Skip to content
#UnblindingResearch Session Two: Formulating research questions and designing a project episode artwork

Mar 23, 2018 28 min

#UnblindingResearch Session Two: Formulating research questions and designing a project

from Take Aurally Hosted by Take Aurally

This episode is a live recording of the second session of #UnblindingResearch held in NUH DREEAM 21st March 2018. Research. THE search. WE search. And...fee search?! This episode covers that moment of inspiration as you embark on research. We cover PICO and how that helps us with our literature search, to formulate a question and how that will begin to shape our methodology. There's even a bit of a history lesson as we discuss Dr James Lind and his work on scurvy in 1747 (because we love a bit of medical history at Take Aurally). The slides for this presentation and the links mentioned are at www.takeaurally.com. Remember you can find Take Aurally and NUH DREEAM on both Facebook and Twitter. The DREEAM Research team can be found @DREEAMResNUH. The next session from #UnblindingResearch is on GCP and Ethics and will be on 18th April.

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

#UnblindingResearch Session Two: Formulating research questions and designing a project

0:00 28:53
of matches

Transcript Auto-generated, may contain errors

Thanks for downloading and welcome to Take Aurelia, the podcast from NUH Dream. This episode is a live recording of the unblinding research session on formulating a research question. As ever, all information is correct at the time of recording. All guidelines mentioned are correct from Nottingham University Hospital's NHS Trust. Other trust guidelines may vary. All views and opinions are the speaker's own. Hello everyone. Hello. Hello.

So, I think you all know me. My name's Jamie. I'm one of the doctors and one of the teaching fellows. Welcome to our second session of Unblinding Research. Thank you to Cecilia for her help with this presentation and for Claudia with her continued support and talking to me in the office earlier last week about what to do. Right then, who wasn't here in our session last month? Cool. People who were here, can you remember what we talked about? What should we talk about?

beautiful yeah so you remember we were talking about uniforms in A&E and how you could do an audit about it you could do research we could do a quality improvement project and what the difference approaches would be so if you have an audit that comes with standards which can be nationwide that could be college from at the RCN can be from the Royal College of Medical Medicine or it could be trust-wide and then you measure your performance against those standards and you measure yourself against those standards. So we use the examples that you could have people, you could audit whether people put things in their top pockets, and measure their performance against the standard. The standard would be that 0% of people should put things in their pockets, and we could go on the shop floor and find out 100% of people are doing things and putting things in their pockets, and so therefore we need to stage an intervention. That could be an educational event, that could be some part of training. You should stop putting things in your top pocket. Once you've done that, you put it in roll call, et cetera, you then measure yourself after that intervention, and you repeat the cycle. And I think we all admitted that that's something that we're very, very bad at, is that next step when we come to repeating the cycle.

We talked a bit about quality improvement projects and about how you have to look at the process when it comes to a quality improvement project. A stage in the patient journey through your department. You plan, do, study and act. So you think of an improvement you can put in place. You study the effect of that. And then you act if there's any benefit. Or actually sometimes with a quality improvement project you can find that you actually do more harm than good if you stop it there and then. And I've done a few quality improvement projects where I've actually found out I've slowed down processes until you have to kill it there and then. And then embedding practice. So going to other departments going, we've done this, this is really good. Putting things out nationwide and you can find those things nationwide spread through the NHS.

And then finally we talked about research, and that's what we're going to build on a little bit today. So the research process. About looking at the literature, about finding out what we know, and also what we don't know. Have people proven things already? We don't want to repeat their work. Have they shown areas of future work that we can build on? What is out there already? What outcome is it that we want? And that's really, really important. We have to be very smart with that outcome. What is that specific outcome that we want to measure? And then that will affect our methodology. So that will then determine what sort of study we're going to perform. Cool.

So that means for our talk today is on formulating research questions and designing a project. And these are our learning objectives which should be there on your reflective performance. So before we carry on a bit further I just want to remind everyone that we need to, this is a friendly environment so if anybody has a question feel free to speak up. I want this to be a nice environment for everybody to learn. Go forward and do research.

