Today, we’re speaking to Dr Jasjot Saund, a public health registrar based in London.

Title of paper: Association of the NHS Health Check with all-cause mortality: a longitudinal cohort study in primary care

Available at: https://doi.org/10.3399/BJGP.2025.0451

The NHS Health Check (NHS-HC) has been shown to improve detection and management of surrogate markers of health, such as blood pressure and cholesterol. The long-term impact, however, is less evidenced, with mortality reductions seen in a UK Biobank cohort and no mortality reduction seen in a Cochrane review of general health checks, including those outside the UK. The current study added real-world evidence from a diverse UK urban population, and a 15-year study period. Survival analysis of GP electronic health record data shows reduced mortality in those who attended an NHS-HC compared with those who did not. These findings should be interpreted cautiously given the observational nature of the study design.

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

00:00:01.440 - 00:00:58.570

Hi and welcome to BJGP Interviews. I'm Nada Khan and I'm the senior research editor of the Journal. Thanks for taking the time today to listen to this podcast.

In today's episode, we're talking to Dr. Jazjot Sound, a public health registrar based in London.

We're here to discuss the paper she's recently published here in the BJGP titled Association of the NHS Health Check with All Cause Mortality, A Longitudinal Cohort Study in Primary Care. So thanks very much for joining us here today, Jazja.

It's really great to talk about this work and I think just before we get into the study itself, we know that NHS health checks are something that most GPs will probably be very familiar with, but maybe we don't always stop and think about the evidence behind them. So, just to start off, I wonder if you can explain to us a little bit about, about what the NHS Health Check program is actually designed to do.

Speaker B

00:00:59.050 - 00:02:16.370

Yeah, absolutely.

So the NHS Health Check program was rolled out in 2009 and it's essentially a population level preventative initiative, focusing particularly on cardiovascular disease, but potentially wider as well.

And so it's specifically targeted at people who are between the ages of 40 and 75 and who don't already have a, a cardiovascular disease or a known cardiovascular disease. And I think that's really important because it's targeting people who are potentially at risk or have undetected disease.

And what this means is, so ultimately the health check itself, it consists of a check that's usually undertaken by primary care, so looking at lifestyle factors, modifiable risk factors, biochemical tests, as well as sort of just other general checks. And importantly, this is embedded into primary care pathways.

So essentially it's ensuring that these patients are followed up if they are determined to be high risk and potentially put on prescriptions or followed up with lifestyle modifications. So that's generally what the health check is at the moment.

Speaker A

00:02:16.690 - 00:02:31.070

Brilliant. Yeah. And I guess the big question behind this paper is, do they actually make a difference?

And I wonder what did we actually know about the impact of NHS health checks before you started this work? So what was out there in the literature already?

Speaker B

00:02:31.550 - 00:04:07.320

Yeah, so there is a, a huge amount of evidence in different parts of the health check. So you might be aware already that it's a multifaceted intervention. So there's, there's the process bits like uptake and invitation.

So there's a big amount of evidence that's been telling us that it's really important that we even just the method of invitation can be really influential on who takes up the health check. But then there's also the specific parts of the check itself.

So how we communicate risk to patients, for example, can have an impact on how people go home and take on lifestyle advice. And then there's the body of evidence that's closer to where this paper sits, which is looking at clinical outcomes.

And there is a whole body of evidence looking at surrogate markers, so things like does it shift the marker on blood pressure or bmi. And then there's the evidence on mortality, for example, or clinical endpoints such as stroke.

And so far the evidence is predominantly positive to say we do see a change in clinical outcomes.

And more recently, there was a paper in 2024 which was using a UK Biobank cohort and shows that there was an impact, a positive impact on disease, end stage disease, as well as mortality. But as the paper sort of addresses, there's some gaps in this evidence, which is what we. What we hoped to fill.

Speaker A

00:04:07.560 - 00:04:25.280

Yeah.

And I guess I wondered if you could just talk us through the data used for this study, because this was a really large real world cohort from an inner city London population, wasn't it? So very different than the kind of population that might, for instance, have taken part in UK Biobank?

Speaker B

00:04:26.240 - 00:05:17.420

Yeah, absolutely.

So we thought it was really important to look at whether or not the health check was having an impact as it was being delivered in a real world population.

And so we were lucky enough to look at a cohort going all the way back to when the health check was implemented in 2009 in Lambeth, which is a really ethnically diverse and socioeconomically diverse cohort. So, for example, we have a population in lambeth that is 40% non white, as opposed to, I think the UK Biobank participants were 96% white.

So there's a real difference.

