A simple and objective metric for verifying the data quality of cancer registries

12 Dec 2016 | News
In order to provide a more objective way of assessing the quality of cancer registry data, scientists at the EU’s Joint Research Centre investigated whether cancer incidence rates obeyed Benford's law, or the law of first digits.

In order to provide a more objective way of assessing the quality of cancer registry data, scientists at the EU’s Joint Research Centre investigated whether cancer incidence rates obeyed Benford's law, or the law of first digits. The study confirmed this is the case, opening up the potential to improve the evaluation of the quality of cancer registry data.

The reliability of population-based cancer registries data looking at the risk and burden of cancer in defined populations, implies a formal evaluation of their quality. This is particularly important for data comparisons across different populations and across different time periods, which is essential for informed policy making.

Traditional approach to assessing the different aspects of data quality, including validity, comparability, and completeness, makes use of a set of indices, which inevitably introduces an element of subjective judgement.

To provide a data-quality assessment based on more formal principles, the JRC Cancer Information group tested the adherence or violation of cancer registry data to Benford's law, which can be used to analyse the frequency distribution of the first digits in a set of numbers.

In many real-life sets of numbers, the probability of the first digit is strongly skewed towards the smallest values. Indeed, numbers which adhere to Benford's law have 1 as the first digit in more than 30 per cent of cases, 2 in 17 per cent, 3 in 12 percent, and so on up to 9, for which the probability of being the first digit is less than 5 per cent.

Not all number sets abide by Benford's law, but those which do must respect its theoretical distribution. This characteristic is exploited in forensic accounting and auditing to detect fraud, for example.

The JRC’s analysis of large data sets of population-based cancer-incidence rates showed adherence to Benford's law, thereby allowing a simple, fast, and objective test for verifying cancer-registry data.

Read more in: E. Crocetti and G. Randi: "Using the Benford's Law as a First Step to Assess the Quality of the Cancer Registry Data", Front Public Health 4 (2016) Article 225, doi:10.3389/fpubh.2016.00225

Read the report

 

 

Never miss an update from Science|Business:   Newsletter sign-up