Aim and Scope
The aim of this CPG is to provide guidance on all aspects of palliative care including physical, psychological, social, and spiritual aspects, for all children aged 0 to 18 years with life-threatening or life-limiting conditions and their parents, brothers and sisters (hereafter referred to as families) throughout the entire palliative trajectory (from palliative diagnosis till after end-of-life), with the ultimate goal to improve quality of paediatric palliative care. This CPG is intended for all health care providers from different specialisms who are involved in paediatric palliative care and for all children aged 0 to 18 years with life-threatening or life-limiting conditions and their families.
The guideline is an update of the first CPG for paediatric palliative care that was published in 2013, which provided recommendations on symptom relief, decision making and organisation of care.
Topic selection
In 2018, an invitational conference among experts in paediatric palliative care was convened to evaluate the first CPG for paediatric palliative care and identify new topics that needed to be addressed. This formed the basis for the online survey that was conducted among 89 health care providers to prioritize new topics. In this survey, professionals were asked to weigh topics that should be included during the revision of the CPG on a 5-point Likert scale that ranged from not important to very important. In addition, another survey was conducted among patient representatives and mostly bereaved parents of children with life-threatening or life-limiting conditions (n = 16) to indicate their priorities towards the identified topics in the invitational conference.
Multiple suggestions for new topics were derived from the results of the invitational conference and surveys. A preliminary topic list was generated which included topics covered in the CPG of 2013 (n = 12) and the suggestions for new topics (n = 8). The core group made a final selection based on practical and financial feasibility, and priorities of professionals, and patient and parent representatives. This resulted in a final list of 16 topics of which five topics were newly added. Figure 1 shows the selection process including reasons for exclusion of topics.
Multidisciplinary guideline development panel
A guideline development panel was established which consisted of 72 experts in paediatric palliative care and nine (bereaved) parents (see representation of patients and their families). Professionals from various disciplines such as paediatricians, paediatric oncologists, neurologists, anaesthesiologists, nurses, psychologists, pharmacists, medical pedagogical care providers and researchers, were included in the expert panel (Appendix A). The Netherlands Comprehensive Cancer Organisation recruited the panel members. Members were either mandated by their professional associations or participated on personal title. All members disclosed conflicts of interest at the start and end of the guideline development process.
Based on the final selection of topics, six main working groups (WGs) were formed. These WGs focused on symptom treatment (WG1), refractory symptom treatment (WG2), advance care planning and shared decision-making (WG3), organisation of care (WG4), psychosocial care (WG5), and loss and bereavement (WG6). Sub-WGs were established for WGs that covered multiple topics. Figure 2 provides a full overview of the (sub)WGs. The members of the expert panel were appointed to the (sub)WGs according to their expertise. Moreover, a core group was established to ensure consistency and transparency throughout the guideline. An overview of the working structure and guideline development process of is shown in appendix B and C.
Representation of patients and their families
Different methods were used to ensure representation of patients and their families. First, we conducted a survey to identify patient priorities for topic selection. Second, two patient representatives joined the core group to ensure the representation of patients and their families during the entire guideline process. Third, a panel consisting of 9 (bereaved) parents of children with life-threatening or life-limiting conditions, was established to review guideline texts and recommendations (Appendix A). The parents were recruited by the Child and Hospital Foundation and attended a short training on guideline development. The panel reviewed the first drafts of all guideline texts and recommendations. Additionally, the panel reviewed the complete concept guideline to ensure their input was incorporated correctly. Lastly, parents were asked to share their experiences during the interactive conference for organisation of care (WG4) (see consensus-based approach).
Formulation of clinical questions
Each (sub-)WG proposed several clinically relevant questions. Questions were developed according to the PICOS format, which defines the patient group, intervention, comparison to the intervention, relevant outcomes, and study design for each clinical question. The core group assessed all clinical questions carefully. If necessary, clinical questions were adjusted. The core group sent the final clinical questions to the (sub-)WGs for approval.
Below we describe the methods used to answer the clinical questions. In Table 1, an overview of the methods used per WG is presented.
