Visual Abstract
Keywords: CKD, patient self-assessment, pediatrics, quality of life
Abstract
Key Points
PRO-Kid is a patient-reported outcome measure of the frequency and burden of symptoms.
Higher PRO-Kid scores are associated with lower Pediatric Quality of Life Inventory scores.
Background
Measuring the burden of symptoms that matter most to children and adolescents with CKD is essential for optimizing patient-centered care. We developed a novel CKD-specific patient-reported outcome measure (PRO-Kid) to assess both frequency and impact of symptoms in children. In this study, we further assessed the validity and internal consistency of PRO-Kid.
Methods
In this multicenter study, children age 8–18 years with stages 3–5 CKD, including those on dialysis, were recruited from five pediatric centers. Children completed the 14-item PRO-Kid questionnaire and the validated Pediatric Quality of Life Inventory (PedsQL 4.0). We explored the dimensionality of the PRO-kid scale using exploratory and confirmatory factor analysis, to either establish that it is a unidimensional construct or identify evidence of subfactors. We then assessed internal consistency (Cronbach alpha) and construct validity (Pearson correlations).
Results
In total, 100 children were included. The median eGFR was 27.4 ml/min per 1.73 m2 (7.43–63.4), and 26 children (26%) were on dialysis. Both the PRO-Kid frequency and the impact scales were unidimensional. Cronbach alpha was high for both the PRO-Kid frequency and impact scales, 0.83 (95% confidence interval [CI], 0.78 to 0.88) and 0.84 (95% CI, 0.80 to 0.89), respectively, showing strong internal consistency. Pearson correlations between PRO-Kid and PedsQL scores were also strong: −0.78 (95% CI, −0.85 to −0.70) for the frequency score and −0.69 (95% CI, −0.78 to −0.56) for the impact score, reflecting the association between poorer quality of life and higher symptom burden.
Conclusions
PRO-Kid is a novel patient-reported symptom burden tool for children age 8–18 years with CKD that correlates strongly in the expected direction with PedsQL, supporting its validity. Future work will evaluate changes in PRO-Kid score with progression of CKD and implementation of the tool into clinical care.
Introduction
The quality of life of school-age children and adolescents with CKD and kidney failure is affected by many symptoms such as fatigue, nausea, anorexia, social isolation, impaired cognitive function, delayed growth, depression, and low self-esteem.1–5 In a recent multinational study, children with CKD prioritized most highly the outcomes that directly affected their lifestyle and sense of normality (e.g., lifestyle restrictions, growth, physical activity) or which reflected immediate health concerns (e.g., survival).6 Outcomes that are most relevant to young patients with CKD are often omitted from clinical trials that evaluate interventions to improve care in this population,7 at least partly because of the absence of validated tools that quantify these patient-reported outcomes. The physical, psychological, and emotional symptoms that affect the quality of life of children with CKD are unique, relative to other chronic illnesses. Therefore, it is likely that a tool developed and validated within the pediatric CKD population would be more relevant when assessing patient-reported outcomes versus imputing from or adapting generic measurement tools developed in different patient populations. In the clinical setting, implementation of routine patient-reported symptoms measures of children with CKD has the potential to improve communication between patients and providers, resulting in opportunities to reduce symptom burden and improve shared decision making and quality of care.
Health care providers commonly focus on quantitative laboratory, imaging, or histopathological measures of kidney health to assess disease progression and determine treatment options. Trial data in adults suggest that life-altering treatment decisions, such as timing of initiation of dialysis, should be guided by severity of symptoms rather than laboratory testing (e.g., eGFR) alone.8,9 Assessing symptom severity and impact of quality of life and functional status remains a great challenge in children with CKD/kidney failure.
