1 00:00:00,600 --> 00:00:04,320 Okay. So could you start by telling me your name and your current title? 2 00:00:05,100 --> 00:00:13,650 So I'm Naomi. Evan. I'm professor of epidemiology here at University of Oxford, and I'm chief scientist for UK Biobank. 3 00:00:13,890 --> 00:00:21,120 Okay, thanks very much. And without telling me your entire life story, but just to give me a bit of background, 4 00:00:21,750 --> 00:00:26,879 just tell me how you got from your first interest in science to where you are now. 5 00:00:26,880 --> 00:00:33,120 How did your career unfold? Well, like most people, it's quite a circuitous route. 6 00:00:34,590 --> 00:00:45,960 I actually did Environmental Science at Universe, my undergraduate university and graduated in a recession, was unemployed, couldn't get a job. 7 00:00:47,250 --> 00:00:52,440 And then just after a year of not doing very much thought, actually, 8 00:00:52,620 --> 00:00:58,890 I read I really need to go back to education and do something that I'm really interested in. 9 00:00:59,310 --> 00:01:08,340 And the only thing that I was really passionate about at university was we did a module on epidemiology and public health. 10 00:01:08,520 --> 00:01:13,920 So I went back and I did a master's in epidemiology at the London School of Hygiene and Tropical Medicine in London. 11 00:01:14,400 --> 00:01:17,640 And I felt like I've come home. 12 00:01:17,670 --> 00:01:22,649 It was just I just absolutely loved it. And I haven't really looked back since then. 13 00:01:22,650 --> 00:01:30,300 I came here in Oxford to do my dphil in the late nineties and became a cancer epidemiologist 14 00:01:30,600 --> 00:01:35,220 with a focus on breast and prostate cancer and how diet and hormones influences cancer. 15 00:01:36,360 --> 00:01:43,169 And then in 2011, I also started to work with the UK Biobank as their senior epidemiologist, 16 00:01:43,170 --> 00:01:49,620 and that was really different because rather than just doing research and analysing data and writing papers, 17 00:01:49,620 --> 00:01:56,640 you actually thinking about how can I help to set up and improve cohort studies? 18 00:01:56,820 --> 00:02:03,600 How can I engage the participants in collecting more valuable data for the global research community? 19 00:02:03,600 --> 00:02:09,810 So it's all about enabling others to use this fantastic resource, which I find hugely satisfying. 20 00:02:10,140 --> 00:02:13,680 Hmm. Well, we'll talk a bit more on that later, but I've got this question I've been asking everybody. 21 00:02:13,860 --> 00:02:20,280 If you could say if there was one question that you could say really kind of get you out of bed in the morning, what would it be? 22 00:02:21,330 --> 00:02:23,580 Oh, for me, it would. 23 00:02:23,940 --> 00:02:35,040 It would be, how can we better measure our lifestyle and particularly our environment to find out more about how that impacts health? 24 00:02:35,460 --> 00:02:39,630 That's what gets me out of bed. And what are the methods, the main methods that you use. 25 00:02:39,990 --> 00:02:47,040 So the main methods for finding out about people's lifestyle factors, we is simple. 26 00:02:47,040 --> 00:02:53,100 We ask them questionnaires about their diet, physical activity, how did they sleep? 27 00:02:53,430 --> 00:03:05,669 And of course, more recently we can now use fancy devices like Fitbits and apps on phones to find out more about their sleeping habits, 28 00:03:05,670 --> 00:03:13,319 their physical activity, and even down to having sleep monitors on their head to measure the activity in 29 00:03:13,320 --> 00:03:19,770 their brain and MRI machines to to to measure that volume of their heart chambers. 30 00:03:19,770 --> 00:03:25,590 I mean, it's just astonishing the amount of information you can collect on individuals now. 31 00:03:26,190 --> 00:03:34,380 So it's very much about data. Oh, it is. I mean, the more data we can collect on individuals and how they interact with their environment, 32 00:03:34,620 --> 00:03:38,670 couple that with information on their genetic profile. 33 00:03:39,150 --> 00:03:46,590 Then suddenly you could open up this wonderful box and find out how your people's genetic profile together with 34 00:03:46,590 --> 00:03:54,120 exposures to their lifestyle and why broader environment interacts to influence disease many years later. 35 00:03:54,630 --> 00:04:01,410 So that's really brought us very neatly to Biobank. So let's for the benefit of people who don't know about it, go back to the beginning. 36 00:04:02,190 --> 00:04:06,030 What is UK Biobank, how was it set up and what's it setting out to do? 37 00:04:06,840 --> 00:04:15,840 UK Biobank is arguably the world's most flexible and accessible biomedical resource 38 00:04:16,110 --> 00:04:21,149 for the global research community to use to perform health related research. 39 00:04:21,150 --> 00:04:30,000 It's in the public interest. So we recruited half a million individuals in the UK about 15 years ago aged 40 to 70, 40 00:04:30,540 --> 00:04:35,669 and we collected all sorts of information on them lifestyle factors, physical measures, 41 00:04:35,670 --> 00:04:42,600 we took biological samples from them and then we're following them up over many decades through linkage 42 00:04:42,600 --> 00:04:47,850 to the medical records and through asking them through questionnaires to find out about their health. 43 00:04:48,390 --> 00:04:59,770 And from that, we've made this huge database available for approved researchers worldwide to access this incredibly valuable dataset to. 44 00:05:01,720 --> 00:05:10,070 For the world's best imaginative minds to mine this dataset to find out about the causes and determinants of a whole range of diseases. 45 00:05:10,370 --> 00:05:14,660 So that's what you can buy books about, in essence. And what's your role as chief scientist? 46 00:05:15,020 --> 00:05:23,270 So my role is I'm largely responsible for making sure we follow up the health of all harmony participants through linkage 47 00:05:23,270 --> 00:05:31,130 to a broad range of medical records and other health related records to make that available to the worldwide community. 48 00:05:31,340 --> 00:05:37,070 And also to talk to the researchers about what would make UK Biobank better. 49 00:05:37,220 --> 00:05:43,100 What's over the next 510 years? So the technology is changing so rapidly. 50 00:05:43,520 --> 00:05:47,750 How can we make the resource even better in the future? 51 00:05:48,020 --> 00:05:54,469 Is that through more different measures that we can perform in the blood samples? 52 00:05:54,470 --> 00:05:57,800 Is that through doing repeat measures on the participants? 53 00:05:58,010 --> 00:06:01,800 Is that through slightly different measures now that the cohort or ageing? 54 00:06:02,840 --> 00:06:07,940 What can we do to better enhance the study to make it really valuable to to 55 00:06:08,090 --> 00:06:12,680 enable researchers to answer the most important research questions in the day? 56 00:06:13,190 --> 00:06:17,090 And this is just a kind of slide thing to cover, but how is it funded? 57 00:06:17,900 --> 00:06:23,600 So it's largely funded by the Wellcome Trust and the Medical Research Council in the UK. 58 00:06:25,040 --> 00:06:31,280 And it was set up and it had when it was set up, it had some additional funding from local governments as well. 59 00:06:31,760 --> 00:06:39,409 And but over the last five years we've also received industry funding to do specific enhancements to the resource. 60 00:06:39,410 --> 00:06:46,129 So for example, a consortium of pharmaceutical companies have funded whole exome sequencing, 61 00:06:46,130 --> 00:06:50,450 which measures the the genome that encodes for the proteins. 62 00:06:50,930 --> 00:06:58,220 And again, a consortium of pharmaceutical companies, together with government and charity, have funded whole genome sequencing. 63 00:06:58,430 --> 00:07:08,980 We have a pharmaceutical consent companies invested in UK Biobank to fund measuring circulating proteins in the blood, 64 00:07:08,990 --> 00:07:17,629 measuring circulating metabolites in the blood. So really, industry is starting to invest in this study to measure proteins, 65 00:07:17,630 --> 00:07:23,480 metabolites and genes in order to really accelerate their drug discovery work. 66 00:07:23,720 --> 00:07:30,290 And all those data are made available to the global research community to further enhance scientific discoveries. 67 00:07:30,660 --> 00:07:37,399 So academic and commercial research, the resources available to both academic and commercial research, it is, yeah. 68 00:07:37,400 --> 00:07:39,740 And it's available in exactly the same terms. 