It can leave you feeling a little bit in black. Everybody wants to do research. It sounds sexy, it looks good on our CV. I do research in this. Everybody wants to be the guy that discovers the next statin, the next whatever. We all want to cure cancer, but it's a bit difficult. And about 12 months ago, when I sat down with my dissertation supervisor and said, I'm interested in this, this was kind of like how I was feeling, very much in the dark. You've sat in meetings and people will say things like, I'm interested in teamwork. Well, that's a lifetime's work. What do you actually want to research? People spend decades researching teamwork. What's the actual, that specific piece of research that you want to do?

So research, see what I did here, is about the search. See what I did? I will. Research is about the search. The search for that inspiration, that moment of knowing what we want to do. And luckily, there is a particular mnemonic that we can use, an abbreviation that we can use, that can help us when we're thinking about posing a research question. And that's PICO. Have you all heard of PICO? What does key stand for? Problem yet? So it can be a population or a problem. And a problem.

And if you think about it, different areas will have different populations and different problems. So we here in Nottingham are a former mining area. We have a lot of respiratory issues around here. Areas of London have a very high Afro-Caribbean population. They see a lot of sickle cell problems, much higher than we see here. So where you are in the country can actually, and possibly your own personal interest and your experiences, can lead us to thinking about that population and that problem. And it's important to try and be as precise as possible with that population and that problem. I'm interested in X miners with COPD, for example, a very clear definition. What does I stand for?

It is intervention, yes. So that's the thing that you're interested in studying, that's the thing that you want to do. So that could be the next brand new wonder drug, it could be an educational event, something or other that you are interested in looking at. That is your R, your intervention. C, comparison or control. So think about it, your control is doing nothing, that's pretty much the status quo. Comparison are the other things that you might be looking at. So you might be examining one drug versus another drug. So your drug A is your intervention, your drug B is C for your comparison. And O,

Outcomes, brilliant. This is, again, very, very important to be very precise with what your outcome is. This is the difference between primary and secondary outcomes. Your primary outcome is what you are interested in finding. I'm interested in finding that if I give this drug to patients with this condition, it will reduce blur. That is my outcome. And that's what's known as your primary outcome. And that's what your research is powered for. So we all see those studies in the well-known journal, The Daily Mail, where it says that statins cure cancer. And that they did a study where they're looking at the effect of statins on cholesterol, and they found out patients in that study had reduced rates of cancer. Their study was not powered to look at cancer.

The primary outcome would be the cholesterol. A secondary outcome would be the cancer rate. So the primary outcome is actually what you're looking for. If you find other things along the way, those are secondary outcomes. We have to be very, very careful when we're interpreting those because the study is not powered for that. And we'll talk about powering in a later session. So that will make sense. And that's very useful to think about when you're reading a paper. You can think about what was their problem, what was their population, what was their intervention, what was their control, what was their comparison, what was their outcome. If you think about the P and the I, that forms your literature search. So when you go forward to going to PubMed or going to whatever online journal, your population problem and your intervention, that's what you're going to search for.

So last time I talked about the person who invented the clinical trial. Do you remember who it was? Florence Nightingale invented all its quality improvement. Who invented clinical trials? It was Dr. James Lind in 1747 when he discovered that vitamin C cures scurvy. Well, citrus fruits cured scurvy. Yeah, absolutely. So if you imagine in 1747, his equivalent then of Wikipedia or PubMed, what was his population and his problem? His population and problem were Royal Navy savers with scurvy. His intervention was citrus fruits. So he'd have been looking for those. That's been his literature search.

