Venue: The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27708-0120

 

Presentation

An application of conjoint analysis to assess women's preferences for modified cervical cancer screening programs

Authors: Colleen McCormick (Johns Hopkins Bloomberg School of Public Health); John F.P. Bridges (Johns Hopkins Bloomberg School of Public Health)

Presenter: John Bridges (Johns Hopkins Bloomberg School of Public Health)

Discussant: Derek S. Brown (RTI International)

Session: Screening

Room: Seminar B

When: Wednesday 10:30 a.m. - noon

Background: The knowledge that persistent infection with high risk types of Human Papillomavirus (HPV) causes cervical cancer provides an opportunity to revise screening programs. However, women and providers have been reticent to adopt previous changes in screening guidelines. Understanding how patients value different aspects of screening can help policy-makers design programs that will be acceptable and highlight where efforts may be necessary to ensure acceptance of a new program.

Methods: An adaptive choice-based conjoint analysis was designed. It was administered to 52 women attending an outpatient gynecology clinic. Their utilities for frequency of screening, sensitivity of the test, specificity of the test, who performs the test and if a speculum is needed, cost, whether they learn their HPV status, and whether a provider discussed their results were determined.

Results: Women placed the highest value on high sensitivity (0.418). They also strongly valued learning their HPV status and having a provider discuss their results with them (utility co-efficients of 0.371 and 0.367, respectively). Low sensitivity was the most strongly disfavored (-0.503), followed by not learning their HPV status (-0.450) and then not having their results explained to them (-0.446). Average importances were calculated, to determine how much each attribute contributed to determining women's preferences. Importance was relatively evenly distributed across all the attributes, but sensitivity of the test was the most important aspect (18.34%) followed by whether the doctor explained the results (16.56%). The cost of the test was the least important attribute (7.51%), followed by frequency (12.42%).

Conclusion: This study is the first to use conjoint analysis to examine women's preferences for cervical cancer screening, and provides insight into what attributes are important to women. Women prefer sensitive tests, learning their HPV status, and having a provider explain their results. Programs that incorporate these attributes are likely to be accepted by most women.