1 00:00:00,180 --> 00:00:10,079 So in managing large clinical trials, a trial of any size or any complexity is is hard work and requires one thing in abundance. 2 00:00:10,080 --> 00:00:12,510 And that's efficient trial management. 3 00:00:12,960 --> 00:00:21,840 It also requires dedication and a lot of passion because a lot of it can be very paperwork orientated unless you've got a passion for the subject. 4 00:00:22,050 --> 00:00:27,840 It can grind you down. But managing international trials requires all of that and a bit more. 5 00:00:28,770 --> 00:00:33,749 So what makes a successful trial? Well, I'm sure I'm talking to the converted. 6 00:00:33,750 --> 00:00:38,040 You all know these things, especially if you're working on evidence based medicine courses, 7 00:00:38,400 --> 00:00:42,990 but also because we all need a burning question, a burning scientific question, 8 00:00:43,050 --> 00:00:44,220 where you've got enough people, 9 00:00:44,220 --> 00:00:51,360 where you've got equipoise and enough uncertainty around the country or around the world to make this something that they want to know the answer to. 10 00:00:52,770 --> 00:01:00,030 Relevant outcome measures, which is a topic of great debate and outcome measures that are acceptable to the participants. 11 00:01:00,030 --> 00:01:10,200 I mean, most of us setting up a trial in mass and the clinicians were very keen on relapsing remitting time to remission, time to relapse. 12 00:01:10,890 --> 00:01:16,260 But when we talk to the most patient groups, all they wanted to know was when we don't have to be in a wheelchair. 13 00:01:16,890 --> 00:01:21,930 And so these outcomes that are relevant to clinicians and outcomes that are relevant to participants, 14 00:01:22,230 --> 00:01:27,450 and hopefully you get something that's relevant to both and the sample size that is feasible. 15 00:01:28,170 --> 00:01:33,900 We do large trials and the PTO motto is The bigger, the better, but it's got to be feasible. 16 00:01:34,320 --> 00:01:39,420 I mean, the incidence of the disease, you can't do a big trial and something that's very rare. 17 00:01:39,840 --> 00:01:50,430 So a feasible sample size and quite often trials fail because the original sample size is just unrealistic, a realistic timeline. 18 00:01:50,580 --> 00:01:56,760 I mean, these days we have got into this sort of 70 days of first patient into a trial, which I think has been plucked out there, 19 00:01:56,760 --> 00:02:03,090 don't see any rationale for it, but it's all to do with realistic timelines, enthusiastic teams. 20 00:02:03,280 --> 00:02:07,260 As I said before, you've got to have a passion for some of this and enough resources. 21 00:02:07,500 --> 00:02:16,320 And you all know we all suffer from a resource dirge and a trial manager. 22 00:02:16,470 --> 00:02:22,830 And I have to say this because I am one and, you know, trial managers can actually make or break a trial. 23 00:02:22,830 --> 00:02:31,410 And the importance of a trial manager was recognised quite a long time ago by the MLC and latterly by the HCA because 24 00:02:31,830 --> 00:02:38,670 they actually did some work around whether a trial was managed by a trial manager or by a team of our clinician. 25 00:02:39,000 --> 00:02:47,459 And the successful trials had somebody that had an overview of the trial and actually knew what was going on in all aspects, 26 00:02:47,460 --> 00:02:54,200 didn't have the answers and couldn't actually provide the answers in the clinical areas or the quality controllers, 27 00:02:54,210 --> 00:02:57,600 but had an overview of what was going on. And as I say there, 28 00:02:57,600 --> 00:03:06,330 a good trial is involved in all aspects of it and it avoids sort of unworkable systems and things that you have to go back and review all the time. 29 00:03:07,590 --> 00:03:15,660 And there are generic job descriptions provided by the NIH and the UK trauma network which really said I'm involved in, 30 00:03:16,560 --> 00:03:22,110 and these are some of the responsibilities that you would expect a trial manager, a good trauma to undertake. 31 00:03:22,440 --> 00:03:27,090 Now, there's also a road, of course, he or she just run around for somebody. 32 00:03:27,160 --> 00:03:32,610 Everyone knows a trial manager, and it's not of my six, so they don't do it on their own. 33 00:03:32,610 --> 00:03:42,419 There's obviously a team, but as I say, it's that overview, it's the big picture and project planning, which is crucial. 34 00:03:42,420 --> 00:03:49,320 I mean, no business whether you were remodelling your house with your open shop school or whatever. 35 00:03:49,470 --> 00:03:52,380 No, nobody would go into it without having a plan. 36 00:03:52,950 --> 00:04:02,580 And, you know, project management plans, the print series, etc., all have these features a clear, objective aim to bring about change. 37 00:04:03,420 --> 00:04:10,140 That's what we do with the trial. We have a clear objective. We aim to bring about change, requiring a team of people to do it. 38 00:04:10,320 --> 00:04:19,050 We require a team of people. We hire statisticians, our clinicians, our health, economics people, etc. to set time scale, 39 00:04:19,580 --> 00:04:25,230 to find resources and tasks that which needs to be completed to a pre-specified standard. 