Covid lockdown and cancer survival - what really happened?

For 2020 as a whole, there were 25,470 new cancer cases identified - 13 per cent fewer than the average for 2018 and 2019.

READ MORE: What's happening with cancer? - waiting times, incidence, and mortality

We also know that the cases which were picked up tended to be found at a later stage.

This was particularly the case for cancers which are targeted through screening, with figures for Scotland previously showing that the first year of the pandemic was associated with a 20% reduction in the number of breast cancers detected at an early stage, as well as 33% and 45% dips respectively in early-stage detections of colorectal and cervical cancers.

Graph showing the drop-off in cancer cases recorded during 2020 compared to the two years prior to the pandemic (Image: PHS)

The question is, did this actually result in poorer survival outcomes for people with cancer in 2020?

Three years on, a report from Public Health Scotland has gone some way to answering that for the first time - but the conclusions are not entirely black and white.

On the one hand, PHS estimated that the one-year survival rate for patients diagnosed with any form of cancer in 2020 was 67.5% versus 71.1% for those diagnosed in 2018 and 2019.

At face value, that seems like a clear-cut detrimental impact of the pandemic on patients' chances of surviving cancer.

However, as PHS goes on to note later in the report, it is "not possible to conclude (from the data presented here) whether the patients who were diagnosed in 2020 actually had poorer outcomes than they would have had in a 'normal' year".

READ MORE: Why are cancer deaths lower in 2022 than before Covid? 

If that seems counterintuitive, it is worth considering that the patients identified as having cancer during 2020 were, by definition, more seriously ill: a larger proportion were patients with symptoms whose disease had been found at a more advanced stage.

As PHS puts it, "the 2020 cohort was a smaller cohort with a higher proportion of more advanced cancers compared with pre-pandemic years".

To some extent then, we are comparing apples and oranges.

In other words, it is not necessarily the case that survival rates worsened per se but that outcomes may have been skewed by the changing profile of the cancer cases detected during 2020.

Notably, as screening programmes resumed in the second half of 2020, the one-year survival estimates for patients diagnosed in that period (who would include patients with early-stage breast, cervical and bowel cancers picked up from screening programmes) returned to levels more in line with pre-pandemic trends.

Known colorectal cancer incidence was slower to return to pre-pandemic levels (Image: PHS)

It is also the case that the "overall survival" figure includes deaths from all causes - not necessarily cancer - and that Covid infections posed a particularly high risk to some cancer patients, especially if they were elderly, immuno-suppressed, or had certain forms of the disease, such as blood cancers.

That said, PHS stresses that its own analysis - after adjusting for background mortality - suggests that direct Covid deaths among cancer patients "had a negligible impact on the lower overall survival estimates for the cohort diagnosed with cancer in 2020".

Caveats aside, however, the PHS report certainly does not rule out the possibility that patients diagnosed with cancer suffered adversely in 2020 - particularly during the first lockdown from April to June, when infection controls on healthcare were tightest.

READ MORE: Glasgow man's bowel cancer 'wiped out' after screening find

It examined trends for all cancers combined, as well as the four most common - breast, lung, prostate, and colorectal - and cervical, because it was affected by the screening pause.

The statistics for colorectal cancer are particularly stark.

For patients diagnosed with bowel cancer in 2020, one-year survival is estimated to have been 6% lower compared to patients in 2018 and 2019, while the age-standardised net survival (which adjusts for year-on-year differences in the age profile of patients and removes other, non-cancer causes from the equation) was down by 5.2%.

Colorectal cancer showed a statistically significant change after adjusting for age and other causes of death. For cervical cancer there were too few deaths in the youngest age group and too few cases in older women for reliable estimates (Image: PHS)

One-year survival took longer to recover to pre-pandemic levels in colorectal cancer compared to other common forms of cancer (Image: PHS)

When one-year survival for patients diagnosed in Q2 of 2020 (peak lockdown) is compared against Q1 (January to the end of March, a period when the NHS was largely unaffected by Covid), the difference is even wider: 13%.

PHS stresses that both these figures appear to be "statistically significant" - in other words, they are unlikely to be down to chance. The picture is less clear or consistent for the other forms of cancer.

Routine screening was paused in March but remained suspended longer for colorectal cancer than for the other screened cancers (until October, compared to July for cervical and August for breast).

Did this have an effect?

Under normal circumstances, all adults aged 50-74 are invited to take part in screening.

Looking at "net survival" (to remove the effect of non-cancer causes of death) the PHS report found that the biggest dip occurred in the 55-64 age group - down by 7.6% for the 2020 cohort compared to 2018/19 patients.

That compares to 1.7% for the 15-44 age group, and 4.1% for the 75-99 age group, neither of whom are eligible for screening.

READ MORE: Deaths in Scotland are on the rise - so what's the cause? 

Other factors - such as access to diagnostic tests - may also be a factor, however.

Waiting lists for colonoscopies have ballooned from 9,897 in March 2019 to 14,506 by March 2023, with the number waiting over 13 weeks climbing nearly four-fold from 1,481 to 5,679.

As for the longer term legacy, that remains a conundrum.

The latest data covering January to April 2023 shows that the age-standardised mortality rate (ASMR) for cancer as a whole in Scotland was 2.1% lower than the five-year average.

That translates to roughly six fewer cancer deaths than expected per 100,000 people living in Scotland, at a time when the overall mortality rate - from all causes - was up by 2.6% (an extra 31 deaths per 100,000).

But as the PHS report illustrates, that may vary depending on the type of cancer involved and - for now - we don't have that kind breakdown.