And so we wanted to ensure that the population we looked at was representative of some of the inner city boroughs that have particular challenges around NHS health check uptake.

Speaker A

00:05:17.660 - 00:05:27.330

Okay, so let's get into what you found. And what was the headline result when you compared people who attended at least one NHS health check with those who didn't?

Speaker B

00:05:28.120 - 00:05:53.320

Yeah, so the headline result actually showed that there was a 32% relative risk reduction associated with the health checks on all cause mortality. So that's looking at any cause of death, but shows quite an impressive risk reduction. So that is a relative risk reduction as opposed to absolute.

So it's important to note that overall the absolute rate of survival was high,.

Speaker A

00:05:53.750 - 00:06:03.670

And that's pretty Striking, actually. So that substantially lower risk of all cause mortality, were you surprised by the size of that association? So 32% is quite big, isn't it?

Speaker B

00:06:03.750 - 00:06:33.490

I think initially I was surprised, but then looking back at the evidence and the data that's already out there, it does fall in line with kind of what we were seeing, particularly in the UK Biobank cohort as well. I think there is sort of evidence that potentially looks at general health checks, which has shown a converse picture.

But as we discussed in the paper, it's really different to how the NHS health check is delivered here. So that's definitely something to bear in mind.

Speaker A

00:06:33.810 - 00:06:45.250

And you also looked at absolute mortality risk over 10 years and the difference seemed to be really pronounced with increasing age. So could you talk us a little bit through what you found there?

Speaker B

00:06:45.650 - 00:07:39.440

Yeah, so we found that although the older population had a. A higher absolute risk reduction, they also had a higher absolute baseline risk of death as well.

So it's really difficult to distinguish whether or not that finding was related to the health check being more beneficial in older patients or whether or not actually we see a bigger absolute reduction, because in our over 65 population there is a higher baseline risk of, of death. I think we, like, when looking back at the data population who was enrolled in 2009 at the youngest possible age were 40.

And even though we had 14 years of follow up, they would only be 57 years old at this point now. So their absolute risk of death is still very low, which might be more difficult to pick up in a study like this.

Speaker A

00:07:39.760 - 00:08:10.860

Yeah, fair enough.

And I guess, you know, whenever we look at observational research, I suppose the question is whether people who attend an NHS health check are somehow different than people who don't. And that's, you know, in parallel to the studies. For instance, you know, just focusing on UK Biobank, it's sort of a different population really.

And I guess people who attend a health check could potentially be more health conscious or more likely to engage with healthcare. Anyway. How did you try to account for that in this work?

Speaker B

00:08:11.100 - 00:09:45.530

Yeah, and that's a really important point to pick up and something that's really challenging in observational data in general. But so, yeah, we really wanted to make sure we tried to adjust or account for this. And so, as you said, some.

The healthy attender bias is something that can come up when people are more likely to attend a health check if they're more health conscious.

So we used a method called propensity weighting, which is similar to propensity matching in that we look at the probability that somebody attends a health check based on their characteristics and use that probability to weight the groups to ensure that the characteristics are balanced between those who attend and those who don't attend. So that was one method, but we also saw in our descriptive analyses that actually it wasn't a uniformly healthy group of attendees.

There was a higher rate of overweight and obesity in the people who attended health checks and there was a higher rate of a family history of coronary heart disease as well. To be extra sure. We also did some statistical analyses which tried to measure the level of unmeasured confounding.

So obviously we can only adjust for what we can measure.

So we wanted to understand how big is the unmeasured confounding, and we found that actually it would need to be twice as big as the things that we had already adjusted for to account for the full picture. So that's quite, quite large.

Speaker A

00:09:45.770 - 00:10:15.510

And I was also, as you were describing, interested in the fact that the NHS health check isn't just a single intervention. So it's a check, but then there's all that stuff that happens afterwards.

So identifying the risk factors, how you communicate that to people and how that impacts things, then they might have follow up or start medication or get referred onwards. And do you think that that's one of the reasons it's difficult to really pin down what part of the program is actually making a difference?

Speaker B

00:10:15.830 - 00:11:04.610

Yeah, I think it's a really complex public health intervention and it does require each aspect to have its own rigorous evaluation as well as the whole picture as a whole, which is where our study sits. But that is on top of evidence behind each facet. But it does then make it quite difficult to understand which is the biggest lever.

But I think far what we're seeing is definitely ensuring uptake in underserved populations, as our study also highlighted that there is a benefit across all ethnicities and imd, like deprivation quintiles as well. But also the follow up is important as well, ensuring that these health checks are embedded in clinical processes also been shown.

Speaker A

00:11:04.850 - 00:11:10.050

Yeah. And I wonder if there are any other main findings from this work that you wanted to highlight.

