Visual Abstract
Keywords: symptoms, quality of life, CKD, clinical epidemiology, cohort studies
Abstract
Background and objectives
Late stages of CKD are characterized by significant symptom burden. This study aimed to identify subgroups within the 5-year trajectories of symptom evolution in patients with CKD and to describe associated patient characteristics and outcomes.
Design, setting, participants, & measurements
Among 2787 participants (66% men) with eGFR <60 ml/min per 1.73 m2 enrolled in the CKD–Renal Epidemiology and Information Network (CKD-REIN) cohort study from July 2013 to May 2016, we assessed symptoms annually using the Kidney Disease Quality of Life-36 (KDQOL-36) questionnaire until December 2020. A total of 9121 measures were reported over follow-up; all participants had symptoms scored for at least one time point. We used a joint latent class-mixed model to distinguish profiles of symptom trajectories.
Results
Patient mean age (±SD) at baseline was 67±13 years, and mean eGFR was 33±13 ml/min per 1.73 m2. The prevalence of each symptom ranged from 24% (chest pain) to 83% (fatigue), and 98% of participants reported at least one symptom. After a median (interquartile range) follow-up of 5.3 (3.4–6.0) years, 690 participants initiated KRT, and 490 died before KRT. We identified two profiles of symptom trajectories: a “worse symptom score and worsening trajectory” in 31% of participants, characterized by a low initial symptom score that worsened more than ten points over time, and a “better symptom score and stable trajectory” in 69% of participants, characterized by a high initial score that remained stable. Participants in the worse symptom score and worsening trajectory group had more risk factors for CKD progression at baseline, worse quality of life, and a higher risk of KRT and death before KRT than other participants.
Conclusions
This study highlights a significant worsening of symptoms in about one third of the participants, whereas the majority reported low symptom severity throughout the study.
Introduction
Late stages of CKD are characterized by significant symptom burden (1,2). Symptoms are a self-reported or subjective perception of an individual’s experience of a physical disturbance or disease (3). From the patients’ perspective, reducing symptoms is more important than extending life (1). Conceptually, symptoms are a multidimensional construct characterized by their prevalence, distress, severity, and frequency (4). The totality of symptoms, including their effect on a patient’s health-related quality of life and ability to participate in life, is described as the “symptom burden” (3).
Quality of life and symptoms are described in various type of CKD population (5): patients with nondialysis-dependent CKD (6), dialysis-dependent CKD, CKD under conservative care (7,8), or living with kidney transplant. In these populations, symptom burden is an important predictor of reduced quality of life (9–11). It is positively correlated with mortality, morbidity (12), reduced adherence to treatment (13), cardiovascular events (14), and disability-related productivity loss (1,10). A comprehensive review found that the average number of symptoms per patient ranges from six to 20 (15). Patients most frequently report fatigue, pain, pruritus, insomnia, and anorexia (9). However, symptoms have been assessed with several instruments (15), so the number of symptoms depends on the instrument used. These symptoms rarely occur in isolation, and growing evidence suggests that CKD symptoms frequently occur in clusters (4,16).
Other than the prevalence of symptoms, their classification into clusters (16,17), and their associated factors (16), we have limited knowledge about how symptoms evolve in patients with different stages of CKD. Few longitudinal studies have investigated symptom evolution in patients with nondialysis-dependent CKD, with inconsistent findings (18–20).
We hypothesized that individuals with eGFR <60 ml/min per 1.73 m2 could be classified into distinct evolution subgroups on the basis of self-reported symptom data from the Kidney Disease Quality of Life-36 questionnaire (KDQOL-36).
For this purpose, we used longitudinal data from the French CKD–Renal Epidemiology and Information Network (CKD-REIN) cohort study to identify subgroups within the 5-year trajectories of symptom evolution in patients with CKD and to describe patient characteristics and outcomes in each subgroup identified.
Materials and Methods
Design and Study Population
The CKD-REIN study is a prospective cohort conducted in 40 nephrology clinics, nationally representative geographically and in terms of facility legal status (public, private not for profit, and private for profit). Between July 2013 and March 2016, a total of 3033 adults with a proven CKD diagnosis, eGFR <60 ml/min per 1.73 m2, and no KRT were included in the cohort during a routine visit to their nephrologist. Participants were followed for 5 years. The study design and patient profile have been described in detail elsewhere (21,22).
In this study, we analyzed data for 2787 participants who provided at least one score for the KDQOL-36 symptom dimension during the follow-up, from inclusion date to December 31, 2020. In total, 2787 participants reported a total of 9121 measures of the symptom dimension (Figure 1). Participants who initiated KRT (dialysis or transplant) or died before KRT were censored.
Figure 1.
Flow chart of study participants. CKD–Renal Epidemiology and Information Network (CKD-REIN) cohort.
Data Collection
Information collected at baseline included sociodemographic data (age, sex, education, marital status, monthly income), clinical data (diabetes mellitus, cardiovascular comorbidities), smoking status, medication prescriptions, body mass index (in kg/m2), and biologic data (hemoglobin, creatinine, and albumin levels; eGFR, urine albumin-creatinine ratio, serum calcium, and potassium levels). These baseline characteristics have been reported elsewhere (21–23). Definitions of operational variables are reported in Supplemental Table 1.
On enrollment and annually, all study participants were asked to complete a self-administered patient questionnaire, including validated instruments regarding health-related quality of life and more specifically CKD symptoms. Using the KDQOL-36 (24,25), we computed five-dimension scores, such as the burden, effects, and symptom scores, and mental component summary (MCS) and physical component summary (PCS) scores. Higher scores indicate better health-related quality of life.
