Background

Colorectal cancer (CRC) is the third most common cancer globally [1]. Cancer stages I–III (i.e. nonmetastatic disease) dominate among CRC cases, with curative surgery being the cornerstone of treatment [2]. Patients with CRC are prone to comorbidities [3]. The impact of surgery, in combination with comorbidities, is found to be highest in the first year following surgery [4]. Most CRC patients are currently managed within enhanced recovery schemes [5], including early discharge to home post-surgery, when physiological functions such as oral intake of nutrients or bowel functions may not be fully restored [6]. About half of anastomotic leakages after bowel resection occur after discharge from hospital, with serious consequences for the patient [7]. Consequently, the period of transition from hospital to home may represent a vulnerable time, prone to issues that can contribute to readmission. Readmission rates for CRC range from 9 to 25% [8] and are deemed markers of quality of care [9].

Following discharge, many CRC patients may struggle with navigating the healthcare system and adopting recommended self-management behaviours. The self-management of CRC includes monitoring health, accessing health information [10] and initiating health behaviour changes, such as exercising more [11]. Moreover, CRC patients may struggle with self-management tasks like finding medical information, monitoring health and interacting with healthcare services, which may result in physical and mental fatigue [10].

eHealth is defined as ‘the use of information and communication technologies (ICT) for health’ [12]. eHealth support deployed post-hospitalisation may promote self-management among people with severe conditions [13]. However, further insight is needed into how a more seamless eHealth service during the transition from inpatient to outpatient care may enable patients to obtain adequate self-management support, feel safe and recover well [14].

There is some evidence that eHealth can support cancer survivors in the self-management of treatment side effects and complications and increase their quality of life (QOL) [15]. Recent reviews of eHealth in the context of CRC populations are sparse. In an overview of reviews on telemedicine (e.g. eHealth) in post-treatment cancer survivorship, none of the 29 included systematic reviews focused on CRC patients only [16]. A systematic review aiming to study eHealth support directed at CRC survivors’ follow-up needs upon discharge from the hospital addressed the interventions’ service content, outcomes and software infrastructure [17]. The findings demonstrated that eHealth was useful for CRC survivors in supporting physiological, psychological and cognitive needs and enabling better symptom management and QOL [17]. Nevertheless, there is a knowledge gap concerning technology acceptance and how patients adhere to eHealth interventions. Adherence is defined as ‘the extent to which a person’s behaviour corresponds with agreed recommendations from a healthcare provider’ [18] (p. 3), but little is known about how eHealth may promote adherence to recommended CRC self-care [19].

This study aimed to (1) explore the user interface, content and delivery mode of CRC eHealth interventions following discharge after surgery, (2) investigate patient adherence to the interventions, (3) establish intervention effects on patient-reported outcome measures (PROMs) and (4) describe patients’ experiences of eHealth follow-up interventions.

Methods

The study was conducted according to Whittemore and Knafl’s five-step framework for integrative reviews [20] and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [21].

Step 1: literature search

Comprehensive literature searches were performed by a university librarian in October 2021 using the Embase, Medline, CINAHL and Cochrane Library databases, as well as by manually searching reference lists. The search terms, limitations and search results are displayed in Table 1.

Table 1 Searches in library electronic databases
Full size table

Step 2: study selection

Endnote™ Version X9 [22] was used to manage the generated records from Search no. 3 (Table 1). After removing duplicates (N=471), a blinded screening of 1373 titles and abstracts was performed using the web application Rayyan [232] and a priori inclusion and exclusion criteria (Table 2). Following the blinded screening, a comparison of the decisions showed discrepancies for 37 records (4.5%), resolved through discussions among the authors. Fifty-nine full-text articles were distributed among the authors and assessed for final inclusion, with conflicting opinions being resolved through discussions among the authors. The results of the study selection process are displayed in a PRISMA flow chart [23].

Table 2 Inclusion and exclusion criteria
Full size table

Step 3: data extraction

To achieve consistency in data extraction, an extraction tool was constructed, including publication identifiers, study design, study context and participants, eHealth program, program adherence and patient outcomes and experiences. Any inconsistencies among co-authors were resolved via the assessment of a second reviewer.

Step 4: critical assessment of articles

The authors used the mixed methods appraisal tool (MMAT) [24] in teams of two to establish the risk of bias in the included studies. Here, the MMAT checklists for randomised controlled trials (RCTs) and non-randomised, descriptive, qualitative and mixed methods were used. Each study was assigned an overall quality score, varying from 25% when one criterion was met to 100% when all criteria were met. The MMAT was used as a summarising tool, with methodological quality considered according to the design of each study. The MMAT score was not used for exclusion decisions [24]. Studies were not excluded based on methodological quality. The strength of evidence was summarised as part of the review’s limitations.

Step 5: data synthesis

To analyse and synthesise data, the ‘framework synthesis approach’ was used, which includes five analytical stages: familiarisation with the data content, identification of themes, indexing, charting and map** and interpretation [25]. Data from the extraction table allowed the authors to familiarise themselves with the findings. Coding of the data was performed by one author according to key issues, concepts and themes, namely the outcomes and practices of eHealth follow-up programs, including content, delivery mode and user interface, patient adherence, impact of eHealth interventions and patient experience. The synthesis of the findings was then reviewed by a second author and finally examined by the co-authors.

