CRC mortality rates

It will be several years before the impact of population screening on CRC mortality becomes observable, and many more years before the full effect is achieved. The timing of a reduction depends on the natural history of the disease, and the ‘lead time’ due to screening (i.e. the time by which screening advances the date of diagnosis) as well as on the time taken to cover the target population. It will also depend on the quality of screening. 

Methods to evaluate the impact of screening on CRC mortality include analyses of population trends, cohort studies (aggregated or individual-based) and case-control studies.

Population trends 

Mortality from CRC has been decreasing in many European countries since the mid 1990’s, (Karim-Kos et al. 2008). Analyses of the routinely produced age-gender specific population rates over time will be subject to limitations due to the dilution of the effect of screening from deaths occurring in cases diagnosed prior to the introduction of screening, and/or at an age below which invitations begin. This can be overcome by use of refined CRC mortality rates in which such deaths are excluded. However, the rates will also be confounded by other factors such as cohort effects on underlying incidence, and by the effects of improvements in treatment and/or the stage of diagnosis of symptomatic disease on survival and mortality. Thus whilst a lack of any reduction in population mortality rates several years after the introduction of a screening programme should be a cause for concern, it is difficult to use such trends to quantify the effect, and attempts to do so should take account of the factors discussed above. 

Cohort studies 

In some settings, the introduction of population screening will have been phased in such a way as to facilitate comparisons of populations invited at different time points. Such a model has been used in Finland (see Ch. 2, Sect. 2.6.4). In the absence of such a system, comparisons can be made between geographical areas (regions invited/not invited to screening) or between the same population in different time periods before and after the introduction of screening. Both types of comparison are liable to possible bias due to underlying differences in the risk in the populations/time-periods. This may – under certain circumstances – be compensated for by including also a comparison group from geographic areas where no invitational program existed from before the introduction of screening. Cohort studies using aggregated data need estimates of incidence in order to avoid dilution effect discussed above. 

These biases can be avoided by individual-based cohort studies in which deaths and cancer registrations are linked to screening histories.

Interval cancers

Interval cancers are those that occur following a negative screening episode, in the interval before the next invitation to screening is due. For faecal occult blood testing interval cancers may occur following a negative FOBT, or following a positive test result with negative further assessment (colonoscopy). Rates of interval cancers reflect both the sensitivity of the screening test (false negatives), and the incidence of newly-arising cases not present at the time of screening. With increasing time since negative test, the rate and proportion of the latter will increase. In the absence of repeat screening, incidence rates would eventually reach the background level again. Rates of interval cancers should therefore be presented by time period (years) since previous screen.

Rates of interval cancers will depend on the underlying incidence in the population. They will also depend on the extent of selection bias, whereby rates in those not participating in screening differ from the general population rates. 

For this reason it is important that (age- and gender-specific) incidence rates in non-responders are also monitored, to allow for the underlying incidence in responders to be estimated. Background incidence rates can be estimated from rates prior to the introduction of screening (although time trends need to be considered) or from areas not covered by the screening programme (when geographic differences need to be considered).  

The interval cancer rate can therefore be expressed as a proportion of the background incidence rate, standardised for age and gender, by dividing the number of interval cancers in the specific age/gender group (I) by the ones expected based on the background incidence for that age/gender group (C), or as a proportion of the background incidence rate adjusted for non-participants (C*). The adjusted rate can be calculated as:

C* = (C – (1 – P) N) / P 

P: participation rate 

N: rate in non-responders  

Uptake (participation) rate

The number of people who have been screened, within a defined time frame following an invitation, as a proportion of all people who are invited to attend for screening.