The outcome was the symptom dimension (score) of the KDQOL-36 questionnaire. We collected 11 symptoms at baseline and at 1, 2, 3, and 5 years. These symptoms were muscle pain and soreness, chest pain, cramps, itchy skin, dry skin, shortness of breath, faintness and dizziness, lack of appetite, fatigue, numbness in hands or feet, nausea, or upset stomach.
Each symptom represents an item, and each item has five possible Likert-type response modalities: not at all, a little, moderately, a lot, or very much bothered. A crude score was computed by considering missing data (if <50%): crude score = mean of nonmissing item × total number of items. This crude score was then normalized on a scale of 0–100, with lower scores representing higher symptom severity.
Statistical Analysis
The baseline characteristics of the participants are described with number (percentage) or mean (±SD) and/or median (interquartile range). Included and nonincluded participants were compared with standard statistical tests (Student t test or Mann–Whitney test for continuous variables and chi-squared or Fisher’s test for categorical variables, depending on sample sizes).
We used the joint latent class-mixed model (JLCMM) implemented in the “lcmm” R package (26) to identify subgroups of symptom score trajectories. The time origin was the date of inclusion into the cohort, and the administrative censoring date was set at December 31, 2020. The JLCMM is an extension of the LCMM to account for potential informative dropout due to KRT or death before KRT, for example (26). The LCMM is itself an extension of the standard linear mixed model to account for heterogeneous populations in terms of trajectories (27). It allows for considering participants with only one symptom measure, to reduce selection bias (27,28). Therefore, we used 2787 participants from the cohort, including 551 participants with a single symptom measure during follow-up. The JLCMM consisted of four joint submodels: a multinomial logistic regression model for estimating the probability of each patient to belong to each latent class, a class-specific mixed model for modeling the symptom score trajectories, and two cause-specific proportional-hazard models to account for informative dropout and estimate the risk of KRT and death before KRT in each latent class. None of the submodels introduced patient characteristics because our aim was to identify latent classes independent of these characteristics. For the class-specific linear mixed model, because the symptom score was not normally distributed, we tried different transformations, and the beta transformation best fitted the data. We also tried different functions of time and chose a natural spline with one interior knot at the median with an unstructured variance-covariance matrix for the random effects. For two-cause specific proportional-hazard models, we tried different baseline hazard functions and selected the Weibull distribution. To choose the optimal number of latent classes, we estimated several JLCMMs with one to five classes each and selected the model that provided the best (1) fit to data according to the Bayesian Information Criterion (Supplemental Table 2), (2) discrimination between classes according to entropy and a posteriori classification table (Supplemental Table 3), and (3) mix of fit and discrimination (Integrated Completed Likelihood criterion) (29). We also paid attention to the clinical relevance of results and avoided choosing a model with a class consisting of few participants (<5% of the total sample). All details for the modeling choices are in Supplemental Item 1.
Once the final model was selected, we compared the following patient characteristics at baseline between each class: age (in years), sex, albumin-creatinine ratio, CKD stages, presence of diabetes mellitus, cardiovascular history, body mass index, Charlson Comorbidity Index, quality of life (PCS, MCS, burden and effect of kidney disease), level of physical activity (Global Physical Activity Questionnaire), depression score (Center for Epidemiologic Studies Depression Scale), anemia, and serum albumin, potassium, and calcium levels. We also compared eGFR slopes across symptom trajectory subgroup using mixed-linear model with random intercept and slope. Follow-up time, trajectory subgroup, and their interaction were the model's fixed effects. Finally, we described the evolution of each symptom item within each trajectory subgroup by showing the average of Likert-type responses transformed into a continuous variable, ranging from 1 (not at all bothered) to 5 (very much bothered), for each symptom item at each time point.
Results
Baseline Characteristics
The baseline characteristics of the 2787 participants are in Table 1. The mean (±SD) age was 67±13 years, and 66% were men. The mean eGFR was 33±13 ml/min per 1.73 m2, and 45% of the participants had CKD stages 4–5. As compared with included participants, the 246 excluded participants were younger and had lower eGFR; they more often had diabetes, anemia, and fewer comorbidities. They also experienced more adverse events (death or KRT) (Supplemental Table 4).
Table 1.
