Preventive Medicine 50 (2010) 81–85
Contents lists available at ScienceDirect
Preventive Medicine j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y p m e d
Follow-up for cervical abnormalities in a managed care plan, 1999–2004☆ V.B. Benard a,⁎, N.D. Berkman b, T. Kuo b, C.K. Martin c, L.C. Richardson a a Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Cancer Prevention and Control, Mailstop K-55, 4770 Buford Hwy NE, Atlanta, GA 30341, USA b RTI International, Research Triangle Park, NC, USA c Center for Health Care Policy and Evaluation, Minneapolis, MN, USA
a r t i c l e
i n f o
Available online 22 November 2009 Keywords: Cervical cancer Claims data Abnormal Pap tests
a b s t r a c t Objective. The objective of this study was to determine the follow-up for women after receiving an abnormal Pap test before and after the updated American Society of Colposcopic and Cervical Pathology (ASCCP) guidelines for management of abnormal cytology. Methods. In 1999 and 2004, women who had been enrolled in a US health care plan for at least 21 months and were between 18 and 70 years of age were included. We calculated differences in type of followup between the time periods before and after ASCCP guideline changes in 2002. Results. Overall, 1.7 million women met study criteria and received at least one Pap test. Overall, 227,802 (14%) women received additional follow-up. Of these women, 73% had a repeat Pap test within 9 months as their ﬁrst follow-up, 13% received colposcopy, and 7% had other events. The proportion of women receiving a repeat Pap test decreased signiﬁcantly during the post-guideline time period. The odds of a woman receiving a colposcopy versus a repeat Pap test were 41% higher in the post-guideline period, after controlling for other variables. Conclusions. Our ﬁndings indicate that for the time period after the ASCCP guidelines changed, more colposcopies and fewer repeat Pap tests were performed as a follow-up of abnormal Pap test. © 2009 Elsevier Inc. All rights reserved.
Introduction Each year in the United States, approximately 66 million women have a Papanicolaou (Pap) test to screen for cervical cancer (Eltoum and Robertson, 2007). Of these, approximately 7% have a cytological abnormality requiring additional follow-up or evaluation (Jones and Davey, 2000). Determining which women with cytological abnormality are at risk for signiﬁcant cervical disease, performing appropriate diagnostic work-ups, and treating cancer precursors present a major public health challenge. In 2002, national evidencebased guidelines were published to assist clinicians in managing these women who are diagnosed with cytological abnormalities in the United States (Wright et al., 2002). Despite the existence of national guidelines for the management of women with abnormal Pap tests, evidence suggests that widespread variation still exists between published guidelines and current physician practices (Benard et al., 2005, 2008). Many primary care
☆ The ﬁndings and conclusions in this report are those of the author(s) and do not necessarily represent the ofﬁcial position of the Centers for Disease Control and Prevention. ⁎ Corresponding author. Epidemiology and Applied Research Branch, Division of Cancer Prevention and Control, Mailstop K-55, NCCDPHP, CDC, 4770 Buford Hwy NE, Atlanta, GA 30341, USA. E-mail address: [email protected]
(V.B. Benard). 0091-7435/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2009.11.005
physicians may be unaware of current guidelines or confused by different screening and management options (Holland-Barkis et al., 2006). Zapka et al. (2005) found that only one-half of a sample of primary physicians perceived cervical cancer guidelines as very useful, an additional one-third reported them as somewhat useful, and less than one-tenth reported that guidelines were not at all useful. Another provider survey of both knowledge and behavior found that more than 80% of all physicians who perform Pap tests reported that at least one national screening guideline was very inﬂuential in their practice; however, fewer than 25% followed guideline-consistent Pap screening recommendations for their practices (Yabroff, in press). Because 85% of the United States population is insured, health plan databases provide a rich and valuable source of data to describe current adherence to national guidelines (Schabert et al., 2008). Interim guidelines for the management of abnormal cervical cytology (Kurman et al., 1994) were updated with evidence-based recommendations in 2002 by the American Society of Colposcopy and Cervical Pathology (ASCCP) (Wright et al., 2002). As noted in Table 1, colposcopy rather than repeat Pap test became the preferred method of follow-up for most cytological abnormalities with the newer guidelines. In this study, we examined cervical cancer screening diagnosis and outcomes after follow-up of an abnormal Pap test within a large, geographically diverse United States health care organization during the period 1999 to 2004 to describe adherence to the updated comprehensive, evidence-based guidelines.