Back then it was believed that scurvy was caused by not enough acid in the diet, which is how he came about thinking about giving citrus fruit, vitamin C, so he was on the right track even though for the wrong reason. So yeah, he'd have been searching for Royal Navy sailors with scurvy and citrus fruits. He wanted to know what's known about the health of Royal Navy sailors, what's known about scurvy, and what do we know about citrus fruits. And if you put all four of those together, that makes your research question. You have your population problem, your intervention, your control comparison, your outcome, put those together, rearrange the words so they make an order, that is your research question. So James Lynn's research question in 1747 was...

do our citrus fruits better at curing the scurvy, Royal Navy sailors with scurvy, compared with, and I think he had seawater, sulfuric acid, cider, and a spicy paste mixed with barley. That was his comparison. And yeah, he had two people, he had ten people with scurvy, divided them into five groups, and he gave two of them citrus fruits, two of them cider, two of them seawater, two of them sulfuric acid, and two of them a spicy paste. And the ones with vitamin C got, the ones with citrus fruits got better. I don't know, one would hope that he cancelled the trial a lot, but yeah. So then that would be his research question. Does that all make sense?

So now it's time for the group work. So you're roughly in two groups, aren't you? So, Cecilia, can you pass on to the group over in the middle? Thank you. So in your groups, I'd like you to think of an area of research that interests you. Try and devise your PICO, work out your research question, and also work out what your literature search would be using the PICO formula. Matthew, would you care to come to the front? Come on, my mother.

Would you like to demonstrate your group's research question? Our intervention was improved training programme via foundation project or programme. Our control would be staff who hadn't previously been through the foundation programme and therefore didn't have improved training. And then the outcomes were looking at potentially staff who stayed past six months post joining the department, with a secondary outcome of then looking at how happy they were working within the department.

And how are you going to measure those in there? We didn't actually discuss this properly, but we probably looked at kind of stacking data to see what happens with people who started some time last year, as the result this year you went through the programme, if there's any difference in the amount of people who stayed. Anything else, Mark? Yeah, we talked about, I know what we're looking at doing now. Some of the qualitative data is looking at questionnaires and things. And the other thing I know that we are looking at is doing interviews.

But looking at that percentage of staff that we've obtained at six months, we'll probably do it a year as well, to see where they are at a year. And also look at their competencies and how much of their competencies. There's quite a lot of everything to look at out there. Thank you. Thank you very much. ITU team. What did you guys come up with?

So our population was patients on the adult intensive care unit at Queen's in Luscombe. And our problem was sleep deprivation in those patients. So our intervention would be giving melatonin. And then our comparison or control would be versus frequency, though not melatonin. And then the outcome would be to assess hours of sleep.

overnight. So our research question we said would be, does administration of the dodolos melatonin improve sleep in patients on the adult mental care unit? And then we kind of came up with a few ideas for how we might assess this because we were saying it might be quite difficult. We said, you know, firstly we could ask the patients whether or not they had a good night's sleep, but that's not very, that's quite subjective. so it wouldn't give us very reliable results potentially. And then another suggestion was that we could get the nursing staff to document every hour whether the patient was asleep or awake to assess how many hours of sleep they'd had overnight. And there was also a suggestion that maybe the patient could have Fitbit. We're getting some private commercial sponsorship. I like it, we're going to come on to do funding. But then they just want to come up with their own consultancy, saying with the Fitbit, often they assess your heart rate. And often online for you, we're giving people medication to drive their heart rate up. So that's a compounding factor.

We were saying we might need some exclusion criteria because lots of the patients are sedated, so it's very difficult to assess whether they're asleep or not. We were saying maybe we should exclude patients who are sedated and have patients who are away because they need to assess their actual hours of sleep. Difficult to assess sleep, isn't it? Yes, I think it is. You could be sat there, you could be lying for eight hours and you're fast asleep, but if you have a scary dream just before you wake up, the nurse will write, you've had a great night's sleep, but you'll go, it's horrible. I dreamt about killer bunnies and things like that.