69 00:07:40,100 --> 00:07:50,570 So academic academics and researchers from the commercial world can access the data at the same cost, the same terms and conditions. 70 00:07:50,870 --> 00:08:00,140 And the only thing that we we really are obliging researchers to do in return is to return all of their results back to UK Biobank. 71 00:08:00,350 --> 00:08:10,580 So whether that's results from assays that they performed on the blood samples or whether they've derived a new variable from the MRI scans, 72 00:08:10,610 --> 00:08:15,769 all that data, plus the research data that they've done on the resource is returned back to you 73 00:08:15,770 --> 00:08:19,750 could Biobank so that others can build on the findings that have been made. 74 00:08:19,760 --> 00:08:27,620 So it's one of the first studies in the world actually to really enforce this really collaborative, 75 00:08:27,620 --> 00:08:33,920 open way of performing science and kind of building on what others have already achieved. 76 00:08:34,910 --> 00:08:36,469 And what's a bit if you had to pick out, 77 00:08:36,470 --> 00:08:43,790 you probably do talk sometimes some of the the more impressive achievements of people who've been using the resource. 78 00:08:43,790 --> 00:08:48,710 What would you pick on? So I would say over the last couple of years, 79 00:08:49,940 --> 00:08:57,139 what we're finding is coming out to the UK Biobank resource is in 2017 we made available genetic 80 00:08:57,140 --> 00:09:01,850 data on half a million participants in the world's largest study at that time to do so. 81 00:09:02,210 --> 00:09:10,590 That led literally overnight to an explosion of research into the genetic determinants of disease. 82 00:09:11,030 --> 00:09:15,500 From that work, that's led to a concept called polygenic risk pools. 83 00:09:15,950 --> 00:09:26,540 So this is where you look at all the genetic variants across an individual's genome, each of which only has a very small increased risk of disease. 84 00:09:26,540 --> 00:09:34,009 But when you add them all together, what you find is the people say in the highest, unfortunately in the highest, 85 00:09:34,010 --> 00:09:41,360 fifth percentile of genetic risk across variants, all their genome have substantially increased risk of, 86 00:09:41,570 --> 00:09:44,210 say, breast cancer or prostate cancer, heart disease. 87 00:09:44,250 --> 00:09:53,300 And you can say the same about pretty much every condition, and that's actually equivalent to many diseases that are monogenic in in order. 88 00:09:53,750 --> 00:09:59,360 And what we found is that if you can use this polygenic risk score idea. 89 00:09:59,960 --> 00:10:06,110 So that is really starting to galvanise the whole area of precision medicine. 90 00:10:06,470 --> 00:10:15,500 So you can imagine, for example, going to your GP, having a blood test done, they'll look at your genetic sequence and they'll say, 91 00:10:15,740 --> 00:10:23,000 well, you've got, you know, a 10% risk of developing heart disease by the age of 30, 75. 92 00:10:23,030 --> 00:10:26,930 Based on your genetic predisposition alone, 93 00:10:27,380 --> 00:10:34,070 we might want to think about perhaps taking preventative measures like taking statins to reduce your cholesterol or doing lifestyle advice. 94 00:10:34,430 --> 00:10:45,110 So you can see how this idea of knowing an individual's genetic profile and how that may influence risk can then start to tailor either 95 00:10:45,110 --> 00:10:52,720 screening programs or targeted treatments or preventative advice to those individuals who are increased genetic risk of a certain disease. 96 00:10:52,740 --> 00:10:56,690 So I think that's by far being today anyway. 97 00:10:56,990 --> 00:11:02,000 And the biggest broad achievement of the resource. 98 00:11:02,300 --> 00:11:06,010 Mm hmm. So let's finally arrive at a kind of boot camp. 99 00:11:06,020 --> 00:11:09,070 Can you remember where you were, what you were doing, or how. 100 00:11:09,110 --> 00:11:12,830 How you first heard that there was a pandemic in the offing? 101 00:11:13,910 --> 00:11:26,239 I remember being in a meeting in London, and it must have been February of 2020, and everybody was whispering and did that before the meeting started. 102 00:11:26,240 --> 00:11:31,040 There was all of you seen the news app, you seen this, and there was somebody show me something on Twitter. 103 00:11:31,040 --> 00:11:35,030 And it was this graph that was just just just going upwards. 104 00:11:35,240 --> 00:11:42,890 And it was it was a number of people in Wuhan who had been diagnosed with this with this strange virus. 105 00:11:43,400 --> 00:11:50,540 And there was all this talk about, oh, well, you know, it might not come over here and we might be alright and then all. 106 00:11:50,540 --> 00:11:55,580 But no it's already happening in Italy and Italy's really if it can happen in Italy it can happen anywhere. 107 00:11:55,910 --> 00:12:00,680 So there was this sense of unease about the rapid spread of something that was 108 00:12:00,680 --> 00:12:06,739 very unknown and we were kind of sitting ducks and there was a real sense that, 109 00:12:06,740 --> 00:12:12,650 Oh my God, there is this wave of infection that is killing people rapidly. 110 00:12:13,400 --> 00:12:16,879 And it was it. I remember it very well. 111 00:12:16,880 --> 00:12:27,260 It was a sense of like cold shivers down your spine, like, oh, my God, this is going to really change things in this country and we are not prepared. 112 00:12:28,650 --> 00:12:32,370 Right? Yes. So having got over the shock, 113 00:12:32,910 --> 00:12:40,440 how soon was it before you and your colleagues felt that this was something that by a bank could contribute to a straightaway? 114 00:12:40,860 --> 00:12:44,339 So it must have been end of March 2020. 115 00:12:44,340 --> 00:12:50,820 So just as we were entering the first lockdown, yeah, the Department of Health and Social Care got in touch with us and said, 116 00:12:51,270 --> 00:12:55,860 could you could you set up a study to find out what's going on in the UK? 117 00:12:56,880 --> 00:13:02,459 You know, we want to know the spread of infection across the UK, different population, 118 00:13:02,460 --> 00:13:06,720 subgroups, disease, how is it by age, by sex, by region and so on. 119 00:13:06,720 --> 00:13:14,220 And more importantly, how long once people who had been infected with the virus, how long do their antibodies last? 120 00:13:14,370 --> 00:13:17,639 Over time. And that was a crucial at that point. 121 00:13:17,640 --> 00:13:27,240 That was a crucial question because nobody knew that once you'd been infected, were you immune forever after that, how long were you protected for? 122 00:13:27,510 --> 00:13:35,129 And knowing the answer to that question had a real impact, of course, on vaccine rollout and how long in space the the vaccine. 123 00:13:35,130 --> 00:13:45,510 So we very quickly mobilised all our resources to set up this Seroprevalence study to find out the extent 124 00:13:45,540 --> 00:13:53,520 of infection across the UK and how long antibodies last full just just using the UK Biobank participants. 125 00:13:53,790 --> 00:13:55,970 Well, so that was another I mean, 126 00:13:55,980 --> 00:14:03,090 I've interviewed Derek Crook and I think I'm beginning to get and I've interviewed Sarah Walker and I I'm beginning to get slightly confused. 127 00:14:03,780 --> 00:14:09,030 It is quite confusing things in Sarah for the first time because you wanted to find 128 00:14:09,030 --> 00:14:14,999 out what was the rate of past infection across different age groups and so on. 129 00:14:15,000 --> 00:14:18,960 All of our participants are now in the sixties, early seventies. 130 00:14:19,470 --> 00:14:26,880 For the first time ever. We got the kids and the grandkids of the part of the existing participants involved 131 00:14:27,240 --> 00:14:33,780 and we were absolutely bowled over with willingness to contribute to the study. 132 00:14:33,780 --> 00:14:38,580 I would think within five days of putting out an email to the participants and saying, 133 00:14:38,790 --> 00:14:42,240 If you've got children who want to you want to involve, let them know. 134 00:14:42,660 --> 00:14:47,190 We had over 100,000 people volunteer to join the study. 135 00:14:47,190 --> 00:14:51,690 We only needed 20,000 to find out really what was going on across the country. 136 00:14:51,690 --> 00:14:55,169 So we were bowled over with support. And yeah, 137 00:14:55,170 --> 00:15:07,049 hopefully it means that the the value of doing this type of epidemiological study was not not just in the forefront in the mind of the participants, 138 00:15:07,050 --> 00:15:15,240 but also their wider families. And, and you must have needed to collaborate with other organisations because you needed that test results presumably. 