Background Children with progressive, non-curable genetic, metabolic, or neurological conditions require specialised care to enhance their quality of life. Prevention and relief of physical symptoms for these children needs to begin at diagnosis, yet, little is known about their patterns of symptoms and functional abilities.
Aim To describe these children's symptoms, as well as how the children's condition affects them physically.
Design Cross-sectional, baseline results from an observational, longitudinal study, Charting the Territory, that followed 275 children and their families.
Setting/participants Seven tertiary care children's hospitals in Canada, 2 in the USA. Families were eligible based on the child's condition. A total of 275 children from 258 families participated.
Results The 3 most common symptoms in these children were pain, sleep problems, and feeding difficulties; on average, they had 3.2 symptoms of concern. There was a pattern of under-reporting of children's symptoms for clinicians compared with parents. Regardless of use of associated medications, pain, feeding and constipation symptoms were often frequent and distressing. Children with a G/J tube had a higher total number of symptoms, and respiratory problems, pain, feeding difficulties and constipation were more likely to occur. They also tended to have frequent and distressing symptoms, and to need extensive mobility modifications which, in turn, were associated with higher numbers of symptoms.
Conclusions These children experience multiple symptoms that have been previously documented individually, but not collectively. Effective interventions are needed to reduce their symptom burden. Future longitudinal analyses will examine which disease-modifying interventions improve, or do not improve, symptom burden.
- Palliative Care
Statistics from Altmetric.com
What is already known on this topic?
Progressive, non-curable genetic, metabolic, or neurological conditions involve impairment/injury of the central nervous system.
Children often have multiple symptoms associated with such impairment, for example, pain, seizures and loss of mobility, and need appropriate interventions.
These children often rely on enteral nutrition, for example, via G/J tube, and require mobility modifications.
What this study adds?
The three most common symptoms in these children are pain, sleep problems and feeding difficulties; on average, they have three symptoms of concern.
Regardless of medication use, seizure, pain, feeding and constipation issues may be frequent and distressing to these children.
Children with G/J tubes, lower levels of function, or extensive mobility modification requirements have a higher total number of symptoms.
Children with progressive genetic, metabolic, or neurological conditions for which there is no cure (PNC) and their families require specialised care to enhance their quality of life. In Canada, about half the annual 2500 childhood non-traumatic deaths occur because of PNCs.1 ,2 These diagnoses account for more than 50% of children who receive paediatric palliative care in North America.1 ,3 ,4 Though individual conditions are rare, they have been broadly categorised within a single group because of their unifying problem: they involve impairment/injury of the central nervous system and are progressive.5 ,6 These children typically display problems seen with such impairment, for example, pain, seizures, and loss of mobility,7 therefore, prevention and relief of physical symptoms needs to begin at diagnosis. Yet, little is known about the patterns of symptoms or possible correlations with functional abilities.
Research in pain and symptom management is a priority in paediatric palliative care,8 and families of children with PNCs report symptoms and issues around medications as two important milestones in a child's illness trajectory.6 Common symptoms reported by clinicians include: pain, breathing problems, feeding difficulties, alertness/interaction changes, sleep problems, seizures and constipation.3 ,7 ,9–22 There is a need to research children's symptoms, but information about symptoms is not always documented even in patient records,4 and the co-occurrence of symptoms is rarely explored. Children may be prescribed multiple medications,3 ,20–24 but little is published about the specific types of medications commonly used by clinicians, and whether or not the prescribed medications alleviate symptoms.
These children usually experience developmental delays and functional losses that increase over time.6 ,9 ,25 There is some evidence that pain behaviour may be different for children with neurodegenerative conditions,25 but the literature is inconsistent regarding whether functional levels affect pain expression.26 ,27 The relationships between functional levels and other symptoms remain unclear, yet loss of mobility and needing modifications/equipment are significant milestones in this population.6
Many children with PNCs receive enteral feeds, for example, via gastrostomy or jejunal (G/J) tubes.6 ,24 Improved nutrition can lead to improved quality of life for the child and family,28 but G/J feeding can result in complications such as aspiration pneumonia, site infection and reflux.22 ,28 Consequently, the children may have related symptoms, for example, pain, feeding or respiratory problems, but little is known about correlations between G/J feeds and symptoms. Further, clinical experience suggests that children with G/J tubes often have restricted mobility, which may contribute to increased symptoms. Exploration of this anecdotal evidence is warranted.
