Page 34 - Screening for Cervical Cancer: Systematic Evidence Review

Chapter II. Methods

• Compared with a reference standard of histology or colposcopy;

• Verified by use of the reference standard within an average of three months interval

from the screening sample; and

• Reported in a fashion that allows completion of a 2x2 table relating the new method

to the reference standard.

In the case of new Pap testing technologies, these strict criteria also required that performance of

the new test could be compared to that of conventional Pap testing.

We screened 48 full articles to determine relevance for updating Key Question 2.

Overall, based on these strict inclusion criteria, only three articles were eligible for inclusion.

Virtually all excluded articles failed to use a clinical reference standard or did so only among

those with a positive test without confirmation of either all or a subset of test negative tests.

Because a secondary goal of our work was to update the Cervical Cytology report, we relaxed

our criteria to include studies that would have met the criteria for that report (which allows for a

cytology reference standard). In total, we abstracted three high-quality articles and five other

articles for the evidence tables and retained five for supplementary information.

For completeness, we added these new studies to the new technologies evidence table

originally produced for the Evaluation of Cervical Cytology. The updated table is reproduced in

its entirety in this report as Evidence Table 2, Appendix D. We did not identify any studies that

report rates of outcomes over time, such as rates of cervical dysplasia or cervical cancer among

cohorts of women receiving screening using conventional cytology compared to either new

technologies alone, or to a system that adds a new technology to conventional cytology. We

found no publications of randomized trials comparing individual outcomes by screening

modality.

31