Table 1
Overview of methods used per WG
| Symptom treatment | Refractory symptom treatment | Advance care planning & shared-decision making | Organisation of care | Psychosocial care | Loss & bereavement |
| WG 1 | WG 2 | WG 3 | WG 4 | WG 5 | WG 6 |
Identification of quantitative studies | | | | | | |
Systematic literature search to identify (SRs of) RCTs and CCTs on paediatric palliative care interventions | X | X | X | X | X | X |
GRADE assessment | X | X | X | X | X | X |
Identification of qualitative studies | | | | | | |
Systematic literature search to identify qualitative studies on barriers and facilitators for advance care planning and shared decision-making | | | X | | | |
GRADE CERQual assessment | | | X | | | |
Identification of additional literature | | | | | | |
Search for clinical practice guidelines on paediatric palliative care, general paediatrics, adult paediatric and/or palliative care | X | X | X | X | X | X |
Search for textbooks on paediatric palliative care | X | X | | | | |
Search for SRs of observational, qualitative, or mixed-method studies* | | | | | | X |
Other methods | | | | | | |
Consensus-based approach: Ideafactory | | | | X | | |
*Studies are derived from the search for (SRs of) RCTS/CCTs on paediatric palliative care interventions |
Identification of evidence – quantitative studies
Systematic literature search
WG1 to WG6 formulated a total 37 clinical questions related to the effect of paediatric palliative care interventions (Appendix D). We conducted a systematic literature search to identify Randomized Controlled Trials (RCTs), Controlled Clinical trials (CCTs) and systematic reviews (SRs) of RCTs on paediatric palliative care interventions. We updated the literature search that was conducted for the former CPG (2013). We searched Ovid MEDLINE and PreMEDLINE, MEDLINE (PubMed), CENTRAL and the Cochrane Database of systematic reviews from January 1, 2010 to January 24, 2020 (initial search October 5, 2018; top-up search, January 24, 2020), using a combination of the search terms “child”, “palliative care”, “randomized controlled trial” and “systematic review” (Appendix E).
The following inclusion criteria were defined: (1) RCTs and CCTs including a study population of at least 10 children, and SRs of RCTs, (2) study population consisting of children aged 0 to 18 with a life-threatening conditions and life-limiting conditions; at least 75% of the study population should be aged 0 to 18 years, (3) paediatric palliative care interventions related to (a) treatment of anxiety and depression, delirium, dyspnoea, haematological symptoms, coughing, skin complaints, nausea and vomiting, pain, neurological symptoms and fatigue, (b) treatment of refractory symptoms, c) advance care planning and shared decision-making, (d) organisation of care, (e) psychosocial care, and (f) loss and bereavement. Only studies published in English or Dutch language were included. Studies that described interventions on complementary or alternative medicine were excluded (Appendix F).
We searched for eligible studies in reference lists of included studies and identified SRs, guidelines, and textbooks. Moreover, we asked WG members to provide eligible studies.
Study selection
The studies were selected through two rounds of screening, title/abstract screening, and full text screening. One independent reviewer screened the titles and/or abstracts of all citations. The full text screening was performed by one independent reviewer. In case of doubt, the citations were discussed in the core group and were included only if there was consensus. The selected citations were distributed among the WGs. When citations were relevant for multiple WGs, they were included in all relevant WGs.
Summary and appraisal of evidence
All studies were summarized in evidence tables. Evidence tables described study characteristics (study type, setting, duration and years), participant characteristics (number and diagnosis of participants, age, and sex), intervention and control characteristics, outcomes and results, and strengths, limitations, and study quality.
We determined individual study quality by assessing risk of bias according to the criteria of Cochrane Risk of Bias tool (10). This tool assesses risk of selection bias, attrition bias, detection bias and performance bias of each study. Risk of bias can be classified as either low, high, or unclear.
We categorized evidence by outcome measures in summary of findings tables for every clinical question. We then formulated conclusions of evidence for each outcome measure. We assessed the quality of the total body of evidence with the Grading Recommendation Assessment Development and Evaluation (GRADE) criteria (11). The GRADE appraisal was performed by two reviewers. Quality of evidence was downgraded if study limitations, inconsistency, indirectness, imprecision, or publication bias were present. Quality of evidence was upgraded if a dose response effect or large magnitude of effect was identified.
Identification of evidence – qualitative studies
Systematic literature search
WG3 (advance care planning and shared decision-making) formulated one clinical question on barriers and facilitators of advance care planning and shared decision-making (appendix D). We performed a systematic literature search to identify qualitative studies on barriers and facilitators for advance care planning and shared decision-making. We updated the literature search that was conducted in the guideline ‘End of life care for infants, children and young people with life-limiting conditions (2016)’ of the National Institute for Health and Care Excellence (NICE) (12). We searched Medline (PubMed) from January 1, 2016 to September 16, 2020 using the search terms “child”, “palliative care”, “advance care planning”, “shared decision-making”, “qualitative study” (Appendix E).
The following inclusion criteria were defined: (1) qualitative studies, mixed-methods observational studies with qualitative data and SRs of qualitative studies, (2) study population consisting of children aged 0 to 18 years old with life-threatening or life-limiting conditions, their parents and health care providers, (3) study outcomes were barriers and facilitators on advance care planning or shared decision-making. Moreover, only studies published in English or Dutch language were included (Appendix F).