40 00:04:25,800 --> 00:04:29,040 Our SOPs guidelines, our approval systems. 41 00:04:29,370 --> 00:04:33,270 So this is what project management is. It's nothing different from running a trial. 42 00:04:33,480 --> 00:04:35,190 In fact, it's essential to running a trial. 43 00:04:36,840 --> 00:04:47,790 The five basic processes in project management of initiation, planning, executing, monitoring and controlling analysis, reporting and closure. 44 00:04:48,690 --> 00:04:52,260 This is what we do. These are our five processes. 45 00:04:54,210 --> 00:04:59,520 So we initiate our sites. We plan in detail the way forward. 46 00:05:00,070 --> 00:05:07,180 We then have the plateau phase where we're executing this, where there's recruitment and the trial sort of goes into a plateau phase. 47 00:05:08,000 --> 00:05:15,270 Monitoring and controlling. So we had our monitoring site visits and the way we control drug dispensation, etc. 48 00:05:15,650 --> 00:05:19,480 Analysis, reporting and closure. All of these should be very familiar with. 49 00:05:20,560 --> 00:05:29,760 But what I don't think a lot of trials people realise, especially people that run them like me, is that it is part of a well known process. 50 00:05:29,770 --> 00:05:37,150 It's just not us being trial managers, it's all part of a project management plan and they are project managers over the world. 51 00:05:37,660 --> 00:05:45,010 So the whole team efforts that you need to actually put this project plan in place is one of collaboration. 52 00:05:45,010 --> 00:05:49,690 And we talk about collaboration all the time in clinical trials and it tends to 53 00:05:50,050 --> 00:05:54,610 sort of err on the side of it's the collaboration with the investigators at sites, 54 00:05:55,000 --> 00:06:02,770 but actually is collaboration between internal or external in your within your team with your statistician, 55 00:06:03,310 --> 00:06:10,060 everybody needs to own it because if everybody owns it, then it becomes real and it will it will deliver. 56 00:06:11,440 --> 00:06:14,890 The trial is more likely to be successful if everybody enjoys it. 57 00:06:15,700 --> 00:06:21,790 If members of the team feel that they're part of it and they're all valued and quite often that is not the case. 58 00:06:21,790 --> 00:06:28,510 There's a very much a hierarchy and people get sort of left behind. 59 00:06:28,750 --> 00:06:34,780 And when it's a team and it's a multidisciplinary team and they own it, it's more likely to succeed. 60 00:06:36,130 --> 00:06:42,310 And this is everything through from when you first had the first idea and the CEO comes and says what you think to this, 61 00:06:42,310 --> 00:06:46,540 and you sit down around the table and you start developing the question right the way through to, 62 00:06:46,780 --> 00:06:53,320 you know, dissemination and the archiving of the data. And everyone needs to know where it is and for how long it's going to be there, etc. 63 00:06:54,040 --> 00:06:59,710 And integrating all the procedures is another thing that I think is so crucial to clinical trials. 64 00:07:00,100 --> 00:07:05,169 Clinical trials shouldn't be something that's put upon people. It should be integrated into their routines. 65 00:07:05,170 --> 00:07:10,540 If it's integrated into their routines, then it will become part and parcel of what they do day by day. 66 00:07:10,840 --> 00:07:14,380 And other than specific forms they might have to sign or complete, 67 00:07:14,680 --> 00:07:20,380 it won't be a chore to them, because if it is a chore, then your recruitment will be less. 68 00:07:20,890 --> 00:07:26,680 And if the data collection follows the clinical progress and the chances are they will also fill in your form. 69 00:07:26,680 --> 00:07:32,559 Because I'm not doing anything different to what they would do normally in their day to day task. 70 00:07:32,560 --> 00:07:37,960 For looking after that patient, we try to minimise extra visits or tests for participants, 71 00:07:38,150 --> 00:07:44,380 purely so they don't have to keep going back to host hospitals or clinics and the costs attached to that. 72 00:07:45,280 --> 00:07:49,960 But if you've got a bunch of clinicians who are very busy or a bunch of nurses very busy, 73 00:07:50,260 --> 00:07:54,399 you don't want them to actually feel that they've got to do extra work. 74 00:07:54,400 --> 00:08:02,740 And usually in our field for nothing and with international trials I run, the resources are even less. 75 00:08:03,010 --> 00:08:09,370 And so therefore we do try to keep everything to a minimum, but enough data to answer our question. 76 00:08:10,360 --> 00:08:16,210 And efficient systems are again, very crucial to the whole running of the trial. 77 00:08:16,510 --> 00:08:21,870 I mean, I started off running trials with index card system and a daisy will typewriter. 