Pediatric palliative care is a heterogeneous field concerning children with various life-threatening or life-shortening conditions from birth until young adulthood [1]. Pediatric palliative care is provided regardless of diagnosis, and the aim is increased quality of life; “Palliative care for children is the active total care of the child’s body, mind and spirit, and also involves giving support to the family” [2]. There is no definite number of children in need of pediatric palliative care. Global estimates suggest that between 113 per 10,000 children in Zimbabwe and 20 per 10,000 children in the United Kingdom are in need of specialized or generalized palliative care [3]. The complexity of the needs of these children and their families makes them dependent on multidisciplinary efforts to manage symptoms and provide psychosocial care and support [1,4]. These children often experience pain related to medical treatment that is frequently worsened in a hospital—a nonfamiliar environment [5]. Previous studies support pediatric palliative care at home, where children and their families are most at ease [4,6], to improve their quality of life [4] and their quality of care [7,8]. Home-based pediatric palliative care should involve a specialized interdisciplinary team that is often organized in specialized care [4,9]. To meet the care needs of these children and their families, the professionals involved must directly communicate not only with the child and his or her family but also with one another [10]. However, communication is suggested to be a core challenge in pediatric palliative care [7].

Electronic health (eHealth) systems facilitate remote communication to provide care at home without requiring that patients or health care personnel (HCP) travel. eHealth is defined as “the use of information and communication technology for health” [11]. The relevance of eHealth in home-based pediatric palliative care has been highlighted, thus suggesting that eHealth can improve patients’ quality of care [7]; however, further research is warranted as the full potential of health technology has not yet been realized [6]. A recent case report suggested how mobile technology provides a platform for affordable and high-quality communication such as through videoconferences [12]. The factors in favor of eHealth in home-based pediatric palliative care are ease of use, patient and clinician satisfaction, and the potential for saving travel time and money for patients and HCP [6,12]. Age-based preferences regarding technology and communication with clinicians should guide the development of new technology [7], particularly relevant in pediatric palliative care [7].

Bradford et al [6] conducted a systematic review in 2013 to summarize the evidence for home-based telehealth in pediatric palliative care. They emphasized the logistical and ethical issues regarding research that involves this vulnerable group by highlighting the importance of research that minimizes patients’ burdens [6]. This emphasis is supported by the conclusions of a general review regarding research on pediatric palliative care [13]. Despite the rapid development of technology in general, the current field of research on eHealth in home-based pediatric palliative care is rather scarce and lags behind. To ensure that home-based pediatric palliative care supported by eHealth is evidence based, we argue that an updated review of evidence published after Bradford et al’s review in 2013 was necessary [6].

This study aimed to identify and review the use of eHealth to communicate and support home-based pediatric palliative care and appraise the methodological quality of the published research.

This systematic mixed methods review is based on a convergent design [14]. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [15], we applied systematic database searches, and we integrated studies regardless of their design, and qualitative and quantitative methods were equally valued.

The scope and aim were developed and discussed within our research group before a protocol was written. The rationale for conducting a mixed methods approach was based on the limited existing evidence on home-based pediatric palliative eHealth, and thus, a single method review would not sufficiently clarify the evidence within the field. The review protocol was published in the International Prospective Register of Systematic Reviews (PROSPERO), ID: CRD42018119051 [16].

To ensure a comprehensive search, we used the sample, phenomenon of interest, design, evaluation, and research type (SPIDER) search tool to identify targets and search terms [17]. Furthermore, 2 research librarians assisted in the development of a search string and tailored each string to individual databases. The systematic search was prepared in November 2017—followed by an updated search in December 2018—in the following databases: Medical Literature Analysis and Retrieval System Online (search string in Multimedia Appendix 1), EMBASE, PsycINFO, Cochrane Library, Cumulative Index to Nursing and Allied Health Literature, Web of Science, and Scopus.

