Nomograms

Overview

Making informed medical decisions relies on accurate predictions of the possible outcomes. Paper-based nomograms provide an excellent medium for easily displaying risk probabilities and do not require a computer or calculator. The coefficients used to create the nomogram can be used to create a computer-based prediction tool. The use of nomograms should theoretically help physicians and patients make better treatment decisions. Providing predicted probabilities to patients should reduce the likelihood of regret of treatment choice, particularly when complications arise. However, nomograms are only as good as the data that were used in their creation, and no nomogram can provide a perfect prediction. Ultimately, the best evaluation of a nomogram is made by validating the prediction accuracy of a nomogram on an external data set and comparing the concordance index with another prediction method that was validated using the same data.

Deriving Outcome Probabilities

All medical decisions are based on the predicted probability of different outcomes. Imagine a 35-year-old patient, who presents to a physician with a 6-month history of cough. A doctor in Chicago may recommend a test for asthma, which is a common cause of chronic cough. If the same patient presented to a clinic in rural Africa, the physician may likely test for tuberculosis. Both physicians may be making sound recommendations based on the predicted probability of disease in their locale. These physicians are making clinical decisions based on the overall probability of disease in the population. These types of decisions are better than arbitrary treatment, but they treat all patients the same.

A more sophisticated method for medical decision making is risk stratification. Physicians will frequently assign patients to different risk groups when making treatment decisions. Risk group assignment will generally provide better predicted probabilities than estimating risk according to the overall population. In the previous cough example, there are a variety of other factors that may affect the predicted risk of tuberculosis (e.g., fever, exposure to tuberculosis, history of tuberculosis vaccine) that physicians are trained to explore. Most of the risk stratification performed in clinical practice is based on rough estimates that simply order patients into different levels of risk, such as “high risk,” “medium risk,” or “low risk.” Nomograms provide precise probability estimates that generally make more accurate assessments of risk.

Another problem with risk stratification arises when continuous variables are turned into categorical variables. Physicians frequently commit dichotomized cutoffs of continuous laboratory values to memory to guide clinical decision making. Imagine a new blood test for tuberculosis called “serum [Page 815]marker A.” Research shows that patients with serum marker A levels greater than 50 are at an increased risk for tuberculosis. In reality, patients with a value of 51 might have very similar risks compared with patients with a value of 49. In contrast, a patient with a value of 49 would be considered to have the same low risk as a patient whose serum level of marker A is 1. Nomograms allow for predictor variables to be maintained as continuous values while allowing numerous risk factors to be considered simultaneously. In addition, more complex models can be constructed that account for interactions.

...