Today, we’re speaking to Marta Berglund, a Research Assistant and PhD candidate at University College London.

Title of paper: Pre-diagnostic primary care consultations and imaging in emergency-diagnosed vs referred lung cancer patients

Available at: https://doi.org/10.3399/BJGP.2025.0369

It has been postulated that emergency diagnoses of cancer (which occurs frequently and confers a poorer prognosis) may relate to suboptimal diagnostic management in primary care, but evidence to support or refute this hypothesis is sparse. We found that emergency-diagnosed patients with lung cancer were less likely to present with relevant respiratory symptoms and had fewer chest imaging investigations before diagnosis compared to patients diagnosed via referred routes, indicating an important role of disease factors in emergency diagnosis.

Transcript

This transcript was generated using AI and has not been reviewed for accuracy. Please be aware it may contain errors or omissions.

Speaker A

00:00:01.200 - 00:01:06.690

Hello and welcome to BJGP Interviews. I'm Nada Khan and I'm one of the Associate editors of the Journal. Thanks for listening to this podcast today.

In today's episode, we're speaking to Marta Bergland. Marta is a research assistant and PhD candidate at university College London.

She's recently published a paper here in the BJGP titled Pre Diagnostic Primary Care Consultations and Imaging in Emergency Diagnosed versus Referred Lung Cancer Patients.

So, hi, Marta, it's really lovely to meet you and it's great to talk to you about cancer diagnosis, which is a really important area for general practice and also a topic we publish on quite a lot here in the bjgp. It's been fascinating reading this paper and it tackles a cancer that we really don't do well with here in the UK in terms of early diagnosis.

But I wonder if you could just start off by telling us a little bit more about lung cancer and how it's actually often diagnosed, which can sometimes lead to some of those poor outcomes.

Speaker B

00:01:07.010 - 00:02:26.970

So, as you said, lung cancer is one of the cancer sites in the UK where the majority or a large proportion of patients are diagnosed through the emergency route, also known as emergency presentations, which is when a patient is diagnosed after they present through an urgent hospital admission or an A and E attendance in the 30 days before diagnosis.

And that could look like someone who has had a persistent cough for, say, two weeks, but didn't actually go to their gp, and then suddenly they have a more severe onset of symptoms like shortness of breath, and then they go to A and E and are referred to a chest X ray and then get diagnosed, which is a pathway that is associated with worse prognosis and worse outcomes after diagnosis. And the more preferred route, if you will, in England to diagnose patients is through primary care.

So through the GP routine referral or the urgent suspected referral route.

And that could look like someone who presents to primary care with cough or dyspneasia again, but then their GP refers them to a chest X ray and then they're diagnosed with lung cancer.

Speaker A

00:02:27.130 - 00:02:45.290

And I guess I wanted to just before we talk about what you found, I wanted to just cover here again, what you mean by this term, that's diagnostic window, because you mentioned that a few times in the paper. But what does this actually mean?

And it relates a bit back to some of what you're talking about, about people presenting with symptoms, isn't that right?

Speaker B

00:02:45.530 - 00:03:09.880

So, for us, it's a measure of healthcare use before diagnosis, and it could be any healthcare use Measure like consultations, symptoms, blood test use, anything like that.

And it's measuring when that changes compared to baseline before diagnosis, which can signal increased healthcare use associated with the subsequent diagnosis.

Speaker A

00:03:10.040 - 00:03:16.840

Okay, so it just, I guess it's what it says on the tin. It's just that window, isn't it, of potentially being able to pick up a change.

Speaker B

00:03:17.490 - 00:03:25.970

Exactly. So the idea is that if there is an increase long before diagnosis, then possibly there is an opportunity to diagnose these patients earlier.

Speaker A

00:03:26.450 - 00:04:09.190

So this was a really big study using the CPRD and this is a database that a lot of the listeners will be familiar with.

And you had a sample of a million patients registered with UK General Practice and then you looked at people diagnosed with cancer and their pre diagnosis rates of consultation like you were talking about, and also chest imaging by the different possible diagnosis routes. So either as an emergency, a routine or an urgent referral. But I really want to move straight to what you found here.