78 00:08:22,030 --> 00:08:25,180 It wasn't efficient, but that was what we had then. 79 00:08:25,690 --> 00:08:32,620 And but now we have the, you know, the benefits of systems, I.T. systems that can be robust, 80 00:08:32,620 --> 00:08:39,460 that produce reports and newsletters and letters to patients, etc., as the click of a button. 81 00:08:40,000 --> 00:08:49,180 And so a good trial has a good, robust, transparent system and something that, you know, will, if you like, service the trial. 82 00:08:49,510 --> 00:08:54,100 You're not there to service the computer, the computer and the system, service the trial. 83 00:08:55,450 --> 00:08:57,250 And this the trials, international trial. 84 00:08:58,300 --> 00:09:04,120 They should take account of all the different different practices and work environments and obviously governance. 85 00:09:04,570 --> 00:09:10,299 But what we found is most important is that all our systems work well for all different 86 00:09:10,300 --> 00:09:15,040 browsers because in the international arena you don't have the latest windows, 87 00:09:15,040 --> 00:09:20,529 whatever. You have something that's still work working on Windows three or whatever, 88 00:09:20,530 --> 00:09:28,030 and they have to be able to use them because otherwise it costs you, you and your funder a fortune to keep updating everything. 89 00:09:29,560 --> 00:09:38,890 And having said that, in this study I've recently been running, they didn't want to buy Microsoft because it was expensive. 90 00:09:39,460 --> 00:09:45,610 And the thought of spending £40 on a disk where they could get it for £2 in the market, 91 00:09:46,010 --> 00:09:55,150 when we pointed out to them that £2 in the market actually is not a microsoft system, it's a black one and it's full of viruses. 92 00:09:55,630 --> 00:09:59,830 It's just that that initial thought of spending £40 is probably a whole. 93 00:09:59,960 --> 00:10:04,340 A month's wages for somebody is completely foreign to them. 94 00:10:04,760 --> 00:10:09,140 So making sure that they can work on their what they've got. Without too much fuss. 95 00:10:09,710 --> 00:10:16,370 And efficient recruitment militaries. A trial succeeds or fails and whether it manages to recruit the pre-specified number of participants. 96 00:10:16,580 --> 00:10:22,430 And as I said earlier, sometimes it's not realistic the the sample size. 97 00:10:23,120 --> 00:10:29,470 So making sure that it's right up front and that you've you've got the centres that have the patients is, 98 00:10:29,510 --> 00:10:34,700 is really crucial and work done in developing those sort of things is time well spent. 99 00:10:35,780 --> 00:10:42,380 But there's very little evidence about, you know, what recruitment strategies we should go for. 100 00:10:42,410 --> 00:10:48,080 I mean, it's we need to recruit 770,000 or whatever patients and off you go. 101 00:10:48,950 --> 00:10:56,870 There's lots of information out there about, you know, good clinical practice and the stats and equipoise and consort, 102 00:10:57,050 --> 00:11:01,100 but we've got no guidance for what recruitment strategies we should use. 103 00:11:01,100 --> 00:11:10,400 And there has been, you know, recently some in the last ten years, some work done on it, but nothing that actually points you to something specific. 104 00:11:10,410 --> 00:11:15,350 So I will start here. This is where we need to start and this is the process we need to take. 105 00:11:16,010 --> 00:11:20,329 And you know, experienced trials managers just have learnt by socket. 106 00:11:20,330 --> 00:11:26,140 And so see really it's like, you know, okay, we'll start off with give them a target and we see how it goes. 107 00:11:26,630 --> 00:11:37,850 And this is where I just get very frustrated that we've got lots of guidelines and sort of Damocles consults, whatever, on how to report a trial. 108 00:11:38,000 --> 00:11:44,120 We've got lots of things when it comes to GCP on how we should set up a trial and get our approvals and permissions. 109 00:11:44,120 --> 00:11:54,110 But there's the bit in the middle that we've got no real model of management and maintaining recruitment over a long period of time, 110 00:11:54,470 --> 00:11:59,450 you know, requires real stamina from everyone. So your trial's got to be interesting. 111 00:11:59,600 --> 00:12:07,010 It's got to have a passion and it's got to have that ownership which will enable them to feel that they can carry on. 112 00:12:07,070 --> 00:12:17,360 And this is what a Cochrane review of clinical conditions participating cities found when they looked at 11 relevant observational studies. 113 00:12:17,360 --> 00:12:26,809 And they looked and they asked the clinician, So what would prevent you recruiting into the trial or what would help you to recruit in the trial? 114 00:12:26,810 --> 00:12:31,760 What would maintain your interest? And the first one was your interest in evidence based practice. 115 00:12:31,760 --> 00:12:36,380 So they want a question that's going to actually hopefully change practice. 