Baseline characteristics of participants according to their subsequent symptom trajectory class, CKD-Renal Epidemiology and Information Network (CKD-REIN) cohort (n=2787)
| Characteristics | Overall | “Worse Symptom Score and Worsening Trajectory” | “Better Symptom Score and Stable Trajectory” |
|---|---|---|---|
| Number of participants | 2787 (100) | 875 (31) | 1912 (69) |
| Age, yr | 67±13 | 68±14 | 67±12 |
| Age group, yr | |||
| 18–44 | 174 (6) | 65 (7) | 109 (6) |
| 45–64 | 785 (28) | 200 (23) | 585 (31) |
| 65–74 | 983 (35) | 292 (33) | 691 (36) |
| ≥75 | 845 (30) | 318 (36) | 527 (28) |
| Sex, men | 1829 (66) | 572 (65) | 1257 (66) |
| Currently married | 1690 (61) | 531 (61) | 1159 (61) |
| BMI, kg/m2 | 29±6 | 29±6 | 29±6 |
| Body weight status | |||
| Underweight | 41 (2) | 18 (2) | 23 (1) |
| Healthy weight | 734 (26) | 235 (27) | 499 (26) |
| Overweight | 1003 (36) | 285 (33) | 718 (38) |
| Obesity | 956 (34) | 318 (36) | 638 (33) |
| Diabetes mellitus | 1172 (42) | 419 (48) | 753 (40) |
| Cardiovascular history | 1457 (52) | 549 (63) | 908 (48) |
| Charlson comorbidity index ≥5 | 2126 (76) | 706 (81) | 1420 (74) |
| PCS score | 42±10 | 38±10 | 43±10 |
| MCS score | 48±7 | 47±8 | 48±7 |
| Burden score | 74±24 | 67±26 | 78±22 |
| Effect of the kidney disease score | 81±18 | 75±19 | 84±16 |
| Symptoms score | 75±16 | 71±17 | 77±16 |
| Depression score (CES-D) | 25±17 | 28±18 | 24±17 |
| Physical activity (GPAQ) | |||
| Intense | 687 (27) | 159 (20) | 528 (31) |
| Moderate | 624 (25) | 187 (23) | 437 (25) |
| Low | 1211 (48) | 457 (57) | 754 (44) |
| eGFR, ml/min per 1.73 m2 | 33±12 | 26±10 | 37±11 |
| eGFR <30 ml/min per 1.73 m2 | 1241 (45) | 633 (72) | 608 (32) |
| Urine albumin-creatinine ratio, mg/g | |||
| <30 | 711 (26) | 117 (13) | 594 (31) |
| 30–299 | 798 (29) | 205 (23) | 593 (31) |
| ≥300 | 1034 (37) | 477 (55) | 557 (29) |
| Serum calcium level, mg/dl | |||
| <8.4 | 63 (2) | 41 (4) | 22 (1) |
| 8.4–10.4 | 2583 (93) | 799 (91) | 1784 (93) |
| >10.4 | 75 (3) | 21 (2) | 54 (3) |
| Serum potassium level, mmol/L | |||
| <3.5 | 73 (3) | 29 (3) | 44 (2) |
| 3.5–5.3 | 2506 (90) | 764 (87) | 1742 (91) |
| >5.3 | 197 (7) | 82 (9) | 115 (6) |
| Serum albumin <4.0 g/dl | 213 (8) | 117 (13) | 96 (5) |
| Anemiaa | 1038 (37) | 477 (55) | 561 (29) |
| Number of drugs | 8±4 | 9±4 | 7±4 |
All data are presented as n (%) or mean±SD. Missing data: no missing data: age, sex, eGFR, symptoms scores; ≤5% of missing data: depression score, marital status, education, BMI, diabetes mellitus, cardiovascular history, Charlson comorbidity index, burden score, effect of the kidney disease score, symptoms score, and anemia; between 5% and 10% of missing data: urine albumin-creatinine ratio and physical activity; ≥10% of missing data: monthly income, PCS score, MCS score, and serum albumin. A higher score indicates the presence of more depression or best quality of life (PCS, MCS, burden, effect). BMI, body mass index; PCS, physical component summary; MCS, mental component summary; CES-D, Center for Epidemiologic Studies Depression Scale; GPAQ, Global Physical Activity Questionnaire.
Hemoglobin level <12.0 g/dl in women and <13.0 g/dl in men.
The mean symptom score was 75±16. The prevalence of symptoms ranged from 24% (chest pain) to 83% (fatigue), and 98% of participants reporting at least one symptom. Fatigue (83%), muscle pain and soreness (82%), cramps (72%), dry skin (60%), and shortness of breath (68%) were the most common. The prevalence of symptoms and their severity are shown in Figure 2.
Figure 2.
Prevalence and severity of 11 symptoms at baseline in study participants, CKD-REIN cohort (n=2787). Kidney Disease Quality of Life-36 (KDQOL-36) symptoms: muscle pain and soreness, chest pain, cramps, itchy skin, dry skin, shortness of breath, faintness and dizziness, lack of appetite, fatigue, numbness in hands or feet, nausea, or upset stomach.
Distinct Profiles of Symptom Score Trajectories and Associated Outcomes Over Time
Over the study follow-up, 31% (n=872), 19% (n=533), 15% (n=410), and 15% (n=421) of the participants had completed five, four, three, and two KDQOL-36 questionnaires, respectively. Between study enrollment and December 2020 (censoring date), 690 participants started KRT and 490 died before KRT over a median (interquartile range) follow-up of 5.3 (3.4–6.0) years. The best JLCMM had two latent classes (Figure 3). The first class, termed “worse symptom score and worsening trajectory,” included 31% (n=875) of participants. It was characterized by a lower initial level of symptom score, reflecting higher symptom severity, which worsened more than ten points over time. The second class, termed “better symptom score and stable trajectory,” included 69% (n=1912) of participants. It was characterized by a high initial symptom score, reflecting lower symptom severity, which remained stable. Observed individual symptoms trajectories of 50 randomly selected participants a posterior classified in each class are shown in Supplemental Figure 1.
Figure 3.
Predicted mean trajectories (dark lines) of symptom score in the two identified latent classes and their confidence intervals (shaded bands). The worse symptom score and worsening trajectory group is represented in red and the better symptom score and stable trajectory in green. CKD-REIN cohort.
As expected, the risk of KRT and death before KRT was higher in participants with “worse symptom score and worsening trajectory” (Figure 4). The evolution of each symptom item within each trajectory subgroup is depicted in Supplemental Figure 2.
Figure 4.
Unadjusted probability of KRT (left panel) and death before KRT (right panel) by class of symptom score trajectory. CKD-REIN cohort.