Can you give us just an overview of how the different people in this analysis were eventually diagnosed with lung cancer? So were there a lot of emergency diagnoses here?

Speaker B

00:04:09.350 - 00:05:46.240

Yes, I believe we had around 30% of patients who were diagnosed through the emergency route, compared to 20 something percent in the urgent referral route and the GP routine referral route. That aligns with the national data in NCRAS and also the Rapid Cancer Registry data. I guess that's what we expected to see.

We found that the majority of patients do present to primary care, which then disproves this hypothesis that has been presented in the literature that patients who are diagnosed through the emergency pathway don't present to primary care at all and therefore there wouldn't really be a chance to intervene and improve these patients diagnostic pathway. I think that is one of the key findings, although it is a simple finding.

Then we also found that there are short term similar diagnostic windows across these routes.

Patients who are diagnosed as emergencies had similar opportunity to intervene earlier as patients from the other routes, just because of the timing at which things changed.

However, we also looked at the rates and those were consistently lower for emergency diagnosed patients, even though the timing at which things change at the lower rates mean that these patients present less frequently. And so because they present less frequently, there are simpler, fewer chances in primary care to also like see warning signs earlier.

Speaker A

00:05:46.480 - 00:05:56.480

Yeah, so you looked at those consultations rates. So is that what you're describing here? So is that what those findings show in terms of potential opportunities for earlier diagnosis?

Speaker B

00:05:57.280 - 00:06:17.190

Yeah, so what I had in mind was mostly the consultations and the consultations with symptoms, but then acknowledging that we measured two different things.

So the timing at which things change, the diagnostic windows as well as the rates of these consultations, how frequently they were occurring for patients by route.

Speaker A

00:06:17.430 - 00:06:27.510

And what you're suggesting is that people who were diagnosed via emergency had lower rates. So that sort of is a bit counterintuitive. So can you talk us through that again a bit?

Speaker B

00:06:27.590 - 00:07:06.880

It's a bit contradicting.

Well, it would seem that it is because these patients do present to primary care and then when things start going wrong, let's say they happen around a similar time as for patients who are diagnosed through the other routes. But what sets the emergency diagnosed patients apart is that they present less frequently.

So they may still have cough and may still go to their gp, but they may not do so as often as someone who's referred on a two week wait, for example, or now urgent suspected referral, which then means that there are fewer chances for gps to pick up on persistent symptoms and then refer those patients.

Speaker A

00:07:07.200 - 00:07:19.780

And I guess just.

Were there any other main findings that you found in terms of sort of the diagnostic window or sort of consultations before diagnosis via the different routes?

Speaker B

00:07:19.940 - 00:09:14.860

Yeah, so I think one of the most interesting ones as well to the overall finding of patients presenting to primary care is that patients presented with non specific symptoms around 10 to five months before diagnosis across the routes, which is still quite a while before they're diagnosed.

So potentially this could mean that something could have been done differently to, for example, refer these patients earlier in like say month nine before diagnosis rather than nine months later. But again, as you said, this is also in lung cancer patients, which is a very difficult cancer site to diagnose early.

And part of that reason is because the symptoms that patients present with are non specific symptoms.

So it's also understandable that it is difficult to make that call based on someone presenting with cough in primary care, which is why there's more like work to be done and we need to better understand where the thresholds are for referral, like how many times someone comes in with the same symptoms, something like that.

But the work does show that that is like a common characteristic that is shared among, like across lung cancer patients, even diagnosed by the different routes.

And then another finding was also similar to the non specific symptoms, was that patients had chest imaging, so chest X rays around six to four months before diagnosis, which again is still a while before they're diagnosed.

And because these are chest X rays, then it could mean that they have negative chest X rays before diagnosis and then are again perhaps referred back to primary care with symptoms or they have an another chest X ray within those months where things are increasing, like month four to zero before diagnosis. Maybe there's a learning to be found from that.

Speaker A

00:09:15.100 - 00:09:32.620

Yeah, absolutely. And I wanted to just touch on those findings around imaging.

And I wonder what thoughts you had about the role of access to chest CTs for GPs, just given what you've described here about chest X rays and potential potentially negative chest X rays as well in this cohort.

