Thinking Differently About Cervical Cancer Screening in High-Risk Populations Beth E. Meyerson, PhD, Richard A. Crosby, PhD, Barbara J. Van Der Pol, PhD, Gregory D. Zimet, PhD C ervical cancer has become endemic in the U.S., affecting about 12,000 women each year with dis- proportionate impact among African-American and Hispanic women. 1 Although use of Paps has dramat- ically reduced the incidence of cervical cancer in the U.S. and other industrialized countries, the level of endemic- ity has persisted for several years—likely evidence of a systems-level failure in public health. 2 There is an emerging recognition that traditional approaches to cervical cancer screening may have reached their limits, necessitating a re- evaluation based on emerging science. Policy conversations should continue to consider evolving diagnostic technology and also distinguish between screening and diagnosis in order to ensure that prevention efforts reach women who are likely at highest risk for cervical cancer. Two decades ago, the fıeld of cervical cancer preven- tion was dramatically altered by the realization that per- sistent infection with high-risk human Papillomavirus (HPV) is the primary oncogenic agent. 3 It is now estab- lished that two HPV types (Types 16 and 18) account for about 70% of cervical cancers, and remaining high-risk types constitute the balance. 4 Establishing the connection between HPV, a sexually transmitted infection, and cer- vical cancer was a critical fırst step. Translating the con- nection to health practice and policy for the reduction of cervical cancer remains challenging and has proven somewhat problematic in the U.S. Several issues warrant consideration. The fırst issue has to do with diagnostic and screening technologies and the policy that supports their use. Once the connection between cervical cancer and HPV was established, the collective task was to identify the relative effectiveness of the diagnostic tools. With the evolution of HPV DNA technology, it is no longer clear that Paps should be used as the primary means for detecting cervi- cal cancer in all settings. 5,6 This evolution of thinking is partially evidenced in the most recent iteration of U.S. cervical cancer screening guidelines, 7,8 as there remains a reliance on Paps as the primary screening tool in age- targeted populations. Caution persists about the use of HPV genotyping as cervical cancer screening, because of questions about test accuracy 9 and likely resolution of high-risk HPV in younger populations. 10 Questions relate to the high sen- sitivity to infection but low specifıcity to disease, which can lead to high rates of false-positives. The concern is that the harm done through false-positive results is po- tentially greater than the benefıts achieved through the higher sensitivity. These issues will be addressed as diag- nostic science evolves. The more important question in- volves how and where such technologies are used. Excessive screening is costly to the individual and to the healthcare system. The risk of false-positive fındings leads to unnecessary colposcopies and psychological bur- den. This has been shown not only in the case of HPV testing but also in the case of cytology when used as a screening tool for cervical cancer. Castle et al. 6 recently reported data suggesting that specifıcity of liquid-based cytology to detect CIN2 (cervical intraepithelial neo- plasia) is about 73%. Thus, repeated use of cytology as a screening tool for cervical cancer may exact a physical, psychological, and fınancial toll for at least one of every four women. Positive and negative predictive values (PPV and NPV) of any screening test are direct functions of HPV prevalence in the tested populations. A public health per- spective grounded in reducing health inequalities would suggest that if screening efforts increased among women most at risk (e.g., HIV-positive women, women newly diagnosed with STDs, and single women who report more than one sex partner in the past 12 months), then PPV estimates for HPV testing will improve. Evidence suggests that carcinogenic HPV strains become undetect- able by current commercially available screening assays from the healthy immune system in about 8 –10 months. 10,11 Thus, it is fair to presume that high-risk HPV would be absent among women who are not immune-compromised From the Department of Applied Health Science (Meyerson) and, the Department of Epidemiology & Biostatistics (Van Der Pol), Indiana Uni- versity, Bloomington, Indiana; Section of Adolescent Medicine (Zimet), Indiana University School of Medicine, Indianapolis, Indiana; and the Department of Health Behavior (Crosby), College of Public Health, Uni- versity of Kentucky, Lexington, Kentucky Address correspondence to: Beth E. Meyerson, PhD, Department of Applied Health Science, School of Health, Physical Education and Recre- ation, Indiana University, 1025 E. 7th street, Room C-006, Bloomington IN 47405. E-mail: bmeyerso@indiana.edu. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2012.04.021 © 2012 American Journal of Preventive Medicine • Published by Elsevier Inc. Am J Prev Med 2012;43(2):221–224 221