Reference: JAMA Intern Med. 2025 Jun 6 early online
Practice Point: Self-collected human papillomavirus (HPV) screening, much like fecal immunochemical testing (FIT), is a good option to improve screening rates in under-resourced populations.
EBM Pearl: With screening trials, keep your eyes peeled for authors reporting on an “enriched” group without being meticulous about stating that in the conclusion.
Our job is to be cynical and pedantic: It’s what they teach us in the EBM Focus Academy Boot Camp™. But in case you aren’t keeping up with the latest on cervical cancer prevention, some low-tech/cost-efficient innovations have our most jaded team members grinning. Aside from undisputed massive reductions in cervical cancer and other HPV-associated cancers in countries that embrace the HPV vaccine—we’re looking at you Australia!—the big news in cervical cancer prevention has been the recognition that screening for oncogenic serotypes of HPV is cheaper and more effective at predicting cancer and precancerous lesions than the traditional cytology-based Pap smear. Furthermore, unlike Pap smears, people can generally test themselves and avoid the emotional and physical discomfort of a pelvic exam. There is good evidence that self-collected “swabbing” for HPV presence and serotype is concordant with physician-visualized sampling, and self-collected testing is just as effective at case-finding at a lower cost and with much better patient satisfaction on an individual level than physician-administered testing.
HPV vaccination and removal of barriers to cancer screening can both be truly lifesaving, and here at the Focus HQ, we were happy to see authors of a recent study from JAMA Internal Medicine, the named PRESTIS trial (Prospective Evaluation of Self-Testing to Increase Screening), examine the question—in their own words—of “How do mailed self-collection kits, with and without patient navigation, compare to standard telephone reminders to increase cervical cancer screening in a safety-net setting?” The PRESTIS team had taken a group of individuals with low income in Houston, Texas (half of whom spoke Spanish as their primary language) who were being seen in health clinics but had not kept up with cervical cancer screening. Investigators phoned them all to remind them to get screened, and for one-third of them, that was the only intervention. A second group also received a home-test kit by mail that included instructions on how to collect their own sample. A third group received the initial phone call and also additional phone check-ins to answer questions and urge participation. In *this population*, the experiment was an astonishing success. Whereas people who only got telephone calls to come to a clinic had a 17% rate of being screened for HPV in 3 months, those who also got the self-testing home kit had a 41% screening rate and those with added check-ins had a 47% success rate for completing screening. That boils down to a screening difference of 24% comparing telephone reminders with self-testing along with a whopping 29% difference comparing telephone calls with guided self-testing!
Dear Reader, if you carefully look at the information above, you have all the clues to figure out where the authors overstated their conclusion. The authors specifically stated that they were comparing telephone calls with self-testing or guided self-testing. But in actuality, they were comparing telephone calls by themselves with telephone calls + self-testing or telephone reminders + guided self-testing.
Why does this detail matter? Because it turns out that although 3,740 patients were identified as being eligible for the program, only 2,577 were reachable by phone, and only those reached by phone were included in the analysis. The 2,577 patients willing and able to pick up the phone are considered an “enriched” group, which puts the analysis at risk of selection bias. Some might call that cherry-picking: If the researchers had mailed test-kits (perhaps with instructions to call the clinic with questions) to patients who did not pick up the phone, they might have been able to legitimately arrive at their stated conclusion. Alas, they did not do this. A true intention-to-treat analysis looking at the magnitude of effect for telephone reminders versus self-testing/guided self-testing should have an additional numerator for those not reached by phone and use a denominator of 3,740—not 2,577. Increasing the denominator reduces the percentage of successful screenings in all reported categories, bringing the screening difference for adding self-testing to telephone calls down from 24% to 16% and for adding guided self-testing to telephone calls down from 29% to 20%. If the authors had kept their conclusion statements to relative numbers between the three groups, they would have also been on firmer ground. By making a claim of magnitude of effect, in this case the difference between groups, they can fairly be critiqued for overstating their conclusion.
No matter, we still think this study itself was useful and that self-testing for high-risk HPV is a wonderful advance.
For more information, see the topic Cervical Cancer Screening in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Dan Randall, MD, MPH, FACP, Senior Deputy Editor at DynaMed. Edited by Alan Ehrlich, MD, FAAFP, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School; Katharine DeGeorge, MD, MS, Senior Deputy Editor at DynaMed and Associate Professor of Family Medicine at the University of Virginia; Gayle Sulik, PhD, Senior Medical Editor and Team Lead for Palliative Care at DynaMed; McKenzie Ferguson, PharmD, BCPS, Senior Clinical Writer at DynaMed; Rich Lamkin, MPH, MPAS, PA-C, Clinical Writer at DynaMed; Matthew Lavoie, BA, Senior Medical Copyeditor at DynaMed; Hannah Ekeh, MA, Senior Associate Editor II at DynaMed; and Jennifer Wallace, BA, Senior Associate Editor at DynaMed.