Speaker B

00:11:10.470 - 00:11:46.530

Yeah, I think I sort of touched on it just briefly. But interestingly, we ran the analysis on premature mortality as well and found that the benefit still held.

And we also ran the analysis on different deprivation quintiles as well as multiple ethnicities and found that the direction of the evidence is the same, showing that health checks do improve mortality, although some of the smaller subgroups didn't reach statistical signals significance. But we think this is likely just because we're looking at small groups and those with the biggest groups did reach significance.

Speaker A

00:11:47.090 - 00:12:13.160

Really interesting.

And I guess overall, I wonder what your feelings are about what the implications are for general practice, because I guess sometimes it can feel like one more thing that has to be fitted into an already pretty busy system. And it's really interesting to get your perspective as a public health doctor as well.

Do you think that this study should change the way that we think about the value of the health track?

Speaker B

00:12:14.440 - 00:13:29.500

So definitely, potentially, yes. I think some of the wheels are already in motion as well.

So this will have implications as a full body of evidence, not just this paper in isolation, at local and hopefully at national level as well. And we're starting to see changes already. So there's been rollouts of digital pilots as well as workplace based health checks as well.

And so I think the evaluation of those are going to be really telling and important to see whether we're still reaching or we can reach the intended outcomes of the health check.

But it also means that potentially these are pulling on those levers, such as increasing uptake and reducing those traditional barriers and ensuring that, yeah, health checks are equitable.

But also I think we're seeing signs in the recently published Cardiovascular Disease Modern Service Framework, which talks about potentially expanding the scope of health checks, so looking at a more cardiovascular, renal, metabolic focus rather than just cardiovascular disease. And this is because these conditions can co occur and particularly in more underserved populations as well as, and you've mentioned.

Speaker A

00:13:29.500 - 00:13:43.020

Equitable access in some underserved communities.

Was there anything that you felt needed to be highlighted in terms of how health checks need to be more equitable or anything, especially given the population that you included here in this study?

Speaker B

00:13:43.820 - 00:14:28.680

Yeah, so interestingly, Lambeth is obviously a very diverse population and interestingly, the uptake was pretty high in our minority ethnic groups and underserved populations.

So that was really positive in knowing that, okay, it is potentially reaching the right people, but there will still be a group of populations of people that will be struggling to access these.

So I think definitely this paper highlighting the fact that the mortality benefit is held across diverse populations is really important to ensure that within the borough and across other boroughs that we do continue to reach those underserved populations.

Speaker A

00:14:28.840 - 00:14:40.440

And I guess finally, for gps listening to this, who might be trying to think about how to take the findings back to their own work, what's the one thing that you want them to take away from this study about NHS health checks?

Speaker B

00:14:40.520 - 00:15:14.260

Yeah, I think. Well, I'm hoping it's reassuring. So I hope that.

I know that there might be some people on the fence about whether or not health checks are beneficial.

So I hope that this can add to the body of evidence that provides reassurance to gps that although it's very difficult to see such a complex public health intervention moving the needle on, you know, small parts of the population, that actually at the overall population level, it is making a difference or there is an association with a reduced mortality.

Speaker A

00:15:14.740 - 00:15:24.440

Yeah. And it's so brilliant to see this work done, as you say, in a real world sample. And I think the findings are really striking, actually.

So it's great to, to hear more about this work.

Speaker B

00:15:24.600 - 00:15:25.240

Thank you.

Speaker A

00:15:25.560 - 00:15:30.760

Yeah. So brilliant. I think that's a great place to wrap things up. But I just wanted to say thanks very much for joining us here today.

Hi and welcome to BJGP Interviews.

I'm Nada Khan and I'm the senior research editor of the Journal.

Thanks for taking the time today to listen to this podcast.

In today's episode, we're talking to Dr. Jazjot Sound, a public health registrar based in London.

We're here to discuss the paper she's recently published here in the BJGP titled Association of the NHS Health Check with All Cause Mortality, A Longitudinal Cohort Study in Primary Care.

So thanks very much for joining us here today, Jazja.

It's really great to talk about this work and I think just before we get into the study itself, we know that NHS health checks are something that most GPs will probably be very familiar with, but maybe we don't always stop and think about the evidence behind them.

So, just to start off, I wonder if you can explain to us a little bit about, about what the NHS Health Check program is actually designed to do.

Yeah, absolutely.

And so it's specifically targeted at people who are between the ages of 40 and 75 and who don't already have a, a cardiovascular disease or a known cardiovascular disease.

And I think that's really important because it's targeting people who are potentially at risk or have undetected disease.

