Putting new guidelines into practice is easier said than done
Steven Goldstein, MD
New York University Medical Center
Is the Pap test still necessary for every woman, every year? No, according to the latest guidelines, but old habits die hard, even for physicians.
And there is little doubt that yearly screening, though not scientifically based, has contributed much to the reduction of cervical cancer incidence and mortality in American women. Our patients and we as providers have long considered a Pap test the cornerstone of the annual gynecologic exam, as we’ve been urged to do for decades by our leading academic institutions. However, the American College of Obstetricians and Gynecologists (ACOG) and the American Cancer Society (ACS) revised their guidelines last year, and no longer support yearly screening for every woman, every year.1,2 The US Preventive Services Task Force (USPSTF) revised its guidelines in accord, with the exception that it found the evidence insufficient to support screening low-risk women more often than every 3 years, at any age.3
Original rationale: “Pap smear prompt”
These organizations, as well as the National Cancer Institutes (NCI), had supported annual Pap testing since the mid-1950s—long before any data suggested whether one screening interval might be better than another.
In fact, part of the original rationale for annual screening was that it would serve as a vehicle to bring women in for their annual gynecologic exam.4
Frequent screening would detect exceedingly few additional cancers, at an exceedingly high cost
We can confidently counsel patients
A previously well-screened woman over age 30 who has no history of dysplasia has an exceedingly small risk of cervical cancer, whether her next Pap test is 1, 2, or 3 years after her last.
How many cancers will we miss?
Miller MG, Sung HY, Sawaya GF, Kearney KA, Kinney W, Hiatt RA. Screening interval and risk of invasive squamous cell cervical cancer. Obstet Gynecol. 2003;101:29-37.
This matched case-control study assessed the odds of being diagnosed with squamous cell cervical cancer when a Pap test is performed 2 or 3 years versus 1 year after a normal Pap. Data from the Kaiser Permanente Medical Care Program in Northern California was used to identify 482 women who were diagnosed with invasive squamous cell cervical cancer between 1983 and 1995, and to compare each woman with 2 control individuals matched for age, race/ethnicity, and length of program membership. An intact cervix and no prior cervical, uterine, or vaginal cancer were required. A woman who had a Pap test within 18 months of her last negative test was half as likely to have invasive cancer as a woman who waited 3 years (31 to 42 months).
The odds ratios for invasive cancer diagnosed by screening at 1, 2, or 3 years were 1.00, 1.72, and 2.06, respectively. The differences between intervals of 2 or 3 years versus 1 year were significant. The odds ratios increased to 2.15 and 3.60, respectively, in women with at least 2 consecutive negative Pap tests prior to diagnosis.
In all analyses, the odds ratios continued to increase as screening intervals were prolonged beyond 3 years.
Increased relative risk and very small absolute risk. The new ACOG and ACS guidelines recommend extending the screening interval only for women over 30 who have been well screened over the previous decade. This study does not break down the relative risks by age, nor does the sample size allow assessment of the risks for women with more than 2 consecutive negative Paps.
The authors note that the age-adjusted incidence of invasive cervical cancer among all Northern California Kaiser Permanente members is only 6.2 per 100,000 women. In this well-screened population, even doubling the relative risk leaves a very small absolute risk of cervical cancer.
How many fruitless interventions?
Sawaya GF, McConnell KJ, Kulasingam SL, et al. Risk of cervical cancer associated with extending the interval between cervical-cancer screenings. N Engl J Med. 2003;349:1501-1509.
If screening were done annually rather than every 3 years, how many additional tests would be needed to diagnose each additional cancer expected to be found? To find out, Sawaya et al applied data from the National Breast and Cervical Cancer Early Detection Program to a Markov model. They studied 32,230 women with 3 successive negative Pap tests, each no more than 36 months apart.
They predicted that, in a theoretical cohort of 100,000 women who had at least 3 consecutive negative Pap tests, screening at 1-year rather than 3-year intervals would uncover 3 additional cancers in women aged 30 to 44, a single additional cancer in women aged 45 to 59, and no additional cancers for women 60 to 64 years of age.
They calculated that, for this theoretical cohort of 100,000 women:
- To find all 3 additional cancers in the 30- to 44-year-old group would require 69,665 Pap tests and 3,861 colposcopies.
- To find the only additional cancer in the 45- to 59-year-old group would require 209,324 Pap tests and 11,502 colposcopies.
As with all modeling studies, Sawaya’s analysis is limited by the assumptions introduced into the model.
- perfect compliance on the part of this cohort of hypothetical patients,
- use of conventional Pap tests only, and
- uniform sensitivity and specificity.