Patient Characteristics in Each Subgroup of Trajectories
As compared with participants in the better symptom score and stable trajectory subgroup, those in the worse symptom score and worsening trajectory subgroup more often had diabetes (48% versus 40%), obesity (36% versus 33%), comorbidities, anemia (55% versus 29%), and hypocalcemia (4% versus 1%); they were less physically active (57% versus 44%) and more often prescribed medications (9±4 versus 7±4) and had lower eGFR (26±10 versus 37±11 ml/min), and worse quality of life (PCS, burden, and effect of the kidney disease) at baseline (Table 1); and they were more symptomatic (Supplemental Figure 3). They also had significantly faster eGFR decline, 3.56 (95% confidence interval [95% CI], 3.80 to 3.33) ml/min per 1.73 m2 per year than their counterparts, 1.15 (95% CI, −1.24 to −1.06) ml/min per 1.73 m2 per year.
Discussion
This prospective study including 2787 individuals with nondependent dialysis CKD and 5-year assessment of patient-reported outcomes highlighted a significant worsening of standardized measures of symptom severity, by >10 points on a scale of 0–100, in 31% of the participants. The other 69% reported low symptom severity that remained stable. This substantial number of participants with an unfavorable symptom trajectory during predialysis care were at high risk for adverse outcomes including faster progression to KRT and death before KRT. To the best of our knowledge, this study is the first to examine symptom trajectories in nondependent dialysis CKD by using the JLCMM function of the “lcmm” R package.
This model is an interesting, advanced statistical tool to identify distinct trajectories (26), taking into account dropouts due to KRT or death before KRT, which are potentially missing data not at random (22). These results also have important implications for routine clinical practice.
Our findings are hardly comparable with those from other longitudinal studies that differ in population selection and in methods used to assess symptoms and estimate their trajectories (18–20). Among 3939 patients enrolled in the Chronic Renal Insufficiency Cohort (CRIC) study, Grams et al. (18) used an unsupervised clustering approach on the basis of all KDQOL-36 scores (burden, effects, and symptom scores, and MCS and PCS) measured over 14 years to estimate individual trajectories and classify patients into groups, the number of which was prespecified. In this population, 10 years younger than ours, with a baseline eGFR higher by 10 ml/min per 1.73 m2, they observed a relatively stable symptom score among the three identified groups, but large differences in baseline symptom score, ranging from 68±16 in the high-risk group to 93±8 in the low-risk group. Of note, in our study, participants with worsening symptom trajectories had lower score at baseline than those who remained stable, 71±17 versus 77±16, respectively. In another CRIC study, Wulczyn et al. (19) focused on six uremic symptoms from the KDQOL (fatigue, anorexia, pruritus, nausea, paresthesia, pain) and exhibited heterogeneity in symptom change, with 42% of individuals experiencing clinically significant worsening of at least one of these symptoms. Similarly, but in a cohort of more severe patients with CKD stage 5 on conservative care, Murtagh et al. (20) identified three symptom trajectories in the last year of life: 50% of the patients followed a stable trajectory, 24% increased, and 21% fluctuated.
In addition, we found that participants included in the worse symptom score and worsening trajectory group had more risk factors for CKD progression, a higher probability of KRT and death before KRT, a faster eGFR, and worse baseline quality of life (PCS, burden and effect, and symptoms of the kidney disease) than those in the better symptom score and stable trajectory group. In the CRIC study, Grams et al. (18) also reported higher cumulative incidence of KRT, cardiovascular disease, heart failure, and death in the group at higher versus lower risk of severe symptoms. Wulczyn et al. (19) also found a nonlinear association between eGFR and uremic symptom severity score: a greater magnitude of symptomatic worsening occurred with eGFR decreases at lower initial eGFR values.
The importance of quality of life and patient experiences with symptoms is being increasingly recognized in the care of patients with chronic illness (19). The Kidney Health Initiative, the US Food and Drug Administration, and the American Society of Nephrology (1,30) have identified a better understanding of the effect of the burden of symptoms on CKD patient outcomes as a priority. In the Standardized Outcomes in Nephrology initiative, patients on hemodialysis rated fatigue as one of the most important aspects of their disease (31). A novel paradigm of this study was the use of the symptom KDQOL-36 dimension to classify CKD patients into evolution trajectories. Knowing when patients are likely to experience rapid functional decline and high symptom burden and become most dependent on health and social services has major implications for the planning and delivery of their care (32).
In addition to the usual monitoring of clinical and biologic parameters, clinicians should monitor the change in symptoms by using the self-reported symptom dimension of KDQOL-36 questionnaire or any other validated questionnaire. Patients with CKD may not report symptoms correctly during medical consultations, and an active investigation by a self-reported questionnaire may provide a better understanding of the change in symptom, promote patient engagement in their treatment (33), and improve the symptom and quality of life management and other patients’ outcomes (34).
Because routine assessment of symptoms is not standard, large gains may be achievable with greater provider appreciation of the importance of symptoms. Duncanson et al. (35) described a protocol for a qualitative study to evaluate the feasibility and acceptability of electronic patient-reported outcome measures (PROMs) data capture and feedback in hemodialysis after the pilot Symptom monitoring with Feedback Trial. In a qualitative study, Aiyegbusi et al. (36) also reported that the renal ePROM system may play a supportive role in the routine clinical management of patients with advanced CKD if the concerns of clinicians and patients can be sufficiently addressed. This study is the first step for standardizing symptom assessment by using self-administered questionnaires. A better assessment of symptom progression, its determinants, and its effect on CKD progression and quality of life should be a research priority. In the future, the use of smartphone-based symptom collection applications may be considered in assessing symptoms and CKD management. This active assessment of symptoms over time will allow for better follow-up and better adherence of patients with CKD to treatment (33).