116 00:12:36,770 --> 00:12:41,540 And if it doesn't change you for, you know, something different, it sometimes just stops it. 117 00:12:42,320 --> 00:12:48,200 I was involved in a trial once where they were doing an AC ICI bypass and actually they stopped the surgeons 118 00:12:48,200 --> 00:12:52,850 doing it because the result of the trial was they were doing harm and there's quite a few trials like that. 119 00:12:52,880 --> 00:12:59,360 They get more and more getting published nowadays and participation in an academic group. 120 00:12:59,360 --> 00:13:06,290 So lots of clinicians would be more inclined to be involved in a trial if they were working with an academic group rather than industry, 121 00:13:06,680 --> 00:13:11,120 if they had extra staff to help them recruit, if they thought the patients were interested. 122 00:13:11,570 --> 00:13:18,590 And that's where I go back to the MF study, where the patients weren't interested in how long it was going to take for them to relapse. 123 00:13:18,590 --> 00:13:22,069 They were more interested in how long before they were in a wheelchair and 124 00:13:22,070 --> 00:13:27,710 actually at their home and they felt comfortable explained in trials to patients. 125 00:13:27,980 --> 00:13:32,639 And the whole issue of explaining randomisation, I've had people say, 126 00:13:32,640 --> 00:13:38,330 if we just flip a coin and then you get whatever treatment, that's probably not the best way forward. 127 00:13:38,900 --> 00:13:47,300 So these were what the pointers that the clinicians gave us in trying to work through recruitment strategies. 128 00:13:47,570 --> 00:13:52,219 And then again they go back to the ownership thing and making sure everyone is credited for 129 00:13:52,220 --> 00:13:57,200 the trial right the way through to from the beginning to authorship and dissemination. 130 00:13:57,620 --> 00:14:01,159 It's very easy for people to get missed off. 131 00:14:01,160 --> 00:14:02,989 Very important people sometimes. 132 00:14:02,990 --> 00:14:09,830 I mean, you've got your clinicians, your investigators, but then you've got the nurses on the ward that get the notes all the time, 133 00:14:10,220 --> 00:14:16,010 or you've got the pharmacists who make sure that the drugs are accounted for and accountable. 134 00:14:16,400 --> 00:14:22,370 And sometimes all those people get missed, missed or for publication, and they don't get that sort of credit. 135 00:14:22,880 --> 00:14:28,870 So developing ownership throughout and making sure you keep track of it is so important, is really so important. 136 00:14:28,880 --> 00:14:33,410 That's how you build up networks that will go on to the next trial and go on to the next trial because 137 00:14:33,410 --> 00:14:40,190 they feel comfortable and they feel valued in the trauma and just guide that we've we've over the years, 138 00:14:40,190 --> 00:14:43,489 we've got three C's for trial management. 139 00:14:43,490 --> 00:14:48,530 And it's obviously collaboration, it's communication, it's creativity and common sense. 140 00:14:48,530 --> 00:14:54,139 Now, the creativity, we can't do so much of it these days because we're a bit tied with regulations, 141 00:14:54,140 --> 00:14:59,330 but we can be a bit creative and the common sense is the one that I think. 142 00:15:00,880 --> 00:15:05,320 Glaze lays over everything because the whole thing about trial management is cool. 143 00:15:05,320 --> 00:15:12,160 Since you don't actually do a trial of a rare disease and expect to do 10,000 patients in two years in two centres. 144 00:15:12,400 --> 00:15:16,870 It's the common sense that goes through everything and the communication. 145 00:15:16,870 --> 00:15:22,960 The second see is again making people feel valued but making sure that you talk to them all the time. 146 00:15:23,020 --> 00:15:27,250 So if they're not recruiting, you don't just go off and say to all the good centres, go, well done. 147 00:15:27,260 --> 00:15:32,050 Thank you very much for doing brilliantly. You say the same to the centre that not recruiting. 148 00:15:32,290 --> 00:15:36,490 I have a conversation to find out why they're not recruiting. Sometimes it's something very, very simple, 149 00:15:37,450 --> 00:15:43,389 something they've misunderstood or something they've been on honest comment somewhere and you've not known it, 150 00:15:43,390 --> 00:15:47,890 so you just sort of just synthesise the hopeless, but actually they're not. 151 00:15:47,920 --> 00:15:54,130 So, you know, how do we get to the point where we can run big trials, which may have many challenges? 152 00:15:54,610 --> 00:16:02,920 And according to the EU directive, we're all should be educated, trained and experienced in the task we will perform in a trial. 153 00:16:03,670 --> 00:16:08,320 Well, for trial managers, that's a bit difficult because there's no specific trial training. 154 00:16:08,620 --> 00:16:16,390 But to be a trial manager and we can't evidence it anywhere we are other than years of experience. 155 00:16:19,000 --> 00:16:25,899 And so I find this very difficult and I'm quite passionate that there should be ways that trial managers 156 00:16:25,900 --> 00:16:33,400 can become qualified and recognised from the top down so that we're not always struggling to sort of, 157 00:16:33,400 --> 00:16:42,969 if you like, justify ourselves. There are no messes around over the last ten years and Mrs. popped up and a lot of 158 00:16:42,970 --> 00:16:48,550 the the stuff in there in the EMRs is about the practical side of running a trial. 