Objective Symptoms and problems (S&P) are under-reported in children in end-of-life care.
To target future interventions, the primary aim was to examine S&P in children in end-of-life care.
Methods All parents, who lost a child under the age of 18 years due to life-limiting diagnoses in the period 2012–2014 in Denmark, were invited to complete a self-administered questionnaire in 2017. In all, 152 (38%) children were represented by 136 mothers and 57 fathers. In the present study, parents’ assessments of S&P during the last month of life were restricted to children aged 3–18 years. Data were analyses by means of descriptive statistics.
Results Children ≥3 years at the time of death were represented by 71 parents (48 mothers and 23 fathers) representing 56 out of the 152 children. Physical fatigue (93%), sleepiness (90%), poor appetite (87%), pain (84%) and nausea (84%) were the five most frequent symptoms reported by the parents. In all, 65% of the parents reported that satisfactory pain relief was obtained and 64% of the parents reported that the healthcare services to a large extent reacted quickly, when the child and/or family needed help. However, 46% of the parents experienced ‘mess-ups’ or sloppy services in the primary ward and 27% experienced that the children suffered from fear of death.
Conclusion According to the parents, children with life-limiting diagnosis are highly symptomatic and have substantial problems during end-of-life care. Our findings indicate that systematic screening of S&P in children should be considered.
- symptoms and symptom management
- terminal care
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
What was already known
Children and adolescents frequently may experience complex symptoms and problems during end-of-life care. However, symptom assessment and management have primarily been investigated in children with cancer or specific and rare diagnoses and not in a national representative population of children with a variety of life-limiting diagnoses.
What are the findings
The results indicate high prevalence of distressing symptom and problems in children ≥3 years of age with a variety of life-limiting diagnoses during the last month of life. Psychological distress also seems to have high impact on suffering in the children and prominent distressing physical symptoms may worsen the dying process.
What is their significance
Systematic symptom assessment, evidence-based treatment programs and organizing multidisciplinary team conferences should be considered even before end-of-life trajectories to achieve better symptom control in the children with life-limiting diagnoses. Likewise, the high prevalence of anxiety and fear of death indicates a potential need for psychological interventions.
There exists evidence that children and adolescents frequently may experience complex symptoms and problems (S&P) during cancer trajectories.1 A retrospective interview survey of 103 parents to children, who died from cancer, found that 89% of the parents felt that their child suffered severely from at least one symptom in the last month of life. The most frequent symptoms were pain, dyspnea and fatigue.2 A recent systematic review found that burdensome symptoms also affected adolescents in disease trajectories of both malignant and nonmalignant origin.3 Although, the children’s families are considering symptom control to be a high priority4 previous research has shown that healthcare professionals are not sufficiently trained in symptom control.5 However, symptom assessment and management has primarily been investigated in children with cancer or specific and rare diagnoses,6 7 and the evidence base for symptom assessment in children with a variety of life-limiting diagnosis in end-of-life care are generally sparse.
In Denmark, specialised paediatric palliative care (PPC) has been a national healthcare priority since 2015. This has included the establishment of both hospices and hospital-based teams. However, in general there seems to be a pronounced lack of systematic symptom assessment in children with life-limiting diagnosis. Therefore, to target future interventions in specialised PPC the primary aim of this study was to investigate the prevalence and intensity of S&P in children≥3 years with life-limiting diagnoses during the last month of life assessed by their parents. The secondary aim was to investigate the healthcare received during the end-of-life care.
Participants and settings
The study is a cross-sectional study with retrospective questions. In all, 951 children 0–18 years of age, who died in Denmark in the period 2012–2014, were identified from the Danish Register of Causes of Death.8 The classification of primary death causes was performed in accordance with the International Classification of Diseases, 10th Revision.9 10 A directory of life-limiting conditions elaborated in Wales was used to identify the children, who potentially could benefit from specialised PPC. The directory was obtained through diagnoses of children cared for in five children’s hospices and a tertiary specialist palliative care service in the UK.11 The diagnoses listed in the directory was matched with the Danish children’s diagnoses and 402 children were identified. The Danish Civil Registration System12 has through a unique civil registration number registered all individuals living in Denmark. Through the civil registration number, the parents of the 402 children were identified. For further details please see previous work.8 13 14 In the present study, parents-proxy for the children were approached and the study population was restricted to parents of children aged 3–18 years during the last month of life. The main reason for this restriction was that it was difficult for the parents to assess many of the S&P in infants and preverbal children. This was reflected in the high proportion of missing answers (ie, completely missing answers or do not know answers) for these items.
