With the advent of the Pap smear and subsequent use of liquid-based cervical cytology, cervical cancer incidence and mortality significantly declined.1 Current local jurisdictional guidelines (Ontario, Canada) recommend screening with cervical cytology once every 3 years for all women between the ages of 21 to 70 years with a personal history of sexual activity.2 Routine cervical screening combined with regular follow-up and guideline-based intervention allows for early detection and treatment of neoplastic cervical changes.
The superiority of organized cancer screening programs compared with opportunistic screening is well-supported in the literature.3–5 In our jurisdiction, an organized cervical cancer screening program is in place for 4.4 million screening eligible women.6 Within this program, a central registry of cytology results from 1998 to present is available. Overall incidence rates of invasive cervical cancer in Ontario have decreased from 21.1/100,000 in 1981 to 12.8/100,000 in 2010.7 However, despite ongoing screening, the rate of decline has started to plateau and cervical cancer remains the third most common cancer among Ontario women between the ages of 20 to 44 years and the second most common cause of cancer-related death.7
Across the 10 Canadian provinces, cervical cancer screening uptake varies between 62.9% and 73.8%.8 A recent Canadian study noted that up to 40% of women who develop cervical cancer are screened with cervical cytology in the 3 years preceding their diagnosis.9 The reason for this is likely multifactorial. Although cervical cytology is an effective screening tool, it does have a significant rate of sampling and interpretation error. A recent large-scale, randomized controlled trial reported the sensitivity of cervical cytology to be as low as 55%, implying that roughly 45% of neoplastic lesions may be missed in a single round of screening.10 Inadequate follow-up of abnormal cytology is also a contributing factor. In our jurisdiction, approximately one sixth of patients identified with low-grade squamous intraepithelial lesion (LSIL) or atypical cells of undetermined significance (ASC-US) on cytology are subsequently lost to follow-up and 25% of women with high-grade cytologic abnormalities do not receive proper follow-up within a year of abnormal cytology.11,12 In addition, cytology has increasingly been shown to be less effective at detecting and preventing adenocarcinoma.13,14
Evaluation of factors contributing to the development of invasive cervical cancer helps evaluate the performance of organized screening programs. This research uses a population-based retrospective cohort study to determine factors associated with the development of invasive cervical malignancy among women enrolled in an organized screening program.
MATERIALS AND METHODS
Study Design and Cohort
The study involved a population-based retrospective cohort design of women in Ontario 21 years or older who were diagnosed with invasive cervical cancer between 2011 and 2014. Diagnosis of invasive cervical cancer was defined as: International Classification of Diseases for Oncology codes C53. Cancer diagnosis date was defined as the index date. Women excluded from the study were those who registered for provincial health care coverage less than 1 year from the date of cancer diagnosis; had missing or incorrect histology; or had missing administrative/demographic information. The characteristics of excluded women are outlined in Table 1 in the Appendix (http://links.lww.com/LGT/A102).
Data Sources and Variables
Information on invasive cervical cancer was obtained from the Ontario Cancer Registry, which records 95% of all cases of cancer in Ontario.15 The Registered Persons Database was used to identify patient age and demographics. Cytobase was used to identify cytology test results, and the Ontario Health Insurance Plan provided information regarding healthcare services delivered by physicians including cervical cytology, colposcopy follow-up, and treatment as well as surgical interventions.
The main study outcome was cervical cytology test history. Screening history was defined as any cervical cytology test done in the 2 to 10 years before cervical cancer diagnosis. Cervical cytology performed within 2 years of cancer diagnosis was excluded to avoid detecting a prevalent cancer case. A secondary outcome examined abnormal cervical cytology follow-up, which was defined as repeat cytology, colposcopy, or treatment received within 1 year of the abnormal screening test.
Additional variables included in the study were the following: number of cervical cytology tests done within 2 to 10 years of diagnosis, age, registration with a family physician, histology of cancer, stage of cervical cancer, index cervical cytology result, specialty of physician who conducted the screening, urban/rural neighborhood, and neighborhood income quintile.