Specific inclusion and exclusion criteria were applied according to the SPIDER framework (Table 1). As eHealth may nevertheless be considered a relatively new field of research, we anticipated smaller studies using noncontrolled designs; therefore, studies were included regardless of their design. We included papers published after February 22, 2012, as our review builds on a previous review [6]. There were no language restrictions, and we exclusively included papers published in peer-reviewed journals. We excluded both letters and editorials.

All citations were assessed independently by 2 researchers according to the inclusion and exclusion criteria set a priori (HH assessed all citations, whereas AW and KR split the citations and each assessed one half). Disagreements were resolved by discussion among the 3 researchers, and no disagreements required an independent researcher. All citations were screened through their titles and abstracts to exclude those that clearly did not meet our inclusion criteria before reading the remaining papers in full to assess their relevance to our aim.

The results of the eligible papers were extracted using the following structured form based on the SPIDER framework [17]: author, country, sample, phenomenon of interest, research type and design, evaluation, and results.

To appraise the methodological quality, we used the standardized checklists according to the designs of our primary studies, which are available from Joanna Briggs Institute [18]. All studies were independently assessed by 2 reviewers (HH and KR), whereas a third researcher (AW) assisted with any disagreements that arose.

The studies were summarized descriptively and in tables. As a first step in the analytical process, all included papers were read in full by the 3 researchers, and the results sections of all papers were extracted as data for our analyses, regardless of their initial design. We transformed all quantitative data into qualitative data [14] and applied line-by-line coding, inspired by the thematic synthesis described by Thomas and Harden [19]. A thematic synthesis has been used in line with the convergent design of mixed methods reviews [20]. The 3 researchers individually coded all the material before discussing the codes and agreeing upon descriptive categories and conceptual themes. The heterogeneity of the designs, interventions, and methodological quality did not allow for any pooled statistics or meta-analyses.

The initial search identified 1642 citations, and the repeated search identified 346 new citations (Figure 1). After duplicates were removed, 1277 titles and abstracts were screened according to the inclusion criteria. The remaining 85 papers were read in full. We contacted 5 authors to clarify the study details and sent 1 reminder to those who did not reply. Two authors replied positively to our requests. We included 7 papers [21-27] in our review (Tables 2 and 3).

The studies applied various designs (Table 3); 1 study aimed for a controlled design [21] but was forced to end recruitment prematurely because of unanticipated patient deaths. The participants were mainly HCP who discussed technology on behalf of children and their families in pediatric palliative care. All studies evaluated an intervention. To grasp the intervention evaluated by Bradford et al [21-24], data were extracted from the primary publication [28]. Likewise, for the paper by Levy [27], the intervention was described elsewhere [29]. Overall, 6 of the 7 papers reported on videoconferences as the primary method for remote communication [21-24,26,27], whereas 1 study evaluated a Web-based tool [25]. One research group in Australia conducted 4 of the included studies [21-24], whereas the 3 remaining studies originated from 3 separate research groups: 1 also in Australia [26] and 2 in the United Kingdom [25,27].

The methodological appraisal revealed several shortcomings (Multimedia Appendix 2). A total of 3 studies [21,25,26] were assessed using the checklist for quasi-experimental designs [30]. The lack of control groups and clarity regarding any comparison increased the likelihood of bias. Furthermore, 2 studies [24,27] were appraised using the qualitative research checklist [31]. We did not find a philosophical perspective in either, although the chosen methodology seemed appropriate. Neither study addressed the researcher’s influence on the research, but both adequately presented the participant’s voices. One study [23] was evaluated through the economic evaluation checklist [32] where all items were satisfactorily covered except for the item covering whether or not the intervention’s clinical effectiveness was established—in this case, videoconferencing in home-based pediatric palliative care. Finally, 1 study [22] was appraised using the checklist for case-control studies [33], where most items were satisfactorily covered. None of the included studies were excluded based on the methodological appraisal.

The qualitative synthesis resulted in the following 3 themes: eHealth at home, technological features, and system for eHealth (Table 4).