Speaker B

00:09:33.180 - 00:10:02.380

Well, I do think that it's also pretty well documented in the literature that chest X rays aren't necessarily the best and most accurate diagnostic test for lung cancer and that improving access to low dose CT in England has helped diagnose lung cancer patients. So I think improving access to chest imaging and CT scans specifically could also present an opportunity to diagnose patients earlier.

Speaker A

00:10:02.830 - 00:10:16.990

And I wonder, just given all this information, what you found in this study, what do you think are the main implications for potentially opportunities to diagnose lung cancer earlier and not via emergency routes for these patients?

Speaker B

00:10:17.390 - 00:11:07.090

Again, I think the picture is complex because as you've also mentioned, the paper doesn't have kind of like a clear finding of something that was very different in the emergency diagnosed route. But actually patients who are diagnosed through the emergency route look quite similar to the, the primary care referred routes.

Then I think what we can do is focus on the things that were similar for all patients, meaning that there's also a similar opportunity to diagnose patients earlier across all routes.

Those things include being more vigilant about the non specific symptoms that patients present with and perhaps having a lower threshold for referral or more thorough follow up. And then I also think improving access to chest CTs could also help diagnose

Speaker A

00:11:07.090 - 00:11:22.290

patients earlier in the paper you touch very briefly on screening.

Is there anything that you want to mention here about sort of potentials for lung cancer screening or what might already be in play in terms of potential policy for screening for lung cancer?

Speaker B

00:11:22.850 - 00:12:16.370

Yeah, I mean, I think lung cancer screening is very important and it's definitely going to change the way things look in terms of how patients are diagnosed. And I think we're already seeing that since I think around like 2022 when the program started being rolled out.

There's around 7% of patients who are diagnosed through screening now, which we can see in the Rapid Cancer Registry data set that's actually publicly available as well.

So with that being said, it is still a bit difficult to know what that means because we need to let it play out for a few more years until we know what that means for diagnoses through the other routes. Hopefully it will mean that some patients who would be diagnosed through an emergency route won't be.

But of course we also don't know to what extent emergency diagnoses are completely avoidable.

Speaker A

00:12:16.610 - 00:12:23.280

And I think it's important to point out probably that the lung cancer screening programs are really targeted at the moment as well, aren't they?

Speaker B

00:12:23.360 - 00:12:50.190

Yeah, exactly. I believe there were patients who are eligible are those aged 55 to 74 and who have a history of smoking.

So for example, in our study that was 38% of patients who are diagnosed as emergency patients. So there's still 62% of emergency diagnosed patients who would not have been eligible for the screening pathway regardless.

Speaker A

00:12:50.910 - 00:13:11.950

And I think what you say here about actually not just focusing on what you found amongst the patients who were diagnosed with lung cancer via the emergency routes, but actually looking at everyone who's diagnosed with lung cancer and trying to improve care for all is really important in terms of extrapolating the findings here. It's been really great talking to you about this work. So thanks to again for joining me.

Speaker B

00:13:11.950 - 00:13:12.590

Thank you.

Speaker A

00:13:13.390 - 00:13:29.310

And thank you all very much for your time here and for listening to this BJGP podcast.

Marta's original research article can be found on bjgp.org and the show notes and podcast audio can be found@bjgplife.com thanks again for listening and bye.

Hello and welcome to BJGP Interviews.

I'm Nada Khan and I'm one of the Associate editors of the Journal.

Thanks for listening to this podcast today.

In today's episode, we're speaking to Marta Bergland.

Marta is a research assistant and PhD candidate at university College London.

She's recently published a paper here in the BJGP titled Pre Diagnostic Primary Care Consultations and Imaging in Emergency Diagnosed versus Referred Lung Cancer Patients.

So, hi, Marta, it's really lovely to meet you and it's great to talk to you about cancer diagnosis, which is a really important area for general practice and also a topic we publish on quite a lot here in the bjgp.

It's been fascinating reading this paper and it tackles a cancer that we really don't do well with here in the UK in terms of early diagnosis.

But I wonder if you could just start off by telling us a little bit more about lung cancer and how it's actually often diagnosed, which can sometimes lead to some of those poor outcomes.