The major strengths of the CKD-REIN prospective study include its large sample size of participants with all types of diagnosed CKD and repeated measurements of several PROMs, such as the symptom dimension of the KDQOL-36 questionnaire over a 5-year follow-up. This cohort included a representative sample of patients under nephrology care. Clinical outcomes such as PROMs were rigorously collected and adjudicated. However, this study also has limitations. First, there is a potential selection bias because the 246 participants excluded had a worse baseline clinical profile than those included. These participants likely had worse symptom scores, thus resulting in possible underestimation of participants classified in the “worse symptom score and worsening trajectory.” Second, the detection of symptom change is closely related to the psychometric properties of the KDQOL-36, including its responsiveness.
However, despite increasing number of shorter symptom measures such as ESAS-r or IPOS-Renal, the KDQOL-36 remains the most widely used instrument in CKD (5) and allows for comparisons between studies (20,37,38). Third, the interval between symptom measurements is wide and may not be clinically meaningful for participants with rapidly worsening symptom trajectories. Fourth, our method to calculate symptom dimension scores in the presence of missing items (if <50%) assumed these were at random. However, it is possible the KDQOL completion rate was lower in participants with a worsening symptom trajectory, which may have led to an underestimation in this subgroup. Finally, <5% of the participants ended CKD care at the nephrology clinic and were censored at their last visit, but the potential risk of attrition bias is likely to be limited.
In conclusion, this study reports substantial heterogeneity in symptom trajectories among patients with CKD not dependent on dialysis, with one third exhibiting a worse symptom score and worsening trajectory of symptoms. The risk of KRT initiation and death before KRT was higher in participants belonging to the worse symptom score and worsening trajectory. This study suggests a systematic assessment of symptoms by a validated questionnaire for planning early therapeutic interventions. There is a need for future research to assess the modifiable factors that affect the unfavorable symptom trajectory and other aspects of quality of life.
Disclosures
C. Ayav reports serving on speakers bureau for Baxter. C. Combe reports consultancy agreements with Rhythm and Travere; reports receiving research funding from Novartis and Sanofi; reports receiving honoraria from Amgen, Fresenius, and Travere; and reports having an advisory or leadership role for Nephrology Dialysis Transplantation. D. Fouque reports having consultancy agreements with Astellas, AstraZeneca, Freseniuskabi, GSK, Lilly, Theradial, and Vifor. L. Le Gall reports serving as a member of the French Society of Nephrology SFNDT. B. Stengel reports receiving research funding from Amgen, AstraZeneca, Fresenius Medical Care, GlaxoSmithKline (GSK), Otsuka, and Vifor Fresenius. Z.A. Massy reports receiving research funding from Amgen, Fresenius Medical Care, GlaxoSmithKline, and Merck Sharp and Dohme-Chibret; reports receiving government support for CKD-REIN Project and Experimental Projects; reports receiving honoraria to the charities or for travel from AstraZeneca, Boehringer, and GSK; and reports serving advisory or leadership roles for Journal of Nephrology, Journal of Renal Nutrition, Kidney International, Nephrology Dialysis Transplantation, and Toxins. R. Pecoits-Filho reports having consultancy agreements with George Clinical; reports receiving research funding from Fresenius Medical Care; reports receiving honoraria from Akebia, AstraZeneca, Bayer, Boehringer, Fresenius Medical Care, and GSK; reports serving an advisory or leadership role for SONG Initiative Executive Committee; reports serving on the Editorial Boards of American Journal of Kidney Diseases, Blood Purification, Hemodialysis International, International Society of Nephrology, Kidney Disease Improving Global Outcomes, Nephrology, and Peritoneal Dialysis International; and reports serving on speakers bureau for AstraZeneca, Bayer, and Boehringer. R. Pecoits-Filho reports being employed by Arbor Research Collaborative for health, which runs the DOPPS studies. Global support for the ongoing DOPPS Programs is provided without restriction on publications by a variety of funders; funding is provided to Arbor Research Collaborative for Health and not to R. Pecoits-Filho directly. For details see https://www.dopps.org/AboutUs/Support.aspx. All remaining authors have nothing to disclose.
Funding
CKD-REIN is funded by the Agence Nationale de la Recherche through the 2010 Cohortes-Investissements d’Avenir program (ANR-IA-COH-2012/3731) and by the 2010 national Programme Hospitalier de Recherche Clinique. CKD-REIN has been supported by a public–private partnership with Fresenius Medical Care and GSK since 2012 and Vifor France since 2018, Sanofi-Genzyme from 2012 to 2015, Baxter and Merck Sharp and Dohme-Chibret (MSD France) from 2012 to 2017, Amgen from 2012 to 2020, Lilly France from 2013 to 2018, Otsuka Pharmaceutical from 2015 to 2020, and AstraZeneca from 2018 to 2021. The Inserm Transfert set up has managed this partnership since 2011.