139 00:15:15,720 --> 00:15:20,700 Yes. So what we did was we sent a a blood, 140 00:15:20,760 --> 00:15:30,059 a tiny blood sampling device which literally you prick the end of your finger and you squeeze it to get a drop of blood into a little test tube. 141 00:15:30,060 --> 00:15:34,800 You wrap it up, pop it in the post. So it was dead simple for people to use. 142 00:15:35,220 --> 00:15:40,920 And then that went to one of our collaborating labs, which is just over the way here in Oxford, 143 00:15:41,160 --> 00:15:49,830 and they set up in double quick time and assay that could measure the main antibody to the spike protein of the corona virus. 144 00:15:49,830 --> 00:15:52,920 So that was very close. That was Derek. Yes. Outfit. 145 00:15:52,980 --> 00:15:57,180 Yeah. So all the samples were first sent to our co-ordinating laboratory. 146 00:15:57,180 --> 00:16:08,610 So a Royal Mail van came every day with these 20,000 samples and we collected samples every single month for six months from 20,000 individuals. 147 00:16:08,610 --> 00:16:16,410 So it was an awful lot of samples that we had to we had to unpack adequate put into little place, 148 00:16:16,590 --> 00:16:21,720 put them on dry ice curry, carrying them down to Oxford's Derek Crookes lab. 149 00:16:21,960 --> 00:16:27,270 They then had to unpack the measured them. It was a huge, huge effort all round. 150 00:16:27,270 --> 00:16:37,920 But we got fantastic data on how long antibodies to coronavirus so we could tell who would be previously infected with the virus, 151 00:16:38,130 --> 00:16:41,520 how long antibodies last lasted over a six month period. 152 00:16:41,790 --> 00:16:50,309 And what we found was, was really clear data actually, that almost everybody had been infected in the first wave. 153 00:16:50,310 --> 00:16:59,760 So March, April, May in 2020, nearly everybody still had antibodies three months later and the vast majority, 154 00:16:59,760 --> 00:17:04,530 88% still had antibodies six months later. 155 00:17:04,920 --> 00:17:10,709 So what we're doing now is we've just sent a final sample and the back end of last year, 156 00:17:10,710 --> 00:17:22,320 so December of 21 to again to all of these 20,000 individuals to find out if their antibodies persist 12 to 18 months after infection, 157 00:17:22,320 --> 00:17:28,170 because that's still an outstanding question. We know antibodies wane because people get reinfected. 158 00:17:28,990 --> 00:17:32,590 But we don't actually know. Does that vary by age? 159 00:17:32,890 --> 00:17:43,060 Does that vary by ethnicity? And just what does that trajectory look like in terms of the waning of of protection, likely protection from infection? 160 00:17:43,060 --> 00:17:53,680 So that's something that we're looking at now. Mm hmm. So that I mean, I, I assume the the awareness study was also looking at serology. 161 00:17:54,850 --> 00:18:00,040 How did your results compare with this? And I mean, is there a difference in your approach? 162 00:18:00,070 --> 00:18:10,719 Well, they're very complimentary. I mean, they they own study looked at households again across the UK and they but they 163 00:18:10,720 --> 00:18:15,400 didn't look at they didn't have intense repeat measures like UK Biobank had. 164 00:18:15,700 --> 00:18:23,680 So we were able to collect multiple blood samples from 20,000 individuals, Onassis, a larger Subsample. 165 00:18:24,190 --> 00:18:27,400 But they did collect blood samples as frequently as we did. 166 00:18:27,970 --> 00:18:30,340 So that's a main difference. But they're they're complementary. 167 00:18:30,340 --> 00:18:36,190 And it's it was really useful when we were going through there in the teeth of the pandemic, if you like, 168 00:18:36,640 --> 00:18:41,590 talking to Onassis and the React study and our study about just, 169 00:18:41,590 --> 00:18:47,890 just how complimentary our findings were, particularly around the spread of the virus across the UK. 170 00:18:48,070 --> 00:18:53,290 You know, the higher rates in elderly people, higher rates in ethnic minorities, 171 00:18:53,290 --> 00:18:57,460 higher rates in urban areas, and the possible reasons why that might be. 172 00:18:59,290 --> 00:19:03,669 So presumably because you're connected to people's health records, you know, 173 00:19:03,670 --> 00:19:09,320 you would get sort of as people in the study became infected, you would you would see that. 174 00:19:09,460 --> 00:19:19,210 Were you able to use that data along with all your environmental and and genetic data to try and answer questions about who was getting infected? 175 00:19:20,200 --> 00:19:23,440 So the short answer was yes, but with some difficulty. 176 00:19:23,620 --> 00:19:34,630 So before the pandemic, we usually obtain health record data on an annual basis cancers, deaths, hospital admissions and so on. 177 00:19:35,020 --> 00:19:38,440 When the pandemic first started, we thought, Hang on a minute. 178 00:19:38,440 --> 00:19:46,270 This is nowhere near frequent enough to enable research into COVID into the determinants of severe COVID 19. 179 00:19:46,630 --> 00:19:53,320 So researchers were coming to us saying, Well, we could look at the what's the genetic predisposition to severe COVID 19? 180 00:19:53,440 --> 00:19:59,349 What are the lifestyle factors related to, you know, a smoking a risk factor? 181 00:19:59,350 --> 00:20:04,960 Is obesity a risk factor? Is ethnicity a risk factor? We could answer all these questions using UK Biobank. 182 00:20:05,380 --> 00:20:18,190 In order to enable that, we had to have rapid updates of health data, including COVID 19 testing data and critical care data from hospitals. 183 00:20:19,480 --> 00:20:28,930 So we we worked very, very well with all the data providers, and we managed to get monthly updates of these data. 184 00:20:28,930 --> 00:20:40,600 And our poor data finally came in. UK Biobank literally worked around the clock to make hospital inpatient data, death data, COVID 19 test data. 185 00:20:40,990 --> 00:20:50,680 And we were able to there was emergency legislation that was introduced called a coping notice that enabled us to obtain primary care data, 186 00:20:50,710 --> 00:20:55,240 so data from GP practices solely for the purposes of COVID 19. 187 00:20:55,540 --> 00:21:01,470 So we had to rapidly put in pipelines in place to process these millions and millions of rows of data from 188 00:21:01,490 --> 00:21:08,110 primary care records and make that available to researchers on a regular basis for COVID 19 research. 189 00:21:08,380 --> 00:21:16,600 And all our hard work was, was we felt vindicated in all of our hard work because, I mean, I think within the first six months, 190 00:21:16,600 --> 00:21:25,899 there were 200 odd papers looking at the genetic determinants and the lifestyle and environmental determinants of COVID 19. 191 00:21:25,900 --> 00:21:30,700 And it made a real difference to our understanding of actually what were the 192 00:21:30,700 --> 00:21:34,360 main risk factors for severe COVID and what were some of the key findings? 193 00:21:34,480 --> 00:21:40,840 So, I mean, obviously, there's genetic predisposition to developing severe COVID 19. 194 00:21:41,440 --> 00:21:50,050 Ethnic minorities have an increased risk of contracting COVID 19 in the first place. 195 00:21:50,290 --> 00:21:57,790 And research that we've done in the Seroprevalence study, because we also found that rates were much higher in black and ethnic minorities. 196 00:21:58,060 --> 00:22:01,210 We then sent out a questionnaire asking about their behaviour. 197 00:22:02,470 --> 00:22:06,340 You know, do do you work in a public facing role? Do you have public transport? 198 00:22:06,340 --> 00:22:12,670 Whereabouts do you live? How many people live in your household? All those questions we thought might be relevant to see if we could. 199 00:22:12,880 --> 00:22:18,220 That could explain why ethnic minorities were more likely to catch COVID. 200 00:22:18,700 --> 00:22:28,020 And what we found was that actually it was the type of job that you had, particularly whether you worked in the NHS or whether. 201 00:22:28,090 --> 00:22:30,910 In a public facing role and where you lived. 202 00:22:31,270 --> 00:22:41,620 So whether you lived in an urban area, and particularly if you lived in the London area of the UK, you're much more likely to develop COVID 19. 203 00:22:41,620 --> 00:22:47,560 And that and that explained two thirds of the excess risk in black and ethnic minorities. 