Researchers in Massachusetts say they have developed a model to assess the life-expectancy of kidney disease patients on dialysis that could help guide discussions about end-of-life care.
The model, which combines scores based on five simple factors, accurately predicted the likelihood of dying within six months in 87% of cases (95% CI 0.82 to 0.92), according to a published study.
The model was validated in a second cohort of patients whose six-month mortality risk was predicted accurately 80% of the time (95% CI 0.73 to 0.88).
Michael J. Germain, MD, of Baystate Medical Center in Springfield, and colleagues wrote, “The integrated prognostic model lends itself to risk stratification of patients, it is more specific and sensitive than any of its components … and it seems to be a considerable improvement over other existing instruments at predicting survival in the dialysis population.”
Unwilling to make patients feel hopeless, and lacking accurate tools for predicting life expectancy, nephrologists rarely have discussions with patients about prognosis and end-of-life care, the authors wrote. However, the Renal Physicians Association and American Society of Nephrology explicitly recommend doctors have such discussions.
“Clinically, this is necessary to allow patients and families to be informed of prognosis and provided with the opportunity to engage in meaningful advance-care planning discussions,” Germain and colleagues wrote.
“Depending on patient symptoms and preferences, the identification of a poor prognosis can lead to a palliative care consultation and hospice referral.”
To that end, researchers developed a model that relies on reviewing a patient’s chart for actuarial data and on the doctors’ response to a so-called “surprise question,” or SQ: “Would I be surprised if this patient died within the next six months?”
A negative answer suggested a lower life expectancy, as did older age, decreased serum albumin, presence of dementia, and presence of peripheral vascular disease. The six-month duration was chosen because it is the maximum life expectancy requirement for hospice eligibility.
The researchers developed the model using prospective data collected between July 2006 and September 2007 from a study 512 kidney disease patients on dialysis at five clinics in western Massachusetts. This allowed them to determine which factors best predicted life expectancy.
They found that the following factors put dialysis patients at increased risk of early mortality:
- Older age (hazard ratio [HR] for a 10-year increase 1.35; 95% CI 1.17 to 1.57)
- Dementia (HR 1.88; 95% CI 1.24 to 2.84)
- Peripheral vascular disease (HR 2.24; 95% CI 1.11 to 4.48)
- Decreased albumin (HR for a 1-U increase 0.27; 95% CI 0.15 to 0.50)
- Answering “No” to the SQ (HR 2.71; 95% CI 1.76 to 4.17)
When combining these factors to predict short-term life expectancy, the model was 87% accurate.
Once they had identified these five factors, the researchers tested the accuracy of their model between September and October of 2008 on 514 kidney disease patients on dialysis at eight New England clinics. In this group of patients, the model was 80% accurate.
The authors cautioned that their study was limited to patients living in New England, and that future studies are need to determine if the predictive model they developed is accurate in other regions and populations.
They also noted that the two groups of patients they studied had different proportions of patients on dialysis for longer than three months. Also, while the percentage of cases with missing information was low, there were some cases of missing data for nearly all the variables.
“Despite all of these considerations, the integrated prognostic model satisfactorily predicted mortality in the validation cohort,” they concluded.
Published online in the Clinical Journal of the American Society of Nephrology, Dec 2009