Supplementary Material
Acknowledgments
We acknowledge the CKD-REIN study coordination staff for their efforts in setting up the CKD-REIN cohort: Dr. Marie Metzger, Dr. Elodie Speyer, Dr. Céline Lange, Dr. Reine Ketchemin, Dr. Natalia Alencar de Pinho, and all of the clinical research associates. We thank CKD-REIN investigators and clinical site staff listed in Supplemental Summary 1, and Mr. NGUEYEN SIME Willy, biostatistician in the Centre d’Investigation Clinique, CHRU Nancy, for his precious help. All legal authorizations were obtained including those from the Comité consultatif sur le traitement de l'information en matière de recherche dans le domaine de la santé (12.360), the Commission nationale de l’informatique et des libertés (DR-2012-469), and from the Kremlin-Bicêtre Comité de protection des personnes (IDRCB 2012-A00902-41). CKD-REIN biological collection is registered in the management application of the COnservation D'Eléments du COrps Humain (2012-1624). The Institut national de la santé et de la recherche médical (Inserm) Institutional Review Board approved the study protocol (IRB00003888). ClinicalTrials.gov: NCT03381950.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
See related editorial, “Can We Turn the Symptom Curve? Symptom Trajectories and Outcomes among Patients with CKD,” on pages 1586–1587.
Author Contributions
C. Ayav, M. Faye, L. Frimat, K. Legrand, and B. Stengel conceptualized the study; M. Faye was responsible for the data curation; M. Faye, K. Leffondré, L. Le Gall, and A.Y. Omorou were responsible for the formal analysis; N. Alencar de Pinho, C. Ayav, C. Combe, D. Fouque, L. Frimat, C. Jacquelinet, M. Laville, S. Liabeuf, R. Pecoits-Filho, Z.A. Massy, E. Speyer, and B. Stengel were responsible for the funding acquisition; N. Alencar de Pinho, C. Ayav, C. Combe, D. Fouque, L. Frimat, C. Jacquelinet, M. Laville, S. Liabeuf, K. Legrand, Z.A. Massy, R. Pecoits-Filho, E. Speyer, and B. Stengel were responsible for the investigation; C. Ayav, M. Faye, L. Frimat, K. Leffondré, L. Le Gall, K. Legrand, and B. Stengel were responsible for the methodology; C. Ayav, L. Frimat, K. Legrand, and B. Stengel were responsible for the project administration; L. Frimat and B. Stengel were responsible for the resources; M. Faye was responsible for the software; C. Ayav, L. Frimat, and K. Legrand provided supervision; C. Ayav, L. Frimat, K. Leffondré, L. Le Gall, K. Legrand, and B. Stengel were responsible for the validation; M. Faye was responsible for the visualization; M. Faye and K. Legrand wrote the original draft; and N. Alencar de Pinho, C. Ayav, C. Combe, M. Faye, L. Frimat, C. Jacquelinet, S. Liabeuf, M. Laville, K. Leffondré, L. Le Gall, Z.A. Massy, A.Y. Omorou, R. Pecoits-Filho, E. Speyer, and B. Stengel reviewed and edited the manuscript.
Supplemental Material
This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.06140522/-/DCSupplemental.
Supplemental Summary 1. CKD-REIN clinical sites and investigators, by region.
Supplemental Table 1. Definitions of the operational variables used in the study.
Supplemental Table 2. Summary table of the statistical criteria for fitting and classification of the different n-class models tested.
Supplemental Table 3. Posterior classification table of the two-class model.
Supplemental Table 4. Comparison of participants included in the analysis (n=2787) and those not included (n=246).
Supplemental Figure 1. Observed individual symptoms trajectories of 50 randomly selected participants a posterior classified in each class.
Supplemental Figure 2. Time course of individual symptoms according to trajectories.
Supplemental Figure 3. Proportion of severe symptoms (a lot or very much) at baseline according to trajectory.
Supplemental Item 1. Main analysis strategies.
References
- 1.Lockwood MB, Lash JP, Pauls H, Chung SY, Samra M, Ryan C, Park C, DeVon H, Bronas UG: Physical symptom cluster subgroups in chronic kidney disease. Nurs Res 69: 100–108, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jager KJ, Ocak G, Drechsler C, Caskey FJ, Evans M, Postorino M, Dekker FW, Wanner C: The EQUAL study: A European study in chronic kidney disease stage 4 patients. Nephrol Dial Transplant Off Publ Eur Dial Transpl Assoc - Eur Ren Assoc 27: 27–31, 2012 [DOI] [PubMed] [Google Scholar]
- 3.Kalantar-Zadeh K, Lockwood MB, Rhee CM, Tantisattamo E, Andreoli S, Balducci A, Laffin P, Harris T, Knight R, Kumaraswami L, Liakopoulos V, Lui S-F, Kumar S, Ng M, Saadi G, Ulasi I, Tong A, Li PK-T: Patient-centred approaches for the management of unpleasant symptoms in kidney disease. Nat Rev Nephrol 18: 185–198, 2022 [DOI] [PubMed] [Google Scholar]
- 4.Almutary H, Douglas C, Bonner A: Multidimensional symptom clusters: An exploratory factor analysis in advanced chronic kidney disease. J Adv Nurs 72: 2389–2400, 2016 [DOI] [PubMed] [Google Scholar]