Overall, there is a paucity of research on which to base best practices for care of children with PNCs.3 ,29 Much of the fairly limited literature focuses on specific conditions.9–15 ,30 ,31 Discussion is often about the treatment of one symptom such as seizures.9 ,16 A majority of the research is retrospective13 ,14 ,30 and often the sample sizes are small.9 ,10 ,15 Though some information has been published, there is little to describe the natural progression of this group as a whole, particularly from a prospective approach. Little information exists about differences in symptoms by physical functioning, for example, by the need for mobility modifications or presence/absence of a G/J tube.
This paper reports baseline results about 275 child participants (from 258 families) in a multisite, longitudinal study, Charting the Territory, that followed children with these conditions, and their families. The focus is on symptoms reported, and differences in selected symptoms by use of medications, G/J tube presence, and/or level of mobility modifications.
The larger study was developed to determine the child's clinical trajectory including symptoms, and the associated emotional, social, physical and spiritual experiences of the family over time. It used quantitative methods, with established instruments and record reviews. Children (0–19 years) and their families were followed for 18–48 months, depending on when they entered the study. Data collection began at baseline and continued through bereavement or until the end of study.1
Families were recruited from July 2009 till October 2012, mostly through referrals from hospital clinics (49%) and hospice/palliative teams/services (43%). Eligibility was based on the child's condition (see appendix A, online only). Of 385 families evaluated for eligibility, 93 did not meet the criteria or declined further contact, 34 could not be contacted, leaving 258 families.
To facilitate continuity and reduce the likelihood of inter-rater discrepancies, parents identified a ‘designated’ parent in their family to answer questions about the child. Designated parents completed baseline questionnaires in person, by mail, or a combination thereof and, subsequently, reported on their child's symptoms monthly via the Internet or, for a few, by telephone. Parents who chose the online method needed to have an existing email account and access to the internet. The software included an email component every month to prompt parents to visit the secure website and complete the symptom questionnaire. A research assistant (RA) called telephone responders to obtain answers.
After enrolment, the site RA collected baseline information regarding the child's diagnosis, clinical condition, medications and procedures undergone. Disease diagnoses were characterised according to a scheme developed in an iterative process by one of the authors (HS) and two study collaborators from Biochemical Diseases/Genetics and Paediatric Neurology.
Information about seven common symptoms in this population was collected at baseline and then monthly. Parents were given definitions for each symptom—pain, respiratory/breathing problems, feeding difficulties, alertness/interaction changes, sleep problems, seizures and constipation—prior to completing the baseline assessment (see appendix B, online only).
The instrument takes 6–10 min to complete and was designed by the research team in consultation with other experienced clinicians, families linked to a paediatric palliative care hospice programme, and a psychometrician. It is a modification of the PediQUEST symptom recording tool, and the revised Memorial Symptom Assessment Scale (7–12 years of age) developed for symptoms in children with cancer.32 Symptoms are tracked for timing of onset (symptom latency15), frequency in the last week, change since first appeared (or since previous month), and extent of distress. Face and content validity were evaluated prior to the start of the study with the assistance of parents of children who received palliative care, and the instrument was pilot tested in clinical practice (HS).
Annual functional assessment: paediatric evaluation of disability inventory (PEDI)
The PEDI33 ,34 was used to assess the children's function at baseline, then annually. It yields information under Parts I (Functional Skills), II (Caregiver Assistance), and III (Modifications) in the domains of self-care, mobility and social function. Scores are summarised into composite scores. Raw scores from Parts I and II can be converted to normative standard scores as well as scaled scores for comparison against norm-age values. Raw scores are reported for this study because children with PNCs tend to have significant disabilities; therefore, it was deemed unreasonable to compare them to ‘norms’. Appendix C (online only) provides details about the reliability and validity of the PEDI, procedures for administering it, and steps taken to ensure RAs were competent in using the instrument.
Data were entered into a database for preprocessing, data cleaning, and determination of scale composites according to the appropriate procedures for the standardised measures. Missing data for outcome measures were linearly imputed, if possible, according to standard procedures for the specific tool. Analyses were performed using the Statistical Package for the Social Sciences (SPSS), V.20, with statistical significance at p<0.05.
Demographics and outcome measures were summarised using descriptive statistics: frequencies and percentages for categorical variables, means and SD for continuous variables. Two-tailed comparison tests included: χ², two-sample t test, Spearman's rs and Pearson r correlations, 1-way and 2-way ANOVA.
Ethics approval for this binational (Canada/USA) study was obtained from nine study settings plus the four universities where the researchers were affiliated. The University of British Columbia and Children's & Women's Health Centre of BC (Certification #H08–00124) approved the central site in Vancouver. On behalf of themselves and their child(ren), parent participants signed a consent form; the ill children were unable to sign consent/assent forms due to their health conditions.
