The cost-effectiveness of community-based screening for oral cancer in high-risk males in the United States: a Markov decision analysis approach

Abstract

Objective: The 2004 US Preventative Services Task Force (USPSTF) guidelines do not recommend routinely screening adults for oral cancer given no proven mortality reduction. A large cluster-randomized controlled screening trial in Kerala, India, in 2005, however, reported a significant reduction in mortality for screened male tobacco and/or alcohol users. In the United States, office-based screening efforts targeting males of high risk (regular use of tobacco and/or alcohol) have been unsuccessful due to poor attendance. Given the newfound screening mortality benefit to this high-risk subpopulation, we sought to ascertain the cost-effectiveness threshold of a yearly, community outreach screening program for males more than 40 years regularly using tobacco and/or alcohol.

Study design: Markov decision analysis model; societal perspective.

Methods: A literature search was performed to determine event probabilities, health utilities, and cost parameters to serve as model inputs. Screen versus No-Screen strategies were modeled using assumptions and published data. The primary outcome was the difference in costs and quality-adjusted life-years (QALYs) between the two cohorts, representing the potential budget for a screening program. One-way sensitivity analysis was performed for several key parameters.

Results: The No-Screen arm was dominated with an incremental cost of $258 and an incremental effectiveness of -0.0414 QALYs. Using the $75,000/QALY metric, the maximum allowable budget for a screening program equals $3,363 ($258 + $3,105) per screened person over a 40-year time course.

Conclusion: Given the significant health benefits and financial savings via early detection in the screened cohort, a community-based screening program targeting high-risk males is likely to be cost-effective.

Figure 1

Simplified Model

Figure 2

Disease Stages Within Each Cohort