Cancer Care Ontario is designated a “prescribed entity” for the purposes of section 45 (1) of the Personal Health Information Protection Act of 2004. As a prescribed entity, Cancer Care Ontario is authorized to collect personal health information from health information custodians without the consent of the patient and to use such personal health information for the purpose of analysis or compiling statistical information with respect to the management, evaluation, or monitoring of the allocation of resources to or planning for all or part of the health system, including the delivery of services. Because this study is in compliance with privacy regulations, ethics review was not required.
Descriptive statistics were calculated for the following: patient and provider characteristics; cytology from the most recent screening test; screen cytology history; and characteristics of women with no follow-up postabnormal cervical cytology.
To assess the effect of screen history, we conducted 2 multivariate logistic regression models. The first model estimated the odds ratio of having a screening test in the last 2 to 10 years. Control variables included in the model were the following: age, cancer stage, income, primary care enrollment (PEM) status, and cancer histology. The second model estimated the odds of having a follow-up within 1 year after abnormal cytology testing. Control variables included in the second model were age, cancer stage, index cytology result, and PEM status. All results are reported considering a level of significance of of less than 0.05. SAS software Version 9.4 was used to perform the analyses (SAS Institute Inc, Cary, NC).
Between 2011 and 2014, there were 2,281 cases of cervical cancer diagnosed in Ontario. Of these, 77 cases were excluded because the women were nonresidents or newcomers to Ontario. An additional 202 Ontario residents were excluded because they were diagnosed with a nonsquamous/nonadenocarcinoma cancer or with a cancer of unknown histology. The remaining 2,002 cases were examined in our study cohort.
There were 1358 cases of squamous cell carcinoma (SCC) and 644 cases of adenocarcinoma. The mean age at diagnosis was 50.3 and 11.7% of cases occurred in women outside of the recommended screening age bracket (>70 years of age). There was an inverse relationship between disease prevalence and income quintile and 23% of women who developed cervical cancer were not registered to a primary care physician. At the time of diagnosis, 973 women (56.7%) had cancer confined to the cervix (International Federation of Gynecology and Obstetrics stage 1A or 1B), while the remainder had locally advanced or distant spread of disease (see Table 1).
Most women (55%) diagnosed with cervical cancer had not received screening for more than 5 years before their diagnosis. Thirty-six percent did not receive a single screening test in the 2 to 10 years prior, whereas 20% received 5 or more screening cytology tests. The mean number of screening cytology tests done in the 2 to 10 years before cancer diagnosis was 2.3 (see Table 2).
Almost a quarter of women (23%) diagnosed with cancer received screening in the 24 to 36 months before their diagnosis. Close to 70% received normal cervical cytology results on this preceding screen. Of those women diagnosed with adenocarcinoma who had been screened between 24 and 36 months before diagnosis, 78.6% received normal cytology reports, whereas only 9.9% had atypical squamous cells of undetermined significance (ASC-US), LSIL, high-grade squamous intraepithelial lesion (HSIL), atypical squamous cells, cannot rule out high-grade lesion (ASC-H), atypical glandular cells (AGCs), or adenocarcinoma in-situ (AIS). These numbers are compared with SCC, where 61.6% of women who were screened between 24 and 36 months before diagnosis received normal results and 24.2% were diagnosed with neoplastic disease (Table 2).
For women presenting with stage 3 and stage 4 disease, 49% and 58.8% had not been screened for more than 10 years before their diagnosis, respectively. Of the women presenting with stage 1A and 1B disease, 54.2% and 55.2%, respectively, had been screened within 5 years of diagnosis (Table 3).
A multivariate analysis was performed to analyze the predictors of screening participation. Women in higher income quintiles, younger than the age of 70 years, and registered to a family physician were more likely to receive screening (p < .01). Screening recency was protective regarding stage at presentation. Women with stages 2 to 4 were one third to one half as likely to have had screening in the 2 to 10 years before diagnosis as compared with those presenting with stage 1. Compared with women diagnosed with SCC, women with adenocarcinoma were twice as likely to have received cervical cytology within 10 years of their diagnosis (see Table 4).
A multivariate analysis was performed to determine factors associated with follow-up within 1 year of abnormal cytology screening. Women with HSIL, ASC-H, AGC, and AIS were 3.4 times more likely to have follow-up than those with low-grade neoplasia (ASC-US/LSIL).