The effectiveness of the programme will depend on the participation rate. In the randomised FOBT trials, uptake at the first round was between 49.5% and 66.8% (Table 3.2); uptake at subsequent rounds varied according to the policy for reinvitation. In a US study that recruited volunteers 75%–78% of subjects invited were screened at least once (Mandel et al. 1993). Reported uptake in population-based programmes ranges from 17.2% to 90.1% at the first round; the range at subsequent rounds is smaller 

N people invited and screened/tested during the time frame* 

N eligible people invited during the time frame*  

* equal to the defined screening interval or reporting period

For flexible sigmoidoscopy, uptake rates in RCTs ranged from 32.4% to 83.5%, again with high rates being associated with recruitment of volunteers or those who had expressed interest in participation). In population-based programmes, uptake rates range from 7% to 55%

Outcomes with faecal occult blood testing (FOBT) for primary screening  

FOBT indicators will vary according to the type of test used and programme policy, and therefore these should be reported.

Inadequate FOBT rate  

The rate of inadequate tests is defined as the proportion of people screened with one or more FOBT returned during the respective time frame (e.g. a 12-month period) none of which were adequate.

Rates of inadequate tests should remain low. They reflect, among other things, the understanding of the people who are using a test and therefore also the quality of the information provided to them. 

In population-based programmes, inadequate gFOBT rates between 0.4% and 4.5%  have been reported. No data are available yet for iFOBT. 

EVALUATION AND INTERPRETATION OF SCREENING OUTCOMES

Age should be recorded as the age of the person at the time of the invitation (for measurement of coverage/participation) or at time of screening (for measurement of screening outcome) for the respective screening round. The outcome of the screening examination for a person should thus be recorded in the same age category throughout a particular screening episode.

Screening performance indicators will also be affected by the background incidence in the target population in the absence of screening. Efforts should therefore be made to document age-gender specific incidence rates in the target population for the period immediately prior to the introduction of the screening programme. 

If high-risk subjects are identified, managed, and/or excluded from the programme and reported separately, this should be stated.

 Performance indicators will also vary according to whether the screen is a prevalent (first) screen for those invited for the first time, an incident (repeat) screen for those previously screened at the routine interval, or a screen for previous non-responders. Indicators at subsequent rounds will vary according to the screening interval.  

Only the first organised screening round will consist entirely of subjects invited and attending for the first time; all additional rounds will comprise subjects falling into each of the categories described above. The cut-off point for separating ‘subsequent regular’ from ‘subsequent irregular’ screening should be established, taking into consideration that most programmes do not succeed in inviting each individual participant at the routine screening interval (e.g. a cut-off point at 30 months for a programme with a 2-year screening interval).

Programme outcome variables

The following outcome variables apply to CRC screening performed with any of the currently available primary screening tests.

Follow-up colonoscopy  

Participants in the group on which diagnostic or therapeutic colonoscopy4 has been performed to follow-up primary screening according to programme policy include participants, the screening endoscopy of which was inadequate or incomplete. Note that each person is counted once regardless of the number of follow-up colonoscopies that were performed. Where more than one colonoscopy or other follow-up investigation is performed, the reported result should be that of the complete diagnostic or therapeutic work-up.

Lesions 

Any lesion removed or biopsied at endoscopy or surgery (whether or not they were diagnosed as adenomas) should be recorded.  

Adenomas 

Pathological specimens removed at endoscopy or surgery that have been reported by a pathologist to be adenomatous should be recorded.

Advanced adenoma

 If it is not possible to collect such details for organisational reasons, the programme should at least focus on collecting and reporting data on adenomas 10 mm in size 

Cancers

 Colorectal cancer diagnosed by the screening programme, or diagnosed as a direct result of participating in the screening programme 

 List of recommended data tables to be produced by CRC screening programmes

1. Targeted 

2. Eligible 

3. Invited 

4. Screened/tested at first screening and at subsequent screening episodes 

5. Inadequate tests 

6. Positive test or screening 

7. Follow-up colonoscopy examination attended (diagnostic assessment and/or treatment) 

8. Negative follow-up colonoscopy examination (diagnostic assessment and/or treatment) 

9. Positive follow-up colonoscopy examination (diagnostic assessment and/or treatment) 

10. Lesion detected (at least one) 

11. Adenoma detected (at least one) 

12. Non-advanced adenoma detected (at least one) 

13. Advanced/high-risk adenoma detected (at least one) 

14. Cancer detected by stage 

EVALUATION AND INTERPRETATION OF SCREENING OUTCOMES

Evaluation and interpretation of screening outcomes are essential to recognising whether a colorectal cancer screening programme is achieving the goals for which it has been established. It is recognised that the context and logistics of screening programmes will differ by country and even by region. For example, the prior existence of a population register facilitates issuing personalised invitations, whereas the absence of such a register may lead to recruitment by open invitation. Many of these contextual differences will affect the measured outcomes.

