Study outcomes were classified using the Triple Aim [23]. The Triple Aim describes an approach to improve health system performance by focusing on the following: 1. Improving the health of populations 2. Improving patient experience (including quality, patient-centeredness, safety, and timeliness of care) 3. Reducing the per capita cost of health care [23] We used the framework by Struijs et al [109,110], who elaborated on this model by breaking down the 3 aims into more concrete dimensions (Textbox 1).
Furthermore, a quality appraisal was conducted for each empirical evaluation study using the Effective Public Health Practice Project Quality Assessment Tool for Quantitative Studies [111]. This tool has been reported to have construct and content validity [112,113]. Furthermore, the tool can be used to gain insight into the quality of different study designs, making it easier to compare the results of the quality appraisal in this review. This tool assesses 6 components: (1) selection bias, (2) study design, (3) confounders, (4) blinding, (5) data collection methods, and (6) withdrawals and dropouts. Each component can be rated as strong, moderate, or weak based on the guidelines for the tool. Based on the ratings of each component, the tool allocates an overall methodological score for the study: strong, moderate, or weak. Textbox 1. Overview of levels in Triple Aim based framework by Struijs et al [109,110]. Finally, an overview of funding sources per article can be found in Multimedia Appendix 4.
Customized data extraction sheets were developed for the intervention characteristics and the study design, quality appraisal, and study outcomes. To ensure consistency in data extraction, one researcher (LvD) independently subtracted the data of each study and a second researcher (LS) subtracted data of a random sample of 15% of these studies. The interrater agreement was 83.5%, which was considered good. Data were narratively synthesized in 2 sections. The first section discusses the intervention characteristics of the identified interventions.
The second section discusses the study design, quality appraisal, and empirical study outcomes. The main characteristics of the interventions are described in the subsequent section to provide an overview of available eHealth interventions in cancer care and their characteristics as described in the scientific literature (the first study aim). The described intervention characteristics are purpose, functional category, type of eHealth, setting, target population, support of health care professionals, and the use of theory.

Intervention Purpose
The included interventions had a broad range of purposes, such as supporting decision-making (eg, decision aids), communicating with health care professionals, monitoring patient-reported outcomes, and participating in online support communities. Almost half of the interventions targeted psychosocial factors (eg, cognitive or sexual functioning and psychological adjustment) or problems (eg, smoking, drinking behavior, depression, and anxiety). Approximately two-thirds of these psychosocial interventions aimed to reduce general psychosocial issues or psychological complaints or foster patients' self-efficacy or disease coping.

Functional Category, Type of eHealth Intervention, and Setting
The interventions had various functions, in some cases, more than one. The most common functions were inform (n=35), self-manage (n=14), treat (n=11), and preventive behavior change (n=7). Most interventions were web applications or web portals (n=34) or mobile apps (n=7). Most of the interventions were used in secondary care (n=32).

Target Population
Approximately half (17/38, 45%) of the interventions targeted the general population of patients with cancer or survivors, whereas others targeted a specific type (

Support of Health Care Professionals and Use of Theory
Support from a health care professional was possible in 55% (21/38) of the interventions. Support comprised, among others, web-based support from a coach [30,31], weekly feedback from a health care provider [32][33][34], and teleconsultation with a health care provider [35,36]. Approximately 60% (23/38) of the interventions were theory-based, using, for example, principles from cognitive behavioral theory and the theory of planned behavior.
More details on the intervention characteristics can be found in Multimedia Appendix 3.
Characteristics of the empirical studies and the study results are described in the subsequent sections to provide an overview of the empirical evidence regarding the impact of eHealth interventions in cancer care on population health, quality of care, and per capita costs, the Triple Aim domains (the second study aim).