Together with patient partners of the Canadians Seeking Solutions and Innovations to Overcome CKD Strategic Patient Oriented Research network, we reviewed the content and structure of existing relevant patient-reported outcome measures (PROMs) and found the tools to either lack the disease specificity needed to capture the symptoms of CKD (e.g., Symptom Screening in Pediatrics Tool, designed for and used in pediatric oncology) or lack the population specificity needed for children (e.g., Edmonton Symptom Assessment System-Revised: Renal, for adults with kidney failure). Given this identified gap and the need for a disease-specific symptom burden assessment tool for children with CKD/kidney failure, we developed a novel CKD-specific patient-reported outcome measure (PRO-Kid).10 This previous work included cognitive interviews of children and adolescents with CKD and parents to identify symptoms that are relevant to youth with CKD and kidney failure and determined the most optimal way to inquire about them while ensuring it is delivered at the appropriate level of understanding. The result was a PROM for the assessment of frequency and impact of CKD/kidney failure symptoms, in the three domains of physical, socio-emotional, and cognitive symptoms: the 14-item PRO-Kid questionnaire.10 The objective of this study was to externally validate PRO-Kid in a Canadian cohort of children with CKD/kidney failure. We hypothesized that higher PRO-Kid scores would be associated with lower Pediatric Quality of Life Inventory (PedsQL) scores. We also hypothesized that there would be two domains to the PRO-Kid tool including physical and emotional symptoms.
Methods
Study Design and Population
This was a multicenter, cross-sectional study of children age 8–18 years with stages 3–5 CKD, including children on peritoneal or hemodialysis. Children were recruited from outpatient clinics and dialysis units from five sites with a Pediatric Nephrology program across Canada (BC Children's Hospital in Vancouver, Alberta Children's Hospital in Calgary, Health Sciences Center Children's Hospital in Winnipeg, McMaster Children's Hospital in Hamilton and the Hospital for Sick Children in Toronto). Children age 8–18 years were selected because they are able to complete self-report questionnaires.11 Patients were excluded if they had an intellectual disability that affected their ability to complete the questionnaire independently, were unable to speak or understand English, or had undergone kidney transplantation. A convenience sample was recruited between June 2021 and February 2023.
This study was approved by the Research Ethics Boards of all participating sites and was conducted in accordance with the Declaration of Helsinki of 1975, as revised in 2013. Informed consent (and assent, as appropriate) was obtained for all study participants before initiating study activities.
Questionnaire Administration
Patients were approached during regularly scheduled clinic visits or dialysis treatments. When possible, the PRO-Kid and PedsQL questionnaires were completed during this clinical encounter. However, patients were also given the option to complete the questionnaires at home. All questionnaires were completed electronically using the survey function of the Research Electronic Data Capture (REDCap12) database.
Clinical Data Collection
Demographic information was collected via self-report or caregiver report. Demographic information included sex, age, self-declared ethnicity, school grade level, location in a rural versus urban setting, and distance from home to treatment center. Clinical information was obtained from the patient's electronic medical record or chart and included primary kidney disease, need for intermittent bladder catheterizations, presence of gastrostomy tube, eGFR, CKD stage based on CKD in children study eGFR,13 and dialysis modality. The bedside CKD in children study equation was used to estimate GFR, which was the clinical standard at the time the study started.13 All data were entered into a password-protected REDCap12 database housed at the Hospital for Sick Children in Toronto.
PRO-Kid and PedsQL
Details of the development of the PRO-Kid questionnaire have been previously described.10 In brief, PRO-Kid is designed as an easy to complete 14-item questionnaire that represents the spectrum of physical, cognitive, and emotional symptoms endorsed most frequently by children and adolescents age 8–18 years with CKD/kidney failure. Each item on the PRO-Kid questionnaire is rated on a five-point scale for the dimensions of frequency and impact (not at all=0, a little=1, medium=2, a lot=3 and always=4). The recall period is 7 days.
The 23-item PedsQL 4.0 Core Module includes: physical functioning (eight items), emotional functioning (five items), social functioning (five items), and school functioning (five items).14 For ease of interpretability, scores are linearly transformed to a 0–100 scale, where higher scores indicate better health-related quality of life.
PRO-Kid Scoring
A methodology analogous to the PedsQL 4.0 scoring was used. Each item was transformed as follows: 0=0, 1=25, 2=50, 3=75, and 4=100. Scores were summed and divided over the number of questions answered to yield a component score for each of frequency and impact. Higher scores indicate higher symptom burden.