159 00:16:49,390 --> 00:16:55,720 But at the moment, much of the collective wisdom of of people like me and there are many more of me 160 00:16:57,070 --> 00:17:00,910 sort of suck it and see trials and we pass it on by word of mouth and just say, 161 00:17:00,910 --> 00:17:02,980 Well, I did this, why don't you try that? 162 00:17:03,430 --> 00:17:15,700 And that is really probably at the detriment to a lot of trials that we don't have a qualified staff who can manage trials to a particular standard, 163 00:17:15,730 --> 00:17:17,140 to a particular model. 164 00:17:18,460 --> 00:17:28,210 And the survey was done just last year by the UK chairmen and of the 460 trial managers, 60% of them did not want a higher degree. 165 00:17:28,240 --> 00:17:33,400 What they wanted was they were seeking, you know, specific training to do their job. 166 00:17:33,850 --> 00:17:40,629 But what they did, what was the recognition for that? And they wanted to be accredited somehow and that's what we were working towards, 167 00:17:40,630 --> 00:17:49,540 some sort of accreditation system where people who don't want to go to a higher degree have some accreditation that can take 168 00:17:49,540 --> 00:17:56,290 forward almost like your sort of your years of experience that you can put in when you you go for a higher education post. 169 00:17:56,950 --> 00:18:03,760 And the other thing is, although intuitively trained managers actually use project management principles and practices, 170 00:18:04,180 --> 00:18:12,819 I think we need to have more evidence that it works in a trial situation and more research on research. 171 00:18:12,820 --> 00:18:19,420 I mean, I've been trying to persuade trauma just to do research on their research for years and in the way they do it. 172 00:18:19,420 --> 00:18:20,680 What works, what doesn't work? 173 00:18:21,130 --> 00:18:27,160 Bits and pieces coming out in the literature from time to time, but not a lot because not a lot of journals are interested. 174 00:18:27,340 --> 00:18:30,580 In fact, it's only trials that I think takes everything that way. 175 00:18:30,970 --> 00:18:34,510 Most of the other journals, we're not interested in that sort of research. 176 00:18:34,510 --> 00:18:38,950 So if it's large or it's international or if it's small and it's local, 177 00:18:39,580 --> 00:18:51,309 trials fail because tried and tested systems have not been documented or evaluated or published so that the next one's coming along behind us. 178 00:18:51,310 --> 00:19:01,209 The next generation can read up in the literature how to do it, what worked, what didn't work, etc. And we've been reinventing the wheel all the time. 179 00:19:01,210 --> 00:19:04,360 And still today somebody said to me, What would you do if? 180 00:19:04,840 --> 00:19:10,930 And I said, Well, I did that in the US at that trial and they did that in that trial and it was who would find out about that. 181 00:19:10,930 --> 00:19:16,780 And if you read the trial publication, there's very little on how we did it. 182 00:19:17,380 --> 00:19:23,110 The methods section will have a lot about maybe the randomisation and the minimisation, 183 00:19:23,620 --> 00:19:31,360 but very little about what challenges we're faced with with recruitment and retention, how it was overcome. 184 00:19:32,230 --> 00:19:35,380 It's not it's just not acceptable by the journals. 185 00:19:36,040 --> 00:19:37,809 And so that's what we carry on doing. 186 00:19:37,810 --> 00:19:45,970 So to improve the successful delivery of trials, we do need management guidelines and we do need robust methods of evaluating what we do. 187 00:19:46,090 --> 00:19:52,750 But international trials do bring added challenges. And, you know, they can be overcome with some good systems, 188 00:19:53,200 --> 00:19:59,680 but some of the obvious differences between a national or local trial, international trial or languages time difference. 189 00:19:59,760 --> 00:20:03,300 Is local regulations and in-country problems. 190 00:20:03,720 --> 00:20:09,060 And they can be many, many. But if you've got a good project plan, 191 00:20:09,300 --> 00:20:16,770 you should have space within your planning to allow you to take take the emergencies as they arrive and deal with them. 192 00:20:17,340 --> 00:20:19,880 And that's what I found with the internationals over the year. 193 00:20:19,890 --> 00:20:25,770 I've got to have a plan almost day by day, week by week, so that when something comes on your desk on a monday morning, 194 00:20:25,770 --> 00:20:29,489 you can move it to Friday because you know that that's within the emergency. 195 00:20:29,490 --> 00:20:36,330 You just left yourself enough time to do what you going to do on Monday to move it for days so you can deal with something else. 196 00:20:36,390 --> 00:20:39,840 But knowing your audience in an international trial is really crucial. 197 00:20:40,500 --> 00:20:46,620 And knowing the countries, the people, traditions and the culture are all so very important. 