We asked WG members to provide eligible studies and searched for eligible studies in identified SRs and guidelines on barriers facilitator for advance care planning and shared decision-making.
Study selection
Both title/abstract screening and full text screening were performed by two independent reviewers. One reviewer performed title/abstract screening and full text screening for all identified citations. For the second review, citations were divided among eight WG members. In case of doubt, citations were discussed in the core group and included only if there was consensus.
Summary and appraisal of evidence
All studies were summarized in evidence tables. Evidence tables described study characteristics (study type, objective, setting, duration, and years), participant characteristics (number and diagnosis of participants, age, sex, ethnicity, religious preference, and level of education), outcomes and results, and strengths, limitations, and study quality.
We determined individual study quality by assessing the methodological limitations according to the criteria of Critical Appraisal Skills Programme (CASP) checklist tool (13). This tool assesses the aim and appropriateness of the qualitative study design, rigor in study design, sample selection, data collection, data analysis and results. Methodological limitations are classified as low, high, or unclear.
We assessed the quality the total body of evidence with the adapted GRADE Confidence in the Evidence from Reviews of Qualitative research (GRADE CERQual) methodology (14). The GRADE CERQual appraisal was performed by two reviewers. Quality of evidence was downgraded if methodological limitations were present or if there was a lack of coherence, relevance or data saturation (15). Quality of each conclusion of evidence was classified as high, moderate, low, or very low.
Identification of additional literature
As the expectation was that the systematic searches would yield little to no evidence, we searched for additional literature.
For all WGs, we searched for guidelines on paediatric palliative care, general paediatrics, and adult palliative care. To identify relevant (inter)national guidelines on paediatric palliative care, general paediatrics, and adult palliative care, we searched multiple databases. We searched the Guideline International Network database from 2010 to January 24, 2020, using the search terms “child” and “palliative care”, to identify guidelines on paediatric palliative care. Furthermore, we searched databases of the NICE, International Paediatric Oncology Guidelines in supportive care network (iPOG), the Dutch Association for Paediatrics, and Pallialine to identify guidelines on paediatric palliative care, general paediatrics, and adult palliative care. Guideline panel members were also asked to supply additional guidelines (Appendix E).
For the selection of guidelines, our first choice was to include (inter)national guidelines on paediatric palliative care. If guidelines on paediatric palliative care were not available, we included guidelines on general paediatrics if deemed relevant such as for topics related to (refractory) symptom treatment. We only included guidelines on adult palliative care, if no (relevant) guidelines on paediatric palliative care or guidelines on general paediatrics were identified. Selected guidelines were used to refine considerations and recommendations (Appendix F).
Moreover, we included two textbooks on paediatric palliative care as additional literature. The textbooks were considered relevant for WG1 (symptom treatment) and WG2 (refractory symptom treatment). Textbooks were used to refine considerations and recommendations.
Lastly, we derived SRs of observational, qualitative, or mixed-method studies from the systematic literature search on paediatric palliative care and through referencing. The inclusion of these SRs was only considered relevant for WG6 (loss and bereavement). We summarized the results of the SRs in evidence tables and translated these into conclusions of evidence. As the results of SRs included multiple studies from multiple study designs, we were not able to determine individual study quality nor the quality of the total body of evidence. The formulated conclusions were used to base recommendations upon.
Consensus-based approach
We found that not all included clinical questions in the revised CPG could be appropriately answered through an evidence-based approach. Clinical questions related to organisation of care (WG4) were considered as highly specific to the Dutch context. Therefore, an Ideafactory was organized. This is an interactive conference with a competitive element that is designed to find the best solutions for problems (formulated as questions). These solutions were used as the basis for formulating the recommendations. The methods and results will be presented in a subsequent manuscript.
Formulation of recommendations
When formulating recommendations, several factors were taken into account: (1) the quality of the evidence (the higher the quality, the more likely it is to formulate a strong recommendation), (2) additional literature, (3) patient perspectives (values and needs), (4) professional perspectives (clinical expertise, values and needs), (5) acceptability (legal and ethical considerations), (6) feasibility (sufficient time, knowledge and manpower) and (7) benefits versus harms of the interventions.
For each clinical question, WG members described the relevant considerations. Decisions were made through group consensus. The strength of each recommendation was graded according to published evidence-based methods (16, 17) (appendix G). Recommendations were categorised as strong to do (green), moderate to do (yellow) or strong not to do (red). A strong recommendation reflected a high degree of certainty. Moderate recommendations have a higher degree of uncertainty, therefore factors such as the clinical expertise, the patients and family’s situation and preferences, feasibility and relevant harms and benefits need to be considered.
All recommendations were supported unanimously by the core group, WG members, and patient and parent representatives