Statistical Analysis
A priori, the sample size required for validation of PRO-Kid was determined as at least five participants per variable on the questionnaire (which would be N=70), with a minimum sample size of 100 participants on the basis of the guidelines for factor analysis.15 Patient and caregiver characteristics were summarized as mean±SD for continuous variables or frequencies and percentages for categorical variables. A total PRO-Kid score for frequency and impact was calculated for each participant as above.
Psychometric Analysis
We first explored the dimensionality of the PRO-Kid frequency scale. Our goal was to either establish that it is a unidimensional construct or find evidence of subfactors. The dimensionality of a scale indicates whether it is a single construct (unidimensional) that can be treated as a single score or whether it comprises distinct subscales (multidimensional), which may or may not be strongly related to each other. Dimensionality can be estimated by examining patterns in the item correlations. To this end, we used exploratory, followed by confirmatory factor analysis (CFA). To estimate the number of latent dimensions we used parallel analysis, which compares the magnitude of the observed correlation matrix eigenvalues against those from a simulated random data set with the same number of variables and observations. If an observed eigenvalue is larger than the corresponding eigenvalue from the simulated data, it is deemed to be real. This is similar to a scree plot analysis but with the simulated eigenvalues overlaid as a null reference.
We further investigated the measurement properties of the PRO-Kid frequency scale with a CFA model, with its dimensionality determined by the above exploratory analyses. The CFA model was used to estimate statistics of model fit (Comparative Fit Index [CFI]; higher=better model fit; Root Mean Square Error of Approximation [RMSEA]; lower=better model fit]; Standardized Root Mean Square Residual [SRMR]: lower=better model fit) standardized item loadings, and modification indices. The magnitude and direction of item loadings were inspected to assess their relevance to the construct and direction of effect. Modification indices were used to add residual correlations between items until satisfactory model fit was attained, as necessary. Thus, our confirmatory model was also exploratory in nature. An independent sample would be required for unbiased estimates of CFA fit statistics. The exact same steps were repeated for PRO-Kid impact scale.
The internal consistency of the PRO-Kid frequency and impact items was evaluated using the Cronbach's alpha (Cα), which is a measure that compares the average covariance between items relative to the average variance. If items relate to the same underlying construct, they should be highly correlated, termed consistent. Thus, higher values of Cα correspond to better consistency; α values >0.9 are considered to indicate excellent, 0.80–0.89 good, and 0.70–0.89 acceptable internal consistency.16 We examined both the overall α statistic and for each item individually, the α statistic that would result if the item were deleted. The latter is a measure of each item's relevance to the construct.
Construct Validity
Construct validity was assessed using convergent validity and was determined by Pearson correlations between the PRO-Kid and the Core Module of the PedsQL4.0, R >0.7 would be considered strong, 0.5–0.7 moderate, and <0.5 weak.17 We considered moderate or strong correlation as an indicator of acceptable convergent validity. We hypothesized that PRO-Kid would have a moderate or strong inverse correlation with PedsQL scores (i.e., more endorsed symptoms would be associated with lower quality of life and vice versa.) We also compared the average and median values of the PRO-Kid total score between CKD stages (3, 4, 5 predialysis and on dialysis). Differences between groups were evaluated using one-way ANOVA.
Analyses were conducted using R version 4.0.5. Psychometric properties of the PRO-Kid instrument were evaluated using the psych18 and lavaan packages.19
Results
A total of 100 children completed the study; 138 potentially eligible participants had been approached, and of those who provided consent, 38 did not complete the study. Reasons for not completing the study included received a kidney transplant before completion of questionnaires (therefore, no longer eligible) or loss of interest in the study. Demographic data on all participating children are summarized in Table 1. The mean age of the participants was 13.7 (SD, 3.14) years, and 63% were male. Congenital anomalies of the kidney and urinary tract were the most common etiology of CKD in our cohort (45%). The median eGFR was 27.4 ml/min per 1.73 m2 (7.43–63.4), and 26 children (26%) were on dialysis. Of those on dialysis, 15 (58%) were on peritoneal dialysis.
Table 1.