V.B. Benard et al. / Preventive Medicine 50 (2010) 81–85
Table 1 Comparison of guidelines for the management of abnormal cervical cytology, 1994 and 2002. Pap test results
NCI interim guidelines 1994
Repeat Pap 4–6 months
LSIL ASC-H HSIL, SqCa
Repeat Pap (Category added 2001) Colposcopy
Repeat Pap 4–6 months HPV DNA test Colposcopy Colposcopy Colposcopy
ASCUC: atypical squamous cells of undetermined signiﬁcance; LSIL: low-grade squamous intraepithelial lesions; ASC-H: atypical squamous cells, cannot rule out high-grade disease; HSIL: high-grade squamous intraepithelial lesions; SqCa: squamous cancer; NCI: National Cancer Institute; ASCCP: American Society of Colposcopy and Cervical Pathology.
Statistical analysis We documented the type and number of follow-up events after the index Pap. We deﬁned time periods as pre-ASCCP (January 1999–April 2002) and post-ASCCP (May 2002–December 2004), before and after the guidelines were updated, to determine whether providers adhered to the changes recommended for follow-up of abnormal Pap test results. We used multivariate modeling to evaluate the descriptive ﬁndings while controlling for other variables that might have inﬂuenced each woman's follow-up. The logistic regression model explored whether women received a colposcopy or a repeat Pap test as their ﬁrst follow-up event, both before and after the guideline change. Control variables included age, ﬁnal diagnosis during the follow-up period, primary insurance coverage, and insurance type, region of the country, and urbanicity. The model also controlled for ofﬁce setting and type of physician who performed the index Pap test. We used SAS 9.1.3 (SAS Institute Inc., Cary, NC) to complete all analyses.
Results Data source We obtained administrative and claims data for women enrolled in a large US health care organization from 1998 through 2005; this database provided 12 months of follow-up for cohorts of women for the study period of 1999 to 2004. Information obtained from administrative enrollment data included each woman's age, region, and insurance product type. Claims data provided procedure, diagnosis, provider specialty, and facility type. As a retrospective analysis of a de-identiﬁed database, the research was found to be exempt from Institutional Review Board review. Women were included in the cohort who were between 18 and 70 years of age and who had cervical cancer screening between January 1, 1999 and December 31, 2004. We assigned women an index date that reﬂected the ﬁrst record of cervical cancer screening during the study period after all eligibility criteria had been satisﬁed (referred to as the “index Pap”). Women who were eligible were enrolled in the health plan coverage for at least 9 months before the index Pap and for at least 12 months after (21 months of continuous health plan coverage), were not pregnant, and did not have an abnormal cervical diagnosis or procedure of interest in the 9-month period before the index Pap (N = 1.7 million). Diagnoses of interest included abnormal Pap test (International Classiﬁcation of Diseases [ICD]-9: 795, 795.x, 795.xx and 079.4), dysplasia (ICD-9: 622.1 and 622.9) and cervical cancer (ICD-9: 233.1, 180.0, 180.1, 180.8 and 180.9). Procedures of interest included cervical screening (ICD-9: 9146, V723, V762, V72.32; Healthcare Common Procedure Coding System (HCPCS): P3000, P3001, Q0091, G0101, G0123, G0124, G0141, G0143–G0145, G0147, G0148; Current Procedural Terminology (CPT): 88,142–88,145, 