And what this means is, so ultimately the health check itself, it consists of a check that's usually undertaken by primary care, so looking at lifestyle factors, modifiable risk factors, biochemical tests, as well as sort of just other general checks.

And importantly, this is embedded into primary care pathways.

So essentially it's ensuring that these patients are followed up if they are determined to be high risk and potentially put on prescriptions or followed up with lifestyle modifications.

So that's generally what the health check is at the moment.

Brilliant.

Yeah.

And I guess the big question behind this paper is, do they actually make a difference?

And I wonder what did we actually know about the impact of NHS health checks before you started this work?

So what was out there in the literature already?

Yeah, so there is a, a huge amount of evidence in different parts of the health check.

So you might be aware already that it's a multifaceted intervention.

So there's, there's the process bits like uptake and invitation.

So there's a big amount of evidence that's been telling us that it's really important that we even just the method of invitation can be really influential on who takes up the health check.

But then there's also the specific parts of the check itself.

So how we communicate risk to patients, for example, can have an impact on how people go home and take on lifestyle advice.

And then there's the body of evidence that's closer to where this paper sits, which is looking at clinical outcomes.

And there is a whole body of evidence looking at surrogate markers, so things like does it shift the marker on blood pressure or bmi.

And then there's the evidence on mortality, for example, or clinical endpoints such as stroke.

And so far the evidence is predominantly positive to say we do see a change in clinical outcomes.

But as the paper sort of addresses, there's some gaps in this evidence, which is what we.

What we hoped to fill.

Yeah.

And I guess I wondered if you could just talk us through the data used for this study, because this was a really large real world cohort from an inner city London population, wasn't it?

So very different than the kind of population that might, for instance, have taken part in UK Biobank?

Yeah, absolutely.

So we thought it was really important to look at whether or not the health check was having an impact as it was being delivered in a real world population.

So, for example, we have a population in lambeth that is 40% non white, as opposed to, I think the UK Biobank participants were 96% white.

So there's a real difference.

And so we wanted to ensure that the population we looked at was representative of some of the inner city boroughs that have particular challenges around NHS health check uptake.

Okay, so let's get into what you found.

And what was the headline result when you compared people who attended at least one NHS health check with those who didn't?

Yeah, so the headline result actually showed that there was a 32% relative risk reduction associated with the health checks on all cause mortality.

So that's looking at any cause of death, but shows quite an impressive risk reduction.

So that is a relative risk reduction as opposed to absolute.

So it's important to note that overall the absolute rate of survival was high,.

And that's pretty Striking, actually.

So that substantially lower risk of all cause mortality, were you surprised by the size of that association?

So 32% is quite big, isn't it?

I think initially I was surprised, but then looking back at the evidence and the data that's already out there, it does fall in line with kind of what we were seeing, particularly in the UK Biobank cohort as well.

I think there is sort of evidence that potentially looks at general health checks, which has shown a converse picture.

But as we discussed in the paper, it's really different to how the NHS health check is delivered here.

So that's definitely something to bear in mind.

And you also looked at absolute mortality risk over 10 years and the difference seemed to be really pronounced with increasing age.

So could you talk us a little bit through what you found there?

Yeah, so we found that although the older population had a.

A higher absolute risk reduction, they also had a higher absolute baseline risk of death as well.

So it's really difficult to distinguish whether or not that finding was related to the health check being more beneficial in older patients or whether or not actually we see a bigger absolute reduction, because in our over 65 population there is a higher baseline risk of, of death.

And even though we had 14 years of follow up, they would only be 57 years old at this point now.

So their absolute risk of death is still very low, which might be more difficult to pick up in a study like this.

Yeah, fair enough.

And I guess, you know, whenever we look at observational research, I suppose the question is whether people who attend an NHS health check are somehow different than people who don't.

And that's, you know, in parallel to the studies.

For instance, you know, just focusing on UK Biobank, it's sort of a different population really.

And I guess people who attend a health check could potentially be more health conscious or more likely to engage with healthcare.

Anyway.

How did you try to account for that in this work?

Yeah, and that's a really important point to pick up and something that's really challenging in observational data in general.

But so, yeah, we really wanted to make sure we tried to adjust or account for this.

And so, as you said, some.

The healthy attender bias is something that can come up when people are more likely to attend a health check if they're more health conscious.

So we used a method called propensity weighting, which is similar to propensity matching in that we look at the probability that somebody attends a health check based on their characteristics and use that probability to weight the groups to ensure that the characteristics are balanced between those who attend and those who don't attend.

So that was one method, but we also saw in our descriptive analyses that actually it wasn't a uniformly healthy group of attendees.

