Performance indicator 6a: Interval cancers
Summary
For the index years 2018, 2019, and 2020, for participants aged 50–74, the interval cancer rates for both first and subsequent screening rounds were higher in the second year (13–24 months) after a negative screening episode than in the first year (0–12 months).
Definition
The number of invasive breast cancers detected in participants aged 50–74 screened through BreastScreen Australia that arise during an interval between 2 screening rounds, per 10,000 participants.
Rationale
The ability of screening mammography to successfully detect invasive breast cancer in women in the target age group can be assessed by considering the relative number of:
- invasive breast cancers detected at a screening episode
- invasive breast cancers diagnosed 0–12 months after a screening episode detected no cancer
- invasive breast cancers diagnosed 13–24 months after a screening episode detected no cancer.
The goal of BreastScreen Australia is to have a high proportion of invasive breast cancers detected within screening episodes and a low proportion diagnosed after a screening episode detected no cancer (interval cancers).
Guide to interpretation
Interval cancer rates are disaggregated into time since screening (0–12 months, 13–24 months, and 0–24 months) and screening round (first and subsequent).
To produce comparable rates from the relatively small number of cases, interval cancer rates are reported by 10-year age groups and aggregated over 3 years.
Interval cancer data are reported per 10,000 participants screened.
The latest data for interval cancers are for participants screened in 2018, 2019, and 2020 (referred to as index years 2018–2020). These are the latest data available because at least 2 years need to have passed since a participant’s last routine screening mammogram in order to know whether they were diagnosed with an interval cancer in that time – though this time period is often longer than 2 years due to time required for cancer registries to be notified of the cancer, and for linkage between the BreastScreen registers and cancer registers to occur.
More information on interval cancers
Interval cancers are invasive breast cancers that are diagnosed after a screening episode that detected no cancer and before the next scheduled screening episode (Kavanagh at al. 1999). For most participants, the next screening episode will occur around 24 months after their previous negative screening episode, as the recommended screening interval for most participants in BreastScreen Australia is 24 months. The exception to this is participants on annual screens, for whom the next screening episode will occur around 12 months after their previous negative screening episode.
An interval cancer may be:
- an aggressive breast cancer that emerges and grows very rapidly in the period between screening episodes
- a breast cancer that, due to the characteristics of the cancer or the breast tissue, is not visible on screening mammography and therefore not able to be detected
- a breast cancer that can be retrospectively detected on the previous screening mammogram.
The first two types of interval cancer described above are true interval cancers, and therefore do not represent any failure in detection; the third represents a failure of the screening process. Through the BreastScreen accreditation process, state and territory BreastScreen programs are required to audit interval cancers. All interval cancers in all state and territory BreastScreen Programs undergo clinical review. On investigation, more than 80% are found to be true interval cancers.
Interval cancers may be detected outside BreastScreen Australia or through BreastScreen Australia, depending on the policies for screening symptomatic women in each state and territory that can affect interval cancer detection rates.
Results
For the index years 2018–2020, for participants aged 50–74, the interval cancer rates for both first and subsequent screening rounds were higher in the 13–24 months after a negative screening episode than in the 0–12 months after a negative screening episode.
For the index years 2018–2020, for participants aged 50–74, in the 0–12 months after a participant’s first negative screening episode, there were 7.3 interval cancers per 10,000 participants. In the 0–12 months after a subsequent screening episode, there were 6.3 interval cancers per 10,000 participants (Table 6.1a and Table 6.1b).
For participants aged 50–74, in the 13–24 months after a participant’s first negative screening episode, there were 11.1 interval cancers per 10,000 participants. In the 13–24 months after a subsequent screening episode, there were 11.5 interval cancers per 10,000 participants (Table 6.1a and Table 6.1b).
For the index years 2018–2020, there were no appreciable differences in the interval cancer rate between the first and subsequent screening rounds for either 0–12 months or 13–24 months after a negative screening episode (apparent differences are considered not statistically significant due to overlapping confidence intervals). This indicates that the likelihood of participants being diagnosed with an interval cancer is similar between the first and subsequent screening rounds for 0–12 months or 13–24 months after a negative screening episode (Table 6.1a and Table 6.1b).
First screening round | 0–12 months | 13–24 months | 0–24 months |
|---|---|---|---|
Number | 174 | 246 | 420 |
Crude rate | 7.3 | 11.1 | 9.1 |
AS rate | 7.5 | 11.3 | 9.3 |
95% CI | 6.1–8.9 | 9.6–13.1 | 8.2–10.5 |
Notes:
- Crude rate is the number of interval cancers detected per 10,000 person-years; age standardised (AS) rate is the number of interval cancers detected per 10,000 person-years, age standardised to the population of participants attending a BreastScreen Australia service in 2008; 95% CI are 95% confidence intervals.
- For information on confidence intervals, see Box A6.1.
Source: AIHW analysis of BreastScreen Australia data.
Subsequent screening rounds | 0–12 months | 13–24 months | 0–24 months |
|---|---|---|---|
Number | 1,559 | 2,631 | 4,190 |
Crude rate | 6.3 | 11.5 | 8.8 |
AS rate | 6.3 | 11.3 | 8.7 |
95% CI | 5.9–6.6 | 10.8–11.8 | 8.4–9.0 |
Notes:
- Crude rate is the number of interval cancers detected per 10,000 person-years; age standardised (AS) rate is the number of interval cancers detected per 10,000 person-years, age standardised to the population of participants attending a BreastScreen Australia service in 2008; 95% CI are 95% confidence intervals.
- For information on confidence intervals, see Box A6.1.
Source: AIHW analysis of BreastScreen Australia data.
Interval cancers by age
For the index years 2018–2020, in the 0–12 months after a negative screening episode, the interval cancer rates for all screening rounds were lower for participants aged 50–74, and higher for participants outside these age groups (Table 6.2). In the 13–24 months after a negative screening round, the interval cancer rates were lower for participants aged 40–49. 50–59, and 60–69, and higher for participants outside these age groups.
Age group (years) | Number (0–12 months) | Crude rate (0–12 months) | Number (13–24 months) | Crude rate (13–24 months) |
|---|---|---|---|---|
40–49 | 279 | 8.5 | 327 | 11.7 |
50–59 | 740 | 6.4 | 1,121 | 10.7 |
60–69 | 701 | 6.3 | 1,254 | 12.0 |
70–74 | 292 | 6.8 | 502 | 12.5 |
75+ | 96 | 10.0 | 141 | 15.8 |
Note: Crude rate is the number of interval cancers detected per 10,000 person-years.
Source: AIHW analysis of BreastScreen Australia data.
Interval cancers by state and territory
For the index years 2018–2020, in the 0–24 months after a negative screening episode, the interval cancer rate for participants aged 50–74 varied across states and territories, from 6.6 to 9.7 per 10,000 participants (for all screening rounds).
Differences in state and territory policies for managing women with symptoms may affect interval cancer rates. For example, in some jurisdictions, women with a negative screening mammogram but who have symptoms are referred for diagnostic follow up outside BreastScreen Australia, rather than being recalled for assessment within BreastScreen Australia. Any cancers found in these women will be counted as ‘interval cancers’, leading to a higher apparent interval cancer rate. On the other hand, jurisdictions that do recall women to assessment if they have symptoms (even in the face of a negative screening mammogram) may have lower apparent interval cancer rates. These differing practices and policies affect the comparability of this indicator across jurisdictions.
Kavanagh AM, Amos AF & Marr GM 1999. The ascertainment and reporting of interval cancers within the BreastScreen Australia Program. Sydney: NHMRC and NBCC.