Study participant characteristics
| Characteristic | Total, N=100 |
|---|---|
| Age, yr | 13.7±3.14 |
| Male, No. (%) | 63 (63) |
| Ethnicity, No. (%) | |
| Black | 4 (4) |
| Indigenous | 9 (9) |
| Missing | 4 (4) |
| Other | 17 (17) |
| South/East Asian | 21 (21) |
| White | 45 (45) |
| Etiology of CKD, No. (%) | |
| CAKUT | 45 (45) |
| Autoimmune | 18 (18) |
| Genetic | 17 (17) |
| Diabetes | 1 (1) |
| Other | 19 (19) |
| eGFR, ml/min per 0.73 m2 | 27.4 (7.43–63.4) |
| CKD stage, No. (%) | |
| 3 | 32 (32) |
| 4 | 29 (29) |
| 5, predialysis | 13 (13) |
| 5, dialysis-dependent | 26 (26) |
| Dialysis modality, No. (%) | |
| Peritoneal | 15 (15) |
| Hemodialysis | 11 (11) |
| Median time on dialysis, yr | 0.7 (0.05–6.85) |
| Urinary catheterizations, No. (%) | 17 (17) |
| Gastrostomy tube in situ, No. (%) | 12 (12) |
| Study site, No. (%) | |
| The Hospital for Sick Children, Toronto | 45 (45) |
| BC Children's Hospital, Vancouver | 23 (23) |
| Health Sciences Center, Children's Hospital of Winnipeg | 21 (21) |
| McMaster Children's Hospital, Hamilton | 8 (8) |
| Alberta Children's Hospital, Calgary | 3 (3) |
| Distance of home from pediatric nephrology center, km, No. (%) | |
| <50 | 49 (49) |
| 50–100 | 30 (30) |
| 150–300 | 8 (8) |
| >300 | 11 (11) |
| Missing | 2 (2) |
CAKUT, congenital anomalies of the kidney and urinary tract.
The median (min–max) component PRO-Kid frequency score was 17.6 (0–57.1), and the median component impact score was 13.0 (0–58.9) (scale 0–100) where higher scores reflect a larger symptom burden. There were no significant differences in median scores by sex for either the frequency (male: 17.9 [8.9–25] versus female: 16.1 [10.7–35.7]; P = 0.3) or impact scores (male: 13.0 [5.4–23.1] versus female: 13.4 [8.9–27.2]; P = 0.4). There were similarly no significant differences in median scores by age group (8–13 versus 13–18 years). The median frequency scores were 19.6 (12.5–25) versus 16.1 (8.9–26.3); P = 0.3, and the median impact scores were 14.3 (8.9–22.9) versus 12.5 (5.4–25); P = 0.7, for the younger and older age groups, respectively.
The median PedsQL score was 72.3 (26.1–100), where higher scores indicate better health-related quality of life. Histograms of the PRO-Kid frequency and impact total scores are presented in Figure 1, A and B. The distribution of the individual raw PRO-Kid frequency and the impact symptom scores is shown in Table 2. The most frequently reported severe PRO-Kid symptom was tiredness (28.5%), followed by sleep difficulties (17%) and problems with concentrating/unable to focus (15%). Feeling left out (12%) was the most frequently endorsed socioemotional symptom.
Figure 1.
Distribution of total PRO-Kid scores of children with stage 3–5 CKD. (A) Histogram of PRO-Kid frequency total score (B) Histogram of PRO-Kid impact total scores.
Table 2.