88,147, 88,148, 88,150–88,158, 88,160–88,162, 88,164–88,167, 88,172–88,175, 88,199), HPV screening (CPT: 87,620–87,622), colposcopy, biopsy, and excision of the cervix (CPT: 57,420, 57,452, 57,454–57,456, 57,460, 57,461, 57,500, 57,505, 57,510, 57,511, 57,513, 57,520–57,522, 57,530, 57,531, 57,540, 57,545, 57,550, 57,555, 57,556; ICD–9: 67.4, 67.11, 67.12, 67.19,: 67.2, 67.32, 67.33), and hysterectomy (ICD-9: 68.3, 68.31, 68.39, 68.4, 68.5, 68.51, 68.59, 68.6, 68.7, 68.9; CPT: 58,200, 58,210, 58,260, 58,262, 58,263, 58,267, 58,270, 58,275, 58,280, 58,285, 58,290–58,294, 58,550, 58,552– 58,554; HCCPCS: S2078). In our analyses, we considered ofﬁce visits as a follow-up to the index Pap test when an abnormal diagnosis code of interest was included on the claim. CPT codes used to specify an ofﬁce visit included 99,201–99,205, 99,211–99,215, 99,241–99,245, 99,271–99,275. Analytic cohort selection We included women in the cohort if they had: (1) a diagnosis of interest including an abnormal Pap test result, dysplasia, or cervical cancer within 12 months after the index Pap or (2) a repeat Pap test within 9 months of the index Pap, or (3) other procedures including a biopsy, a colposcopy, a conization, or an excision of the cervix (which as a group we will refer to as colposcopy) within 12 months of the index Pap test. Based on this deﬁnition, a woman would have been included because her index Pap test was abnormal (N = 227,802) and follow-up was needed. To improve speciﬁcity, we deﬁned new cases of dysplasia and cervical cancer only when we found both a diagnosis code and a procedure (biopsy, colposcopy, conization or excision of the cervix) within 1 year following the index screening.
Table 2 provides the characteristics of the 227,802 women in the cohort who required follow-up after an index Pap test before and Table 2 Characteristics of women who had been enrolled in a US health care plan by guideline period. Characteristics
Age (years) 18–29 30–39 40–49 50–64 65–70 Region Northeast South Midwest West Urban/Rural code Urban Suburban Rural Insurance type HMO PPO/POS Other Primary coveragea Commercial Medicare Medicaid Setting for index Pap Ofﬁce/Clinic Hospital Laboratory Other Physician type for index Pap OB/GYN Family practitioner Internist Other Final diagnosis Invasive cervical cancer Noninvasive cervical cancer Dysplasia Abnormal Pap No diagnosis
21,905 29,079 39,387 31,853 3644
17.4 23.1 31.3 25.3 2.9
23,025 22,492 29,515 24,390 2512
22.6 22.1 29.0 23.9 2.5
11,082 47,864 58,329 8593
8.8 38.0 46.3 6.83
16,892 41,032 27,901 16,109
16.6 40.2 27.4 15.8
84,177 32,164 9527
66.9 25.6 7.6
74,388 21,254 6292
73.0 20.8 6.2
115,793 6601 3474
92.0 5.2 2.8
28,482 63, 203 10,249
27.9 62.0 10.1
117,206 4307 4355
93.1 3.4 3.5
96,462 2697 2771
94.6 2.7 2.7
93,629 15,314 16,713 212
74.4 12.2 13.3 0.2
73,780 15,305 11,096 1753
72.4 15.0 10.9 1.7
75,156 20,298 6482 23,932
59.7 16.1 5.2 19.0
54,599 16,952 6453 23,930
53.6 16.6 6.3 23.5
365 1402 10,710 37,070 76,321
0.3 1.1 8.5 29.5 60.6
291 1397 10,440 36,369 53,437
0.3 1.4 10.2 35.7 52.4
All characteristics are signiﬁcant at the p b 0.05 because of the large sample size (N = 227,802). Pre-ASCCP = 1999–April 2002; post-ASCCP = May 2002–2004. Guidelines changed in 2002. ASCCP: American Society of Colposcopy and Cervical Pathology; HMO: health maintenance organization; PPO: preferred provider organization; POS: point of service plan; OB/GYN: obstetrician/gynecologist. a Four women had unknown insurance.