204 00:22:47,950 --> 00:22:54,909 So, you know, it's and it was less to do with the number of people in the household and so on. 205 00:22:54,910 --> 00:23:00,550 So that that was quite interesting. The social and economic determinants of of clothing. 206 00:23:00,730 --> 00:23:05,200 So I overturned a lot of assumptions people had been making about multigenerational households. 207 00:23:05,410 --> 00:23:07,630 Exactly. Exactly. Yes. 208 00:23:07,780 --> 00:23:16,180 And then research coming out from the broader research community, looking at all of the data that we had in UK Biobank, you know, 209 00:23:16,390 --> 00:23:24,520 found, you know, put, put to bed the fact that Vitamin D was protective for COVID 19, that was a big hypothesis that was out there. 210 00:23:24,520 --> 00:23:29,740 And actually there was no real evidence to suggest that it found that obesity was a 211 00:23:30,190 --> 00:23:37,209 strong risk factor for developing severe COVID 19 and also underlying health condition. 212 00:23:37,210 --> 00:23:44,740 So if you had diabetes or depressed or even depression and cardiovascular disease, 213 00:23:44,740 --> 00:23:48,910 you're more likely to develop severe COVID 19 and subsequent complications. 214 00:23:49,720 --> 00:23:59,540 And also, if you had previous vaccines, too, like diphtheria or tetanus, whether they were more likely to protect you from subsequent COVID 19. 215 00:23:59,560 --> 00:24:05,500 So some really interesting research into the determinants of the disease coming out of the resource. 216 00:24:05,560 --> 00:24:14,410 Mm hmm. And were any of those findings converted into policy changes on the fly, as it were? 217 00:24:14,620 --> 00:24:21,370 Well, I think it certainly led to a greater understanding of the likely causes of COVID 19. 218 00:24:21,580 --> 00:24:28,690 And there was, I think, in the general public consciousness, people were aware that if you were A, B, C, 219 00:24:28,720 --> 00:24:36,430 and if you were ethnic minority and if you had underlying co-morbidities that you were somehow more vulnerable to COVID 19, 220 00:24:36,730 --> 00:24:43,450 whether that led to direct changes in policy in real time is is less clear. 221 00:24:43,630 --> 00:24:53,410 I mean, the policy work that came out of our Seroprevalence study in the 2000, I think really did lead to some changes because it helped to influence. 222 00:24:53,620 --> 00:24:54,009 Okay. 223 00:24:54,010 --> 00:25:03,500 Well, if we know antibodies last for at least six months in majority people, that actually has an impact on how often you're going to vaccine people. 224 00:25:03,520 --> 00:25:09,700 So that was, I think, a main contribution of in terms of policy and what people were doing on the ground. 225 00:25:10,120 --> 00:25:16,630 Mm hmm. So another study you did was one that you call the antibody self-test or self-testing. 226 00:25:16,660 --> 00:25:25,900 Yes. So once we'd we finished our seroprevalence study, we were making all the electronic health records available to the research community. 227 00:25:26,140 --> 00:25:31,210 They would get in on doing this fantastic research into finding the determinants of COVID 19. 228 00:25:31,630 --> 00:25:38,770 We then moved our attention to think about what about long COVID, you know, and quite early on in the pandemic, 229 00:25:39,280 --> 00:25:48,280 there were growing anxieties that this wasn't just a respiratory illness, but it had real systemic in some people, 230 00:25:48,280 --> 00:25:58,269 real systemic effects, brain fog, complete exhaustion as what are the risk, the cough and the fever, you know, 231 00:25:58,270 --> 00:26:06,250 muscle aches that lasted for months, loss of sense of taste and smell that lasted that did last for months in some individuals. 232 00:26:07,330 --> 00:26:13,630 So it seemed to be a more systemic virus with with with with wider health consequences. 233 00:26:14,140 --> 00:26:21,880 So we thought UK Biobank would be the ideal resource to look at that because it's a longitudinal study that follows that people over time. 234 00:26:22,420 --> 00:26:32,620 So what we did was we wanted to find out can objectively is possible who had developed covi who had been exposed to 235 00:26:32,620 --> 00:26:42,579 the virus because we knew from our serology work that about 25% of the population had antibodies to coronavirus, 236 00:26:42,580 --> 00:26:45,630 but they never thought they'd had COVID as many as 25. 237 00:26:45,640 --> 00:26:54,730 Yes, yes. And 40% of those who were positive didn't have the three main symptoms that were eligible for testing way back in the first wave. 238 00:26:55,180 --> 00:27:03,489 So there were there's thousands of people walking around, certainly in 2020, who had contracted the virus, 239 00:27:03,490 --> 00:27:09,010 but either weren't eligible for testing because it didn't have the main symptoms or they didn't have any symptoms at all. 240 00:27:09,760 --> 00:27:16,390 So we then sent out a self-test antibody lateral flow tests. 241 00:27:16,600 --> 00:27:19,060 So these were little devices you sent in the post, 242 00:27:19,450 --> 00:27:27,960 you just put a finger prick of blood onto the little cassette and within 15 minutes it tells you whether you got. 243 00:27:28,040 --> 00:27:31,130 Antibodies to to the virus or not. 244 00:27:31,340 --> 00:27:36,140 And so these were provided by the Department of Health and Social Care. 245 00:27:36,380 --> 00:27:40,730 And Amazon kindly donated, distributing all the kits for free. 246 00:27:42,500 --> 00:27:49,200 How did that relationship come about? Well, we already had an a kind of an ongoing relationship with Amazon for some of our other projects. 247 00:27:49,200 --> 00:27:57,770 So and they just wanted to help. There was a real sense of what can we do to help figure out what's going on with COVID, particularly long COVID. 248 00:27:58,130 --> 00:28:04,400 And so Amazon were like, Yeah, we'll deliver we'll deliver your your kits to half a million participants. 249 00:28:05,000 --> 00:28:08,270 So we developed a really close working relationship with them. 250 00:28:08,270 --> 00:28:18,620 About every participant got one as well. So we invited every participant for to have one of these case and over 200,000 actually had a kit, 251 00:28:18,800 --> 00:28:22,970 did the test and gave us a result which was which was fantastic. 252 00:28:24,210 --> 00:28:33,470 Unfortunately for us, this was all this all happened in 2021, just as the first vaccines were starting to be rolled out. 253 00:28:34,160 --> 00:28:38,810 And the kits that we sent out, they test for antibodies to the virus. 254 00:28:39,080 --> 00:28:45,980 They couldn't distinguish between antibodies produced from infection and antibodies produced from the vaccine. 255 00:28:46,400 --> 00:28:53,390 So we were we were really up at the clock was ticking to get these kits out before people would react. 256 00:28:54,050 --> 00:29:04,900 And I won't bore you with the details, but we had huge delays getting MHRA approval to use these tests in the home for research purposes. 257 00:29:04,910 --> 00:29:08,120 It has to go through a long bureaucratic. Because it involves blood. 258 00:29:08,390 --> 00:29:11,660 Because it involves blood. It's a medical device. 259 00:29:11,660 --> 00:29:14,960 So it needs to be approved by the the authorities. 260 00:29:15,230 --> 00:29:21,500 And it was being used it wasn't being used by a health professional. It has been used by Joe Bloggs in the home for research purposes. 261 00:29:21,740 --> 00:29:29,390 That took months to get approved. Meanwhile, people are starting to get jabbed and we knew that the test couldn't differentiate between the two. 262 00:29:29,930 --> 00:29:34,670 So what we had to do was ask people, have they been vaccinated? 263 00:29:34,700 --> 00:29:38,720 If so, when and if they had a positive antibody result. 264 00:29:38,990 --> 00:29:46,580 We then had to send them a another test that got sent to a different laboratory, 265 00:29:46,580 --> 00:29:51,710 which could tell whether the antibodies was from the vaccine or from previous infection. 266 00:29:52,550 --> 00:29:57,740 So it was a hugely complex and frustrating project, 267 00:29:57,950 --> 00:30:08,929 but at the end of it we've got data back from over 200,000 individuals on whether they were previously infected with the virus, 268 00:30:08,930 --> 00:30:18,709 irrespective of whether they had symptoms or not. Incorporating that data into UK Biobank that will enable really large scale 269 00:30:18,710 --> 00:30:25,310 epidemiological research into the long term health effects of being infected, 270 00:30:26,060 --> 00:30:32,330 regardless of whether you had symptoms or not. So some individuals were completely asymptomatic, others were hospitalised. 