The researchers created a classic scoring algorithm — nomogram — and boosting algorithm to compare predictive accuracy. The study authors examined 649 patients from 2002 to 2014 and used a multivariate Cox model to determine factors associated with survival. The factors they found associated with decreased survival were:
1. Older age
2. Poor performance status
3. Primary cancer type
4. Spine metastasis
5. Lung and/or liver metastasis
6. Brain metastasis
7. Systemic therapy for cancer before a surgical procedure (chemotherapy, immunotherapy, hormone therapy)
8. Higher white blood cell count
9. Lower hemoglobin levels
The researchers found “the nomogram was more reliable at estimating survival on the test data sets, with an accuracy of 0.75 (30 days), 0.73 (90 days) and 0.75 days (365 days).”
The boosting algorithm was comparable to the nomogram for testing data sets, but the researchers found the nomogram was easier to apply and can be more useful as a surgical decision-making aid.
More articles on spine surgery:
Spine Hospital of Louisiana honored with Press Ganey award: 3 highlights
26 latest statistics & notes on neurosurgeon salary—Hourly, annual & regional pay
Is race an independent predictor of unplanned 30-day readmission after spine surgery? 5 observations