V.B. Benard et al. / Preventive Medicine 50 (2010) 81–85 Table 3 Final diagnosis of women who had been enrolled in a US health care plan by number of follow-up events and guideline period.
Invasive cancer Noninvasive cancer Dysplasia Abnormal Pap No diagnosis Overall
One follow-up event
Two or more follow-up events
4 47 1370 16841 48585 66,847
1.4 3.4 13.1 65.3 90.1 73.2
358 1359 9516 12028 7621 30,882 p b 0.01
98.1 96.9 88.8 36.9 10.0 25.4
287 1350 9070 8962 4852 24,521
98.6 96.6 86.9 34.7 9.1 26.8
7 1.9 43 3.1 1194 11.1 20,539 63.1 68,700 90.0 90,483 74.5 p b 0.001
There were 4503 in pre-ASCCP and 10,566 in post-ASCCP with no follow-up event of interest that was omitted from the table. p-value testing difference between one follow-up event (and two or more follow-up events) pre- vs. post-ASCCP. Pre-ASCCP = 1999–April 2002; post-ASCCP = May 2002–2004. Guidelines changed in 2002. ASCCP: American Society of Colposcopy and Cervical Pathology.
after the ASCCP guidelines changed. Over half of the women in this study population were 40 years of age or older. The percentage of women living in the Midwest declined in the post-ASCCP time period, with a corresponding increase in the other three regions. Prior to 2002, the overwhelming majority of women had HMO type insurance coverage which changed to PPO/POS coverage. More than 70% of the index Pap tests were performed in an ofﬁce or clinic setting and over half were performed by a gynecologist. There were 656 invasive and 2799 noninvasive cancers diagnosed from 1999 to 2004. As expected, the more severe the diagnosis, the more likely the women were to have 2 or more follow-up events performed (Table 3). Of the women with an abnormal index Pap test, the majority in both time periods had only 1 follow-up event (63% and 65%). Fig. 1 presents the ﬁrst follow-up event after the index Pap by cohort year. The probability of follow-up with a repeat Pap test decreased each year, with the largest decrease from 2002 to 2003; the probability of follow-up with colposcopy increased over the years with the exception of 2004. The proportion of women with no followup also increased over time. The percentage of women with repeat Pap tests decreased signiﬁcantly from 80% to 65% in the pre- vs. post-ASCCP guideline period (p b 0.0001) (Table 4). This decrease occurred in every age group. For both time periods, the percentage of follow-up with a repeat Pap test increased with age. Overall, follow-up with colposcopy increased from 12% to 14% between the two time periods (p b 0.0001). Again, this increase occurred in every age group except women aged
65–70 years; women under age 30 years were most likely to be followed-up with colposcopy. The multivariate logistic regression model compared the probability that women received a colposcopy with the probability that they received a repeat Pap test as the ﬁrst follow-up event following the index Pap test (Table 5). We estimated the odds of a woman receiving a colposcopy to be 41% higher in the post- versus pre-guideline change period, after controlling for variables that are associated with type of follow-up. We also found that the odds of a woman between the ages of 18 and 29 receiving a colposcopy as her ﬁrst follow-up procedure, rather than a repeat Pap test, were 17% higher than for a woman between the ages of 30 and 39 years. Similarly, women whose index Pap test was performed by a gynecologist were more likely to be examined by colposcopy than women seen by other types of clinicians. Also, the more serious the ﬁnal diagnosis, the more likely the follow-up event was colposcopy rather than a repeat Pap test. Discussion This study examines follow-up events of Pap tests in an insured population from 1999 through 2004 using insurance claims data obtained from a large national health care organization. Over this period, many changes have occurred in the prevention and