Distribution of individual PRO-Kid symptom scores
| PRO-Kid Symptom | Score Distributions | |||||
|---|---|---|---|---|---|---|
| Frequency | Impact | |||||
| None (0) | Mild-Moderate (1–2) | Severe (3–4) | None (0) | Mild-Moderate (1–2) | Severe (3–4) | |
| No. (%) | No. (%) | No. (%) | No. (%) | No. (%) | No. (%) | |
| Sad | 40 (40) | 58 (58) | 2 (2) | 55 (55) | 38 (38) | 6 (6) |
| Scared/worried | 40 (40) | 54 (54) | 6 (6) | 48 (48) | 41 (41) | 10 (10) |
| Left out | 60 (60) | 27 (27) | 12 (12) | 62 (63) | 21 (21) | 14 (14) |
| Tired | 15 (15) | 55 (56) | 28 (28) | 31 (32) | 49 (51) | 16 (16) |
| Nausea | 63 (63) | 34 (34) | 3 (3) | 66 (68) | 27 (27) | 4 (4) |
| Poor appetite | 41 (41) | 45 (45) | 12 (12) | 53 (55) | 31 (32) | 12 (12) |
| Poor sleep | 49 (49) | 33 (33) | 17 (17) | 53 (54) | 29 (29) | 15 (15) |
| Not able to focus | 40 (40) | 44 (44) | 15 (15) | 46 (47) | 40 (41) | 11 (11) |
| Constipation | 70 (70) | 26 (26) | 4 (4) | 72 (75) | 18 (19) | 5 (5) |
| Itching | 52 (52) | 43 (43) | 5 (5) | 61 (62) | 34 (34) | 3 (3) |
| Headache | 57 (57) | 35 (35) | 8 (8) | 62 (63) | 27 (27) | 9 (9) |
| Pain | 62 (62) | 33 (33) | 5 (5) | 62 (64) | 26 (27) | 9 (9) |
| Trouble breathing | 79 (79) | 20 (20) | 1 (1) | 81 (83) | 13 (13) | 4 (4) |
| Change in taste of food | 76 (76) | 21 (21) | 3 (3) | 84 (87) | 11 (11) | 2 (2) |
Psychometric Analysis
The parallel analysis suggested three factors for both PRO-Kid frequency and impact, but two of the eigenvalues were very close to random noise and followed a clear, sharp drop in the scree plot. Thus, in combination with theoretical considerations of the construct and its contents, a single underlying factor was defensible, and the instruments were treated as unidimensional. A CFA model with one latent factor showed good fit to the data for the frequency and impact instruments. For PRO-Kid frequency, CFA model fit was CFI=0.993, RMSEA=0.032, and SRMR=0.105, after adding a residual correlation between the items tired and sleep (ρ=0.48), as suggested by CFA modification indices. For PRO-Kid impact, CFA model fit was CFI=0.987, RMSEA=0.045, and SRMR=0.117, without residual correlations between items. All items were strongly associated with the latent factors in both CFA models (all fully standardized loadings ≥0.40). This was further evidence for the unidimensionality of the two instruments and suggested there were no ill-fitting items that should be removed. Cα was high for both the PRO-Kid frequency and impact, 0.83 (95% confidence interval [CI], 0.78 to 0.88) and 0.84 (95% CI, 0.79 to 0.88), respectively, showing strong internal consistency. Each item showed good relevance to the construct according to the deletion statistic.
As shown in Figure 2, the difference between the mean and median values of the PedsQL score (A) and the PRO-Kid frequency total score (B) between CKD stages were not statistically significant.
Figure 2.
PedsQL and PRO-Kid total scores according to stage of CKD. (A) Box plot of PedsQL total score versus CKD stage. (B) Box plot of PRO-Kid frequency total score versus CKD stage. PedsQL, Pediatric Quality of Life Inventory.
Construct Validity
Pearson correlations between PRO-Kid and PedsQL scores were strong: R −0.78 (95% CI, −0.85 to −0.70) for the frequency score (Figure 3A) and R −0.69 (95% CI, −0.78 to −0.56) for the impact score (Figure 3B), reflecting poorer quality-of-life association with higher symptoms burden.
Figure 3.
Correlation between PRO-Kid scores and PedsQL scores in children with CKD. (A) Inverse relationship between PRO-Kid impact total score and PedsQL. (B) Inverse relationship between PRO-Kid total frequency score and PedsQL.
Discussion
Emerging studies have shown that children with CKD want increased involvement in treatment decision making.20 A pediatric CKD-specific PRO has the potential to provide children and adolescents with CKD and kidney failure a voice to communicate their symptoms and the impact on their daily life in a standardized way to better inform patient-centered decision making in CKD care. In this validation study, we found that our novel patient-reported symptom burden tool for children 8–18 years, PRO-Kid, appears to be unidimensional, and no items stand out as being a poor fit. We showed the internal reliability of the PRO-kid frequency and impact instruments is high and correlate strongly in the expected direction with PedsQL, supporting its construct validity.
