Results During the study period, , unique patients completed a total of , surveys overall completion rate Scores were similar across dialysis modalities. Ceiling effects were observed for SPKD. This question survey instrument was published in [ 3 ], based upon a longer KDQOL instrument first developed in [ 4 ]. Here, we sought to understand the survey response rates, distribution of component scores, and distribution of responses to individual items among a large, nationally representative contemporary cohort of US patients with ESRD. We also sought to examine correlations between component scores, individual item responses, and fluid status.
|Published (Last):||21 May 2015|
|PDF File Size:||1.34 Mb|
|ePub File Size:||1.47 Mb|
|Price:||Free* [*Free Regsitration Required]|
Received Aug 29; Accepted Dec This article has been cited by other articles in PMC. Associated Data Data Availability Statement The data that support the findings of this study are available from DaVita Clinical Research but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available.
Data are potentially available from the authors upon reasonable request and with permission of DaVita Clinical Research. Abstract Background For older adults receiving dialysis, health-related quality of life is not often considered in prognostication of death or future hospitalizations.
To determine if routine health-related quality of life measures may be useful for prognostication, the objective of this study is to determine the extent of association of Kidney Disease Quality of Life KDQOL subscales with adverse outcomes in older adults receiving dialysis.
We adjusted for sociodemographic variables, hemodialysis access type, laboratory values, and Charlson index. Cohort members with a SF physical component summary PCS in the lowest quintile had an increased adjusted risk of death [hazard ratio HR , 1.
Cohort members with a SF mental component summary in the lowest quintile had an increased risk of hospitalization HR, 1. In adjusted analyses, there was no association between the symptoms of kidney disease, effects of kidney disease, and burden of kidney disease subscales with time to death or first hospitalization.
Competing risk models showed similar HRs. Conclusions Among the KDQOL subscales, the SF PCS demonstrates the strongest association with both death and future hospitalizations in older adults receiving hemodialysis Further research is needed to assess the value this subscale may add to prognostication. Electronic supplementary material The online version of this article Keywords: Geriatric nephrology, Health-related quality of life, Competing risks, SF, Prognostication Background Health-related quality of life is increasingly recognized as an important patient-centered outcome for older adults with kidney disease, many of whom have limited life-expectancy and a significant symptom burden [ 1 ].
Because health-related quality of life instruments assess self-rated health through items related to physical health, mental health, symptoms and limitations, another potential use for health-related quality of life assessment may be prognostication of adverse outcomes in these patients. Such prognostic information would help clinicians identify a subset of patients at increased risk for death and hospitalization who may benefit from interventions both to reduce the risk of these outcomes and to prepare for treatment decisions that may lie ahead [ 2 , 3 ].
Several studies in general population cohorts of older adults, older adults with specific chronic conditions, and cohorts of kidney disease in adults of all ages demonstrate the relationship between health-related quality of life and survival [ 4 — 18 ].
To our knowledge, prior studies have not examined the association of health-related quality of life with mortality or hospitalizations in cohorts limited to prevalent older adults receiving maintenance dialysis. While many of the available prognostic tools in older adults with kidney disease include markers suggestive of functional disability e. Because functional status often determines health-related quality of life and is a stronger predictor of adverse outcomes than disease-specific measures in older adults in the general population [ 24 ], we hypothesized that health-related quality of life from the commonly administered Kidney Disease Quality of Life instrument KDQOL may be useful for prognostication.
Towards testing this hypothesis, we first need to determine the extent of association of KDQOL subscale scores with adverse outcomes in older adults. Thus, we conducted a cohort study to determine the strength of the association between subscales on the KDQOL and mortality and hospitalization in a cohort of older adults receiving maintenance dialysis.
For each cohort member, we received non-identifiable patient-level clinical data including dates of clinical events, comorbidities, and laboratory data from January 1, to December 31, We selected the cohort size and follow-up period to accommodate financial resource constraints.
Dialysis unit social workers either supplied a paper copy of the KDQOL for self-administration or they helped patients complete it during dialysis if they were unable to self-administer.
Of the patients included in this cohort, we excluded one patient from analyses because multiple KDQOL assessments with inconsistent responses were documented in a single day. An additional 6. Because patients receiving peritoneal dialysis PD often differ from hemodialysis patients, we excluded an additional cohort members who were receiving PD, resulting in final analytic sample size of Exposure variables The KDQOL is a item health-related quality of life instrument adapted from the original item KDQOL, an instrument principally developed to measure quality of life of dialysis patients [ 25 ].
Because there is no currently accepted overall KDQOL score that incorporates all of its subscales [ 26 ], and to be consistent with prior studies [ 6 ], we calculated scores for the following five subscales of the KDQOL separately: 1 SF physical component summary PCS , 2 SF mental component summary MCS , 3 Burden of kidney disease, 4 Symptoms of kidney disease and 5 and Effects of kidney disease.
KDQOL subscale scores ranged from 0 to , and lower scores indicated worse self-reported quality of life. Covariates We identified the following baseline characteristics present on the date of KDQOL administration in age, gender, race e. Outcomes Outcomes of interest were time to death and time to first hospitalization and were measured from date of KDQOL administration through event date.
We defined first hospitalization as the date of a hospitalization occurring in Among with complete KDQOL data, five deaths and 37 hospitalizations occurred before the documented date of KDQOL administration, so cohort members with these events were not included in survival analyses. Cox analyses were repeated, modeling individual subscales in isolation. We used Lowess curve deviation for horizontal of Martingale residuals to assess linearity of continuous covariates.
We assessed categorical covariates for violation of the proportional hazards assumptions i. Because serum albumin did not meet the proportional hazards assumption, we performed stratified Cox models to allow the form of the underlying hazard function to vary across levels of albumin. For the Cox regression models, subjects were censored at the date of the event death or first hospitalization after KDQOL administration or at the end of the study period, December 31, We evaluated goodness of fit using Schoenfeld residuals, Cox-Snell residuals, and differences in fit due to observation deletion.
We assessed the joint relevance of including the KDQOL subscales in models for death and hospitalization by comparing fit between models with and without the subscales to obtain a global likelihood ratio LR statistic [ 32 ].
For the competing risk analysis, KDQOL assessment was treated as study start date, first hospitalization post assessment as event of interest, and death as competing event. We used the Fine and Gray method to estimate unadjusted and multi-adjusted sub-distribution relative hazards of hospitalization [ 34 , 35 ]. Surviving participants with no hospitalizations during the study period were right censored. All analyses were performed using Stata version SE Results Cohort characteristics Of the patients in the analytic cohort, at baseline, the average age was Table 1.
Kidney Disease Quality of Life Instrument (KDQOL)
Kidney Disease Quality of Life instrument™ - 36 items (KDQOL-36™ Survey)