Assessment and data collection
A modified version of the questionnaire ‘To lose a child’ was used. It was developed and has been used in bereaved parents, who lost a child due to malignancies in Sweden.15 Sweden and Denmark are similar in language, culture and healthcare systems, which influenced our choice. The original questionnaire contained 129 questions and covered demographics, the parent’s perceptions on the child’s S&P during the last month of life, perceptions regarding the healthcare received and the parent’s well-being at present. The questionnaire was translated from Swedish into Danish according to a revised version of the European Organization for Research and Treatment of Cancer manual.16 As the questionnaire was originally developed for children with cancer the wording was changed in some of the questions and specific questions related to cancer were removed. Thus, our questionnaire included 122 questions. The questionnaire has formerly been face to face validated by our research group.15 For further study details please see previous work.8 13 14 Thus, this article is a part of a larger material and the data presented in the present study comprised the parents’ perceptions regarding S&P affecting the child’s well-being during the last month of life as well as the healthcare services received. The selected questions cover 19 different S&Ps of both physical and mental nature (figure 1). The following five-point scale was used to assess the over-all S&P intensity: not at all, to less extent, to some extent, to a large extent and do not know. In all, 21 questions describe the parent’s perspectives on the healthcare received during end-of-life care of their child. The questions cover symptom relief in the children and access to interdisciplinary healthcare professionals (figures 2–4). For all these questions, the same five-point scale as mentioned above was used. The parents were enrolled in the period April to June 2017.
Descriptive statistics were used and presented as percentages with 95% CIs calculated using the Wilson score method (adjusted to account for clustering within children). All analyses were performed in SAS V.9.4.
Characteristics of the sample
The parents of 402 children were invited to participate. Totally, 152 (38%) children were represented by 136 mothers and 57 fathers. Hereby, 193 parents of children under the age of 18 years completed the nation-wide questionnaire survey . In the present study, the study population was restricted to children≥3 years at the time of death This group was represented by 71/193 parents (48 mothers and 23 fathers) representing 56/152 children. In the group of parents to children ≥3 years 71% of the respondents were married, and 84% shared the custody of the child at the time of the child’s illness and all respondents were biological parents. Mean age of the parents at the time of the survey were 46.4 (SD 6.82) ranging from 29 to 66 years of age. Furthermore, 75% of the parents had tertiary education.
The three most prevalent main causes of death among children ≥3 years were neoplasms (52%), diseases of the nervous system (15%) and congenital malformations, deformations and chromosomal abnormalities (13%). The last 20% covers deaths caused by for example, metabolic diseases, neoplasms, heart and lung diseases.
S&P in the children
The most prevalent S&P (less to a large extent) reported by the parents were physical fatigue (93%), sleepiness in the daytime (90%), a poor appetite (87%), pain (84%) and nausea (84%) (figure 1).
Furthermore, the most prevalent S&P (to a large extent) were physical fatigue (70%), a poor appetite (55%), sleepiness in daytime (54%), reduced mobility (54%) and difficulty swallowing (42%).
Sixty-five per cent and 47% of the parents experienced that relief of pain and other physical symptoms, respectively, were achieved to a large extent in the children during the last month of life. However, 25% of the parents observed at least on one occasion that pain lasted ≥4 hours in the children despite treatment attempts by the staff (Data not shown). Further, anxiety and other mental symptoms were not or only to some extent reported to be relieved in 42% and 55% of the children, respectively (figure 2).
In all, 38% of the parents reported that the children received medications to treat anxiety or depression during end-of-life care. Twenty-seven per cent of the parents reported that the child felt fearful of death. According to the parents, 62% of the children could talk or communicate during end-of-life care (data not shown).