271 00:30:32,600 --> 00:30:37,760 So it will enable research into the effects of long COVID across the full disease spectrum. 272 00:30:38,000 --> 00:30:46,489 So obviously this research will take months, years, as you know, long COVID plays out over the ensuing months. 273 00:30:46,490 --> 00:30:53,270 But that data, coupled with the vaccination data, which we just put it into the resource, will be, I think, 274 00:30:53,360 --> 00:31:00,560 a unique biomedical dataset for people to work on the effects of corona virus infection for months and years to come, actually. 275 00:31:00,590 --> 00:31:05,210 Mm hmm. And now you mentioned that it's very much a systemic condition. 276 00:31:05,390 --> 00:31:09,020 And one of the things it affects when it affects multiple organs. 277 00:31:09,500 --> 00:31:13,850 And you've got scans of some of your partners. We have, yeah. 278 00:31:13,880 --> 00:31:25,200 So another way that we thought we could enable research into Long-covid was by taking advantage of our ongoing imaging study. 279 00:31:25,430 --> 00:31:35,120 So before the pandemic, where we're embarking on a hugely ambitious programme to perform MRI scans on 100,000 participants, 280 00:31:35,120 --> 00:31:41,209 of which we're halfway through. So by the end of 2019, we'd imaged 50,000 participants. 281 00:31:41,210 --> 00:31:50,870 So this involves MRI of the brain heart body optimum full body, DEXA scan, carotid ultrasound is a four hour visit, 282 00:31:50,870 --> 00:32:00,050 is very intensive for the participants, but it provides just unique, invaluable data on the physiology and the structure and function of organs. 283 00:32:00,890 --> 00:32:05,390 So 50,000 participants had already undergone an image, an assessment before the pandemic. 284 00:32:05,870 --> 00:32:15,859 So we thought, well, why not bring some of these participants back for a repeat scan, half of who we know have been infected with coronavirus. 285 00:32:15,860 --> 00:32:21,530 And we can tell that either from the self-test and lateral flow antibody kit that we sent 286 00:32:21,530 --> 00:32:26,360 to them or that they been diagnosed with COVID from linkage to medical health records. 287 00:32:26,960 --> 00:32:33,740 And then we'll bring back. So there are cases, if you like, cases with COVID 19 and will match them to controls. 288 00:32:34,010 --> 00:32:41,240 So matched similarly to age sex where they have that imaging assessment when they have their imaging assessment done. 289 00:32:42,020 --> 00:32:45,530 And we managed to invite back 2000 individuals. 290 00:32:46,280 --> 00:32:53,389 So this is the largest study in the world of the effect of coronavirus on internal organs. 291 00:32:53,390 --> 00:33:01,250 And it's the only study in the world that's got imaging scans of participants before 292 00:33:01,250 --> 00:33:06,050 they were infected with the virus and after they were infected with the virus. 293 00:33:06,290 --> 00:33:15,019 So it enables research into the likely direct effect of coronavirus on changes in internal organs. 294 00:33:15,020 --> 00:33:22,310 So for example, you could look at the effect of corona virus in changes in the heart, 295 00:33:23,780 --> 00:33:28,760 in the heart structure or the or the or the heart function in some way. 296 00:33:29,750 --> 00:33:35,569 Actually, SARS is unique because it's the only state in the world that's got imaging 297 00:33:35,570 --> 00:33:39,230 scans before and after infection and again across the full disease spectrum. 298 00:33:39,410 --> 00:33:46,670 Most of the studies they have, lots of studies have looked at the effect of COVID on imaging internal organs, 299 00:33:46,670 --> 00:33:53,780 but they tend to recruited people who are hospitalised with the disease and of course they've only got the scans after they were infected. 300 00:33:54,440 --> 00:33:58,759 So you can't see severe cases and they're particularly severe cases. 301 00:33:58,760 --> 00:34:08,510 So that's right. So it is impossible to look at the effect of all of of the virus because you don't know what the baseline was. 302 00:34:08,520 --> 00:34:11,990 So you don't know what their function was before they were infected. 303 00:34:12,260 --> 00:34:19,700 So we we were thrilled to be able to to perform this study last year. 304 00:34:19,940 --> 00:34:25,730 And as I said, we we managed to get 2000 individuals come back despite all the COVID restrictions. 305 00:34:25,740 --> 00:34:31,100 They were willing to come back for a second visit. And now that data is available to the research community. 306 00:34:31,100 --> 00:34:34,970 And already some really interesting findings are starting to come out of that. 307 00:34:35,000 --> 00:34:37,240 Can you pick back a few of them? Yes. 308 00:34:37,250 --> 00:34:49,010 So that the first finding to come out of this unique study is the effect, the likely effect of corona virus infection on the brain. 309 00:34:50,000 --> 00:35:01,310 So what researchers in based here in Oxford have shown is that those individuals who are infected with coronavirus, 310 00:35:01,340 --> 00:35:04,370 most of whom developed mild symptoms, 311 00:35:05,720 --> 00:35:18,140 they had a greater decline in grey matter in the brain, in all different areas of the brain, but particularly in the olfactory region. 312 00:35:18,150 --> 00:35:22,130 So the region of the brain related to sense of smell and taste. 313 00:35:23,060 --> 00:35:27,060 So this decline in grey matter volume, the significance of grey matter. 314 00:35:27,090 --> 00:35:30,080 So the significance of grey matter is, 315 00:35:30,230 --> 00:35:36,440 is this is a part of the brain that's really responsible for how it works or the neural connections in the brain. 316 00:35:36,830 --> 00:35:41,660 So if you've got a reduction in in in that part of the brain, 317 00:35:42,410 --> 00:35:53,180 it suggests that your connections are perhaps even perhaps slower or or that there's that then there may be, 318 00:35:53,900 --> 00:35:59,150 at least over the short term, some reduction in in how that area of the brain works. 319 00:35:59,390 --> 00:36:05,270 So the fact that one of the main symptoms is loss of sense of taste and smell, 320 00:36:05,570 --> 00:36:12,350 it was quite interesting to see that actually there was a corresponding reduction in the volume of the brain in this in this area. 321 00:36:12,830 --> 00:36:20,450 And also the researchers found that there was there was likely to be an increase 322 00:36:20,450 --> 00:36:24,350 in the time it took participants to perform some cognitive function tests. 323 00:36:24,560 --> 00:36:32,020 So as well as having an MRI, they also performed a range of different tests to measure thinking and memory and so on. 324 00:36:32,990 --> 00:36:37,160 And while these differences were not big, they they were there. 325 00:36:37,880 --> 00:36:44,900 So, of course, what we don't know is how long these these differences will last in the brain is usually plastic. 326 00:36:44,900 --> 00:36:47,300 It can recover. We know that. 327 00:36:48,260 --> 00:36:57,680 So the good news is that we hope that these are not permanent reductions in the brain and not permanent changes in cognitive function. 328 00:36:58,820 --> 00:37:03,710 And, of course, you know, only further MRI scans will really be able to answer that. 329 00:37:04,010 --> 00:37:07,639 But very interesting that you can actually see the effect of the virus. 330 00:37:07,640 --> 00:37:17,030 Even in mild cases, it really does seem to have an effect on the structure and possibly the function of the brain, 331 00:37:17,030 --> 00:37:19,190 at least over a couple of months period. 332 00:37:20,360 --> 00:37:27,050 And I'm wondering whether that's something you might repeat to see whether the electron variant and that the H2 variant. 333 00:37:27,880 --> 00:37:31,180 Act any differently from the previous variants that people have? 334 00:37:31,720 --> 00:37:35,890 Yeah, I mean, at the moment we don't have that granular information about variants. 335 00:37:36,080 --> 00:37:43,450 Maybe it's not everybody who has a PCR test for for COVID has their virus sequenced. 336 00:37:43,480 --> 00:37:51,310 Oh, I see. Yeah. And actually, we don't yet have half the information in the UK Biobank resource. 337 00:37:52,370 --> 00:37:56,650 There are there are work underway to try and marry the two different data sets up. 338 00:37:57,370 --> 00:38:00,040 So at the moment, that type of research is impossible. 339 00:38:00,220 --> 00:38:06,730 But there's there've been waves of the different variants on this side just from the time a person becomes infected. 