Does melatonin improve sleep? Is that something that's known? So we use it. I haven't actually looked into whether it actually helps, but we do use it quite a lot. So there, just through thinking about our PICO, we can start to revise our question and then also think about our methodology and then how that might lead us open to potential questions about how useful our study is.

you guys were talking about comparing to a previous group of staff. You want that group that you're comparing against to be as close to your current group as possible, so in terms of their demographics and in terms of the time that you're doing the study. And you've already spoken a little bit about, you know, you can be asleep, but there's actually good quality sleep, you know, would you want a mixed methodology approach, what the nurse thinks, plus what the patient thinks, etc. Would you want to get commercial, get Fitbit in? I think you've got apps as well that measure how good you were sleeping, your REM sleep and everything, yeah. They can record you while you sleep, it's all very creepy. Right, so well done, guys. So for the last bit of this, we're going to have a look about...

The other thing with research is that we can sometimes feel like we're alone, that you're the person doing that research. And that's certainly not a thing you should ever feel like. You should always feel like you have support. So you're going to like this bit as well. Research, we-search. This is what I think about. If you remember nothing else, you'll remember those slides. No. Wait a minute. One second, guys. Turn around. So if you guys were starting some research, where would you go?

starting some research, you're doing your thing on, you know, you're doing some studies looking at vanity, so if you want to look at sleep cycles in ICU, where would you go? Anywhere closer to home? You want to speak a bit louder? Yeah, absolutely. Who do you get in touch with? You get in touch with R&I. You get in touch with the R&I department, our lovely research innovation team, who will, you know, I'm thinking about doing this research. You should never start it alone. And the earlier you get other people involved, the absolute best it is. Okay. Your email address is through the Trust Outlook.

On a very basic level, I'm thinking about doing this study, what can we do guys? So, you all heard that every trial must have a sponsor? Is that a thing that rings a bell to people? What does the sponsor do? It's not about money. Even though they're called sponsor, it's nothing to do with funding. The sponsor is the person or the organisation responsible for your trial's governance. Is it ethical? Is it following the right guidance? The trust can be a sponsor. When you have an audit, that can sometimes be one person being your sponsor. A larger study, you have a whole organisation being your sponsor. And you go about getting a sponsor by getting in touch with research and innovation and saying, well, there is this trial we want to do. And this is something that I'm thinking about this study. This is my PICO. What's the best methodology for me to find my outcome? There are other national services as well.

The Research Design Service, RDS, a great resource. Basically, coming up to them, this is my PICO, this is what I'm interested in, can you help me design my research? NHS funded, obviously for the good of the NHS, so it has to be clinically relevant, you can't come in talking about chocolate or something. But if it's a useful study and they're interested in it, they will help you with research. But then what do we do about funding? Who do we go for funding? For your research? People not from R&I. Where do you go for money?

You can go through the charity pot, beautiful. So, the trust, I believe, is it once a year, Claudia? I think it's twice a year. Is it twice a year? I'm not going to show it up. But basically, if you search on the intranet, it's not going to let me come up, but if you search for pump priming, they have a pot of about £100,000, which they like to divide up between 10 pilot projects every year. So you get £10,000 to come in. I want this little study that's going to prove a point of principle, be a pilot study, something that we might then be able to take nationally and be bigger. What do you think? And there's a form that you can fill in. It's all nice and simple. They want to keep it nice and short. And if you win, you get £10,000, and you can then spend that to run your project. Because in running a project, you need equipment, you need time, and you need staff. And with that money, you can use that to help get your project off the ground. That's known as pump priming.

Because you're priming the pump. I don't know. I'm not a plumber. I don't know. I was told about it. I wasn't just randomly searching phrases. It's really good. And basically, the online form is there. And they call it competition. But you just send off and go, I'm interested in this. Can you give me £10,000? Charity funding is another option.

Another option nationwide is the National Institute for Health Research. And so this is NHS money again, comes through the Clinical Research Network, and this is money that is given to fund clinical research to help further our NHS care, patient care. Basically, if you go to their website, it goes to take through all the different steps. Now, because there's NHS money, there's a lot of money here, there are particular areas that they're more interested in others than others. And so they have a variety of different funding programmes that you have to find and see where your study fits in under that. So things like efficiency and mechanism, health technology assessment, down here the systematic reviews as well, that's where your Cochrane reviews come in as well. So you have to find your area. You can't just come in and go, I'm interested in this. You have to find an area that they're already interested in and see where your project fits in with that.