Healthcare services received
In all, 42% of the parents reported that the children did not have access to psychosocial support by a social worker or psychologist, 49% did not have access to play therapy and 25% did not have access to dietary advice. However, 64% of the parents reported that the healthcare services to a large extent reacted quickly, when the child and/or the family needed support. In all, adequate access to physicians and nurses were reported by 58% and 69%, respectively, and 89% felt that their child got a hospital bed when needed. Figure 3 shows whose who have no or only to some extent access to various healthcare services during the last month of life.
In total, 27% of the parents reported that the staff to a large extent took initiative to introduce supportive counselling and 79% reported that the staff were medically competent. However, 46% of the parents experienced “mess-ups” or sloppy services in the primary ward. To what extent the children had access to various healthcare services during the last month of life is shown in figure 4.
Our results indicate high prevalence of distressing S&Ps in children ≥3 years of age with a variety of life-limiting diagnoses during the last month of life. To our knowledge, there is limited knowledge describing symptoms across diagnoses in children with life-limiting diagnoses and the parent’s perspectives on the professional healthcare services received during end-of-life care. In our study, physical fatigue and sleepiness in daytime were some of the most pronounced S&Ps in the children reported by the parents. Previous research in children with cancer has demonstrated that fatigue was pervasive and distressing to both children and their parents. It disrupts the energy needed for healthy experiences during pivotal developmental periods and is among others associated with sleep disturbances,17 18 which may negatively influence quality of life.19 20 Our findings are based on parental report, which may not fully reflect the child’s experience, particularly for a symptom as subjective and multidimensional as fatigue. However, as our study was based on reporting from the child’s last month of life fatigue closer to death is known to be more prevalent and may in some cases even be considered to provide relief of suffering.21 In children with cancer both acute and chronic pain states have also been identified as highly prevalent and distressing.22 23 In our study, pain was reported by parents from less to a large extent in 84% of the children. The parents reported that nearly one-third of the children did not experience that pain relief was achieved to a large extent and one fourth of the children had experienced excruciating pain for 4 hours or longer, which could not be relieved by the staff. In a previous work based on the same questionnaire data we found that 11% of the parents did not realise or understood that their child was going to die until the event occurred, and 19% only realised that the child’s death was imminent a few hours before they died.14 If the parents had realised that the child’s death was imminent they might have been more proactive in ensuring that the symptoms were more thoroughly treated for example, higher doses of medications or had encouraged the healthcare professionals to be less cautious in increasing doses to achieve better symptom control. However, four out of five parents reported that they found the staff medically competent. In turn, the fact that nearly half of the parents experienced ‘mess-ups’ or sloppy services in the primary ward deserves further study.
A study from a children’s hospice in UK identified pain, spasms, seizures, vomiting and unidentified distress as symptoms, which caused the children most anxiety.24 Also, nausea, which was prevalent in our study (84%), has been demonstrated to induce anxiety, which in turn may cause low quality of life, in children and adolescents with cancer.25 26 In our study, anxiety was not or only to some extent reported to be relieved in 42% of the children and 38% of children received medications to treat anxiety or depression during end-of-life care. The prominent prevalence of anxiety may likely be associated with increased distress during the child’s end-of-life trajectory. Of course, also existential issues may cause anxiety as one-third of the children in our study feared death according to the parents’ reporting. Thus, psychological distress seems to have high impact on suffering in the children and the dying process may deteriorate under the influence of distressing physical symptoms.27 However, the children in this survey were not clinically diagnosed and are exclusively based on parent-proxy report; however; the high prevalence of anxiety and fear of death indicates a potential need for psychological interventions.