Study Outcomes
Most studies measured at least one dimension within either the population health or quality of care domain (23 and 24 studies, respectively).
Three studies measured at least one dimension within the per capita costs domain (Table 2 and Multimedia Appendix 7 [21,30,32,). An overview of the domains and dimensions measured per study can be found in Multimedia Appendix 8 [21,30,32,. The outcomes are described by dimension in subsequent sections. Unless stated otherwise, significant between-group differences were described by comparing the intervention and control groups. h: Course of supportive care needs n.s. Patient-physician interaction over time n.s. Self-efficacy n.s. Personal control n.s. Patient activation n.s. In the IC, 78% activated their account and 52% used the intervention as intended.
Study 1 [41] h: The loss to follow up in the IC was 36%. l: OncoCompass is likely to be equally effective on utilities and not more expensive than usual care. Study 2 [42] e: Physical health after 12 months ITT and protocol analysis n.s. g: After 6 weeks: Depression (GI vs GWL group and Total glioma group vs non-CNS cancer group) n.s. Fatigue (GI vs GWL group) sig*. After 12 weeks: depression n.s. Fatigue n.s. Other measures (GI vs GWL group) n.s. h: Most patients said they had benefitted from participating (73% glioma; 67% non-CNS), and the program was useful (92% in both groups) and informative (86% glioma; 92% non-CNS). The participation rate Everything under control (Alles onder controle) [31] was 40%. The adherence of the IC was 85% for the introduction and 77%, 52%, 40%, 37%, and 35% for modules 1 through 5, respectively. h: Satisfaction with information sig*. Involvement n.s. Decisional conflict n.s. Knowledge scores n.s. Subjective knowledge sig**. Objective knowledge n.s.
Prostate cancer decision aid (Prostaatkanker keuzehulp) [45] g: Fatigue severity sig*. Psychic complaints n.s. Positive and negative affect n.s. h: The proportion of participants who dropped out before completing 6 weeks of the protocol was 18% in the AAF condition, 38% in the eMBCT, and 6% in the psychoeducation condition.

Study 2 [53]
EvaOnline e: Sexual functioning n.s. HRQoL n.s. g: At T1: Both IC groups' (guided and self-managed) perceived impact of HF and NS sig**. Guided group overall levels of menopausal symptoms sig**. Both IC groups sleep quality sig**. Guided hot flush frequency sig. Guided group night sweats frequency sig**. Psychological distress n.s. h: Minimum compliance rate was 90.6% for the guided and 78.8% for the self-managed IC's. Study 1 [21] l: The guided and self-managed iCBT are cost-effective. Self-managed iCBT is the most cost-effective strategy.

Study 2 [55]
Home-based exercise intervention c: Self-reported physical activity at 6 months sig*. BMI at 6 months n.s. Mean absolute VO 2 peak at 6 months n.s. Aerobic fitness at 6 months sig.
h: 16 (84%) patients evaluated the physical exercise program as good or excellent, and 4 as moderately or sufficiently satisfactory. Mean adherence was 79%.  My-GMC [58] b: Symptom burden n.s. Teleconsultation for patients receiving palliative home care [36] PCT studies Transmural oncological support h: The average score of all patients for the monitoring function was 8.0 (10-point scale). The average score rated by 7 GPs of the electronic health information support system was 5.6 (10-point scale). The participation rate was 66%. All patients used the system.

Study 1 [43]
e: After the intervention: 5 of the 22 QoL subscales (state anxiety, fear related to specific head and neck problems, physical self-efficacy, perceived abilities in swallowing and food intake, and general physical complaints) sig. At 3 months: 1 subscale (physical self-efficacy) sig*. Other subscales n.s.
h: The participation rate in the IC was 66%, and 35 out of 39 patients completed all questionnaires.

Study 2 [44]
Before-and-after design studies g: Total number of "pain registrations" in the medical records sig*. Home monitoring tool for adequate pain treatment [54] a Triple Aim domains: a=health outcomes, b=disease burden, c=behavioral and physiological factors, d=Participation, e=Functioning and quality of life, f=Patient safety, g=Effectivity, h=Responsiveness, I=Timeliness, j=Support, k=Accessibility, l=Costs of care, m=Volume, n=Organizational costs, o=Productivity loss. b sig=significant positive between-group difference in favor of IC, P value unknown; sig*=significant positive between-group difference in favor of IC, α≤.05; sig ** =significant positive between-group difference in favor of IC, α≤.01; ns=nonsignificant between-group difference in favor of IC.
c MT=controlling for multiple testing or comparisons; d ITT=intention-to-treat analysis.

Per Capita Costs
Three studies measured a dimension within the domain per capita costs [42,54,55]. Two studies [42,55] measured the dimension costs of care, and both found through economic evaluation that the intervention was likely to be equally cost-effective compared with care as usual. One study [54] measured the dimension volume and did not find significant effects. None of the studies measured the dimensions organizational costs or productivity loss.