So, as you said, lung cancer is one of the cancer sites in the UK where the majority or a large proportion of patients are diagnosed through the emergency route, also known as emergency presentations, which is when a patient is diagnosed after they present through an urgent hospital admission or an A and E attendance in the 30 days before diagnosis.

And that could look like someone who has had a persistent cough for, say, two weeks, but didn't actually go to their gp, and then suddenly they have a more severe onset of symptoms like shortness of breath, and then they go to A and E and are referred to a chest X ray and then get diagnosed, which is a pathway that is associated with worse prognosis and worse outcomes after diagnosis.

And the more preferred route, if you will, in England to diagnose patients is through primary care.

So through the GP routine referral or the urgent suspected referral route.

And that could look like someone who presents to primary care with cough or dyspneasia again, but then their GP refers them to a chest X ray and then they're diagnosed with lung cancer.

And I guess I wanted to just before we talk about what you found, I wanted to just cover here again, what you mean by this term, that's diagnostic window, because you mentioned that a few times in the paper.

But what does this actually mean?

And it relates a bit back to some of what you're talking about, about people presenting with symptoms, isn't that right?

So, for us, it's a measure of healthcare use before diagnosis, and it could be any healthcare use Measure like consultations, symptoms, blood test use, anything like that.

And it's measuring when that changes compared to baseline before diagnosis, which can signal increased healthcare use associated with the subsequent diagnosis.

Okay, so it just, I guess it's what it says on the tin.

It's just that window, isn't it, of potentially being able to pick up a change.

Exactly.

So the idea is that if there is an increase long before diagnosis, then possibly there is an opportunity to diagnose these patients earlier.

So this was a really big study using the CPRD and this is a database that a lot of the listeners will be familiar with.

And you had a sample of a million patients registered with UK General Practice and then you looked at people diagnosed with cancer and their pre diagnosis rates of consultation like you were talking about, and also chest imaging by the different possible diagnosis routes.

So either as an emergency, a routine or an urgent referral.

But I really want to move straight to what you found here.

Can you give us just an overview of how the different people in this analysis were eventually diagnosed with lung cancer?

So were there a lot of emergency diagnoses here?

Yes, I believe we had around 30% of patients who were diagnosed through the emergency route, compared to 20 something percent in the urgent referral route and the GP routine referral route.

That aligns with the national data in NCRAS and also the Rapid Cancer Registry data.

I guess that's what we expected to see.

We found that the majority of patients do present to primary care, which then disproves this hypothesis that has been presented in the literature that patients who are diagnosed through the emergency pathway don't present to primary care at all and therefore there wouldn't really be a chance to intervene and improve these patients diagnostic pathway.

I think that is one of the key findings, although it is a simple finding.

Then we also found that there are short term similar diagnostic windows across these routes.

Patients who are diagnosed as emergencies had similar opportunity to intervene earlier as patients from the other routes, just because of the timing at which things changed.

However, we also looked at the rates and those were consistently lower for emergency diagnosed patients, even though the timing at which things change at the lower rates mean that these patients present less frequently.

And so because they present less frequently, there are simpler, fewer chances in primary care to also like see warning signs earlier.

Yeah, so you looked at those consultations rates.

So is that what you're describing here?

So is that what those findings show in terms of potential opportunities for earlier diagnosis?

Yeah, so what I had in mind was mostly the consultations and the consultations with symptoms, but then acknowledging that we measured two different things.

So the timing at which things change, the diagnostic windows as well as the rates of these consultations, how frequently they were occurring for patients by route.

And what you're suggesting is that people who were diagnosed via emergency had lower rates.

So that sort of is a bit counterintuitive.

So can you talk us through that again a bit?

It's a bit contradicting.

Well, it would seem that it is because these patients do present to primary care and then when things start going wrong, let's say they happen around a similar time as for patients who are diagnosed through the other routes.

But what sets the emergency diagnosed patients apart is that they present less frequently.

So they may still have cough and may still go to their gp, but they may not do so as often as someone who's referred on a two week wait, for example, or now urgent suspected referral, which then means that there are fewer chances for gps to pick up on persistent symptoms and then refer those patients.

And I guess just.