340 00:38:06,730 --> 00:38:10,840 You've got to you've got a good guess at what you've got. You've got a good guess. 341 00:38:10,960 --> 00:38:18,100 Yeah. So as you say, there are you could depending on when they contracted the disease or when they were infected, 342 00:38:18,100 --> 00:38:21,610 you've got a pretty good guess at what variant they most likely had. 343 00:38:22,900 --> 00:38:29,860 Yeah. So it would be very interesting to see whether Long-covid work as to when people were infected. 344 00:38:29,860 --> 00:38:33,760 Does that. Well, we know it changes the symptoms that they get acutely. 345 00:38:33,910 --> 00:38:38,950 Does that change the downstream long COVID consequences as well? 346 00:38:38,950 --> 00:38:45,280 And that's something that we don't know yet and that hopefully over the coming months and years, we'll will be able to answer that. 347 00:38:45,410 --> 00:38:52,390 Mm hmm. So do you have any in-house research programmes, or is this all other people coming to you and saying, Can we use your data? 348 00:38:52,510 --> 00:39:00,760 The answer is this is well. So our job at UK Biobank is to build the resource to enable other researchers to use it. 349 00:39:00,910 --> 00:39:04,750 So by and large, we don't do the research ourselves. 350 00:39:04,750 --> 00:39:08,740 We have a small epidemiological and analyst team here. 351 00:39:09,730 --> 00:39:17,530 And, you know, sometimes we do do our own research, but we're we're we're treated as researchers just like anybody else. 352 00:39:17,530 --> 00:39:20,770 So we have no preferential or exclusive access to the data. 353 00:39:21,010 --> 00:39:28,260 So if we're particularly interested in some research question, we might ourselves put in an application to use the data and answer that. 354 00:39:28,540 --> 00:39:32,800 But essentially our job but you could buy a case is is to build the resource. 355 00:39:37,350 --> 00:39:41,070 So, yes, I think I've got all those. 356 00:39:42,070 --> 00:39:48,130 I've covered that. We've talked about health data linkage and we talked about Non-covid. 357 00:39:49,220 --> 00:39:52,430 Um. Yeah. So. 358 00:39:52,520 --> 00:39:58,009 So how do you think this work will help to plan future strategy for this pandemic? 359 00:39:58,010 --> 00:40:01,850 And I guess we I think we have to expect there will be future pandemics. 360 00:40:03,380 --> 00:40:09,530 I think. What? Well, I'd like to think maybe we didn't do that. 361 00:40:09,800 --> 00:40:16,400 Oh, I'm so sorry. Thank you. To not me along the way. 362 00:40:18,680 --> 00:40:25,190 Okay. So what? I think the pandemic, if there's any silver lining to the pandemic, 363 00:40:25,520 --> 00:40:32,300 is that everybody became familiar with epidemiology and the importance of epidemiology. 364 00:40:32,420 --> 00:40:38,600 I couldn't believe that people were talking on the dinner table about the R number and T cells. 365 00:40:38,630 --> 00:40:48,590 And to me, having a wider public discourse about science can only be a good thing. 366 00:40:48,620 --> 00:40:51,680 I mean, everybody became armchair epidemiologist overnight. 367 00:40:52,100 --> 00:41:00,550 And I just think that's wonderful. And I just really hope that stays in the public consciousness and the govt and you know, 368 00:41:00,560 --> 00:41:10,250 the government as well about the importance of large scale population based epidemiological studies because they're very expensive to run. 369 00:41:10,700 --> 00:41:16,220 They take 20, 30 years to really come to maturity because you're following people over decades. 370 00:41:16,700 --> 00:41:27,520 But actually they're crucial to find out about the determinants of disease and particularly find determinants of of of pandemics. 371 00:41:27,530 --> 00:41:33,140 So when, you know, coronavirus came along, you could buy it and was already up and running. 372 00:41:33,350 --> 00:41:38,840 If we had to build a study from scratch, we'd still be there recruiting people two years later. 373 00:41:39,170 --> 00:41:51,079 So being able to tap into existing resources and existing studies to suddenly focus on on trying to understand the distribution 374 00:41:51,080 --> 00:41:59,300 of the disease and then the determinants of the disease can only really be done with existing large scale population studies. 375 00:41:59,330 --> 00:42:10,129 So I hope the epidemiology has kind of earned its its place at the table now and is is 376 00:42:10,130 --> 00:42:15,350 recognised as being absolutely crucial to public health in this country and globally. 377 00:42:16,430 --> 00:42:21,440 I suppose one thing we didn't talk about is another potential consequence of having a dataset 378 00:42:21,440 --> 00:42:27,200 like this is that it may actually point to where the particular the genetic information, 379 00:42:27,530 --> 00:42:36,410 where there might be targets that pharmaceutical companies or researchers can focus on to try and try and make some new treatments. 380 00:42:36,590 --> 00:42:47,930 Absolutely. I mean, the whole reason why pharmaceutical industry have invested and it's millions of dollars to perform whole 381 00:42:47,930 --> 00:42:54,139 exome and whole genome sequencing on half a million participants is to accelerate drug discovery work. 382 00:42:54,140 --> 00:43:03,110 Because if if you find it an association between a particular genetic variation variant and a disease, 383 00:43:04,580 --> 00:43:08,960 you're you've got a good bet that that genetic variant is somehow related. 384 00:43:08,990 --> 00:43:12,080 So if you can find a drug target, then that's good. 385 00:43:12,230 --> 00:43:18,260 That's built around that function of the gene that that's going to massively accelerate your drug discovery work. 386 00:43:18,350 --> 00:43:27,499 So absolutely. I mean, the resource is already those data already finding that there are genetic variants that are 387 00:43:27,500 --> 00:43:33,110 related from the exome sequencing data that relates to a much lower risk of developing obesity, 388 00:43:33,260 --> 00:43:36,409 which is kind of the holy grail of drug companies. 389 00:43:36,410 --> 00:43:42,680 They of course, you want a magic bullet to produce a drug to reduce rates of obesity. 390 00:43:43,010 --> 00:43:50,750 And so the exome sequencing data has really shone a light as to potential drug targets that can do exactly that. 391 00:43:51,920 --> 00:43:59,120 And the same is true for, you know, all sorts of diseases, including COVID and the diseases that will come in the future for sure. 392 00:44:00,020 --> 00:44:06,120 Now, you talked about the incredible public awareness. Where you personally involved in media appearances? 393 00:44:06,140 --> 00:44:19,730 Yes. Yes. It was very interesting because the the media obviously want to want to have an individual story. 394 00:44:19,970 --> 00:44:22,760 And epidemiology is all about the population. 395 00:44:23,060 --> 00:44:32,960 And so there's a little bit of tension between actually what you find on a population level and may not be directly relevant to an individual. 396 00:44:34,430 --> 00:44:43,100 And of course, the media, what very simple messages and what we found with the pandemic, it's not so straight forward. 397 00:44:43,400 --> 00:44:51,470 You know, even even with the public messaging around COVID, when they try to really drill it down to. 398 00:44:51,550 --> 00:44:54,400 Three words, you know, face both hands. 399 00:44:54,880 --> 00:45:02,110 You know, it, you know, it makes you realise that in order to get across a public health message, it really needs to be very simple. 400 00:45:02,230 --> 00:45:04,180 And everybody needs to follow the rules. 401 00:45:04,430 --> 00:45:13,210 And start as soon as you start introducing nuance or error margins or maybe they switch off, that's not what they want to hear. 402 00:45:13,420 --> 00:45:21,489 So there is a natural tension between scientists wants to err on the side of uncertainty 403 00:45:21,490 --> 00:45:28,030 and models and predictions with some caveats to the media who want it in black and white. 404 00:45:28,060 --> 00:45:32,920 This is going to happen or this isn't going to happen with safe or we know. So that's been quite an interesting journey. 405 00:45:33,360 --> 00:45:39,080 Hmm. Yes. I mean, I could have gone away from what you said earlier and said. Everybody's brain is going to decline. 406 00:45:40,450 --> 00:45:45,609 We'd all be vegetables. But yes. 407 00:45:45,610 --> 00:45:53,650 And I think the message there was really actually, you know, the reduction in in the grey matter, the brain is actually pretty small. 