A final multivariate analysis examined factors associated with the receipt of a normal cytology report on the screening test preceding cancer diagnosis. Women with adenocarcinoma were 2.2 times more likely to have received a normal cytology test than those diagnosed with squamous cell. In addition, women between the ages of 70 and 79 years were 4.6 times more likely to have had a false-negative cytology test than those between the ages of 21 and 29 years.
The purpose of this study was to determine factors associated with the diagnosis of invasive cervical cancer in the setting of an organized cervical cancer screening program. This study evaluates a cohort of 2,002 women who were diagnosed with cervical cancer between 2011 and 2014.
Seventy-seven percent of women in our cohort were registered to a primary care physician. Despite access to primary care, it is concerning that only 45% of women received cervical screening in the 2 to 5 years before their cancer diagnosis. It is even more concerning that 36% did not receive any screening at all in the preceding 2 to 10 years.
Family physicians play an integral role in the provision of cancer screening and women registered to a primary care physician are 60% more likely to receive cervical cancer screening.16 However, despite this important responsibility, there are still some primary care practitioners who do not provide screening. A recently published systematic review regarding physician characteristics and cancer screening found that female and white physicians tended to screen more than male and nonwhite physicians. In addition, gynecologists tended to screen more than internists and family physicians.17
A study in our jurisdiction of Ontario, Canada, found that medical school training location influenced future cervical cancer screening delivery. Physicians trained in the Caribbean/Latin America, Middle East/North Africa, South Asia, and Western Europe were less likely to perform cervical screening than Canadian graduates.18 It is important to ensure that graduates of appropriate residency programs are comfortable performing cervical screening or understand the need for referral to a physician who will complete screening.
Patient willingness to undergo cervical screening is also a barrier to screening receipt. A population-based survey in Britain determined barriers to cervical screening among women. Women who exhibited the attitude that they had “other more important things to worry about than screening,” those who thought their risk was low based on sexual behavior and those who assessed the risks and benefits and decided it's “not worth me going for screening” were more likely to decline screening.19
A survey carried out in the United States studied nonfinancial barriers to cervical cancer screening among women with health insurance and access to a physician. When compared with women who received regular screening, the following characteristics were associated with poor screening uptake: ages 60 to 65 years, smoking history, obesity, and annual income less than US $10,000. These women were also more likely to have chronic conditions such as chronic obstructive pulmonary disease, depression, and diabetes.20
Much research exists addressing strategies to improve screening participation; however, most studies have only shown modest gains.21,22 In 2014, the Ontario Cervical Cancer Screening Program instituted a patient directed correspondence program using screening invitation letters sent at timely intervals. This intervention yielded a positive effect with an odds ratio of 1.8 in favor of women receiving screening upon receipt of the letter.23
In addition to population-based screening efforts, interventions targeted at physician screening rates are equally important. Various strategies exist to improve primary care screening delivery, with physician reminder systems having been shown to be most effective.24,25 Most recently, the Primary Care Screening Activity Report was implemented to improve screening in Ontario. The Primary Care Screening Activity Report functions as an audit and feedback tool, providing primary care physicians with information detailing which of their enrolled patients are due or overdue for cervical, breast, and colorectal screening. This tool was shown to provide a small increase in screening participation when accessed by physicians (odds ratio = 1.07).26
Ongoing research has found that human papillomavirus (HPV) self-sampling in underscreened women may be an effective tool to increase screening uptake among women who may not otherwise receive screening; however, more research is required to determine how to effectively implement such testing.27,28
The performance of cervical cytology as a screening tool is also an important component in the success of cervical cancer prevention. In our cohort, 65% of women had at least 1 cervical cytology test between 2 and 10 years of their cancer diagnosis, and on average, women received 2.3 screening tests in this time frame. Furthermore, 70% of cytology tests performed between 2 to 3 and 3 to 5 years before diagnosis were read as normal. Interestingly, less than 10% of cytology results found high-grade neoplasia on testing done between 24 and 36 months before diagnosis.
Given the multiyear natural history of cervical neoplasia, it is concerning that our study revealed a high proportion of normal cervical cytology tests received in the 2 to 5 years before cancer diagnosis. This raises the possibility that the cytology missed an existing cervical neoplasia or malignancy. In addition, we note that those women diagnosed with adenocarcinoma were 2-fold more likely to have had a normal cytology result.