detection of cervical cancer. These include the reporting system (Kurman et al., 2003), new technologies (Kulasingam et al., 2002), increased nationwide screening (Solomon et al., 2007) and updated management guidelines (Wright et al., 2002). These advances reﬂect greater understanding of the natural history of HPV infection and its role in the development of cervical cancer. Through these changes, colposcopy became the preferred method of follow-up for most cytological abnormalities (Wright et al., 2002). Our ﬁndings indicate that after the management guidelines changed, more colposcopies and fewer repeat Pap tests were performed as a follow-up of a presumed abnormal Pap test. Also noted over time was an increase in the number of women with no follow-up and an increase in the use of the HPV DNA test, suggesting that increased screening intervals and new technologies may also play a role in this cohort of women. However, even with the 2002 guideline changes and improved technology to detect cytological abnormalities, as seen in these data an abnormal Pap test signals the ﬁrst step in a much longer diagnostic process that can entail months of follow-up which can be costly to both the health care system and the patient (Insinga et al., 2008). In this study, we found 7.5% of all Pap tests to be abnormal based on our assumptions which are consistent with the literature on a national level (Jones and Davey, 2000). The majority of women repeated a Pap test as their ﬁrst follow-up event and if a second follow-up diagnostic
Fig. 1. First follow-up event of all women after abnormal Pap by year (N = 227,802).
V.B. Benard et al. / Preventive Medicine 50 (2010) 81–85
event occurred, a repeat Pap test was the method most often chosen. In addition, we found that over half of the abnormalities had no diagnosis of interest after the follow-up events. The type of follow-up a woman received was highly dependent on her age. We found that the percentage of women receiving colposcopy following an abnormal Pap test decreased with increasing age in both the pre- and post-guideline period; the youngest age group (b30 years of age) was more likely to undergo colposcopy as the ﬁrst follow-up option. This pattern of follow-up in younger women has been noted in other studies (Benard et al., 2005; Benard et al., 2008). However, cervical cancer is generally found in an older population with the peak rate noted in 40- to 49-year-old women (Saraiya et al., 2007). Younger women are more likely to have a high clearance of HPV infections and regression of Pap abnormalities to normal and therefore could have a more conservative follow-up approach. The direction of our ﬁndings in relation to follow-up based on guideline changes is similar to those from a study conducted of underserved women served by a national screening program with low-grade Pap test abnormalities across time. In that study, the percentage of providers who followed the recommended guidelines with colposcopy increased by 9% from before the guideline change to afterwards (Benard et al., 2008). The percentage of follow-up with colposcopy after a low-grade abnormality ranged from 53% to 75% depending on age and time period, whereas the percentage with follow-up with a Pap test ranged from 7% to 26%, again depending on age and time. In our study of members of a managed care plan, we documented that fewer colposcopies were performed (5% to 24% depending on age and time period) and that more repeat Pap tests were done (63% to 83% depending on age and time period). Several other studies have assessed follow-up of Pap tests using retrospective laboratory data. In a study from 2001 through 2003, all categories of squamous cell abnormalities from a hospital database were evaluated to determine follow-up with biopsy (Elsheikh et al., 2006). Of 129,911 tests, 6% were coded as abnormal. Of these abnormal Pap tests, 18.7% reported