The effectiveness of a programme is a function of the quality of its individual components. Success of the programme is measured not only by its impact on public health, but also by its organisation, implementation, and acceptability. 

To determine whether a programme has been effective with regard to its impact on morbidity and mortality requires continuous follow-up of the target population over an extended time-frame. Therefore early-performance indicators using standard definitions, available early in the lifetime of a screening programme are essential to measure the quality of the programme and its potential longer-term impact

A key component in the evaluation of population screening programmes is data collection. Colorectal cancer screening can be performed using various tests or techniques. Data collection necessary for evaluation can be common to all tests or specific to particular tests. The examples given in these Guidelines refer to in vitro stool tests based on detection of faecal occult blood (FOBT) that are currently the most widely used, and to endoscopic tests i.e. flexible sigmoidoscopy (FS) or colonoscopy (CS). In the text, gFOBT refers to guaiac-based FOBTs, and iFOBT to immunological FOBTs. 

It should be noted that in a setting where opportunistic screening (for example by colonoscopy) has been taking place for some time, the uptake and performance of an organised programme may differ markedly from those in a setting where no such screening has been taking place. The majority of the values of the indicators described below will relate to the latter setting. 

Data items necessary for evaluation  

Programme conditions 

Programme type 

As mentioned above, the organisational aspects of a screening programme influence the evaluation and interpretation of screening outcomes. Population-based programmes are recommended because they require an infrastructure that is conducive to implementation of quality assurance and evaluation, such as through linkage of screening data and cancer registry data (Karsa et al. 2010). It is therefore important to document the type of programme (population-based or non-population-based) and to describe the sources of population data used for identification and invitation of the eligible target population (e.g. population registry). Data on screening outcomes should be linked with data from other registries in order to monitor and evaluate the programme. 

Primary screening test 

Currently only the faecal occult blood test (FOBT) is recommended by the EU for CRC screening. However endoscopic screening programmes with flexible sigmoidoscopy (FS) or colonoscopy (CS) as primary screening tests are currently running in a number of Member States. Given the potential impact of the type of primary screening test or tests used in a programme on the respective results and performance, the type of primary screening test should always be indicated when documenting results and reporting. 

Population base 

A screening programme is population based when every member of the target population in the area designated to be served by the programme is known to the programme, and when the eligible members of the target population are individually invited to participate. 

The availability and reliability of target population data will depend on the existence, quality and accessibility of population registers in the region where the programme is being set up. Population registers are not always available and demographic data for identifying the target population might be obtained from various sources, e.g. census data, electoral registers, private or statutory health care registers or health insurance funds registers. The choice of the target population database for issuing invitations will depend on the completeness of the database and on the individuals or variables included, e.g. electoral registers might not include eligible foreigners or dates of birth.

A database consisting of individual records (one record per person for each screening episode) is essential in order to produce results on organisational aspects of the programme (coverage, participation) and screening performance. The data collected should respect a logical order and follow the development of the screening process (identification of person [date of birth, gender], date of invitation, date of reminder, date of test, test results, date of the examination performed during assessment, results, colonoscopy date, results, adverse effects, treatment). The location in the bowel of any detected lesions or cancers (Tumour site) should also be recorded [Rectum, sigmoid, descending colon (distal colon) transverse colon, splenic flexure, ascending colon]. 

Each variable should be precisely defined. All data collected for each round should be kept and updated information should not overwrite data provided during preceding rounds. All information on the timing of events during each screening episode, including invitation history, should be recorded as calendar dates. This ensures maximal flexibility of the database for future evaluation efforts and participation in multi-centre studies. It also permits distinguishing between the first and subsequent screening episodes and between participants with different patterns of attendance