198 00:20:46,620 --> 00:20:54,570 If you're going to actually get the recruitment in knowing that the participants understand what they're entering into. 199 00:20:55,080 --> 00:21:03,360 So spending time talking to the local investigators, the nurses and the participants is valuable time spent in the development time that you, 200 00:21:03,360 --> 00:21:08,710 you know, save time as the trial progresses when you're not having to unpick misunderstandings, etc. 201 00:21:08,730 --> 00:21:14,879 And also, flexibility is key. So I'm not saying that we give away from our sop's or whatever, 202 00:21:14,880 --> 00:21:24,660 but being able to think about how flexible you can be in an international trial, again, is important upfront because things will happen. 203 00:21:25,170 --> 00:21:31,980 You know, we in the trial I've just been running have had all sorts of things like, you know, civil war, floods, earthquakes. 204 00:21:33,360 --> 00:21:38,610 But you can be flexible in some things. You can allow consent to be taken with a thumbprint. 205 00:21:39,420 --> 00:21:47,730 You can allow consent by a spouse providing that you know, this upfront and when it goes to their local ethics committee, that's what's agreed. 206 00:21:47,940 --> 00:21:50,850 But that's where the time in the development is really important. 207 00:21:51,690 --> 00:21:56,219 And again, encouraging the local ownerships, because the results must be applicable within that country. 208 00:21:56,220 --> 00:22:06,450 You know, if they never do a laparoscopy, what is the point in proving laparoscopy works for whatever procedure works in whatever procedure, 209 00:22:06,450 --> 00:22:10,700 if they haven't got a lack of scope and we've had that sort of it's happened. 210 00:22:10,710 --> 00:22:16,650 So I just want to run you through this trial, which I'm just bringing to an end and clear out the back. 211 00:22:16,650 --> 00:22:29,400 There will probably be saying if she did she so describe this is a caesarean section is one of the most common operations carried out in the world. 212 00:22:29,530 --> 00:22:37,020 I think some of you might be surprised to know that, you know, in developing countries, as many as 25% of births are by caesarean section. 213 00:22:37,320 --> 00:22:42,480 And in places like Sudan, it's 59%. And then likewise in Brazil. 214 00:22:42,960 --> 00:22:47,010 So it's an operation that's carried out daily, which you think is very safe, 215 00:22:47,490 --> 00:22:52,410 but it's done in all sorts of different ways and a variety of surgical techniques, 216 00:22:52,740 --> 00:22:56,870 none of which have ever been proven to be one better than other are used. 217 00:22:57,330 --> 00:23:03,810 So it was a distinct lack of good quality of evidence when we started to look at this issue. 218 00:23:04,230 --> 00:23:13,180 So but we did realise that not only was it a huge population we had to look at, we also realised that, you know, 219 00:23:13,200 --> 00:23:18,809 if you had a tiny difference it would make a lot of difference to a lot of women and a lot of babies. 220 00:23:18,810 --> 00:23:27,810 So the trial design was a two by two by two by two by two factorial fractional factorial design trial, which is very difficult. 221 00:23:27,840 --> 00:23:33,510 It's two pairs to five intervention pairs and it's called non-regular because 222 00:23:33,510 --> 00:23:38,730 we only randomised them to three of the five and that's why it's fractional. 223 00:23:39,180 --> 00:23:47,310 And it was done like this because these are the five main techniques used to enter a woman's womb and retrieve a baby. 224 00:23:47,610 --> 00:23:57,570 So these five techniques were the main ones and we only randomised to three four for compliance reasons to give the clinicians some clinical freedom. 225 00:23:58,050 --> 00:24:07,500 So for instance, if a woman, if a site rather only ever done a single layer closure because that was their protocol, 226 00:24:07,680 --> 00:24:13,200 we didn't want them to differ from their protocol. We will carry on doing what they usually do because that was their protocol. 227 00:24:13,620 --> 00:24:17,130 If they could do the others then they could be randomised to the other three. 228 00:24:17,640 --> 00:24:21,840 We had 90 sites in seven countries and eight regional trial offices. 229 00:24:22,380 --> 00:24:27,570 I have to say the five by two not making fractional factorial design has never been used in medicine before. 230 00:24:27,570 --> 00:24:33,510 It's only ever been used in agriculture. So this is the very first trial ever to be done like this. 231 00:24:33,520 --> 00:24:36,750 It was published in The Lancet last May, May 13. 232 00:24:37,150 --> 00:24:40,379 We had eight regional trial offices, which was also a first for me, 233 00:24:40,380 --> 00:24:48,540 because what we did was we wanted to set up trial offices in every country so that they could do their own data cleaning. 234 00:24:48,720 --> 00:24:53,549 They could we could teach them how to run a good local trial. 