408 00:45:53,680 --> 00:45:57,730 On average, it was about 2%, which in the grand scheme of things is quite small. 409 00:45:57,880 --> 00:46:03,820 We don't really know how clinically relevant that is in terms of your longer future health, 410 00:46:03,820 --> 00:46:07,060 because obviously we need to follow people up to find out that out. 411 00:46:07,600 --> 00:46:17,390 And also, given how plastic we know the brain is, you know, is it may be that in six months time that difference is no longer there. 412 00:46:17,410 --> 00:46:25,870 So this year, you could make a big splash of a headline going, Oh my God, everybody who's got COVID has got smaller brain. 413 00:46:26,260 --> 00:46:29,890 Whereas actually, the realities is quite different. 414 00:46:30,030 --> 00:46:38,760 Mm hmm. And Adam, did you have any help with Make It work did you do get have any training or some time ago? 415 00:46:38,770 --> 00:46:44,560 Yes. So few years ago, you know, we talked to the media quite a lot about UK Biobank more broadly. 416 00:46:44,770 --> 00:46:48,640 Yeah. So I'm just well through it now. 417 00:46:51,940 --> 00:46:56,560 Yeah. So how does this work? Raised new questions that you're interested in exploring in the future? 418 00:46:57,730 --> 00:47:02,860 Well, for me, it's really raised the importance, something I've always been interested in, 419 00:47:02,860 --> 00:47:12,010 but it is even made it higher on my priority list, if you like, is about how important infectious diseases still are. 420 00:47:12,370 --> 00:47:16,120 You know, we've been focus I've been focusing on cancer all my career. 421 00:47:16,480 --> 00:47:21,129 You think of cancer. Cardiovascular disease is the two main important things. 422 00:47:21,130 --> 00:47:26,620 And actually, because we can we've got linkage to cancer registries and linkage to hospital inpatient data, 423 00:47:26,680 --> 00:47:37,590 they're quite easy to study other diseases like COVID, but also things that you go to your GP about but don't make it into hospital. 424 00:47:37,600 --> 00:47:44,260 So things like, you know, arthritis, asthma, diabetes, actually eczema, 425 00:47:44,560 --> 00:47:52,000 they've been relatively under-researched because as epidemiologists we haven't had access to the data about who's got all of these things. 426 00:47:52,390 --> 00:47:57,980 So in my mind, one thing what I'm trying to say is, 427 00:47:57,980 --> 00:48:06,310 is I've become very much more interested in the role of infectious diseases and how that impacts chronic disease. 428 00:48:06,580 --> 00:48:15,760 So in terms of long-covid, it may be that, you know, coronavirus in some individuals may lead to longer term health consequences. 429 00:48:15,760 --> 00:48:24,040 We know other viruses cause cancer, HPV, Epstein-Barr virus, hepatitis C, we know that they cause cancer. 430 00:48:25,480 --> 00:48:29,740 And, you know, we now know that EBV can cause multiple sclerosis. 431 00:48:30,040 --> 00:48:38,440 So the role of infections is not just in their wacky infectious disease symptoms is actually they may cause chronic longer term disease down the line. 432 00:48:38,770 --> 00:48:45,249 So I become very interested in the role of infections and in longer term chronic disease. 433 00:48:45,250 --> 00:48:51,069 And I think coronavirus has just kind of made me more aware that actually we 434 00:48:51,070 --> 00:48:55,360 need to look at these as exposures as well as outcomes in their own right. 435 00:48:55,360 --> 00:49:03,790 Yes, yes. Yep. So things like just think of things like urinary tract infections, which are incredibly common but regarded as a bit trivial. 436 00:49:04,980 --> 00:49:09,990 Yeah. Exactly. Yeah. So what other health what are the health impacts of those? 437 00:49:09,990 --> 00:49:16,440 You've had duties five, ten years later, and it's only by having thousands and thousands of people, 438 00:49:16,800 --> 00:49:20,940 you can really try to tease out those small effects because they will be small. 439 00:49:22,970 --> 00:49:29,120 And how did the first lockdown impact on what you were able to do? 440 00:49:29,730 --> 00:49:34,460 So we haven't really talked before. And you said your colleagues were very busy, but it must have changed your life as well. 441 00:49:34,850 --> 00:49:41,089 Well, it's changed all of our lives, didn't it? I mean, we it was it was quite interesting. 442 00:49:41,090 --> 00:49:46,600 We had a new member of staff and her first day in the office was the end of March. 443 00:49:46,610 --> 00:49:51,559 It was it was our last day in the office was her first day in the office. 444 00:49:51,560 --> 00:49:56,990 And she came to work because, you know, she wanted to work in a big team and lots of group chats. 445 00:49:56,990 --> 00:50:02,450 And actually, when you do that for a day, because we're we work from home and literally overnight, 446 00:50:02,780 --> 00:50:08,239 we all had to be set up with, you know, VPN systems that we can log in and do our work. 447 00:50:08,240 --> 00:50:16,460 We all had our Microsoft teams and you're suddenly I was working from, you know, a tiny box room or spare bedroom. 448 00:50:16,910 --> 00:50:23,330 And, you know, suddenly you're working for 14 hours a day on calls. 449 00:50:23,330 --> 00:50:25,580 And but actually it was. 450 00:50:26,790 --> 00:50:36,809 I look back on it quite fondly because we were all working frenetically towards a common goal that we got to set at this studying. 451 00:50:36,810 --> 00:50:40,200 We've got to do it in double quick time. And it was all hands on deck. 452 00:50:40,380 --> 00:50:43,670 And we all felt we were doing something important. 453 00:50:43,680 --> 00:50:53,669 And it kind of we all pooled together across the whole organisation and we all worked really, really, 454 00:50:53,670 --> 00:51:00,120 really well together because normally in your day to day things, you don't have that real sense of urgency. 455 00:51:00,120 --> 00:51:03,990 You've done your work in a slightly different things, slightly different timelines. 456 00:51:04,230 --> 00:51:10,110 And it was the first time I've ever experienced this kind of all hands on deck mentality. 457 00:51:11,550 --> 00:51:18,750 And we were all working silly hours and things, you know, messages flying around everywhere as to how to do things. 458 00:51:18,750 --> 00:51:24,950 And it was it was hugely rewarding. It was exhausting, but it was hugely rewarding for all of us, I think, involved. 459 00:51:24,960 --> 00:51:29,970 And do you think it helped to support your own wellbeing, the fact that you knew you were doing something? 460 00:51:30,180 --> 00:51:40,030 Oh, for sure. For sure. I mean, you know, lots of my friends who were on furlough who, you know, picking up hobbies that they that they, 461 00:51:40,170 --> 00:51:49,080 you know, put down years ago and and worrying about money and having to homeschool their children. 462 00:51:50,400 --> 00:51:56,250 And I think if your your we were lucky in that we were focussed on one task and it was 463 00:51:56,250 --> 00:52:02,180 just down the rabbit hole with that and everything else was kind of pushed to one side. 464 00:52:02,190 --> 00:52:11,700 So I think that kind of bunker mentality kind of did you didn't have time to think about anything else or how scary it was or or what have you. 465 00:52:11,700 --> 00:52:16,589 But I mean, some call it I mean, these, you know, millions of people were were affected, 466 00:52:16,590 --> 00:52:23,190 both in terms of loved ones being really ill, but also having to homeschool young children while they're trying to work. 467 00:52:23,520 --> 00:52:27,930 You know, I can't tell you the amount of times we had Zoom calls and there were little kids, 468 00:52:28,350 --> 00:52:35,340 you know, under the desk or popping up to say hello and hugely challenging, but good fun. 469 00:52:35,610 --> 00:52:43,860 Mm hmm. And one thing a number of people I spoke to said is that, I mean, academic life can be quite competitive. 470 00:52:43,860 --> 00:52:47,160 And you it you know, you're constantly worrying about getting on it's paper out and so on. 471 00:52:47,400 --> 00:52:53,129 I mean, you already work in a fairly collaborative way, but was this noticeably more collaborative? 472 00:52:53,130 --> 00:52:58,240 And do you think that's a lesson for. Well, it's definitely more collaborative. 473 00:52:58,260 --> 00:53:03,480 I mean, the fact that I mean, we've been trying to get access to primary care data for years. 