a follow-up biopsy. Another study assessing the follow-up of low-grade abnormalities reported that 32% and 7% were followed with colposcopy (Alanen et al., 1998). These percentages are more consistent with our ﬁndings. Insurance claims-based analyses can provide insights into the care that large cohorts of women receive, but the known limitation is that these data are collected to meet the needs of providers and payers, not those of researchers. [AU1]Since the results only represent women in our cohort, they may not be generalizable to the insured population. Information on race and ethnicity was not available in this dataset and there could be differences in follow-up by race and ethnicity that we were not able to address. Detailed Table 4 Type of ﬁrst follow-up event after abnormal Pap test for women who had been enrolled in a US health care plan by guideline period and age group. Age group
Age (years) 18–29 30–39 40–49 50–64 65–70 Overall
5478 3489 3312 2162 138 14,579
23.8 15.5 11.2 8.9 5.5 14.3
15,850 22,705 31,830 26,635 3056 100,076 p b 0.0001
72.4 78.1 80.8 83.6 83.9 79.5
12,424 14,107 19,913 17,816 1970 66,230
54.0 62.7 67.5 73.1 78.4 65.0
3883 17.7 3800 13.1 4287 10.9 2715 8.5 225 6.2 14,910 11.9 p b 0.0001
Variable Age group (years) 18–29 30–39 40–49 50–64 65–70 Guideline change Post Pre Insurance coverage Commercial Medicare Medicaid Product type HMO PPO/POS Other Site of index Pap Inpatient Outpatient Ofﬁce visit Lab Other Provider of index Pap OB/GYN Family practitioner Internist Other Location Urban Small urban Rural Region of country Northeast South Midwest West Final diagnosis Invasive cancer Noninvasive cancer Dysplasia Abnormal Pap No diagnosis
1.17 Ref. 1.04 1.08 1.13
1.00–1.09 1.03–1.14 0.80–1.59
Ref. 0.76 1.02
Ref. 0.93 0.95
0.60 1.01 Ref. 0.97 0.74
0.43–0.85 0.96–1.06 0.90–1.06 0.61–0.90
Ref. 0.67 0.70 0.61
0.64–0.71 0.64–0.75 0.57–0.65
Ref. 0.96 0.93
Ref. 1.04 1.04 0.95
0.99–1.10 0.98–1.10 0.88–1.01
106.68 101.15 74.85 5.86 Ref.
87.08–130.68 91.48–111.84 71.58–78.27 5.64–6.09
CI: conﬁdence interval; OB/GYN: obstetrician/gynecologist; HMO: health maintenance organization; PPO: preferred provider plan; POS: point of service plan. a Bolded odds ratios are statistically signiﬁcant at p b 0.05.
diagnostic codes distinguishing the abnormality of the Pap test were not available in these claims data. For that reason, we were not able to separate the follow-up procedures received by women with a speciﬁed Pap result to determine the appropriate method of followup based on national guidelines. However, even with our assumption of an abnormal Pap test, our ﬁndings are consistent with those of other studies as noted above.
Repeat Pap test
Table 5 Odds ratios of colposcopy versus repeat Pap test as ﬁrst follow-up event for women who had been enrolled in a US health care plan calculated from logistic regression resultsa.
p-value testing difference between colposcopy (and repeat Pap test) pre- vs. postASCCP. Pre-ASCCP = 1999–April 2002; post-ASCCP = May 2002–2004. Guidelines changed in 2002. ASCCP: American Society of Colposcopy and Cervical Pathology.
Conclusion Our analysis among women with follow-up after an abnormal Pap test provides a window on the utilization of health care services across the United States in a multitude of clinical settings. These data are needed for conducting policy assessments to evaluate new medical technologies and care guidelines, and to determine whether providers are in fact changing practices as guidelines are updated. From our ﬁndings, the follow-up of an abnormal Pap test with colposcopy has increased, however, not as much as one would expect based on the recommended guidelines. As organizations continue to revise guidelines based on the evidence, these large geographically diverse health care databases can help to determine the gaps between the scientiﬁc approach and the clinical practice.