In an adult study of over a thousand patients on dialysis, the Edmonton Symptom Assessment System Revised: Renal improved patient and provider symptoms awareness, particularly for under-recognized psychosocial symptoms, and empowered patients to raise issues with providers.21 In pediatric oncology, a web-based application is being trialed that enables symptom screening using a validated PROM similar to PRO-Kid and access to clinical practice guidelines for symptom management.22 Our validated pediatric CKD-specific PRO-Kid tool has the potential to similarly improve patient-provider communication and empower children and adolescents to become a partner in their care.
PRO-Kid may also inform clinical decision making surrounding treatment initiation or disease management, standardized evaluation of existing and future treatments, and enable longitudinal assessment of quality of care. While symptoms can be subjective, PRO-Kid quantitatively measures pediatric patients' perspectives of CKD, which can be incorporated into both clinical care and research.23 Important clinical decisions such as timing of initiation of KRT are guided by the presence and severity of uremic symptoms.24 Randomized trials in adults show no benefit to early dialysis initiation on the basis of estimated kidney function, and suggest that instead, the presence and severity of symptoms should guide this decision.9,25 A cross-sectional survey of Canadian pediatric nephrologists by our research group showed significant practice variation regarding timing of dialysis initiation,24 a reflection of the subjectivity of symptoms in the clinical setting, and lack of objective symptom assessment tools for this population. In other clinical specialties, including oncology, evidence-based guidelines have been developed to manage symptoms identified by PROs.26
Young study participants were able to complete the PRO-Kid questionnaire with ease, supporting the feasibility of future implementation of this questionnaire in outpatient clinics. For a tool designed for use in a busy clinical setting with multiple competing priorities, the simplicity with which the PRO-Kid questionnaire can be administered and completed is important. Another potential facilitator as we look to implement PRO-Kid in clinical care is its electronic format which allows completion of the questionnaire in a variety of settings (e.g., at home before clinic or in the clinic waiting area). As a result of the coronavirus disease 2019 pandemic during the development phase of PRO-Kid, we had to pivot to an electronic platform delivered through REDCap that allows capturing the scores in electronic medical records.
The strengths of our study include a representative and diverse sample of children and adolescents in Canada with CKD/kidney failure and high levels of data completeness enabling a detailed analysis. Furthermore, patients participated in the study while receiving their routine care, demonstrating the potential feasibility of implementation of this tool in clinical care. One limitation is our relatively small sample size; however, we reached our target enrollment for validation. The lack of significant differences between PRO-Kid scores in the expected direction by CKD stage could raise concern about its construct validity. However, in support of PRO-Kid score validity, PedsQL scores by CKD stage were congruent with the direction of change of PRO-Kid scores (Figure 2B). We hypothesize that children with stage 3 CKD may be in an important transition from the asymptomatic phase of CKD to a more symptomatic phase. Children with stage 4 CKD may have adapted to their symptoms or their symptoms may be better addressed and managed with more intensive follow-up and multidisciplinary care. This is also a cross-sectional study, and individual children have different thresholds when reporting symptoms and are affected differently by their symptoms. This study was not designed or powered to assess these differences. Further research, including a prospective study is required to better understand the symptom profiles of children at different stages of CKD. Finally, we did not conduct test re-test reliability but assessed consistency and concurrent validity.
PRO-Kid is a novel patient-reported symptoms burden tool for children 8–18 years of age with CKD that correlates strongly in the expected direction with PedsQL, supporting its validity. Future work will focus on the sensitivity of PRO-Kid score in prospective measurement of change in symptoms burden of CKD within individuals, with progression of disease, as well as validation of modified versions of PRO-Kid for younger children (2–7 years), PRO-Kid in other languages, and implementation of the tool into clinical care.
Acknowledgments
We would like to thank Ms. Banke Oketola from the Children's Hospital Research Institute of Manitoba for all her efforts in working with the participating sites to complete this validation study. We would also like to acknowledge and thank our patient and parent partners—Kelly Loverock, and Christabel Agyei Gyamfi—for their invaluable support and feedback in preparation for implementation of PRO-Kid into clinical care.
Footnotes
See related Patient Voice, “Patient and Caregiver Perspective on PRO-Kid Quality of Life Tool,” on pages 821–822.
Disclosures
Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/CJN/B948.