Principal Findings
This systematic review is the first to provide an overview of eHealth interventions in Dutch cancer care and use the Triple Aim framework to examine the empirical evidence of these interventions on population health, quality of care, and per capita costs (the Triple Aim domains). The review focused on Dutch cancer care; however, the results are also relevant to other Western countries involved in digital care for patients with and survivors of cancer. A total of 38 interventions were identified, and the results showed that most eHealth interventions targeted psychosocial factors or problems. In addition, interventions were aimed at many different target groups, including the general population of patients with and survivors of cancer, patients with a specific type of cancer, or patients who experienced a specific problem, such as cancer-related fatigue or smoking behavior. Few interventions were tailored to age, gender, or disease severity. The most common intervention types studied were web portals or web applications. These function to inform and facilitate self-management. Other types of interventions (eg, electronic health records or video communication tools), functions (eg, communication or diagnosis), and target outcomes (eg, communication with health care professionals or access to electronic health records) were rarely found.
Most outcome measures could be related to the Triple Aim domains population health and quality of care, whereas the per capita costs domain was largely neglected. Within the population health domain, mixed effects were found regarding the impact of eHealth on functioning and quality of life. Most studies measuring behavioral and physiological factors found positive effects. More specifically, there was preliminary evidence for the positive effects of eHealth interventions on physical activity and aerobic fitness. None of the studies considered the dimension participation, including outcome measures such as social inclusion. Within the quality of care domain, eHealth interventions seemed effective in increasing sleep quality and decreasing fatigue, in line with a meta-analysis showing that eHealth interventions effectively manage fatigue in highly fatigued cancer survivors [114]. Findings in terms of positive and negative adjustment to cancer and psychological complaints were inconsistent. One of the measures that was not considered was accessibility, which is worthy of mention as there is increasing global awareness that eHealth should be equally accessible to different populations [115]. The per capita cost dimension was largely neglected in the evaluation studies; only 3 studies considered dimensions within this domain.
This study yielded several interesting findings. With 38 interventions in Dutch cancer care, there appears to be a wide range of eHealth interventions for patients with and survivors of cancer. It seems valuable that most interventions targeting psychosocial factors or problems were aimed at general psychosocial issues, psychological complaints, patients' self-efficacy, and disease coping. Recent research shows that almost all cancer survivors are affected by fatigue [116], 1 in 2 patients with cancer is significantly distressed, and 47% have problems getting around [117]. In contrast, few interventions focused on pain from cancer, which is experienced by half of the patients with cancer during active treatment and 65% of the patients with advanced disease [118]. Some common symptoms of active treatment, such as vomiting, nausea, and constipation [119], were not considered. The lack of tailored interventions according to age, gender, or disease severity is noteworthy as subgroups within these categories are likely to have different preferences and needs. For example, older patients may find it more challenging to use eHealth interventions [120]. In addition, patients in different stages of the disease may have different needs as far as information and support are concerned [14].
We found that most interventions consisted of a specific type (web portals or web applications), function (information provision or facilitation of self-management), and target outcome (psychosocial factors or problems). We assume that besides the interventions we identified, more eHealth interventions are being developed and used by patients with or survivors of cancer. These interventions are likely to be designed or evaluated for a broader target population than patients with and survivors of cancer alone. For example, multiple studies have evaluated the general use of electronic health records and patient portals in academic hospitals without targeting a specific patient population [121][122][123][124]. Our search strategy included only patients with or survivors of cancer as a critical criterion; therefore, our search results did not include these interventions. As a result, the number of interventions available for patients with and survivors of cancer may be more significant and versatile than the results of this review.

