Coming soon to you: Direct access imaging for suspected cancer…Maybe

How to improve timelines for cancer diagnosis has been a hot topic for years. The Long Term Plan (LTP) for cancer announced in 2019 boldly set out to increase the number of patients diagnosed with early stage 1 or 2 cancers from the current ~50% to 75% within 10 years. I think many of us were somewhat skeptical at the time as to whether this was an achievable aim, but then COVID hit and all bets were off. Indeed policy experts have warned (BMJ 2021;374:n2352) that it could take 10 years to clear the current backlog of patients on cancer pathways, and some experts have declared the cancer LTP now not fit for purpose (BMJ 2022;379:o2694).

To improve early diagnosis of cancer there will inevitably need to be significant changes across the whole cancer pathway (as well as substantial investment) but one area that could improve time to diagnosis is GP direct access to imaging. And in November 2022 NHSE announced plans to do exactly that, with a promise of better access to CT, MRI and ultrasound scanning for primary care, as reviewed recently in the BMJ (BMJ 2023;380:e074766).

It’s worth going back a step first to review how direct access to imaging for GPs might fit in and why it may be important. The urgent referral pathway for suspected cancer (AKA the 2 week wait/2WW in England) has been a double edged sword. People who have ‘red flag’ or alarm symptoms or signs that meet the 3% PPV threshold for the pathway clearly have quicker and better access to imaging and specialist assessment. And the latest data analysis (BJGP 2022; 72 (714): 34-37) shows the pathway has had a positive impact - the cancer detection rate (the percentage of new cancer cases needing treatment that have been referred via the 2WW pathway) has increased from 42.3% to 53.5% between 2009/10 to 2019/20. This means more clinically meaningful cancers will have been detected promptly through the 2WW system leading to earlier treatment. The flip side is we have created a two tier system for cancer diagnosis, whereby those on the 2WW system get prompt access and treatment, but those that don’t fulfil the the 2WW criteria (still the best part of 50% of all cancers) are waiting even longer for their diagnosis. This gap has been heightened with the now substantial increased waits for ‘routine’ hospital appointments for patients, some of whom will have cancer. So allowing GPs direct access to imaging may help mitigate some of that widening inequality, and speed up diagnosis for some patients. 

So is there any evidence GP direct access (DA) to imaging works? The evidence is limited, but a systematic review from 2018 was supportive of the idea (BJGP 2018; 68 (674): e594-e603). The evidence was generally of low quality, but analysis of 60 papers showed that conversion rates (that is the proportion of patients diagnosed with cancer who had imaging) were similar between GP DA imaging and those that first saw a specialist, countering the argument that GP direct access to imaging would lead to ‘over-referral’. Time from referral to testing was significantly quicker via the GP DA route, rather than via specialist referral route, and both patient and GP satisfaction was consistently high for direct access pathways. 

So there is both some underlying evidence to support GP direct access to imaging and good rationale as to how it would fit into our current cancer pathways, by speeding up diagnosis for some people who fall outside the 2WW pathways. Indeed this is very much the aim as set out by NHSE that we would have access to CT, ultrasound and brain MRI on an urgent basis for people who don’t fit urgent cancer referral pathways, but who have ‘concerning symptoms’, although that crucial wording of ‘concerning symptoms’ has yet to be defined. The other argument to support this strategy, as discussed in the BMJ (BMJ 2023;380:e074766) is to improve equality of access to imaging. Some NHS areas already have good access to imaging from primary care, both CT and MRI, whereas others have none. 

But the BMJ finishes with a cautionary note that could make or break such a proposal, with the massive elephants in the room - kit and workforce. A sobering stat bomb is that among the 38 OECD countries we lie 34th in terms of number of CT scanners per million population. Per million population most OECD countries have double the number of scanners that we do, e.g. France, Sprain, and many e.g. Germany, Denmark have 3-4 times the number of scanners. Indeed only Mexico, Columbia and Costa Rica have fewer CT scanners per million population than us. Then there is the workforce issue - patients have to be able to access their GP first (increasingly difficult) and then we need radiographers to do the scan (UK estimated vacancy rate >10%) and radiologist to report the scans (current workforce shortfall estimated at >1600). So let’s hope that appropriate investment is made to allow for what should, in theory, be a very positive pathway for direct access to cancer imaging by GPs.