- 5.Fletcher BR, Damery S, Aiyegbusi OL, Anderson N, Calvert M, Cockwell P, Ferguson J, Horton M, Paap MCS, Sidey-Gibbons C, Slade A, Turner N, Kyte D: Symptom burden and health-related quality of life in chronic kidney disease: A global systematic review and meta-analysis. PLoS Med 19: e1003954, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yapa HE, Purtell L, Chambers S, Bonner A: The relationship between chronic kidney disease, symptoms and health-related quality of life: A systematic review. J Ren Care 46: 74–84, 2020 [DOI] [PubMed] [Google Scholar]
- 7.Verberne WR, van den Wittenboer ID, Voorend CGN, Abrahams AC, van Buren M, Dekker FW, van Jaarsveld BC, van Loon IN, Mooijaart SP, Ocak G, van Delden JJM, Bos WJW: Health-related quality of life and symptoms of conservative care versus dialysis in patients with end-stage kidney disease: A systematic review. Nephrol Dial Transplant 36: 1418–1433, 2021 [DOI] [PubMed] [Google Scholar]
- 8.Ren Q, Shi Q, Ma T, Wang J, Li Q, Li X: Quality of life, symptoms, and sleep quality of elderly with end-stage renal disease receiving conservative management: A systematic review. Health Qual Life Outcomes 17: 78, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Almutary H, Bonner A, Douglas C: Which patients with chronic kidney disease have the greatest symptom burden? A comparative study of advanced CKD stage and dialysis modality: which patients with chronic kidney disease have the greatest symptom burden? A comparative study of advanced CKD stage and dialysis modality. J Ren Care 42: 73–82, 2016 [DOI] [PubMed] [Google Scholar]
- 10.Abdel-Kader K, Unruh ML, Weisbord SD: Symptom burden, depression, and quality of life in chronic and end-stage kidney disease. Clin J Am Soc Nephrol 4: 1057–1064, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Almutary H, Douglas C, Bonner A: Towards a symptom cluster model in chronic kidney disease: A structural equation approach. J Adv Nurs 73: 2450–2461, 2017 [DOI] [PubMed] [Google Scholar]
- 12.Amro A, Waldum B, von der Lippe N, Brekke FB, Dammen T, Miaskowski C, Os I: Symptom clusters predict mortality among dialysis patients in Norway: A prospective observational cohort study. J Pain Symptom Manage 49: 27–35, 2015 [DOI] [PubMed] [Google Scholar]
- 13.Davison SN, Jhangri GS: The impact of chronic pain on depression, sleep, and the desire to withdraw from dialysis in hemodialysis patients. J Pain Symptom Manage 30: 465–473, 2005 [DOI] [PubMed] [Google Scholar]
- 14.Porter AC, Lash JP, Xie D, Pan Q, DeLuca J, Kanthety R, Kusek JW, Lora CM, Nessel L, Ricardo AC, Wright Nunes J, Fischer MJ; CRIC Study Investigators : Predictors and outcomes of health-related quality of life in adults with CKD. Clin J Am Soc Nephrol 11: 1154–1162, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Almutary H, Bonner A, Douglas C: Symptom burden in chronic kidney disease: A review of recent literature. J Ren Care 39: 140–150, 2013 [DOI] [PubMed] [Google Scholar]
- 16.Lockwood MB, Chung S, Puzantian H, Bronas UG, Ryan CJ, Park C, DeVon HA: Symptom cluster science in chronic kidney disease: A literature review. West J Nurs Res 41: 1056–1091, 2019 [DOI] [PubMed] [Google Scholar]
- 17.Thong MSY, van Dijk S, Noordzij M, Boeschoten EW, Krediet RT, Dekker FW, Kaptein AA, Netherlands Co-operative Study on the Adequacy Study Group: Symptom clusters in incident dialysis patients: Associations with clinical variables and quality of life. Nephrol Dial Transplant Off Publ Eur Dial Transpl Assoc - Eur Ren Assoc 24: 225–230, 2009 [DOI] [PubMed] [Google Scholar]
- 18.Grams ME, Surapaneni A, Appel LJ, Lash JP, Hsu J, Rosas SE, Fink JC, Scialla JJ, Sondheimer J, Hsu C, Cheung AK, Jaar BG, Navaneethan S, Cohen DL, Schrauben S, Rao P, Feldman HI: Clinical events and patient-reported outcome measures during CKD progression: Findings from the CRIC study. 27 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wulczyn KE, Zhao SH, Rhee EP, Kalim S, Shafi T: Trajectories of uremic symptom severity and kidney function in patients with chronic kidney disease. Clin J Am Soc Nephrol 17: 496–506, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Murtagh FEM, Sheerin NS, Addington-Hall J, Higginson IJ: Trajectories of illness in stage 5 chronic kidney disease: A longitudinal study of patient symptoms and concerns in the last year of life. Clin J Am Soc Nephrol 6: 1580–1590, 2011 [DOI] [PubMed] [Google Scholar]
- 21.Stengel B, Combe C, Jacquelinet C, Briançon S, Fouque D, Laville M, Frimat L, Pascal C, Herpe Y-E, Deleuze J-F, Schanstra J, Pisoni RL, Robinson BM, Massy ZA: The French Chronic Kidney Disease-Renal Epidemiology and Information Network (CKD-REIN) cohort study. Nephrol Dial Transplant 29: 1500–1507, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Stengel B, Metzger M, Combe C, Jacquelinet C, Briançon S, Ayav C, Fouque D, Laville M, Frimat L, Pascal C, Herpe Y-E, Morel P, Deleuze J-F, Schanstra JP, Lange C, Legrand K, Speyer E, Liabeuf S, Robinson BM, Massy ZA: Risk profile, quality of life and care of patients with moderate and advanced CKD: The French CKD-REIN Cohort Study. Nephrol Dial Transplant 34: 277–286, 2019 [DOI] [PubMed] [Google Scholar]
