An official website of the United States government
A description of the formulas used for calculating Incidence Rates.
Polynomial Model
\(logit(I(x,k)) = a_{k} + b_{1}x + b_{2}x^{2} + b_{3}x^{3} + b_{4}x^{4} + b_{5}x^{5} + b_{6}x^{6} \)
where
- \(a_k\) is the logit of incidence of the k-th birth cohort,
- \(x \) is the age for which incidence is being calculated,
- \(b_1, b_2, . . ,b_6 \)are the estimated parameters for each of the polynomial terms.
The predictions used for this method have been considered suitable for many cancer sites, particularly those in which tumor progression and growth are modulated by hormonal factors.
Exponential Model
\(I(x,k) = exp(a_k)x^{b}\)
where \(a_k\) is a categorical birth cohort variable, \(x\) is the current age, and \(b\) is the slope parameter. The validity of this function has been previously determined to have a biological rationale for a general class of cancers, given the multistage theory of carcinogenesis.