Were there any other main findings that you found in terms of sort of the diagnostic window or sort of consultations before diagnosis via the different routes?

Yeah, so I think one of the most interesting ones as well to the overall finding of patients presenting to primary care is that patients presented with non specific symptoms around 10 to five months before diagnosis across the routes, which is still quite a while before they're diagnosed.

So potentially this could mean that something could have been done differently to, for example, refer these patients earlier in like say month nine before diagnosis rather than nine months later.

But again, as you said, this is also in lung cancer patients, which is a very difficult cancer site to diagnose early.

And part of that reason is because the symptoms that patients present with are non specific symptoms.

So it's also understandable that it is difficult to make that call based on someone presenting with cough in primary care, which is why there's more like work to be done and we need to better understand where the thresholds are for referral, like how many times someone comes in with the same symptoms, something like that.

But the work does show that that is like a common characteristic that is shared among, like across lung cancer patients, even diagnosed by the different routes.

And then another finding was also similar to the non specific symptoms, was that patients had chest imaging, so chest X rays around six to four months before diagnosis, which again is still a while before they're diagnosed.

And because these are chest X rays, then it could mean that they have negative chest X rays before diagnosis and then are again perhaps referred back to primary care with symptoms or they have an another chest X ray within those months where things are increasing, like month four to zero before diagnosis.

Maybe there's a learning to be found from that.

Yeah, absolutely.

And I wanted to just touch on those findings around imaging.

And I wonder what thoughts you had about the role of access to chest CTs for GPs, just given what you've described here about chest X rays and potential potentially negative chest X rays as well in this cohort.

Well, I do think that it's also pretty well documented in the literature that chest X rays aren't necessarily the best and most accurate diagnostic test for lung cancer and that improving access to low dose CT in England has helped diagnose lung cancer patients.

So I think improving access to chest imaging and CT scans specifically could also present an opportunity to diagnose patients earlier.

And I wonder, just given all this information, what you found in this study, what do you think are the main implications for potentially opportunities to diagnose lung cancer earlier and not via emergency routes for these patients?

Again, I think the picture is complex because as you've also mentioned, the paper doesn't have kind of like a clear finding of something that was very different in the emergency diagnosed route.

But actually patients who are diagnosed through the emergency route look quite similar to the, the primary care referred routes.

Then I think what we can do is focus on the things that were similar for all patients, meaning that there's also a similar opportunity to diagnose patients earlier across all routes.

Those things include being more vigilant about the non specific symptoms that patients present with and perhaps having a lower threshold for referral or more thorough follow up.

And then I also think improving access to chest CTs could also help diagnose

patients earlier in the paper you touch very briefly on screening.

Is there anything that you want to mention here about sort of potentials for lung cancer screening or what might already be in play in terms of potential policy for screening for lung cancer?

Yeah, I mean, I think lung cancer screening is very important and it's definitely going to change the way things look in terms of how patients are diagnosed.

There's around 7% of patients who are diagnosed through screening now, which we can see in the Rapid Cancer Registry data set that's actually publicly available as well.

So with that being said, it is still a bit difficult to know what that means because we need to let it play out for a few more years until we know what that means for diagnoses through the other routes.

Hopefully it will mean that some patients who would be diagnosed through an emergency route won't be.

But of course we also don't know to what extent emergency diagnoses are completely avoidable.

And I think it's important to point out probably that the lung cancer screening programs are really targeted at the moment as well, aren't they?

Yeah, exactly.

I believe there were patients who are eligible are those aged 55 to 74 and who have a history of smoking.

So for example, in our study that was 38% of patients who are diagnosed as emergency patients.

So there's still 62% of emergency diagnosed patients who would not have been eligible for the screening pathway regardless.

And I think what you say here about actually not just focusing on what you found amongst the patients who were diagnosed with lung cancer via the emergency routes, but actually looking at everyone who's diagnosed with lung cancer and trying to improve care for all is really important in terms of extrapolating the findings here.

It's been really great talking to you about this work.

So thanks to again for joining me.

Thank you.

And thank you all very much for your time here and for listening to this BJGP podcast.

Marta's original research article can be found on bjgp.org and the show notes and podcast audio can be found@bjgplife.com thanks again for listening and bye.