RenderX
Another interesting finding is that the results of the evaluation of study outcomes are mainly in line with the literature. For example, several meta-analyses have been conducted to examine the effect of eHealth on the quality of life of patients with or survivors of cancer do find a statistically significant effect [114,125], while others do not [126,127]. These mixed findings, which we also found in the review, can be explained by the fact that quality of life is a multidimensional variable influenced by multiple factors [128]. The current inconsistent findings for psychological complaints and adjustment to cancer were also found in a previous meta-review, which found inconsistent results for the effect of eHealth on psychological well-being, depression, and anxiety in patients with cancer [14]. When interpreting the study results, it is important to remember that many eHealth interventions are not implemented in daily practice. In addition, many expected benefits of such interventions are not realized in daily clinical practice [129,130] as they are not being used as intended [131,132]. The latter has several root causes such as lack of trust and digital literacy [133]. The suboptimal use of eHealth interventions in daily practice is a significant problem that future research needs to address.
Finally, it is notable that some domains and dimensions are primarily omitted from the studies, such as per capita costs and participation. The scarcity of per capita cost-related study outcomes is in line with previous research on the effectiveness of eHealth interventions in cancer detection, treatment, and survivorship care [134]. As health care costs are increasing in most countries, organizations are actively trying to develop solutions to curb health care expenditures while maintaining access to and harnessing the quality and safety of health care [135]. Digital health care is often viewed as a solution to increasing health care costs. Evaluating eHealth interventions is relevant for adequate resource allocation decisions and designing services for competing health interventions and limited resources. Participation is also an essential theme for eHealth because eHealth interventions can either foster social inclusion or create new risks of social exclusion (eg, for digitally illiterate patients) [136]. In future studies, it will be essential to consider the needs of patients at risk of social exclusion when developing and evaluating eHealth interventions.

Limitations
This review had some limitations. First, this review may not have included all available eHealth interventions, as not all available interventions have been scientifically evaluated. Gray literature and ongoing studies in trial registries were not included in this review, nor were experts consulted nor the authors contacted. Second, the Triple Aim framework used in this review provides a comprehensive overview of the domains and dimensions. However, creating an objective distinction between different dimensions was not always possible. For example, an outcome measure such as improved sleep quality could be classified as effectiveness or behavioral or physiological factors. Hence, categorizing outcome measures into different dimensions was, to some extent, subjective. Third, for each category of study outcomes, we examined only a small number of studies that evaluated the impact of the intervention on the outcome. Publication bias was not investigated in this study. Therefore, we should be cautious about the conclusions drawn regarding the impact of eHealth interventions on certain subdimensions. Finally, the study protocol was not registered.

Future Research
Future research should examine the dimensions of the Triple Aim that have rarely or not been taken into account in previous research, such as participation and accessibility. Furthermore, studies should examine in further detail what explains the mixed results for studies measuring specific dimensions such as functioning and quality of life. This could be done, for example, in experimental studies examining the effect of particular intervention characteristics on the Triple Aim domains. Further research is needed to increase our understanding of how different intervention characteristics influence intervention outcomes and the underlying causal mechanisms that cause an intervention to be effective. Interventions aimed at coping with pain were rarely found. eHealth interventions such as digital training to develop pain coping skills and pain management apps custom-made for patients with cancer have proven feasible and effective in decreasing pain [137,138]. Future research should explore the potential of such interventions in the Dutch context. Furthermore, this review may be repeated in other countries to compare the intervention characteristics and outcomes of eHealth interventions in cancer care internationally, facilitating learning and sharing best practices. Finally, this review focused on specific eHealth interventions in cancer care. Research on the structural embedding of eHealth interventions in care processes is essential for optimally deploying these interventions. Therefore, future research can examine local care pathways to identify new possibilities for eHealth to address challenges and needs across existing care pathways. Potentially, these insights may lead to new care pathways to optimize cancer care quality.

Conclusions
Most of the 38 interventions in this review included eHealth interventions for patients with or survivors of cancer in the Dutch health care system consisting of a specific type (web portals or web applications), function (information provision and facilitation of self-management), and target outcome (psychosocial factors or problems). Almost none of the interventions were tailored to the needs of patients with or survivors of cancer based on age group, gender, or disease severity. The Triple Aim domains population health and quality of care have been studied thoroughly, whereas the domain per capita costs is understudied. Most of the included evaluation studies were assigned a moderate quality appraisal score, and selection bias was likely present in most studies. Our results indicate that eHealth could benefit patients and survivors by improving sleep quality, reducing fatigue, and increasing physical activity. Further research is needed to fully understand the effect of eHealth on aspects such as participation (in the form of social inclusion), accessibility, and the effect on quality of life, patient behavior, physiological health, psychological well-being, and per capita costs. Finally, more economic evaluation of eHealth interventions is required. Overall, continuing a holistic evaluation of eHealth interventions in cancer care will be critical to improve population health, enhance the quality of care, and decrease per capita costs.