Few meaningful differences were detected by study site across all analyses, therefore, results are reported for the whole sample. On average, the 258 families had 2.5 (SD 1.4) living children, range 1–10; 283/574 of the total number of children had a PNC, with two such children enrolled from 17 families. Prior to baseline, a total of 20 children in 16 families had died, 50% from a PNC; of the remainder, five were miscarriages. Most families (n=169, 65.5%) received some form of palliative care.
Families had been engaged with the medical system for over 6 years on average (table 1). They typically waited almost a year between bringing the child to a physician and receiving a diagnosis; though many children still did not have a diagnosis (n=48, 17.5%). About 30% of parents learned prenatally (n=33) or perinatally (n=51) that the child had a problem. Others initiated investigation because of concerns such as the child not meeting developmental milestones (n=54, 19.6%). The child's average age at which parents began the diagnostic process was 12.1 months (SD 25.5). Clinicians typically employed multiple methods for confirming diagnoses.
Records indicated that all children underwent at least one surgery or interventional radiology procedure (table 2). Most had multiple assessments and many took several types of medications (mean 3.5, SD 2.3). Sixty-one percent (n=168) received nourishment other than orally, and 13.1% (n=36) used a ventilator. Many children (n=125, 45.5%) had an abnormal EEG.
As detailed in table 3, symptoms were common, with about half the children experiencing any of the symptoms except for difficulties with alertness. Parents reported an average of 3.2 (SD 1.9) symptoms at baseline, whereas clinicians documented fewer (mean 2, SD 1.7). Correlations between parental and clinician reports of symptoms were positive, and weak to moderately strong; all were statistically significant (p=<0.001).
Parents reported frequency and change for symptoms, as well as the distress they ascribed to the child (table 4). Only two symptoms showed statistically significant differences by conditions: change in seizures (χ2=35.79, p =0.02), and distress from feeding difficulties (χ2=38.26, p=0.03).
The only statistically significant difference between 4 selected symptoms and their associated medication use (table 5) was frequency for seizures (χ2=11.71, p=<0.01), though distress was present for the majority who were receiving seizure medications. Regardless of whether medications were used, parents also reported children experiencing and being distressed by pain, feeding difficulties and constipation.
Children with G/J tubes had more symptoms (mean 3.67, SD 1.74) than children without (mean 2.68, SD 2; p=<0.001). Differences were seen in breathing, pain, feeding and constipation symptoms depending on G/J tube presence (table 6). Despite mostly statistically insignificant results, G/J tube presence was generally consistent with higher levels of symptom frequency and distress (table 7).
Table 8 shows that average functional levels were all on the low end of the respective subscales, indicating that children could provide little self-care (mean 14.7, SD 19.6), had difficulty in getting around (mean 14.6, SD 19.2), and were limited in social function (mean 16.9, SD 17.8). All functional and caregiver assistance subscales showed statistically significant differences by conditions (p=≤0.001).
As shown in table 9, children who scored lower on functional skills had a higher total number of symptoms; children with G/J tubes were more likely to require extensive mobility modifications; and G/J tube presence and level of mobility modification were associated with higher numbers of symptoms, though their interaction was not statistically significant.
Results from this study are supported by the literature in a number of ways, including the plethora of medications given to these children.3 ,16 ,17 One of the most striking findings is the children's large symptom experience. The children had, on average, just over 3 symptoms of concern, and almost every symptom appeared in at least half the children. This finding in children with metabolic, chromosomal, or neurological conditions parallels findings about symptom burden in children with cancer.35 ,36
Symptom management is a hallmark of paediatric palliative care, and it is not surprising that many of the children were being treated for multiple symptoms. However, symptoms were often not well controlled, and children frequently experienced and were distressed by symptoms, whether or not they were receiving associated medications. Where children were receiving an indicated medication, the treatment was often not effective. These findings highlight the difficulties in managing symptoms, especially seizures, in this population, and suggest that further research is needed to identify optimal management.
Though there were significant correlations between parent and clinician reports of children's symptoms, there was a pattern of under-reporting for clinicians compared with parents, a finding that is consistent with some paediatric palliative care literature.7 ,21 ,37 ,38 There are many potential reasons for this finding, including missing documentation.4 One reason may be that neurological symptoms especially are difficult to deal with in this population.23 Another is that clinicians sometimes report only those symptoms that they feel confident in treating7 or they do not assess for a symptom because of the child's impaired cognition,21 and, finally, sometimes clinicians disbelieve parents, especially if a child is non-verbal.38 Clinicians whose approach is to question why they should not believe a parent rather than why they should, and who view symptoms as multidimensional experiences that have meaning for patients and their families rather than simply as side effects, may be less likely to under-report.7 ,38 Also, many anxieties/difficulties accompany parents dealing with PNCs.6 If a symptom causes parental concern then parents may report it more frequently.39 Therefore, clinicians should ask parents why a symptom is of concern, so they can address the parents’ needs. From a clinical perspective, it is equally important to be cognisant of the frequency and the severity of symptoms as each may lead to a different understanding of the nature of a symptom, and, therefore, may contribute to identifying the most appropriate intervention.