There was a higher rate of overweight and obesity in the people who attended health checks and there was a higher rate of a family history of coronary heart disease as well.

To be extra sure.

We also did some statistical analyses which tried to measure the level of unmeasured confounding.

So obviously we can only adjust for what we can measure.

So we wanted to understand how big is the unmeasured confounding, and we found that actually it would need to be twice as big as the things that we had already adjusted for to account for the full picture.

So that's quite, quite large.

And I was also, as you were describing, interested in the fact that the NHS health check isn't just a single intervention.

So it's a check, but then there's all that stuff that happens afterwards.

So identifying the risk factors, how you communicate that to people and how that impacts things, then they might have follow up or start medication or get referred onwards.

And do you think that that's one of the reasons it's difficult to really pin down what part of the program is actually making a difference?

Yeah, I think it's a really complex public health intervention and it does require each aspect to have its own rigorous evaluation as well as the whole picture as a whole, which is where our study sits.

But that is on top of evidence behind each facet.

But it does then make it quite difficult to understand which is the biggest lever.

But I think far what we're seeing is definitely ensuring uptake in underserved populations, as our study also highlighted that there is a benefit across all ethnicities and imd, like deprivation quintiles as well.

But also the follow up is important as well, ensuring that these health checks are embedded in clinical processes also been shown.

Yeah.

And I wonder if there are any other main findings from this work that you wanted to highlight.

Yeah, I think I sort of touched on it just briefly.

But interestingly, we ran the analysis on premature mortality as well and found that the benefit still held.

And we also ran the analysis on different deprivation quintiles as well as multiple ethnicities and found that the direction of the evidence is the same, showing that health checks do improve mortality, although some of the smaller subgroups didn't reach statistical signals significance.

But we think this is likely just because we're looking at small groups and those with the biggest groups did reach significance.

Really interesting.

And I guess overall, I wonder what your feelings are about what the implications are for general practice, because I guess sometimes it can feel like one more thing that has to be fitted into an already pretty busy system.

And it's really interesting to get your perspective as a public health doctor as well.

Do you think that this study should change the way that we think about the value of the health track?

So definitely, potentially, yes.

I think some of the wheels are already in motion as well.

So this will have implications as a full body of evidence, not just this paper in isolation, at local and hopefully at national level as well.

And we're starting to see changes already.

So there's been rollouts of digital pilots as well as workplace based health checks as well.

And so I think the evaluation of those are going to be really telling and important to see whether we're still reaching or we can reach the intended outcomes of the health check.

But it also means that potentially these are pulling on those levers, such as increasing uptake and reducing those traditional barriers and ensuring that, yeah, health checks are equitable.

But also I think we're seeing signs in the recently published Cardiovascular Disease Modern Service Framework, which talks about potentially expanding the scope of health checks, so looking at a more cardiovascular, renal, metabolic focus rather than just cardiovascular disease.

And this is because these conditions can co occur and particularly in more underserved populations as well as, and you've mentioned.

Equitable access in some underserved communities.

Was there anything that you felt needed to be highlighted in terms of how health checks need to be more equitable or anything, especially given the population that you included here in this study?

Yeah, so interestingly, Lambeth is obviously a very diverse population and interestingly, the uptake was pretty high in our minority ethnic groups and underserved populations.

So that was really positive in knowing that, okay, it is potentially reaching the right people, but there will still be a group of populations of people that will be struggling to access these.

So I think definitely this paper highlighting the fact that the mortality benefit is held across diverse populations is really important to ensure that within the borough and across other boroughs that we do continue to reach those underserved populations.

And I guess finally, for gps listening to this, who might be trying to think about how to take the findings back to their own work, what's the one thing that you want them to take away from this study about NHS health checks?

Yeah, I think.

Well, I'm hoping it's reassuring.

So I hope that.

I know that there might be some people on the fence about whether or not health checks are beneficial.

So I hope that this can add to the body of evidence that provides reassurance to gps that although it's very difficult to see such a complex public health intervention moving the needle on, you know, small parts of the population, that actually at the overall population level, it is making a difference or there is an association with a reduced mortality.

Yeah.

And it's so brilliant to see this work done, as you say, in a real world sample.

And I think the findings are really striking, actually.

So it's great to, to hear more about this work.

Thank you.

Yeah.

So brilliant.

I think that's a great place to wrap things up.

But I just wanted to say thanks very much for joining us here today.

Oh, thanks.

And you know, it's really lovely to.

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

Charles Shot's original research article can be found on bjgp.org and the show notes and podcast audio [email protected] thanks again and bye.