This systematic mixed methods review summarizes the research in eHealth for communication and support in home-based pediatric palliative care. We identified merely 7 eligible studies and developed 3 themes to describe our findings according to our aim, including eHealth at home, technological features, and system for eHealth. Generally, HCP’s voices were stronger than those of the patients and their families. eHealth systems were perceived as both a support and a possible intrusion into the home for patients and their families. Furthermore, eHealth systems needed to be easy to use and effective to facilitate communication and support.

Despite our inclusion criteria being open for all designs, we nevertheless did not identify a larger number of eligible studies. The lack of research is among the main findings of this systematic review; 4 papers included in our review were based on the same research project and sample [21-24] and were performed by the same research group who also performed the only existing systematic review on eHealth and home-based pediatric palliative care conducted before ours [6]. This review identified merely 6 eligible pediatric studies. A Cochrane review on eHealth interventions to support mental health in children with long-term physical diseases also identified a limited number of studies of low methodological quality [34], which underlines our finding that research in this area, particularly intervention studies, is scarce. Several explanations can be offered for this lack of research. Previous discussions suggest that research in this vulnerable patient group is challenging to conduct [6]—a finding that was emphasized in 2 studies included in our review [21,25].

Studies in palliative care—particularly randomized controlled trials (RCTs)—place ethical demands on research design. In controlled trials, it is important to ensure that patients receive the best care regardless of the group to which they are randomized. Randomizing patients to a waiting list is one alternative design strategy that provides all participants with the possibility to receive intervention. However, each patient’s life expectancy is limited, and any delay may equate with a patient not receiving an intervention at all. Participant recruitment and attrition are among the major barriers in pediatric palliative care studies [13], which became apparent in 1 study that was required to prematurely abandon its recruitment because of patient deaths [21]. Moreover, attrition poses obvious consequences for follow-up. Timing the measurement is challenging because of the unpredictable development of an illness. A pragmatic solution to the ethical and practical issues related to recruiting and assessing children in palliative care and their families involves recruiting HCP instead. Letting HCP reflect on the efficacy and usefulness of eHealth provides the research field with at least some information; however, eHealth consultations narrated by HCP do not represent the subjective views of children and their parents. The 2 qualitative studies [24,27] that provided this review with the richest data exclusively included HCP. Consequently, the results are largely based on expressions from HCP that represent their experiences and interpretations of the families’ views. As a result, the unique and lived experiences of each child and his or her family are lacking, thus representing a major gap in the research field.

Conducting rigorous eHealth intervention studies has also been demonstrated as methodologically challenging [35-37]. Software is intended to change and progress, which is not compatible with long-lasting, standardized, randomized trials. It has been argued that eHealth interventions are complex interventions that benefit from alternative evaluation designs [35]; nevertheless, most studies in this field are RCTs. In their previous review, Bradford et al [6] highlighted the need for alternative designs in pediatric palliative care evaluation, and our findings confirm that this need persists. Previous research has called for a pediatric palliative care study framework to establish methods to increase recruitment and decrease attrition, while simultaneously maintaining ethical issues [13]—a demand that remains highly relevant. The lack of a methodic consensus on the evaluation methods in eHealth studies adds to the difficulties associated with conducting pediatric palliative care–centered research.

Although the results of this review are characterized by few and diverse eligible studies, the findings complement existing knowledge that has been summarized by Bradford et al’s review (conducted before ours) [6]. Previous research suggests that eHealth must be a feasible means to provide information, education, and support [38,39], but the barriers associated with establishing a holistic and integrated, permanent eHealth system service seem to remain. Our review suggests that the reluctance toward eHealth technology mainly originates from HCP and to a lesser extent reflects the barriers described by the affected children and their families. The disadvantages of eHealth are related to its technological features, although the organizational structure of the health care system within which eHealth technology is placed can largely reduce these disadvantages. For eHealth technology to be integrated into standard care, health care services and HCP must acknowledge the system. HCP’s knowledge and perceptions as well as the culture within health care services will, namely, pose consequences for the adoption and utilization of eHealth services. If HCP are reluctant and prefer telephone calls for remote consultations, the integration of eHealth technology for communication and support in home-based pediatric palliative care will not be facilitated. The perceptions of children and families are additionally crucial, and positive experiences with eHealth can facilitate the use of eHealth in home-based pediatric palliative care.