- 23.Couchoud C, Stengel B, Landais P, Aldigier J-C, de Cornelissen F, Dabot C, Maheut H, Joyeux V, Kessler M, Labeeuw M, Isnard H, Jacquelinet C: The renal epidemiology and information network (REIN): A new registry for end-stage renal disease in France. Nephrol Dial Transplant Off Publ Eur Dial Transpl Assoc - Eur Ren Assoc 21: 411–418, 2006 [DOI] [PubMed] [Google Scholar]
- 24.Hays RD, Kallich JD, Mapes DL, Coons SJ, Carter WB: Development of the kidney disease quality of life (KDQOL) instrument. Qual Life Res 3: 329–338, 1994 [DOI] [PubMed] [Google Scholar]
- 25.Boini S, Frimat L, Kessler M, Briançon S, Thilly N: Predialysis therapeutic care and health-related quality of life at dialysis onset (The pharmacoepidemiologic AVENIR study). Health Qual Life Outcomes 9: 7, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Proust-Lima C, Philipps V, Liquet B: Estimation of extended mixed models using latent classes and latent processes: The R Package lcmm. J Stat Softw 78: 1–56, 2017 [Google Scholar]
- 27.Boucquemont J, Loubère L, Metzger M, Combe C, Stengel B, Leffondre K; the NephroTest Study Group : Identifying subgroups of renal function trajectories. Nephrol Dial Transplant 32: ii185–ii193, 2017 [DOI] [PubMed] [Google Scholar]
- 28.Thiébaut R, Walker S: When it is better to estimate a slope with only one point. QJM 101: 821–824, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Nylund KL, Asparouhov T, Muthén BO: Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Struct Equ Model Multidiscip J 14: 535–569, 2007 [Google Scholar]
- 30.Flythe JE, Hilliard T, Lumby E, Castillo G, Orazi J, Abdel- Rahman EM, Pai AB, Rivara MB, St Peter WL, Weisbord SD, Wilkie CM, Mehrotra R; Kidney Health Initiative Prioritizing Symptoms of ESRD Patients for Developing Therapeutic Interventions Stakeholder Meeting Participants : Fostering innovation in symptom management among hemodialysis patients: Paths forward for insomnia, muscle cramps, and fatigue. Clin J Am Soc Nephrol 14: 150–160, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Evangelidis N, Sautenet B, Madero M, Tong A, Ashuntantang G, Sanabria LC, de Boer IH, Fung S, Gallego D, Levey AS, Levin A, Lorca E, Okpechi IG, Rossignol P, Sola L, Usherwood T, Wheeler DC, Cho Y, Howell M, Guha C, Scholes-Robertson N, Widders K, Gonzalez AM, Teixeira-Pinto A, Viecelli AK, Bernier-Jean A, Anumudu S, Dunn L, Wilkie M, Craig JC; SONG-CKD Investigators : Standardised Outcomes in Nephrology - Chronic Kidney Disease (SONG-CKD): A protocol for establishing a core outcome set for adults with chronic kidney disease who do not require kidney replacement therapy. Trials 22: 612, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Murtagh FEM, Murphy E, Sheerin NS: Illness trajectories: An important concept in the management of kidney failure. Nephrol Dial Transplant 23: 3746–3748, 2008 [DOI] [PubMed] [Google Scholar]
- 33.Greenhalgh J: The applications of PROs in clinical practice: What are they, do they work, and why? Qual Life Res 18: 115–123, 2009 [DOI] [PubMed] [Google Scholar]
- 34.Etkind SN, Daveson BA, Kwok W, Witt J, Bausewein C, Higginson IJ, Murtagh FEM: Capture, transfer, and feedback of patient-centered outcomes data in palliative care populations: does it make a difference? A systematic review. J Pain Symptom Manage 49: 611–624, 2015 [DOI] [PubMed] [Google Scholar]
- 35.Duncanson E, Bennett PN, Viecelli A, Dansie K, Handke W, Tong A, Palmer S, Jesudason S, McDonald SP, Morton RL; Symptom monitoring WIth Feedback Trial (SWIFT) Investigators : Feasibility and acceptability of e-PROMs data capture and feedback among patients receiving haemodialysis in the Symptom monitoring WIth Feedback Trial (SWIFT) pilot: Protocol for a qualitative study in Australia. BMJ Open 10: e039014, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Aiyegbusi OL, Kyte D, Cockwell P, Marshall T, Dutton M, Walmsley-Allen N, Slade A, McMullan C, Calvert M: Patient and clinician perspectives on electronic patient-reported outcome measures in the management of advanced CKD: A qualitative study. Am J Kidney Dis 74: 167–178, 2019 [DOI] [PubMed] [Google Scholar]
- 37.Breckenridge K, Bekker HL, Gibbons E, van der Veer SN, Abbott D, Briançon S, Cullen R, Garneata L, Jager KJ, Lønning K, Metcalfe W, Morton RL, Murtagh FEM, Prutz K, Robertson S, Rychlik I, Schon S, Sharp L, Speyer E, Tentori F, Caskey FJ: How to routinely collect data on patient-reported outcome and experience measures in renal registries in Europe: An expert consensus meeting. Nephrol Dial Transplant Off Publ Eur Dial Transpl Assoc - Eur Ren Assoc 30: 1605–1614, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Morton RL, Lioufas N, Dansie K, Palmer SC, Jose MD, Raj R, Salmon A, Sypek M, Tong A, Ludlow M, Boudville N, McDonald S: Use of patient-reported outcome measures and patient-reported experience measures in renal units in Australia and New Zealand: A cross-sectional survey study. Nephrology (Carlton) 25: 14–21, 2020 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.