Funding
This project was supported by an Operating Grant from the Children's Hospital Research Institute of Manitoba (OG2019-10) as well as the Canadian Institutes of Health Research—Canadians Seeking Solutions and Innovations to Overcome CKD Strategic Patient Oriented Research network.
Author Contributions
Conceptualization: Allison B. Dart, Mina Matsuda-Abedini, Michael Zappitelli.
Data curation: Allison B. Dart, Mina Matsuda-Abedini.
Formal analysis: Brenden Dufault, Kimberley Widger.
Funding acquisition: Allison B. Dart, Mina Matsuda-Abedini, Michael Zappitelli.
Investigation: Rahul Chanchlani, Allison B. Dart, Janis M. Dionne, Mina Matsuda-Abedini, Michael Zappitelli.
Methodology: Allison B. Dart, Sara N. Davison, Brenden Dufault, Mina Matsuda-Abedini, Adam Rapoport, Kimberley Widger.
Project administration: Rahul Chanchlani, Allison B. Dart, Janis M. Dionne, Mina Matsuda-Abedini, Susan Samuel, Michael Zappitelli.
Resources: Allison B. Dart, Mina Matsuda-Abedini, Michael Zappitelli.
Software: Brenden Dufault, Veronica Ka Wai Lai.
Supervision: Mina Matsuda-Abedini.
Validation: Allison B. Dart, Sara N. Davison, Brenden Dufault, Mina Matsuda-Abedini, Adam Rapoport, Kimberley Widger.
Visualization: Allison B. Dart, Brenden Dufault, Mina Matsuda-Abedini.
Writing – original draft: Mina Matsuda-Abedini.
Writing – review & editing: Ke Fan Bei, Rahul Chanchlani, Allison B. Dart, Sara N Davison, Janis M. Dionne, Brenden Dufault, Veronica Ka Wai Lai, Mina Matsuda-Abedini, Adam Rapoport, Susan Samuel, Kimberley Widger, Michael Zappitelli.
Data Sharing Statement
All data are included in the manuscript and/or supporting information.
References
- 1.Tjaden L, Tong A, Henning P, Groothoff J, Craig JC. Children's experiences of dialysis: a systematic review of qualitative studies. Arch Dis Child. 2012;97(5):395–402. doi: 10.1136/archdischild-2011-300639 [DOI] [PubMed] [Google Scholar]
- 2.Tong A Wong G McTaggart S, et al. Quality of life of young adults and adolescents with chronic kidney disease. J Pediatr. 2013;163(4):1179–1185.e5. doi: 10.1016/j.jpeds.2013.04.066 [DOI] [PubMed] [Google Scholar]
- 3.Roumelioti ME Wentz A Schneider MF, et al. Sleep and fatigue symptoms in children and adolescents with CKD: a cross-sectional analysis from the chronic kidney disease in children (CKiD) study. Am J Kidney Dis. 2010;55(2):269–280. doi: 10.1053/j.ajkd.2009.09.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Goldstein SL, Graham N, Burwinkle T, Warady B, Farrah R, Varni JW. Health-related quality of life in pediatric patients with ESRD. Pediatr Nephrol. 2006;21(6):846–850. doi: 10.1007/s00467-006-0081-y [DOI] [PubMed] [Google Scholar]
- 5.Varni JW, Limbers CA, Burwinkle TM. Impaired health-related quality of life in children and adolescents with chronic conditions: a comparative analysis of 10 disease clusters and 33 disease categories/severities utilizing the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes. 2007;5:43. doi: 10.1186/1477-7525-5-43 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hanson CS Gutman T Craig JC, et al. Identifying important outcomes for young people with CKD and their caregivers: a nominal group technique study. Am J Kidney Dis. 2019;74(1):82–94. doi: 10.1053/j.ajkd.2018.12.040 [DOI] [PubMed] [Google Scholar]
- 7.Chong LSH Sautenet B Tong A, et al. Range and heterogeneity of outcomes in randomized trials of pediatric chronic kidney disease. J Pediatr. 2017;186:110–117.e11. doi: 10.1016/j.jpeds.2017.03.034 [DOI] [PubMed] [Google Scholar]
- 8.Preka E, Rees L. Should we abandon GFR in the decision to initiate chronic dialysis? Pediatr Nephrol. 2020;35(9):1593–1600. doi: 10.1007/s00467-019-04333-4 [DOI] [PubMed] [Google Scholar]