Other jurisdictions have found similar limitations with the use of cervical cytology as a screening test. Leyden et al.29 examined the utility of cervical cytology within an organized screening program, finding that when excluding the unscreened population, 32% of invasive disease was due to failure of cervical cytology to detect an abnormality, whereas 13% was due to inadequate abnormal cytology follow-up.
A significant amount of research surrounding HPV primary screening has found it to be more sensitive than cervical cytology for detecting precursor lesions. Specifically, in the Canadian Cervical Cancer Screening Trial, cervical cytology had a reported sensitivity of only 55.4%, whereas HPV test sensitivity was 94.6%.10 The recently published HPV FOCAL trial which compared HPV testing at 48-month intervals to cervical cytology performance at 24 and 48 months found that HPV testing in the first round of screening detected more cervical intraepithelial neoplases (CIN) 3+ cases with a risk ratio of 1.61 (95% CI = 1.09–2.37). At the completion of the study at 48 months, there were significantly less cases of CIN 3+ in the HPV versus cytology screened arm with a risk ratio of 0.42 (95% CI = 0.25–0.69). Interestingly, the CIN 3+ risk ratio for HPV or cytology negative screen tests was 0.25, supporting the role of HPV testing in primary screening over that of cytology.30
In Ontario, the cervical cancer screening guidelines are supportive of HPV primary screening and the Ontario Cervical Cancer Screening Program is now in the process of implementing primary HPV screening in the organized setting. This process may detect precancerous lesions that were previously missed in the cervical cytology-based screening system. Furthermore, in our jurisdiction, a school-based HPV vaccination program has been in place since 2007 offering primary prevention to children 12 years and older.
Other international comprehensive screening programs in Australia, the United States, Scotland, Sweden, and the Netherlands are implementing HPV primary cervical cancer screening as well. In Australia, where a robust HPV vaccination and screening program exists, it is estimated that cervical cancer will be eradicated by 2035.31
The strengths of this article include a large cohort of women in a jurisdiction within an organized screening program who developed cervical cancer. The databases used allowed access into information on physician access, screening history, screening results, and demographic data.
The limitations of this article include no access to the clinical chart, and as such, it is not possible to comment on the interaction between physicians and patients including the offer/decline of cervical screening and symptoms that may be related to cervical malignancy. In addition, our cytology registry contains approximately 85% to 90% of cytology results, so it is possible that we underestimated the screening rate in the study population; however, those numbers would be very small.
This study evaluated the cervical cancer screening histories of women in Ontario diagnosed with cervical cancer. Thirty-six percent of women were not screened in the 2 to 10 years before their diagnosis. The remainder did receive screening with cervical cytology during this interval but developed cervical cancer nevertheless. This study underscores the challenges associated with screening participation and discusses the limitations of cervical cytology as a screening tool.
Findings from this analysis demonstrate opportunities for improvement within current screening programs. Despite the presence of an organized screening program, many women are still not being screened according to guidelines, representing missed opportunities for early detection and treatment of preinvasive disease. Women with low socioeconomic status and those not enrolled with a primary care provider are particularly at risk. Directed efforts aimed at improving screening participation and retention in these populations are needed to improve health equity in our communities.
In addition, findings from this review demonstrate the need for a more sensitive cervical cancer screening tool. Persistently stable rates of cervical cancer and the poor sensitivity of conventional cervical cytology, particularly with respect to preventing adenocarcinoma, are of significant concern. The molecular-based HPV test has been shown to be superior in the detection and prevention of all types of cervical cancer. This knowledge, in conjunction with the findings from our review, supports the movement toward HPV molecular testing for primary cervical cancer screening.
1. Liu S, Semenciw R, Probert A, et al. Cervical cancer in Canada: changing patterns in incidence and mortality. Int J Gynecol Cancer
2. Murphy J, Kennedy EB, Dunn S, et al. Cervical screening: a guideline for clinical practice in Ontario. J Obstet Gynaecol Can
3. Benedet JL, Anderson GH, Matistic JP. A comprehensive program for cervical cancer detection and management. Am J Obstet Gynecol
4. Laara E, Day NE, Hakama M. Trends in mortality from cervical cancer in the Nordic countries: association with organised screening programmes. Lancet
5. Wingo PA, Ries LA, Rosenberg HM, et al. Cancer incidence and mortality, 1973-1995: a report card for the U.S. Cancer
8. Canadian Partnership Against Cancer. Cervical Cancer Screening in Canada
. Toronto, ON: Canadian Partnership Against Cancer; 2016.