The increased pattern of symptom burden when a G/J tube is present and/or when there are extensive mobility modifications indicates that clinicians need to be alert to assessing and managing symptoms in these children. We do not know whether feeding devices or mobility tools increase symptoms directly, but both interventions are indicative of children with more severe conditions and, consequently, more symptom burden. Therefore, extra attention to symptoms in children with feeding devices and/or mobility modifications is warranted.
A main limitation of this report is its cross-sectional, primarily descriptive nature, but these are only the first results from a longitudinal study. Future analyses of data, for example examining correlations among variables and over time, will provide a better understanding of changing outcomes as the child's illness trajectory unfolds. Another limitation is that parents reported on behalf of their child. However, this population of children with PNCs is typically non-communicative, and parents usually speak for their child in everyday situations. Therefore, it may be reasonable to expect parents rather than these children to provide information.
A limitation of the PEDI in a population with a high level of neurological impairment is that it does not indicate other functional limitations that can impact health, such as respiratory decline and earlier death correlating with a child's inability to hold his head up when prone. However, an important point to note is that a functional assessment can highlight a child's deterioration over time, or it can point out when an intervention may be warranted. Routine evaluation of functioning over time would document a condition's trajectory, and would also identify places for clinical intervention, as well as provide comparative data about an individual child over time.
Paediatric palliative care is an expanding field, yet the paucity of research means that clinicians have little evidence on which to base their practice. This baseline report from a unique longitudinal study offers detailed information about symptoms in children with progressive, non-curable genetic, metabolic, or neurological conditions. The large sample size lends credence to findings and provides a solid foundation for understanding trajectories over time. Future longitudinal analyses will examine which disease-modifying interventions improve, or do not improve, symptom burden.
The authors would like to thank all the children and families who participated in the study. We also thank the research assistants in each study setting for their contribution to data collection. We appreciate that the terminology is not settled regarding the most appropriate term to use when reporting on children with the types of conditions in this study. Various terms, such as life-threatening, life-limiting, life-shortening, and complex medical problems, have all been proposed and often one is preferred over another depending on one's country. For the purposes of this paper, we have used the term ‘progressive, non-curable conditions’ because it reduces ambiguity and is acceptable to this journal.
Review history and Supplementary material
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
Collaborators We thank our site lead clinician collaborators who enabled access to families and acted as lead investigators in the various settings: Sharron Spicer – Calgary, AB, Canada (Alberta Children's Hospital); Helly Goez/Dawn Davies – Edmonton, AB, Canada (Stollery Children's Hospital); Adam Rapoport – Toronto, ON, Canada (Hospital for Sick Children); Christina Vadeboncoeur – Ottawa, ON, Canada (Children's Hospital of Eastern Ontario); Stephen Liben – Montreal, QC, Canada (Montreal Children's Hospital); Marie-Claude Gregoire – Halifax, NS, Canada (IWK Health Centre); Scott Schwantes – St Paul's, MN, USA (Gillette's Children's Specialty Healthcare); Stefan J Friedrichsdorf – Minneapolis, MN, USA (Children's Hospitals and Clinics of Minnesota).
Contributors GA: conceptualised and designed the study, analysed and interpreted the data, revised the resubmitted manuscript, and approved the resubmitted manuscript. SC: conceptualised and designed the study, revised the resubmitted manuscript, and approved the resubmitted manuscript. BD: conceptualised and designed the study, revised the resubmitted manuscript, and approved the resubmitted manuscript. LF: Feichtinger analysed and interpreted the data, revised the resubmitted manuscript, and approved the resubmitted manuscript. HS: conceptualised and designed the study, analysed and interpreted the data, revised the resubmitted manuscript, and approved the resubmitted manuscript. MS: analysed and interpreted the data, revised the resubmitted manuscript, and approved the resubmitted manuscript. RS: conceptualised and designed the study, analysed and interpreted the data, drafted and revised the resubmitted manuscript, and approved the resubmitted manuscript.
Funding External funding was secured from the Canadian Institutes of Health Research, Canada; MOP 89984.
Competing interests None.
Ethics approval University of British Columbia and Children's & Women's Health Centre of BC (Study Number H08-00124).
Provenance and peer review Not commissioned; externally peer reviewed.
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.