Kidney Disease Quality of Life Subscales and Risk of Death or Hospitalization in Older Patients

For older patients with kidney disease, many of whom have limited life expectancy and a significant burden of symptoms and comorbidities, health-related quality of life is increasingly being recognized as an important patient-centered outcome. Instruments designed to measure health-related quality of life include items related to physical health, mental health, symptoms, and limitations. In older patients with kidney disease, prognostication of adverse outcomes in that patient population may be possible utilizing health-related quality of life assessment instruments.

There are few data available on the association of health-related quality of life and mortality or hospitalizations in cohorts limited to older adults receiving maintenance dialysis. Rasheeda K. Hall, MD, and colleagues conducted a longitudinal cohort study designed to determine the extent of the association of Kidney Disease Quality of Life-36 (KDQOL-36) subscales with adverse outcomes in older adults on maintenance dialysis. The researchers sought to test the hypothesis that health-related quality of life from the KDQOL-36 may be useful for prognostication. Results of the study were reported online in BMC Nephrology [doi:10.1186/s12882-017-0801-5].

The cohort included 3500 adults ≥75 years of age receiving dialysis in the United States in 2012 and 2013. Cox and Fine and Gray models were used to evaluate the association of KDQOL-36 subscales with the risk of hospitalization and death. The models were adjusted for sociodemographic variables, hemodialysis access type, laboratory values, and Charlson index scores.

The analytic cohort included 3132 patients. At baseline, average age was 80.5 years, 50.1% (n=1570) were male, and 22.9% (n=675) were dual eligible Medicare and Medicaid beneficiaries. Mean time since initiation of dialysis was 5.98 years and average Charlson comorbidity index was 7.4. The majority of the cohort (64.5%, n=2018) had an arteriovenous fistula. Mean KDQOL-36 subscale scores were 345 for the 12-item Short Form Health Survey (SF-12) physical component score; 50.9 for the SF-12 mental component score; 78.5 for symptoms/problems; 74.3 for effects of kidney disease; and 52.6 for burden of kidney disease. No one in the cohort achieved the maximum score for SF-12 physical component or mental component score.

The cohort was stratified into quintiles based on scores for five subscales of the KDQOL-36: (1) SF-12 physical component summary; (2) SF-12 mental component summary; (3) burden of kidney disease; (4) symptoms of kidney disease; and (5) effects of kidney disease. Scores of patients in the first quintile were the lowest, those in the fifth quintile were the highest.

During the study period, 880 individuals died following completion of the KDQOL-36 (median follow-up was 512 days). The risk of death was higher among individuals in the first quintile (relative to the fifth quintile) for all five subscales in unadjusted analyses. The associations were attenuated following adjustment for sociodemographic characteristics, time since initiation of dialysis, Charlson index, access type, and laboratory values.

When all five KDQOL-36 subscales were combined in a model with covariates, only the participants in the first quintile maintained a significantly higher hazard of death (hazard ratio [HR], 1.55; 95% confidence interval [CI], 1.19-2.03). Race, gender, time since initiation of dialysis, hemoglobin, and Charlson index also maintained significant associations with time to death.

Of the 3132 study participants, 64.6% (n=2023) were hospitalized at least once after the date of completion of the KDQOL-36 over the study period (median follow-up of 203 days). Among those, 36.1% (n=730) later died during the observation period. However, 92.2% (n=673) died at least 7 days after the hospitalization. The average length of stay was 5.6 days.

In unadjusted analyses, risk of hospitalization was higher in participants in the first quintile for all five KDQOL-36 subscales. The associations were attenuated following adjustment for sociodemographic characteristics, time since initiation of dialysis, Charlson index, access type, and laboratory values. When all five subscales were combined in a model with covariates, the higher hazard of hospitalization was maintained in participants in the first quintile for the SF-12 mental component summary (HR, 1.39; 95% CI, 1.17-1.65) and the SF-12 physical component summary (HR, 1.29; 95% CI, 1.09-1.54).

The researchers cited some limitations to the study, including deriving the cohort from a single large dialysis organization, perhaps limiting the generalizability of the findings to all patients ≥75 years of age worldwide, particularly those with shorter time since initiation of dialysis and those in the United States receiving dialysis in non-profit dialysis organizations or federal programs. The cohort included only those patients who completed the KDQOL-36, potentially limiting the generalizability of the findings to those who are able to complete the KDQOL-36. The available data did not include all measured confounders identified in previous studies, such as new medical diagnoses, medications, blood pressure, weight gain, lean body mass, depression, social support, level of care, and prior hospitalizations. In addition, there were insufficient data on the most common reasons for hospitalizations. Finally, the analyses did not capture serial KDQOL-36 data, preventing the researchers from assessing the association of worsening quality of life or functional status with mortality and hospitalizations.

In conclusion, the researchers said, “We report that among KDQOL-36 subscales, SF-12 physical component summary has the potential to enhance prognostication of survival and future hospitalizations in older adults receiving dialysis. If validated in other studies, this finding may support routine quality of life assessments and integration of health-related quality of life and clinical data into prediction tools that ultimately enhance risk stratification and shared decision-making for older adults receiving maintenance dialysis.”

Takeaway Points

  1. Prediction tools regarding risk of death and hospitalizations among older adults receiving maintenance dialysis often do not consider health-related quality of life in that patient population.
  2. Researchers conducted a longitudinal study of 3500 adults ≥75 years of age to examine the extent of association of subscales of the Kidney Disease Quality of Life (KDQOL-36) subscales with adverse outcomes in older patients on dialysis.
  3. The strongest association with mortality and future hospitalizations was seen with the 12-item Short Form Health Survey physical component summary.