Estimating the completeness of prevalence based on cancer registry data

Area of Science:

• Epidemiology
• Biostatistics
• Cancer Research

Background:

• Cancer registries are crucial for epidemiological studies.
• Prevalence data from registries can be biased due to incomplete historical case inclusion.
• This bias affects the accuracy of cancer burden estimates.

Purpose of the Study:

• To estimate the relevance of incompleteness bias in cancer registry data.
• To develop a method for quantifying the completeness of registry-based prevalence estimates.
• To provide an index for evaluating the quality of cancer prevalence data.

Main Methods:

• Modeling incidence and relative survival as parametric functions for various cancer types.
• Computing prevalence estimates under different disease reversibility assumptions.
• Deriving an analytical ratio of observed-to-total prevalence as a completeness index.

Main Results:

• The study provides an analytical evaluation of the completeness index.
• The index is shown to be a function of disease characteristics and data collection parameters.
• This method allows for the quantification of bias in registry data.

Conclusions:

• Incompleteness bias is a significant issue in cancer registry prevalence data.
• The proposed index offers a quantitative measure to assess and correct for this bias.
• Accurate cancer prevalence estimation requires accounting for historical data limitations.