Regulations meant to safeguard patients can end up withholding viable implementation of technology, and current regulations must be updated to meet the needs of a new generation of technology and users [40]. Concerns related to security and privacy in eHealth technology might be a barrier toward its development and implementation; interestingly, the integration of eHealth systems with ehealth records is not discussed in some studies, although previous research has highlighted these challenges [39,41,42].

New technology produces limitless possibilities, but unless this development is guided by patients’ needs, such technology is less likely to end up as viable and feasible for patients and HCP. Thus, the assessment of users’ needs and process evaluations are crucial in the development and evaluation of eHealth systems for communication and support in pediatric palliative care. It is ethically difficult to justify evidence that informs new services wherein the significant part is not included. This review determined that eHealth was evaluated by HCP, or objectively through medical chart notes after consultations, or (less so) through children’s and families’ self-reports. HCP and parents tend to underreport the frequency and severity of symptoms compared with self-report by the affected children [43]. Although HCP mainly focus on physical symptoms, children, siblings, and parents often suffer from psychological symptoms that are not always acknowledged by HCP [44]. The lack of research on users’ needs is alarming as every child and family is unique and the subjective experiences of both are crucial for individualizing care and optimizing their quality of life.

Although we determined that children’s voices were absent, this review indicates that eHealth technology may potentially support communication between these children and HCP without their parents’ presence and subsequently facilitate the child’s autonomy. This detail is particularly important as previous research suggests that a child wants to be actively included in both his or her own care [45,46] and any decisions related to his or her health and care [47-49].

The participants’ demographic characteristics suggest that the gender perspective should be addressed in pediatric palliative care as most children and their parents were females; boys and fathers were largely absent. Merely 2 industrialized countries were represented, and the premises for eHealth in home-based pediatric palliative care might differ between industrialized and developing countries, most often as pediatric palliative care is frequently lacking in the latter [50]. However, differences may also exist within a health care system. A lack of stable internet access, necessary equipment, and digital capabilities among users may create diverse conditions for eHealth interventions. These factors may increase health inequalities.

Conducting systematic reviews to synthesize evidence from qualitative and quantitative methods represents an emerging field of research. Several approaches exist [14,18]; however, agreement as to which method is more appropriate is lacking. In this review, we applied rigorous search strategies, appraised eligible studies according to checklists, and analyzed our findings using well-established methods. Thus, we are confident that we have identified and synthesized existing and relevant evidence, although other method choices may have possibly provided us with alternative results.

The wide use of eHealth and health technology terms posed consequences for our systematic search. Internet, Web-based, and app are terms we experienced as challenging to handle during the search process. In this context, the abbreviation PC can represent both personal computer and palliative care, and the result was such that, in search strings, we would miss out the combination of the two. Owing to few relevant search results and the risk of excluding relevant studies, we decided to search with broader terms in addition to these narrower terms.

The scarce amount of research in the area involving eHealth-supported, home-based pediatric palliative care and the methodological and ethical challenges involved affected the conclusions that could be drawn from this mixed methods review. The results in the primary studies were mainly based on information from HCP. For eHealth to complement pediatric palliative care at home, we need research that identifies the needs and wishes of both children and their families. eHealth poses many possible advantages and can play an important role in home-based pediatric palliative care. If measures are not taken to establish a consensus on satisfactory research methods, then eHealth technology may be implemented without undergoing proper evaluation.

The findings of this review can specially inform future research through the need for a prioritization of research within eHealth to support home-based pediatric palliative care, because of the limited knowledge regarding the affected children and their families’ needs and wishes concerning eHealth. There is a need to develop research strategies to reduce unnecessary burdens on the children and their caregivers and simultaneously strive to optimize the study design.