Talking about research
Back in 2016*, Ruth interviewed Professor Arri Coomarasamy , Consultant Gynaecologist and Sub-specialist in Reproductive Medicine and Surgery at Birmingham Women’s Hospital.
* N.B. Results of several of the research trials mentioned below have been published since this conversation, and and are reported on in the links.
RBA: Arri, most people who’ve been through miscarriage wish that we knew more about why it happens and how we might prevent it. Is that your experience too?
AC: When I see patients who have had a miscarriage, they generally have two questions: why did it happen? How can we stop it happening again? We are developing a reasonable understanding of why miscarriages happen, although it may be difficult to work out the reasons for a specific couple.
RBA: I know you think research can help, especially trials of certain treatments. Can you say something about that?
AC: The question about how we may prevent miscarriage is the tougher one to answer, although it is possibly the most important question for the couple. Some early studies sometimes suggest a benefit with this or that medicine, but when we eventually study that medicine carefully, we may find that it doesn’t work, or even worse, it is actually harmful.
RBA: Can you give me an example?
AC: A good example of this is aspirin. We use it for women with sticky blood conditions such as antiphospholipid syndrome (APS), and it works to reduce miscarriage in those women. However, even when there is no evidence of a ‘sticky blood’ condition, many clinicians have recommended it and women have diligently taken aspirin in the hope that it will reduce the risk of miscarriage. But when this question was properly researched, it was clear that there was no major difference between the results for those taking aspirin and those taking the dummy tablets. In fact, one study showed that there were slightly more miscarriages amongst the women who took the aspirin.
Only research can uncover such important findings that help guide treatment for women with miscarriage.
RBA: Are you saying that aspirin can increase the risk of miscarriage in women who don’t have APS or other sticky blood conditions?
AC: In some cases, yes. The difference was very small but it shows how a treatment given with the best of intentions could achieve exactly the opposite of what was expected. However, I must emphasise: for women with sticky blood conditions, the available evidence is that they will benefit from aspirin and heparin, and if their doctors prescribe these, then it would be a good idea to take them.
However, if there is no evidence of a sticky blood condition, women are well advised not to take aspirin, unless their doctor prescribes it to them for some other reason.
RBA: That’s really important information, Arri – thank you for highlighting it.
To go back to the question of research. I imagine some of it, like trying to find out the causes of miscarriage, doesn’t directly involve patients – that is, they don’t take an active part in the research except, perhaps, for allowing blood or tissue samples to be analysed.
Arri: This is correct Ruth. We often do what is known as ‘basic science’ or ‘bench’ research in the laboratories to understand the science behind why miscarriages occur and how a treatment may work. This research does not normally involve patients directly, but relies on the blood or tissue samples.
Such research is really important to find out potential new therapies, which can then be tested in clinical trials. As an example of this, we are currently studying the blood of women with thyroid antibodies to see if there is an immunological cause for miscarriage in these patients. If we unearth important findings here, this may help us develop and test potentially important therapies.
RBA: That’s good to know. And you are also involved in some clinical research trials where you are looking at the effects of certain treatments – like aspirin. Some of them, like the PRISM (progesterone) and TABLET (thyroxine) trials, are randomised controlled trials, or RCTs. Can you explain what these terms mean?
Arri: Sure. It is really important to evaluate treatments in high quality studies, so that patients and their doctors can be confident that the treatment does really work. The highest quality studies for doing this are called randomised controlled studies, or RCTs.
PRISM and TABLET are randomised controlled studies. In PRISM, women with early pregnancy bleeding will receive either progesterone or a placebo – that is, something that looks like the real treatment but doesn’t have any active ingredients. Neither the doctor nor the patient knows if the patient is taking progesterone or placebo tablets. Only after the study has been completed will we know which treatment a patient was taking and whether that treatment made a difference.
This is important to avoid any ‘biases’, so that if the study shows that progesterone reduces the miscarriage rate, then we can be confident that it does really work. Equally, if the PRISM trial shows progesterone does not work, then we can be confident that it does not work. Patients and doctors need to have confidence in the treatment that they receive and provide, and RCTs gives everyone confidence in the results of a study.
RBA: You’re saying that if patient or doctor knows whether they are having the real treatment or the placebo, then that could cause a bias, make a difference to the outcome. Why? How?
Arri: It is possible that the doctor could consciously influence the outcome of a study, say by promoting one treatment over another or over the placebo. The doctor may offer additional care for patients in one group, and this could influence and prejudice the results.
In fact, there doesn’t even need to be a conscious influence. There is evidence that if we know who is in the treatment group, we may even unconsciously influence the findings and let all sorts of biases creep in. Patients and doctors need trustworthy evidence, and a study which is not properly randomised or blinded may not produce trustworthy evidence.
RBA: Hmm. Thinking about it, though, if I had had repeated miscarriages and nobody knew why – AND if there’s a treatment which is being tried out – I don’t know that I’d want to risk being given the placebo drug and possibly missing out on a good treatment. Why shouldn’t I just go out and find someone who’ll prescribe the treatment?
Arri: This is a very reasonable concern. However, you must then remember that the whole reason for doing a study is because we do not genuinely know if a treatment works! If we thought it did, then we would be prescribing it.
Sometimes we think we know the answer, but then you only have to look at the aspirin story to realise that rushing in and taking a treatment without good evidence can actually cause harm. In fact, if it weren’t for the women who willingly took part in that research, we may have unwittingly continued to cause harm.
RBA: I suppose there might also be people who are anxious about taking the treatment in case of possible harm or side-effects.
Arri: This is true. However, any clinical study goes through rigorous assessment to ensure the study is ethical and does not cause harm to the patients. Moreover, as the study is conducted, there is continual assessment to ensure there is no emerging evidence of serious harm to mothers or babies. This task is carried out by an independent body called the Data Monitoring Committee. This should give reassurance to patients in a study.
RBA: I’ve just got one more question, Arri, going back to the idea of the psychological aspects of being in a trial. I understand that there is some evidence that just taking part in a clinical trial can improve outcomes, whether women are taking the treatment or the placebo. It might be to do with having extra care and monitoring, perhaps. Is this really true?
Arri: This is correct, although there is some controversy over this. What is true is that patients taking part in trials do tend to get additional care and monitoring, and there is of course some evidence that enhanced care can have a positive impact in reducing miscarriage risks. I certainly think that trials such as PROMISE and TABLET have very little scope for harm, while carrying a potential chance for benefit.
RBA: So overall, it looks like taking part in clinical trials on miscarriage can be a really positive thing to do. It helps us to work out which treatments are effective and which should be avoided because they either don’t work or are positively harmful. That’s a great gift to others who go through the sadness of miscarriage.
What’s more, it looks like even if the treatment doesn’t turn out to be effective, just being in the trial can itself be helpful. Now that’s a bonus.
Thank you very much, Arri.