474 00:53:04,230 --> 00:53:11,610 And overnight it happened because there was emergency legislation to copy notes that basically told the 475 00:53:11,610 --> 00:53:17,970 system's suppliers to make the primary care data available for UK Biobank for COVID 19 research only. 476 00:53:18,450 --> 00:53:22,080 So you know where there's a will, there's a way. 477 00:53:22,620 --> 00:53:30,510 And so, yeah, huge collaboration at all levels and a real sense of working on it together. 478 00:53:31,290 --> 00:53:37,349 And everybody from the Department of Health and Social Care, from Amazon, from other academic colleagues, 479 00:53:37,350 --> 00:53:45,210 just wanted to do anything to help and really cut through the red tape to try and speed things up. 480 00:53:45,450 --> 00:53:53,310 So I really hope that, you know, translates into more cooperation in the future. 481 00:53:53,640 --> 00:53:58,410 And I think that having pushed that primary care data door open, you might be able to hang on to it. 482 00:53:58,860 --> 00:54:00,899 I really hope so. I really hope so. 483 00:54:00,900 --> 00:54:13,230 You know, I mean, it's silly, really, because the government made the primary care data available only for COVID 19 research because it's a pandemic. 484 00:54:13,620 --> 00:54:18,120 And yet we've got this more people died of cardiovascular disease than COVID. 485 00:54:19,020 --> 00:54:25,290 We've got pandemics of cancer and pandemics of cardiovascular disease and dementia going on all the time. 486 00:54:25,590 --> 00:54:28,950 So the fact that they made these data available only for COVID 19, 487 00:54:28,950 --> 00:54:35,190 because we been in an emergency where we're in a dementia emergency, we were in a cancer emergency. 488 00:54:35,490 --> 00:54:40,110 So why can't you make the data available for that? To me, it just doesn't make any sense. 489 00:54:40,530 --> 00:54:46,410 And I'm hoping that once we've shown that the data can be used responsibly, 490 00:54:47,280 --> 00:54:54,630 that it is of real scientific added value to find out the determinants of disease. 491 00:54:54,930 --> 00:55:03,360 Then they will be will have an easier path to try and make the data available for research into all conditions, not just COVID 19. 492 00:55:03,450 --> 00:55:13,320 Mm hmm. So were you also involved in or involved in any way in the in the kind of safety regime for the department 493 00:55:13,860 --> 00:55:18,240 deciding when it was safe to come back and had to come back and that kind of thing for UK Biobank. 494 00:55:19,170 --> 00:55:26,160 Yeah. I mean, we have, you know, a chief operating officer and health and safety, you know, that was really. 495 00:55:26,240 --> 00:55:34,520 They're very mixed. But yeah, you know, we're we're we're we'll have a say in what works for us, 496 00:55:34,520 --> 00:55:40,040 how we feel safe when we should be going back in kind of with safety measures, 497 00:55:40,040 --> 00:55:46,040 particularly for participants who are, you know, some of whom are quite elderly, 498 00:55:46,250 --> 00:55:52,100 coming in for imaging assessments, how to really make sure they feel safe and that is safe for them. 499 00:55:52,100 --> 00:55:54,230 So we just have a lot of work around that. 500 00:55:54,680 --> 00:56:01,669 I didn't have a problem getting PPE and that kind of thing that you needed in order to keep everybody safe knowing the image and sensor. 501 00:56:01,670 --> 00:56:03,770 Yes. No, that was that was that was fine. 502 00:56:03,770 --> 00:56:14,179 But I think that was because we actually we did the co-lead imaging study in 2021 when the PPE was the the pipeline scan setup. 503 00:56:14,180 --> 00:56:21,110 It was really in the first wave of the pandemic. Then getting your hand on gloves and masks and and gowns and stuff was a real problem. 504 00:56:21,350 --> 00:56:26,680 Right. And so as an FDA epidemiologist, you deal with risk all the time. 505 00:56:26,690 --> 00:56:30,800 And we were obviously very aware of the risks that were around to do. 506 00:56:31,010 --> 00:56:35,670 Did you personally feel threatened by that, by by the risk of actually getting the vaccine? 507 00:56:35,870 --> 00:56:44,630 Did you get the disease? I've never had never had COVID, although it's as we speak, the rates are probably the highest they've ever been. 508 00:56:45,230 --> 00:56:53,030 Um, and I've, I've been very, very fortunate in no one in my immediate family has been really ill with. 509 00:56:53,150 --> 00:56:56,209 I haven't lost anybody to it unlike others. 510 00:56:56,210 --> 00:57:02,930 So I think I recognise a very creepy when I've had a very easy time of it. 511 00:57:03,440 --> 00:57:13,280 Um, and I've also noticed that those of us who've had an easy time of it may get a bit blasé about it and you know, 512 00:57:13,430 --> 00:57:19,489 individuals who've lost loved ones who have been really ill before, I say like no put your mask on. 513 00:57:19,490 --> 00:57:26,180 No, you know, two metres. So I think you just have to be mindful that everybody's had slightly different experiences in their, 514 00:57:26,510 --> 00:57:30,889 their personal experiences may actually be very, very different to your own. 515 00:57:30,890 --> 00:57:35,810 But so far, personally, I've been very, very fortunate. 516 00:57:39,090 --> 00:57:48,920 So, yes, this is the last question. Has your experience of the pandemic and everything with it changed your attitude or your approach to your work? 517 00:57:48,930 --> 00:57:51,930 And how would you like things to change in the future? 518 00:57:53,820 --> 00:58:03,650 Well, so obviously we overnight we had to move to Homeworking and it was more collaborative. 519 00:58:03,660 --> 00:58:08,010 We were talking to many more people than we would otherwise do. Everybody wanted to help. 520 00:58:09,160 --> 00:58:14,130 And now that we're we're kind of coming out the other side. 521 00:58:15,480 --> 00:58:23,910 It has been hugely complicated about what does New Normal look like for not just us here, 522 00:58:23,910 --> 00:58:28,890 but across the country, about working from home, coming into the office? 523 00:58:29,280 --> 00:58:37,200 It's a very emotional subject and, you know, really having to think about, well, what is optimal for our team. 524 00:58:38,310 --> 00:58:42,310 And we're still figuring out what works best. 525 00:58:42,330 --> 00:58:49,380 You know, we, we recognise that actually getting everybody into the office five days a week is maybe not, 526 00:58:49,770 --> 00:58:52,979 not best, for it is not optimal for everybody's way of working. 527 00:58:52,980 --> 00:59:00,719 They've got long commutes. Most of us don't live in Oxford, you know, most of us have some caring responsibility. 528 00:59:00,720 --> 00:59:13,230 So actually, can we get the best? Can we learn from COVID and get the best out of our staff by having some flexibility and reaching a new normal? 529 00:59:13,230 --> 00:59:15,030 That looks a little bit different. 530 00:59:16,020 --> 00:59:26,700 But quite what that looks like and making sure this seems to be fair for everybody is more difficult than we originally thought. 531 00:59:28,350 --> 00:59:32,190 So there's no question of just going back to how everything was before. 532 00:59:33,780 --> 00:59:39,629 I think the world's moved on. I think you're right. I don't think we should be saying right. 533 00:59:39,630 --> 00:59:40,470 It's over now. 534 00:59:40,740 --> 00:59:52,139 Everybody back to coming into the office five days a week, because I think we've all recognised that we can be just as productive working from home. 535 00:59:52,140 --> 00:59:56,250 And actually sometimes if you want to focus on something and you don't want distractions, 536 00:59:56,250 --> 01:00:04,170 it's better to work from home no matter what your job title is and you can have a better quality of life. 537 01:00:04,560 --> 01:00:07,890 You don't have to sit in a car for 2 hours a day to get to the office. 538 01:00:08,340 --> 01:00:10,290 And yes, we recognise that we, 539 01:00:10,560 --> 01:00:18,870 we have to come in for meetings and have more days and the collaboration and don't undervalue the importance of a quick chat around the cattle. 540 01:00:20,160 --> 01:00:30,090 But equally there is just as much value as putting aside a day to really focus on a specific task, and you can do that at home just as well. 541 01:00:30,420 --> 01:00:38,700 So it's managing those. I think people expect very different things now post COVID, they expect to be able to have some flexibility. 542 01:00:38,910 --> 01:00:43,469 They expect to be able to work from home a bit more. And I totally understand that. 543 01:00:43,470 --> 01:00:52,020 And I think we should be trying to optimise the way people work and think a bit more creatively about what's best for people. 544 01:00:53,020 --> 01:00:55,420 It's lovely. Thank you very much. Thank you.