9. Mema SC, Nation J, Yang H, et al. Screening history in 313 cases of invasive cancer: a retrospective review of cervical cancer screening
in Alberta, Canada. J Low Genit Tract Dis
10. Mayrand MH, Duarte-Franco E, Rodrigues I, et al. Human papillomavirus DNA versus papanicolaou screening tests for cervical cancer. N Engl J Med
11. Kupets R, Lu Y, Vicus D, et al. Are there flaws in the follow-up of women with low-grade cervical dysplasia in Ontario? J Obstet Gynaecol Can
12. Kupets R, Paszat L. How are women with high grade Pap smear abnormalities managed? A population based study. Gynecol Oncol
13. Adegoke O, Kulasingam S, Virnig B. Cervical cancer trends in the United States: a 35-year population-based analysis. J Womens Health (Larchmt)
14. Bray F, Carstensen B, Møller H, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev
15. Clarke EA, Marrett LD, Kreiger N. Cancer registration in Ontario: a computer approach. IARC Sci Publ
16. Tavasoli SM, Pefoyo AJ, Hader J, et al. Impact of invitation and reminder letters on cervical cancer screening
participation rates in an organized screening program. Prev Med
17. Neugut AI, MacLean SA, Dai WF, et al. Physician characteristics and decisions regarding cancer screening: a systematic review. Popul Health Manag
2018. doi: 10.1089/pop.2017.0206.
18. Lofters AK, Ng R, Lobb R. Primary care physician characteristics associated with cancer screening: A retrospective cohort study in Ontario, Canada. Cancer Med
2015. doi: 10.1002/cam4.358.
19. Bennett KF, Waller J, Chorley AJ, et al. Barriers to cervical screening and interest in self-sampling among women who actively decline screening. J Med Screen
2018. doi: 10.1177/0969141318767471.
20. Crawford A, Benard V, King J, et al. Understanding barriers to cervical cancer screening
in women with access to care, behavioral risk factor surveillance system, 2014. Prev Chronic Dis
21. Everett T, Bryant A, Griffin MF, et al. Interventions targeted at women to encourage the uptake of cervical screening. Cochrane Database Syst Rev
22. Eaker S, Adami HO, Granath F, et al. A large population-based randomized controlled trial to increase attendance at screening for cervical cancer. Cancer Epidemiol Biomarkers Prev
23. Tsoa E, Kone Pefoyo AJ, Tsiplova K, et al. Evaluation of recall and reminder letters on retention rates in an organized cervical screening program. J Obstet Gynaecol Can
24. Jonah L, Pefoyo AK, Lee A, et al. Evaluation of the effect of an audit and feedback reporting tool on screening participation: The Primary Care Screening Activity Report (PCSAR). Prev Med
25. Vink MA, Bogaards JA, Van Kemenade FJ, et al. Clinical progression of high-grade cervical intraepithelial neoplasia: estimating the time to preclinical cervical cancer from doubly censored national registry data. Am J Epidemiol
26. Hatem F, Wilbur DC. High grade squamous cervical lesions following negative papanicolaou smears: false-negative cervical cytology
or rapid progression. Diagn Cytopathol
27. Arbyn M, Castle PE. Offering self-sampling kits for HPV testing
to reach women who do not attend in the regular cervical cancer screening
program. Cancer Epidemiol Biomarkers Prev
28. Madzima TR, Vahabi M, Lofters A. Emerging role of HPV self-sampling in cervical cancer screening
for hard-to-reach women: focused literature review. Can Fam Physician
29. Leyden WA, Manos MM, Geiger AM, et al. Cervical cancer in women with comprehensive health care access: attributable factors in the screening process. J Natl Cancer Inst
30. Ogilvie GS, Van Niekerk D, Krajden M, et al. Effect of screening with primary cervical HPV testing
vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial. JAMA
31. Hall MT, Simms KT, Lew JB, et al. The projected timeframe until cervical cancer elimination in Australia: a modelling study. Lancet Public Health