In the absence of cancer registry data, is it sensible to assess incidence using hospital separation records?
2006

Using Hospital Records to Estimate Cancer Incidence

Sample size: 100000 publication Evidence: moderate

Author Information

Author(s): Moyra E Brackley, Margaret J Penning, Mary L Lesperance

Primary Institution: University of Victoria

Hypothesis

Can first hospitalisation act as a proxy measure for cancer incidence in the absence of registry data?

Conclusion

First hospital separation may be considered a proxy for incidence with reference to colorectal cancer since 1995.

Supporting Evidence

  • First hospitalisation counts consistently overestimate registry incidence counts.
  • Sensitivity for colorectal cancer is 84.2%, indicating a strong correlation.
  • Kappa values for agreement between hospital and registry data are consistently high.

Takeaway

The study looks at whether hospital records can help us understand how many people get cancer when we don't have complete data. It finds that for colorectal cancer, hospital records can be a good estimate.

Methodology

Data were drawn from the British Columbia Linked Health Data resource, including hospital records and cancer registry data from 1990 to 1999.

Potential Biases

Potential biases due to missing data and reliance on hospital records.

Limitations

The study did not have access to an independent gold standard for validation.

Participant Demographics

Data represents individuals from the Vancouver Island Health Authority in British Columbia.

Statistical Information

P-Value

p = 0.42 for lung cancer, p = 0.56 for colorectal cancer

Statistical Significance

p = 0.42 for lung cancer, p = 0.56 for colorectal cancer

Digital Object Identifier (DOI)

10.1186/1475-9276-5-12

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication