Recent trends in breast, cervical, and colorectal cancer screening test utilization in Canada, using self-reported data from 2008 and 2012

Abstract

In Canada, self-reported data from the Canadian Community Health Survey 2008 and 2012 provide an opportunity to examine overall utilization of breast, cervical, and colorectal cancer screening tests for both programmatic and opportunistic screening.

Among women 50–74 years of age, utilization of screening mammography was stable (62.0% in 2008 and 63.0% in 2012). Pap test utilization for women 25–69 years of age remained high and stable across Canada in 2008 and 2012 (78.9% in 2012). The percentage of individuals 50–74 years of age who reporting having at least 1 fecal test within the preceding 2 years increased in 2012 (to 23.0% from 16.9% in 2008), but remains low.

Stable rates of screening mammography utilization (about 30%) were reported in 2008 and 2012 among women 40–49 years of age, a group for which population-based screening is not recommended. Although declining over time, cervical cancer screening rates were high for women less than 25 years of age (for whom screening is not recommended). Interestingly, an increased percentage of women 70–74 years of age reported having a Pap test.

In 2012, a smaller percentage of women 50–69 years of age reported having no screening test (5.9% vs. 8.5% in 2008), and more women reported having the three types of cancer screening tests (19.0% vs. 13.2%).

Efforts to encourage use of screening within the recommended average-risk age groups are needed, and education for stakeholders about the possible harms of screening outside those age groups has to continue.

INTRODUCTION

Observational data and randomized controlled trials have demonstrated the effectiveness of screening, when applied in certain average-risk age groups, for reducing mortality from breast, cervical, and colorectal cancers^1–7. However, the success of screening depends on high participation rates in the recommended age groups. It is also important to determine screening uptake in people outside the recommended age groups, because those individuals might not benefit from screening and could be subject to more potential harms, such as false-positive screening results^8–10.

National evidence-based screening guidelines for breast, cervical, and colorectal cancer screening—which define recommended age groups, types of tests, and intervals for average-risk populations—are published and periodically updated by the Canadian Task Force on Preventive Health Care (ctfphc)^8–11 with support from the Public Health Agency of Canada (Table i). Currently, the ctfphc recommends breast cancer screening with mammography every 2–3 years for women 50–74 years of age, cervical cancer screening with a Pap test every 3 years for women 25–69 years of age, and colorectal cancer screening with a fecal test every 1–2 years or flexible sigmoidoscopy (interval not specified) for people 50 years of age and older. Because of ongoing revisions to the ctfphc’s colorectal cancer screening guidelines, the present report also considers the recommendations issued by the Canadian Association of Gastroenterologists, which include fecal testing every 2 years for people 50–74 years of age¹².

TABLE I.

Guideline recommendations related to cancer screening in average-risk populations from the Canadian Task Force on Preventive Health Care (CTFPHC)¹¹

Cancer site	Recommendation (CTFPHC grade)
Recommendations for screening
Breast cancer	□ For women 50–69 years of age, routine screening with mammography is recommended every 2–3 years (weak recommendation,a moderate-quality evidence). □ For women 70–74 years of age, routine screening with mammography is recommended every 2–3 years (weak recommendation,a low-quality evidence)8.
Cervical cancer	□ For women 25–29 years of age, routine screening for cervical cancer is recommended every 3 years (weak recommendation,a moderate-quality evidence). □ For women 30–69 years of age, routine screening for cervical cancer is recommended every 3 years (strong recommendation,b high-quality evidence)9.
Colorectal cancer	□ There is good evidence to support the inclusion of annual or biennial fecal occult blood testing (“A” recommendation) and fair evidence to include flexible sigmoidoscopy (“B” recommendation) in the periodic health examinations of asymptomatic individuals more than 50 years of age10.
Recommendations against screening
Breast cancer	□ For women 40–49 years of age, routine screening with mammography is not recommended (weak recommendation,a moderate-quality evidence)8. □ Screening at more than 75 years of age is also considered to be screening outside the current guidelines. □ Women undergoing mammography at interval greater than once every 2–3 years are considered to be screening outside the current guidelines.
Cervical cancer	□ For women less than 20 years of age, routine screening for cervical cancer is not recommended (strong recommendation,b high-quality evidence). □ For women 20–24 years of age, routine screening for cervical cancer is not recommended (weak recommendation,a moderate-quality evidence). □ For women 70 years of age and older who have been adequately screened (that is, 3 successive negative Pap tests in the preceding 10 years), it is recommended that routine screening for cervical cancer can cease (weak recommendation,a low-quality evidence)9. □ Women who have undergone more than 1 Pap test every 3 years are considered to be screening outside of the current guidelines.
Colorectal cancer	□ There is insufficient evidence to make recommendations about whether either or both of fecal occult blood testing and sigmoidoscopy should be performed (“C” recommendation). □ There is insufficient evidence to include or exclude colonoscopy as an initial screen in the periodic health examination (“C” recommendation)10 □ Individuals less than 50 years of age who have screened for colorectal cancer using fecal occult blood testing are considered to be screening outside of the current guidelines. □ Individuals who have screened for colorectal cancer using fecal occult blood testing more than once every 1–2 years are considered to be screening outside the current guidelines.

^aMost people in this situation would want the suggested course of action, but many would not¹¹. Different choices will be appropriate for individual patients, and clinicians must help each patient arrive at a management decision. Policymaking will require substantial debate.

^bMost individuals in this situation would want the recommended course of action¹¹. Most individuals should receive the intervention. Adopt as policy.

In Canada, cancer screening strategies and programs are the responsibility of the 10 provincial and 3 territorial health authorities. The scope of organized screening programs has varied across cancer sites. Breast cancer screening has a long history (since the 1990s) of being highly organized. Cervical cancer screening is only partly organized in most provinces and territories. Colorectal cancer screening programs using various methods of fecal testing have been gradually implemented since 2007. Opportunistic screening for all cancer sites also occurs outside of programs. Given that both types of screening exist in Canada, self-reported survey data provide a complete picture with which to measure screening test utilization within and outside of organized programs.

The aim of the present project was to compare recent trends in screening test utilization with the national guidelines issued by ctfphc. Reported screening behaviours are considered for the average-risk population, both within and outside the recommended age groups.

METHODS

Self-reported population data from the 2008 and 2012 versions of the Canadian Community Health Survey (cchs)¹³, an ongoing national cross-sectional survey that collects health information into a database administered by Statistics Canada (http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=3226), were obtained to calculate screening test utilization for mammography (during the preceding 2 years), Pap testing (during the preceding 3 years), fecal testing (during the preceding 2 years), and endoscopy [both sigmoidoscopy and colonoscopy (during the preceding 5 years)]. The proportions of the eligible population that had undergone the various tests for screening purposes were calculated using the indicator definitions presented in Table ii by cancer site. Rates were calculated for people at average risk within and outside the recommended age groups for each cancer screening site. “Average risk” is defined for each cancer site as those people who do not have the risk factors listed in the exclusion criteria for the analyses (Table ii).

TABLE II.

Indicators for the study analysis, data from the Canadian Community Health Survey, 2008 and 2013

	Cancer site (screening test)	Age groups (years)
Breast (mammography)
Indicator	Women who have undergone mammography within a 2-year period	<40; 40–49; 50–69; 70–74; 75+
Numerator	Women (n) who reported mammography screening within the preceding 2 years
Inclusion/exclusion criteria	A woman was deemed relevant if she underwent mammography for reasons of family history, regular check-up or routine screening, age, and use of hormone replacement therapy, and not for a lump, breast problem, follow-up to breast cancer treatment, or other reasons.
Denominator	All women respondents (n)
Cervix (Pap test)
Indicator	Women who have had at least 1 Pap test within a 3-year period	<20; 20–24; 25–29; 30–34; 35–39; 40–44; 45–49; 50–54; 55–59; 60–64; 65–69; 70+; and 20–69a; 21–69a ; 25–69a; 30–69a
Numerator	Women (n) who reported Pap testing within the preceding 3 years
Inclusion/exclusion criteria	Hysterectomy correction was applied by using the answer to a question that was asked of all respondent women about whether they had or had not undergone hysterectomy
Denominator	All women respondents (n), corrected for hysterectomya
Colon and rectum (fecal test, endoscopy)
Indicator 1	Individuals (both men and women) who have had a fecal test within a 2-year period	<50; 50–74; 75+50–74
Numerator	Individuals (n) reporting a fecal test within the preceding 2 years
Inclusion/exclusion criteria	Individuals were deemed relevant if they reported a screening test or tests for reasons of regular check-up or routine screening, age, and race, but not for family history, follow-up of a problem or colorectal cancer treatment, or other reasons.
Denominator	All respondents (n)
Indicator 2	Distribution of individuals who have had only a fecal test within a 2-year period, or only endoscopy within a 5-year period, or both a fecal test within a 2-year period and endoscopy within a 5-year period
Numerator A	Individuals (n) reporting a fecal test within the preceding 2 years and no endoscopy within 5 years
Numerator B	Individuals (n) reporting an endoscopy within 5 years and no fecal test within the preceding 2 yearsb
Numerator C	Number of individuals (n) reporting both a fecal test in the preceding 2 years and endoscopy within 5 yearsb
Inclusion/exclusion criteria	Individuals were deemed relevant if they reported a screening test or tests for reasons of regular check-up or routine screening, age, and race, but not for family history, follow-up of a problem or colorectal cancer treatment, or other reasons.
Denominator	Pool of the three types of tests

^aStandardized to year 2011 because of the large span of age groups.

^bIndividuals who underwent endoscopy as a result of a positive fecal test were excluded from the numerator.

The subset of women 50–69 years of age were also examined to estimate the proportion reporting screening across multiple cancer screening sites (breast, cervical, colorectal). The criteria already indicated for individual screening test utilization were used to calculate rates based on the 2008 and 2012 cchs.

RESULTS

Table iii summarizes the trends in national screening test utilization, 2008 to 2012, by age group.

TABLE III.

Trends in utilization of selected cancer screening tests in Canada, by age group, 2008 and 2012

Cancer screening		Age group (years)	Canadian Community Health Survey respondents
Site	Test		2008			2012
			Rate (%)	95% CI		Rate (%)	95% CI
				Lower	Upper		Lower	Upper
Breast	Mammographya	35–39	7.5	5.4	9.6	4.9	3.5	6.3
		40–49	31.0	28.6	33.4	29.0	26.3	31.7
		50–69	61.0	59.4	62.6	62.0	60.3	63.7
		70–74	52.0	48.6	55.4	54.0	50.9	57.1
		75+	25.0	22.7	27.3	24.0	21.8	26.2
Cervix	Pap testb	18–19	40.0	35.1	44.9	29.0	24.5	33.5
		20–24	74.0	70.6	77.4	64.0	60.2	67.8
		25–69c	78.9	76.9	80.9	78.6	76.6	80.6
		70–74	49.0	44.3	53.7	49.0	44.8	53.2
		75+	24.0	20.7	27.3	19.2	16.3	22.1
Colon and rectum	Fecal testa	35–49	2.8	2.4	3.2	2.7	2.1	3.3
		50–74	16.9	16.0	17.8	23.0	22.0	24.0
		75+	14.5	13.2	15.8	18.6	17.1	20.1

^aDuring the preceding 2 years.

^bDuring the preceding 3 years.

^cAge-standardized to the 2011 Canadian population because of the large span of age groups.

Screening in the Recommended Age Groups

Screening test utilization rates were significantly higher in the age groups for whom screening is recommended than in the age groups below and above the recommended groups. Rates of screening mammography and Pap testing in 2012 were similar to rates in 2008 (Table iii). In 2012, the screening mammography rate was 62.0% in women 50–69 years of age and 54.0% in women 70–74 years of age, and the rate of Pap testing was 78.6% in women 25–69 years of age. Rates of fecal testing for all individuals (men and women) in the recommended age group of 50–74 were much lower than rates of mammography and Pap testing in women, but the increase to 23.0% in 2012 [95% confidence interval (ci): 22.0% to 24.0%] from 16.9% in 2008 (95% ci: 16.0% to 17.8%) was statistically significant. Data about the use of the two testing options (fecal or endoscopy) showed that 23.0% of people had undergone either a fecal test or endoscopy in 2012. Of that 23.0%, 64.1% underwent fecal testing alone, 21.5% underwent endoscopy alone, and 14.3% underwent both tests (Table iv).

TABLE IV.

Distribution of individuals in Canada 50–74 years of age reporting a colorectal cancer screening test, by test type, 2008 and 2012

Test	Canadian Community Health Survey respondents (%)
	2008	2012
Fecal test alonea	66.7	64.1
Endoscopy aloneb	21.5	21.5
Fecal test and endoscopy	11.7	14.3

^aDuring the preceding 2 years.

^bDuring the preceding 5 years.

Screening Outside the Recommended Age Groups

In 2008 and 2012, screening mammography rates for women 40–49 years of age were stable at about 30%. Among women younger than those targeted in the guideline recommendations for cervical screening, rates of screening declined significantly over time, but remained high at 64.0% (95% ci: 60.2% to 67.8%) for women 20–24 years of age and 29.0% (95% ci: 24.5% to 33.5%) for women 18–19 years of age (Table iii). Fecal test utilization in people less than 50 years of age was extremely low in both years.

Rates of screening test utilization for all three cancer screening sites were significantly lower in the age groups older than the recommended age groups. In 2012, mammography screening was reported by 24.0% (95% ci: 21.8% to 26.2%) of women 75 years of age and older and by 49.0% (95% ci: 44.8% to 53.2%) of women 70–74 years of age. Pap testing in the preceding 3 years was reported by 19.2% (95% ci: 16.3% to 22.1%) of women 75 years of age and older (Table iii). In individuals 75 years of age and older, the fecal test rate was 18.6% (95% ci: 17.1% to 20.1%), and the endoscopy rate was 10.7% (95% ci: 9.3% to 12.1%) (data not shown).

Utilization of Screening Tests Across Multiple Cancer Sites

In 2012, significantly fewer women 50–69 years of age reported undergoing no screening for any of the three cancer sites [5.9% (95% ci: 5.1% to 6.7%) compared with 8.5% in 2008 [95% ci: 7.4% to 9.6%)], and significantly more women reported undergoing screening for all three cancer sites [19.0% (95% ci: 17.5% to 20.5%) compared with 13.2% in 2008 (95% ci: 12.2% to 14.2)] (data not shown). The distribution of screening test utilization among women who had reported a screening test for at least one cancer site indicated that, from 2008 to 2012, the proportion of women reporting all three screening tests increased to 22.8% from 16.2% (Figure 1).

FIGURE 1.

Distribution of cancer site screening tests, by type, in Canadian women who reported utilization of at least one screening test, 2008 and 2012. Women were deemed to have had at least one test if they reported a Pap test during the preceding 3 years; mammography during the preceding 2 years for family history, regular check-up or routine screening, age, or use of hormone replacement therapy, but not for lump, breast problem, follow-up to breast cancer treatment, or other reasons; a fecal test or endoscopy during the preceding 5 years, for regular check-up or routine screening, age, and race, but not for family history, follow-up for a problem, follow-up of colorectal cancer treatment, or other reasons. If a respondent reported both a fecal test and endoscopy, that respondent was considered to have had one test. Endoscopy as follow-up to a fecal test was excluded. Data source: Statistics Canada, Canadian Community Health Survey. Denominator: All female respondents 50–69 years of age who reported having had at least one of three tests (Pap test, mammography, or fecal test or endoscopy).

DISCUSSION AND CONCLUSIONS

In Canada, overall utilization of mammography in 2012 for women 50–69 and 70–74 years of age was 62.0% and 54.0% respectively. Utilization of Pap testing in 2012 for women 25–69 years of age was higher, at 79.0%. The overall utilization of fecal testing for colorectal cancer screening in individuals 50–74 years of age was 23.0%. Unlike screening for breast and cervical cancer, which were stable in 2008 and 2012, screening by fecal testing showed a notable rate increase over time. Colorectal screening rates have not yet stabilized, and it will be interesting to see when and whether they will reach rates that are similar to those for the other cancer screening sites. A study from Spain demonstrated that, in 2007, overall participation was 67.0% for breast cancer screening and 69.0% for cervical cancer screening; colorectal cancer screening varied across the 6 regions in which screening was available, ranging from 17.2% to 42.3%¹³.

Analyses of colorectal cancer screening have to consider multiple modalities, and in 2012, the number of people reporting the use of both fecal testing and endoscopy increased. That combined use of tests, as well as screening in populations for whom screening is not recommended, could affect resources, increase the cost to the health care system, and reduce access to services for target populations and symptomatic individuals.

The present study yields an initial snapshot of the utilization of screening tests in the age groups for whom population-based screening is and is not recommended. Utilization of screening outside the recommended guidelines might result in less or no benefit to those who are screened and could increase the likelihood of potential harms, including over-screening and unnecessary diagnostic procedures for false-positive results. However, careful interpretation of the data is required, because self-reported survey data might include individuals with an increased risk of cancer and those undergoing testing for diagnostic reasons. Although comparing behaviours against the most recent guidelines is common practice, earlier recommendations should also be considered when interpreting data collected during the years before release of the newest guidelines. In the ctfphc guidelines before 2013, Pap testing was recommended for women less than 25 years of age. Similarly, before 2011, breast cancer screening was not recommended for women 70–74 years of age, and no clear direction had been articulated for screening women 40–49 years of age. Informed by the results reported here, future studies should investigate mechanisms for improving participation by examining the multifactorial barriers and facilitators to participation according to current guidelines.

It was encouraging to note that, of women 50–69 years of age reporting screening for single or multiple cancer sites, only 6% in 2012 reported no screening at all. Although testing for colorectal cancer is increasing, only 19% of women are currently up-to-date with all three screening tests analyzed here. Because an increase in colorectal cancer screening, together with Pap testing and mammography, was noted for female cchs respondents in 2012, and because screening rates for breast and cervical cancer are relatively stable, it is likely that women have added colorectal cancer screening to their previously established screening routines. That observation confirms the importance of reaching underserved populations and engaging them in screening for at least one cancer site. Additional efforts to encourage use of screening within the recommended average-risk age groups are needed, as is continued education for health care providers and the public about the possible harms of screening outside those age groups.

A limitation of the present study is its use of self-reported data, which could be affected by biases related to the accuracy of recall about screening history and to the social desirability of certain responses, potentially resulting in inaccurate reporting of screening^14,15. However, that limitation would be similar for all respondents. Many studies have concluded that self-reporting is fairly accurate, showing good or fair concordance and overall agreement with administrative health data^15–18. Estimates from an earlier study on breast cancer screening in Canada¹⁹ (based on administrative data), of the proportion of women 50–69 years of age who underwent bilateral mammography within and outside organized programs in a 24-month period, closely resemble the cchs self-reported screening rates. Another limitation is that the cchs questions might not capture enough information to determine whether respondents are truly at average risk. Despite those limitations, the cchs offers the advantages of providing data for all provinces and territories in Canada for the same period and of allowing for the same methodology (inclusions, exclusions, and other adjustments) to be applied for all provinces.

CONFLICT OF INTEREST DISCLOSURES

We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.