V.B. Benard et al. / Preventive Medicine 50 (2010) 81–85
References Alanen, K.W., Elit, L.M., Molinaro, P.A., McLachlin, C.M., 1998. Assessment of cytologic follow-up as the recommended management for patients with atypical squamous cells of undetermined signiﬁcance or low grade squamous intraepithelial lesions. Cancer 84, 5–10. Benard, V.B., Howe, W., Saraiya, M.S., Helsel, W., Lawson, H., 2008. Assessment of follow-up for low-grade cytological abnormalities in the national breast and cervical cancer early detection program, 2000–2005. J. Low Genit. Tract. Dis. 12, 300–306. Benard, V.B., Lawson, H.W., Eheman, C.R., Anderson, C., Helsel, W., 2005. Adherence to guidelines for follow-up of low-grade cytologic abnormalities among medically underserved women. Obstet. Gynecol. 105, 1323–1328. Elsheikh, T.M., Kirkpatrick, J.L., Wu, H.H., 2006. The signiﬁcance of low-grade squamous intraepithelial lesion, cannot exclude high-grade squamous intraepithelial lesion as a distinct squamous abnormality category in Pap tests. Cancer 108, 277–281. Eltoum, I.A., Roberson, J., 2007. Impact of HPV testing, HPV vaccine development and changing screening frequency on national Pap test volume: projections from the National Health Interview Survey (NHIS). Cancer 111, 34–40. Holland-Barkis, P., Forjuoh, S.N., Couchman, G.R., Capen, C., Rascoe, T.G., Reis, M.D., 2006. Primary care physicians' awareness and adherence to cervical cancer screening guidelines in Texas. Prev. Med. 42, 140–145. Insinga, R.P., Ye, X., Singhal, P.K., Carides, G.W., 2008. Healthcare resource use and costs associated with cervical, vaginal and vulvar cancers in a large U.S. health plan. Gynecol. Oncol. 111, 188–196.
Jones, B.A., Davey, D.D., 2000. Quality management in gynecologic cytology using interlaboratory comparison. Arch. Pathol. Lab. Med. 124, 672–681. Kulasingam, S.L., Hughes, J.P., Kiviat, N.B., et al., 2002. Evaluation of human papillomavirus testing in primary screening for cervical abnormalities. JAMA 288, 1749–1757. Kurman, R.J., Henson, D.E., Herbst, A.L., Noller, K.L., Schiffman, M.H., 1994. Interim guidelines for management of abnormal cervical cytology. The 1992 National Cancer Institute Workshop. JAMA 271, 1866–1869. Kurman, R., Solomon, D., Nayar, R., Davey, D., Wilbur, D., 2003. The Bethesda System for Reporting Cervical Cytology: Deﬁnitions, Criteria, and Explanatory Notes. Springer, New York, NY. Saraiya, M., Ahmed, F., Krishnan, S., et al., 2007. Cervical cancer incidence in a prevaccine era in the United States, 1998–2002. Obstet. Gynecol. 109, 360–370. Schabert, V.F., Ye, X., Insinga, R.P., Singhal, P.K., Riedel, A.A., 2008. Five-year routine cervical cancer screening rates and intervals in a US health plan. Curr. Med. Res. Opin. 24, 2429–2435. Solomon, D., Breen, N., McNeal, T., 2007. Cervical cancer screening rates in the United States and the potential impact of implementation of screening guidelines. CA Cancer J. Clin. 57, 105–111. Wright Jr., T.C., Cox, J.T., Massad, L.S., Twiggs, L.B., Wilkinson, E.J., 2002. 2001 Consensus guidelines for the management of women with cervical cytological abnormalities. JAMA 287, 2120–2129. Yabroff, R., Saraiya, M., Meisser, H., et al., in press. Primary care physicians' Pap test screening recommendations and practices: ﬁndings from a national survey. Arch. Intern. Med. 151, 6023-611. Zapka, J.G., Puleo, E., Taplin, S., Solberg, L.I., Mouchawar, J., Somkin, C., Geiger, A.M., Yood, M.U., 2005. Breast and cervical screening: clinicians' views on health plan guidelines and implementation efforts. JNCI Monographs 35, 46–54.