What they also do is they have calls, so they call out for research. So if you search on there, basically they're saying, so closing date at the end of this month, they're interested in something that comes in under Public Health Research Program. So you would click on there, you'd see what they're interested in and see whether you've got some research that you can go with that, can you get a pot of money from there. Or, you can do what you were going to do, you can go to a commercial company and say, could I have some money to do my sleep project? And that does happen. So you have clinical investigators, say you're the CI for the project, who has a relationship with Fitbit. They might take you out for dinner, they might email you and go, actually, you know what, we'll give you some money to help you with your study and you can use Fitbit in your study. Any potential problems in what I've just said there? Marvellous. What happens if you find out that Fitbits kill patients?

Marvelous, no they won't. So, do you think they'll just go into a relationship with anybody? No. So there's things called clinical trial agreements, that they will make you sign, make the trust sign, and they will sign as well, detailing out the terms of the arrangement you're going into. This is why you have to involve research and innovation, because they have an amazing infrastructure. If you've never been up there, go there, it is beautiful, and they have windows, they have lights, which we don't doubt have here in anything.

But they have a lawyer who will sit there and go through those things and have a look at it because they will say, under these terms, we will not publish. And this is why, one of the reasons why, is it double the rate of positive to negative trials that get published? It's something ridiculous. You're more likely to get published if you prove a positive than if you get a negative. That is one of those reasons. But the National Institute of Health Research loves it because for every pound of commercial money that comes in, that's a pound of taxpayer money that doesn't have to go on that trial. So they do like it. And they do actually encourage a mixed funding pot.

So we have to live with it. This is why you will see when you come to write your ITU sleep deprivation, you will say our clinical investigator has a relationship with Fitbit. Excellent. Does that all make sense? Cool. I know we've covered quite a bit. PPI. We've all seen those adverts. Have you been miss old PPI? In this instance, PPI is a good thing. Do you know what it stands for? PPI. PPI is something to be encouraged. This is your patient and public involvement.

This is another reason to speak to R&I nice and early because you have designated members of the public who have personal experience who can help you out. If you are creating a leaflet for the public, there is no good you writing out a leaflet that you think is all very good for a layperson and then bring that to R&I and then they show it to their PPI guy and they go, I can't understand a word of this, this isn't it. This is why get them involved, get PPI involvement. You're much more likely to get funding if you have PPI involvement. Make it nice and patient and public friendly. Especially if you're involving them in any sort of survey or that qualitative stuff, if you're using leaflets, if you're using an app, anything like that. It's not always a bad thing.

So in terms of links, so if you guys want to jot down or take a photo, this is a link to the National Institute of Health Research that we put up there. Just have a play around, have a look about all those things, there's a lot there you can spend quite a while on that. And I wasn't thinking of research and we-search, I was exploring there for a good hour or so. Quite proud of that. And then for the research designers as well there. I'm not going to put these out on the internet, but so...

I was going to put the R&I email address up there, but I'm not going to put that out on the internet. But if you look on the Trust web address, you'll find these guides. As ever, this will be available through take-off. But yeah, so we will put this out. All the slides will be up there as well. So yeah, please have a listen. Marvellous. Our next session is 18th of April, when we'll be talking about GCP and ethics. We'll get on very deep, what we'll talk about. So save the date, 18th of April. Thank you very much. We haven't finished early. Thank you very much, guys. Enjoy wee-searching together.

That was the Take Orally podcast session from Unblinding Research on formulating a research question. You can find the blog entry and the slides for this session as well as the links mentioned at takeorally.com. Remember that the next session from Unblinding Research is on the 18th of April and will be on GCP and ethics.

Frequently asked questions

How long is this episode of Take Aurally?

This episode is 28 minutes long.

When was this Take Aurally episode published?

This episode was published on March 23, 2018.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this Take Aurally episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.