Results

After the full-text assessment of the 30 records, four were excluded based on the eligibility criteria resulting in a total of 26 included papers [26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51] (Fig. 1).

Fig. 1
figure 1

PRISMA flowchart of study selection process

Full size image

Risk of bias

Among the 12 RCTs, two achieved full scores (7/7 points) [30, 51], four achieved 6/7 points [35, 42, 46, 49], five scored 5/7 points [38, 40, 43, 47, 48] and one scored 4/7 points [28]. Across the MMAT domains, seven of 12 RCTs did not report the blinding of outcome assessors. Nearly all the non-randomised studies scored 5/6 points, while one achieved 4/6 points [26]. Here, the less reported item referred to administration of the intervention as intended (5/8 studies). Only one of the three descriptive achieved the full score of 7 points [45] as the authors of the other two did not report on sample representativeness, risk of non-response bias and appropriate statistical analysis [33, 44]. The only mixed-methods study scored 5/7 points as it lacked reporting on sample representativeness and an adequate rationale for using a mixed-methods design to address the research question [36]. Both qualitative studies [31, 50] demonstrated good methodological quality, scoring 7/7 points.

Overview of study characteristics

The studies were published between 2012 and 2022 (Table 3), and most were of European origin. Three studies were performed by the same Swedish research team [31,32,33], while three Dutch studies involved the same eHealth application (i.e. Oncokompas) [47,48,49]. Fourteen studies applied RCT or quasi-experimental study methods, seven used observational designs, two were qualitative, two used mixed methods, and one used a case-study design. The study populations ranged from 1 to 756 participants (median number of participants, n=118). In one study, presented in two publications, the CRC patient population accounted for 25% of the participants [47, 48]. All studies recruited adult CRC patients (18–81 years of age). Median age, based on 21 of 26 studies that provided information on mean or median age was 65 years of age [26,27,28,29,30,31,32,33,34,35,36,37,38,39, 41,42,43,44, 46,47,48,49, 51]. In four studies, most of the participants were female [29, 30, 35, 41, 42]. Only two studies addressed the importance of a diverse sample as to provide eHealth services to demographically (e.g. education and income) and geographically (e.g. rural areas) diverse groups [42, 46]. In all the studies, patients were enrolled during the post-operative care trajectory. In eight studies, patients received the eHealth intervention during adjuvant chemotherapy [26, 28,29,30,31,32,33, 38, 39].

Table 3 A summary of findings on included studies’ origin, publication year, study design, study participants, patient outcomes, patient adherence, findings and quality assessment score
Full size table

Results of data analysis

eHealth interventions’ delivery mode, user interface and content

The modes of eHealth intervention delivery included telephone (n=14) [26, 28, 61] that depends on motivational support from healthcare professionals [62]. We found that eHealth interventions containing information, monitoring and real-time communication from healthcare professionals improved CRC patients’ engagement in PA. The monitoring of behaviour is the cornerstone of a health behaviour change and is often associated with a positive result [63]. In addition, eHealth may facilitate participation for cancer patients who lack access to or cannot conveniently access PA programs in their community [64].

Strengths and limitations

This review clearly describes the methods and outlines the process of data identification and selection as well as steps to synthesise the results from individual studies and evaluate the evidence, all of which create a robust and meaningful review. The inclusion of studies with different study designs enabled a more comprehensive approach to meeting the study aims. On the other hand, even though we employed a rigorous literature search overseen by a highly experienced librarian and used a digital sorting tool for the screening of records, relevant records may have been missed. We did not exclude inadequately reported studies as doing so would not affect the findings in any meaningful way [64].

Conclusion

In this review, we identified 26 studies of eHealth interventions following the discharge of patients from the hospital after curative surgery for CRC. eHealth interventions upon hospital discharge can offer support during a critical period. This review demonstrated that eHealth interventions were mainly telephone-based, delivering education, counselling or support and monitoring symptoms or health behaviours. However, there was a lack of focus on CRC patients’ adherence to eHealth. More research is needed on adherence to eHealth programs and its relationship with the implementation of eHealth in CRC populations.

eHealth follow-up may mitigate anxiety and depression in CRC patients, while the proof of its impact on other psychological morbidities or QOL is less clear. We also did not find strong evidence of the ameliorating effects of eHealth programs regarding the side effects of cancer treatment. eHealth interventions may have a positive influence on CRC patients’ PA behaviours regardless of the user interface, but the combination of technology and human interaction appears important. In general, remote, digital follow-up was experienced as positive, accessible and usable and as an improvement to healthcare services delivery.

This review can inform future intervention research on discharge planning in CRC care. In addition, it may support clinicians working towards ensuring the uneventful and swift recovery of CRC patients. Furthermore, the findings may have value in the development of eHealth services for other cancer patient populations.