Discrepancy between self-report and parent-proxy report of children’s S&Ps and health-related quality of life has been repeatedly acknowledged in the literature as ‘the proxy problem’. To our knowledge, bereaved parents have not been studied as parent-proxies in previous studies. However, other studies in parents to children with other diseases, such as, epilepsy, Attention deficit hyperactivity disorder (ADHD) and diabetes, have indicated that parents can serve as valid and reliable proxies for children and adolescents assessing symptoms and health-related quality of life.28 29 However, a previous study found that parents reported quality of life lower than self-report in children with cerebral palsy.30 Although, studies have shown differences regarding agreement between children or adolescents and parents at the individual level, they have shown that parents may provide a valid substitute for the children, when larger samples are investigated.31 32 Thus, parent report may add valuable and accessible information concerning children, who are unable to self-report due to lack of reading, linguistic and cognitive skills needed to respond to symptom and health-related quality of life measures.33 In addition, parents can also be perceived as appropriate proxy respondents to provide information concerning the child’s illness and health-related quality of life. For children living with a chronic illness, parents are often the primary healthcare provider and decision-makers, thereby influencing the quality of care33 34 by being pertinently and continuously aware of their child’s physical and psychological well-being and needs.35 As such parents become responsible for the monitoring of symptoms, adjusting of medications and organising of healthcare interventions and are, therefore, viewed as suitable, reliable and useful proxy reporters.29 32–34 However, on the other hand, we must be aware that parents’ observations may be influenced by their own physical and emotional well-being and the parent and child dyadic relationship may limit the accuracy of parents as proxy reporters for interpreting the child’s behaviour. Parent and child dyadic relationship may become strained or altered as the child become more independent with age and/or when the severity and management of the illness changes.36 Parent variables such as health, education, culture, ethnicity and socioeconomic status may also affect the agreement between parent and child measures.
It is indeed a sensitive task to investigate a sample of bereaved parents and ethical concerns are prevailing. However, previous research have demonstrated that bereaved parents felt positively affected by participation after the loss of their child; both personally and for the opportunity to help others.15 37
Strengths and limitations
The data used for this study were acquired through national registers and a questionnaire survey. The strength of using registers is a unique opportunity to combine data on an individual level. In this study, the registers allowed us to combine the decreased children with their living parents and, thus, include the parents in the survey.38 Non-response surveys have been made formerly to identify reasons for lack of response. The results of non-response has been published in previous work.8 13 14
Moreover, it is known from previous research that females are more likely than men to participate in questionnaire surveys.39 40 This finding was also reflected in our survey, as we had an over-representation of mothers. However, the effect of clustering has been considered to allow for the likelihood that parents of the same child possibly responded similarly.
Finally, recall bias may influence reporting during a devastating and chaotic period 3–5 years ago in the parents’ life and fluctuations of S&Ps during different stages of the individual illness trajectories as well as cultural, geographical and socioeconomic backgrounds, could not be depicted with the present assessment tools. Likewise, this study did not take cognitive dysfunctions in the children into account but relied on the parent’s over all assessment.
Our study indicated that according to parents reporting, children≥3 years with life-limiting diagnosis are highly symptomatic and have substantial problems during the last month of life. The five most frequent symptoms in the children reported by the parents were physical fatigue, sleepiness, poor appetite, pain and nausea and one-fourth of the parents observed that pain in the children lasted for 4 hours or more. However, the parents experienced that relief of pain and other physical symptoms were achieved to a large extent in their children during end-of-life-care. In the process of developing specialised PPC in Denmark, our findings should be considered for implementing future systematic assessment and treatment programs. In addition, organizing multidisciplinary team conferences could be considered to achieve better symptom control in the children. Further, a group of healthcare professionals of varied disciplines and roles should work closely together in the teams towards a common goal for providing optimal treatment and care for the patient and the family targeting physical, social, psychological and spiritual dimensions of suffering. Finally, the multidisciplinary approach makes it possible to provide care through the many transitions, wishes and needs that are inherent in the trajectories of children with life-limiting diagnoses.
We thank all bereaved parents who generously shared their experiences. Without you, this study had not been possible.
Contributors CL conceptualised and designed the study, carried out the initial analyses, drafted the manuscript and revised the manuscript. PS, MO conceptualised and designed the study, supervised the study and critically reviewed the manuscript. Senior advisor, OE, conceptualised and supervised the study, carried out the analyses and critically reviewed the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Funding This study was funded by Helsefonden (Grant number 2016-6).
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval The study protocol was approved by the Danish Ethics Committee (H-16021831) and the Danish Data Protection Agency (2012-58-0004).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information. Any requests for data sharing should be put in writing to the corresponding author.