235 00:24:53,550 --> 00:24:59,330 So in fact, we got eight local trials and it's that capacity building for the next. 236 00:24:59,420 --> 00:25:03,260 Generation in those countries, which was very important to me. 237 00:25:04,370 --> 00:25:11,090 So we ended up with 54 tribal members in the truly regional offices that we worked with daily. 238 00:25:11,360 --> 00:25:18,200 So we recruited between May oh seven and December ten, and we recruited 50,935 women. 239 00:25:19,480 --> 00:25:22,460 And it was a bit it was on target. 240 00:25:22,500 --> 00:25:29,960 We extended for about three months purely because we had a few difficulties at the beginning and some of our countries had to start late. 241 00:25:31,550 --> 00:25:37,670 President Bhutto went and got assassinated on the day we were flying to Pakistan to start off the Pakistan sites. 242 00:25:37,970 --> 00:25:41,480 So we couldn't go. So, unfortunately, Pakistan starting six months late. 243 00:25:41,480 --> 00:25:44,800 So therefore, we had to switch the recruitment back to the end. 244 00:25:44,810 --> 00:25:49,850 And this is where they were. So Asia, Africa and South America. 245 00:25:49,940 --> 00:25:53,930 And the primary outcome was death or composite of maternal infection. 246 00:25:54,850 --> 00:26:02,690 And you'll see that the components of the return infection and blood transfusion of more than one unit and there's a story to that, 247 00:26:02,690 --> 00:26:06,920 too, is the transfusion for postpartum haemorrhage. 248 00:26:07,400 --> 00:26:11,030 And it's more than one unit, because we after the first interim analysis, 249 00:26:11,030 --> 00:26:19,729 there was a big spike and the GMC asked us to look at blood transfusions and we realised that 75% of our blood transfusions 250 00:26:19,730 --> 00:26:26,270 were coming from one country because they all take one with them when they go in for operation and we have to use it. 251 00:26:26,280 --> 00:26:34,700 It is not ethical to take it home. So in Pakistan, every woman is given one unit of blood after her caesarian section, 252 00:26:34,700 --> 00:26:43,340 which is probably no bad thing because a lot of them are anaemic, but they take that blood in in donated by a relative and they don't take it home. 253 00:26:43,760 --> 00:26:51,050 So we then changed. We had an amendment to our protocol and we changed it to more than one unit so that we didn't have this big scare. 254 00:26:51,230 --> 00:26:56,930 Okay. So some of the challenges we faced, randomisation was done by web and telephone. 255 00:26:57,110 --> 00:27:01,820 And as you can see, the web and telephone lines, the connectivity was was amazing. 256 00:27:02,060 --> 00:27:09,050 They had hanging out windows anywhere they could get a line in to get Web Internet connection. 257 00:27:09,440 --> 00:27:13,069 But they managed 24 seven randomisation during the day time. 258 00:27:13,070 --> 00:27:16,729 It was done over the web, in the trials offices in the night time. 259 00:27:16,730 --> 00:27:19,760 It went through to our office here in Oxford. 260 00:27:20,220 --> 00:27:23,600 It was a voice recognition system in three languages. 261 00:27:23,600 --> 00:27:35,030 It was in Spanish, English and Indian English because our Asian site said, Well, we can we can sort of understand the English, 262 00:27:35,390 --> 00:27:39,130 but can I investigate investigator come into his English because then we'll understand it better. 263 00:27:39,140 --> 00:27:42,680 So he came, did a tape for us. So we had Indian English as well. 264 00:27:43,880 --> 00:27:53,780 And apparently the people in Pakistan understood that the CIA was English and of course very different to the UK. 265 00:27:54,110 --> 00:28:01,880 This is Ghana 16 bed ward and probably about four to a bed, as you can see, under the beds as well. 266 00:28:02,720 --> 00:28:11,570 But they're very jolly and they're also happy and it makes you feel quite humble to go there and work there and then 267 00:28:11,570 --> 00:28:17,780 come back and see what we've got whinge about because they're all very happy with all these lovely babies and you know, 268 00:28:17,930 --> 00:28:26,510 the nurses are all jolly and so we go there and we come back and they can't get GP appointment for next Friday. 269 00:28:28,400 --> 00:28:35,990 So this was our recruitment and as I say, we, we started in May oh seven and we official start was September. 270 00:28:35,990 --> 00:28:42,620 So because we had a running period where we expected to get up, as you say, we went from 12 to 25 to 50 to 75%. 271 00:28:42,860 --> 00:28:49,399 So we we hope to reach our 400 a month, which was our target by September, which we did. 272 00:28:49,400 --> 00:28:55,370 We had some dips and a lot of our dips with the religious holidays, Ramadan and things like that. 273 00:28:55,370 --> 00:29:01,420 And we got to it again, you see, that's knowing your culture. It's not going into a panic when suddenly in August, the first year, 274 00:29:01,520 --> 00:29:06,889 recruitment drops off because, you know, the Sudan and Pakistan, etc., the Ramadan. 275 00:29:06,890 --> 00:29:10,310 So they're not going to recruit. And these were the five intervention pairs. 