- 9.Cooper BA Branley P Bulfone L, et al.; IDEAL Study. A randomized, controlled trial of early versus late initiation of dialysis. N Engl J Med. 2010;363(7):609–619. doi: 10.1056/NEJMoa1000552 [DOI] [PubMed] [Google Scholar]
- 10.Jawa NA Rapoport A Widger K, et al. Development of a patient-reported outcome measure for the assessment of symptom burden in pediatric chronic kidney disease (PRO-Kid). Pediatr Nephrol. 2022;37(6):1377–1386. doi: 10.1007/s00467-021-05269-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Riley AW. Evidence that school-age children can self-report on their health. Ambul Pediatr. 2004;4(suppl l):371–376. doi: 10.1367/A03-178R.1 [DOI] [PubMed] [Google Scholar]
- 12.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schwartz GJ Munoz A Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20(3):629–637. doi: 10.1681/ASN.2008030287 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Varni JW, Seid M, Kurtin PS. PedsQL 4.0: reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care. 2001;39(8):800–812. doi: 10.1097/00005650-200108000-00006 [DOI] [PubMed] [Google Scholar]
- 15.Gorsuch RL. Factor Analysis. Erlbaum; 1983. [Google Scholar]
- 16.Tavakol M, Dennick R. Making sense of Cronbach's alpha. Int J Med Educ. 2011;2:53–55. doi: 10.5116/ijme.4dfb.8dfd [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hinkle DEWW, Jurs SG. Applied Statistics for the Behavioral Sciences, 5th ed. Houghton Mifflin College Division; 2003. [Google Scholar]
- 18.Revelle W. Procedures for Personality and Psychological Research. Northwestern University; 2020. https://CRAN.R-project.org/package=psychVersion=2.1.3 [Google Scholar]
- 19.Rosseel Y. Lavaan: an R package for structural equation modeling. J Stat Softw. 2012;48(2):1–36. doi: 10.18637/jss.v048.i02 [DOI] [Google Scholar]
- 20.Gutman T Hanson CS Bernays S, et al. Child and parental perspectives on communication and decision making in pediatric CKD: a focus group study. Am J Kidney Dis. 2018;72(4):547–559. doi: 10.1053/j.ajkd.2018.05.005 [DOI] [PubMed] [Google Scholar]
- 21.Evans JM Glazer A Lum R, et al. Implementing a patient-reported outcome measure for hemodialysis patients in routine clinical care: perspectives of patients and providers on ESAS-r:renal. Clin J Am Soc Nephrol. 2020;15(9):1299–1309. doi: 10.2215/CJN.01840220 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cook S Vettese E Tomlinson GA, et al. Feasibility of a randomized controlled trial of symptom screening and feedback to healthcare providers compared with standard of care using the SPARK platform. Support Care Cancer. 2020;28(6):2729–2734. doi: 10.1007/s00520-019-05115-2 [DOI] [PubMed] [Google Scholar]
- 23.Selewski DT Thompson A Kovacs S, et al. Patient-reported outcomes in glomerular disease. Clin J Am Soc Nephrol. 2017;12(1):140–148. doi: 10.2215/CJN.13231215 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Saban JA Zappitelli M Samuel SM, et al. Perceptions of pediatric nephrologists regarding timing of dialysis initiation in children in Canada. Can J Kidney Health Dis. 2016;3:31. doi: 10.1186/s40697-016-0123-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nesrallah GE Mustafa RA Clark WF, et al.; Canadian Society of Nephrology. Canadian Society of Nephrology 2014 clinical practice guideline for timing the initiation of chronic dialysis. CMAJ. 2014;186(2):112–117. doi: 10.1503/cmaj.130363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kwekkeboom KL, Wieben A, Stevens J, Tostrud L, Montgomery K. Guideline-recommended symptom management strategies that cross over two or more cancer symptoms. Oncol Nurs Forum. 2020;47(5):498–511. doi: 10.1188/20.ONF.498-511 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data are included in the manuscript and/or supporting information.
