[Editor’s Note: The University of Illinois has updated its news release on this study to address some of our criticisms. The revised release, which can be accessed here, features a new headline and a more extensive discussion of the study’s methods and limitations. The original release that we reviewed has been archived and is accessible here or from the “Read Original Release” link above.]
This release reports on a study that analyzed the Medicare billing records of elderly women diagnosed with cervical cancer during the 1990s and a matched control group of non-cancer patients — 1,200 patient records in all. It attempts to find a link between screening for cancer past age 65 and a reduced risk of cervical cancer. It states that women past 65 who continued to be screened by pap smears showed a reduced risk of cervical cancer, but it offers no information about the degree of risk reduction. Nor does it address concerns over possible harms from the screening or the cost of screening for women who are now told the screening is unnecessary because of their age. The headline claims a causal link that the study design cannot prove.
If this research showed that women who are now told not to get pap smears as a screening for possible cervical cancer should actually continue being screened beyond age 65, then that would be an important public health change. Unfortunately, the release offered little detail to support that claim. Screening is not a harmless activity and while it is invaluable in some cases for some patients, screening is not always beneficial. Stories and releases that suggest that are not serving the public well.
The release includes the statement, “Rosenblatt said there needs to be a more thorough cost-benefit analysis of conducting the screenings in elderly women,” but that’s hardly enough to earn a satisfactory rating in this category. Because these screenings are not recommended by physicians, insurance companies may not pay for them. The cost of getting regular screenings may be prohibitive for elderly women on a fixed income. Readers could benefit from knowing the average cost of such screening including the cumulative costs of multiple screenings over the years, screenings that are not now recommended beyond the age of 65.
This release offers no data to explain in detail the degree of benefits from continuing screening, other than to say, “We found that the newly diagnosed cervical cancer group was less likely to have had a Pap smear, compared with the control group,” and “These results imply that routine Pap tests are beneficial for preventing malignant cervical cancer in women over 65.” Neither of these statements — nor anything else in the release — quantifies the reduction in risk that screening beyond age 65 would bring.
Some actual numbers, given in percentages or statistics, would have been welcome here.
While the release does say that for some women, “screenings may be unnecessarily invasive,” and that “There needs to be further study of the benefits and risks of doing Pap smears in the elderly,” it does not mention the risk of both false positives (over-diagnosis) or false negatives (under-diagnosis) and their impact on elderly women, nor does it acknowledge that the screening carries risks of infection, among others.
While the lede of the release explains that the research, “confirms a link between routine Pap smear screenings and a lower risk of developing cervical cancer in women over age 65,” that cautious statement is preceded by a headline shouting “Routine screenings prevent cervical cancer in elderly women.” Unfortunately, this was a case-control study, an observational study, that cannot show causality, regardless of what the headline argues. If the first absolute statement the reader sees makes a definitive claim, caveats later in a story cannot effectively correct for that kind of exaggeration.
This release does not appear to be disease mongering. It does seem to encourage a proliferation of Pap smears.
The release makes no mention of the funding for this project, nor any statement regarding any possible conflict of interest. The release does, however, link to an abstract of the formal journal paper which offers information that the study was funded by the University of Illinois, so it would have been easy enough to include that in the release.
Pap smears are the routine technique for screening for cervical cancer so alternatives are not applicable in the case of this release.
It’s clear from the release that Pap smear cervical cancer screenings are readily available through private practices, public clinics and hospitals.
The release doesn’t make an explicit claim of novelty but is clear that the study challenges conventional thinking regarding the frequency of cervical cancer screens. Since current guidelines recommend against such screening for women beyond the age of 65, with some exceptions, research that actually suggests a reduction in cervical cancer risk by screening would seem novel enough to warrant a release. Unfortunately, this release offers scant detail which would support such a finding.
The release’s headline suggesting a causal link between reduced cervical cancer risk and screening pushes the envelope too far but the rest of the release, while scant on detail, is more conservative in its claims, warranting a satisfactory rating.