276 00:29:11,120 --> 00:29:16,429 And one of my jobs and one of the questions I was at interview was, okay, Barbara, 277 00:29:16,430 --> 00:29:20,660 if you got five intervention pairs and they start to deviate, what would you do change them? 278 00:29:21,320 --> 00:29:23,330 So I said would probably change the intervention. 279 00:29:23,330 --> 00:29:32,990 See that could change them through another side and about May, May nine as up to one and five started to deviate out. 280 00:29:33,430 --> 00:29:36,530 As you can see, I sit there and started to deviate out. 281 00:29:36,860 --> 00:29:45,079 So because we knew our sites and because we knew that the level of recruitment and because we had asked the questions early, 282 00:29:45,080 --> 00:29:50,270 early on, what would you be prepared to do? We could change two of our sites to a different intervention. 283 00:29:50,270 --> 00:29:56,780 So they stopped what they were doing and they started to do the other two or three intervention pairs, which as you could say brought it back. 284 00:29:56,780 --> 00:30:05,620 So at the end they were. I sleep together and we had about over 9000 in each pair, which was what we were aiming for. 285 00:30:05,680 --> 00:30:08,680 But that was a challenge, monitoring this every week. 286 00:30:08,680 --> 00:30:14,740 And, you know, down here thinking, oh, crikey, they're getting a bit wide, they're getting a bit wide. 287 00:30:15,880 --> 00:30:21,130 And these were uncontrollable events. We couldn't actually build this into our project plan at all. 288 00:30:21,520 --> 00:30:26,590 So. Assassination of Bhutto. Civil war in Sudan, which split the country in half. 289 00:30:28,000 --> 00:30:29,350 Tribal conflicts in Kenya. 290 00:30:29,350 --> 00:30:39,280 Which meant that a lot of our women would come from the areas around Nairobi and they'd all been taken off of the Rift Valley into camps. 291 00:30:39,940 --> 00:30:43,030 Terrorism and flooding in Pakistan. Heatwaves in India. 292 00:30:43,330 --> 00:30:47,290 In fact, we were there setting up the site when that road was melting in Delhi. 293 00:30:47,290 --> 00:30:52,140 It was the hottest it ever been, an earthquake in Chile and all these things, 294 00:30:52,150 --> 00:30:55,660 you know, you have to sort of take in your stride because what can you do? 295 00:30:55,690 --> 00:30:57,640 I mean, we're not going to stop. 296 00:30:57,910 --> 00:31:06,760 What we did worry about, we worry very much about we did worry about the tribal conflicts and the civil war in Sudan and the flooding in Pakistan, 297 00:31:06,910 --> 00:31:12,520 which moved women away. And we were very worried because we were doing at least three year follow up. 298 00:31:13,030 --> 00:31:20,800 So we did worry about attrition when it comes to the fallout because we thought a lot of our women would be so far away. 299 00:31:20,950 --> 00:31:26,439 But having said that, technology has taken over in these countries big time and mobile phones. 300 00:31:26,440 --> 00:31:30,550 Everybody everybody from the four year old onwards got a mobile phone. 301 00:31:30,880 --> 00:31:38,320 So providing you collect enough mobile phone numbers, the chances are you will actually be able to find your woman in the end. 302 00:31:39,310 --> 00:31:49,480 So follow up. We still found out 13,226, which is 84%, which we thought wasn't bad given what had gone on before. 303 00:31:50,020 --> 00:31:55,360 And we're at the moment writing this paper, hopefully published in The Lancet this year. 304 00:31:55,960 --> 00:32:03,070 But some other statistics, this is where we travelled, so you might find this a bit interesting. 305 00:32:03,280 --> 00:32:08,110 The equivalent of 17 times around the world and 28 miles per woman. 306 00:32:09,070 --> 00:32:10,000 And we survived. 307 00:32:10,180 --> 00:32:20,650 But we only survive with around about 800 anti-malaria tablets between the three of us, 20 different hotels, eight passports, all this sort of stats. 308 00:32:20,890 --> 00:32:33,010 This decision is not it's not interested in that. We were and our paper is about 103 Cronus team members because this is because the Eiffel Tower 309 00:32:33,250 --> 00:32:40,540 and we're trying to find somewhere to accommodate all this is what we've been doing for 35 years. 310 00:32:41,260 --> 00:32:44,950 I find it very challenging. I find it very, very interesting. 311 00:32:45,280 --> 00:32:54,070 It's a job I'll always go but for. And as I say, it's been humbling, but also it's been exciting and it's been fun. 312 00:32:54,460 --> 00:32:59,440 And this is, I think, something that is getting lost in clinical trials. 313 00:32:59,740 --> 00:33:04,450 And I think it needs to come back because every job you work in everywhere has to have a bit of fun. 314 00:33:04,720 --> 00:33:11,950 And if you can enjoy it with your experiences with your investigators and the variety of people that you work with on clinical trials, 315 00:33:12,220 --> 00:33:16,870 if you can enjoy it and have fun, then there'll be a good trial and it'll be one worth doing. 316 00:33:18,670 --> 00:33:19,060 Thank you.