Predicting factors in the last week of survival in elderly patients with terminal cancer: a prospective study in southern Taiwan.

BACKGROUND/PURPOSE: The proportion of elderly persons among terminal cancer patients is around 60% in Taiwan. Prediction of the last week of survival can be significant in palliative care units for planning patient management, fulfilling patients' wishes at the end of life, and meeting the needs of relatives. We aimed to find out the predictive factors for the last week of survival in elderly terminal cancer patients.

METHODS

We prospectively observed items based on previous research that we anticipated might influence survival among 459 consecutive elderly patients. Within 24 hours of admission, we collected data including clinical symptoms and signs, demographic information, and biochemical test results. Multivariate logistic regression analysis was performed to identify factors associated with patient mortality within 7 days.

RESULTS

Multivariate analysis indicated that the following factors were predictive of death: higher score of Eastern Cooperative Oncology Group (ECOG) performance status (OR, 2.018; p < 0.001), primary liver cancer (OR, 1.968; p = 0.024), male gender (OR, 2.885; p < 0.001), lower extremity muscle power (OR, 0.722; p = 0.0266), lower systolic blood pressure (OR, 0.985; p = 0.011), higher heart rate (OR, 1.017; p = 0.016), higher hemoglobin (OR, 1.216; p = 0.003), and higher blood urea nitrogen (BUN) (OR, 1.028; p < 0.001).

CONCLUSION

We propose that the probability of survival of less than a week can be predicted by our formula. This formula, which includes factors of demographic information (male gender, and presence of liver cancer), clinical signs (lower systolic blood pressure, higher heart rate, and lower mean extremity muscle power), and biochemical tests (elevated BUN, and higher hemoglobin), may help improve survival prediction in terminal elderly cancer patients.