Seven reasons to question headlines about the benefits of epidural

As is often the case, some journalists and commentators sought click-generating headlines and summarised the research without having an understanding of the deeper issues relating to evidence.

But there are also some confusing aspects to the research itself.

So I am going to share some of my thoughts on this study in this blog post, to help you understand what some of the issues are, both in relation to this piece of research and more widely.

The all-important caveat

I need to add the ever-important caveat before I begin.

Neither I nor this blog post is ‘against’ epidurals.

I believe that everyone should be able to weigh up the pros and cons of any intervention in order to make the decisions that are right for them. That’s the aim of all of my books.

However, the reporting of this particular study has thus far been very biased towards the promotion of epidurals, which does not allow for informed decision making.

What concerns me the most is that the findings don’t quite match the headlines or support the conclusions.

So I have written this to offer the other side of the picture, and to explain why it’s always worth looking more deeply before making health- and birth-related decisions.

1. The findings don’t quite match the headlines

The headlines of the articles that discussed this study note how epidurals ‘cut the risk of serious complications by 35%.’

And the authors wrote that, “In this population based cohort study encompassing 567 216 births in Scotland, epidural analgesia during labour was associated with a 35% risk reduction in SMM and 54% risk reduction in SMM plus critical care admission across all births.” (Kearns et al 2024).

That sounds like epidurals make a significant positive difference, doesn’t it?

What did the results actually show?

When I looked at the crude rates of severe maternal morbidity, I was surprised – given the headlines – to see that there was no difference between those who did and did not have an epidural (see table 2).

Let me put one of the key results into another format for you, so you can see the absolute risk, or the actual number of women who (a) experienced severe maternal mortality and (b) were admitted to critical care units. The crude number is the finding before the authors analysed the results in different ways.

Without an epidural

In this study, 442,192 women gave birth without an epidural.

Of those women, 1885 experienced SMM and 750 were admitted to a critical care unit.

This means that, of the women who did not have an epidural,

1 in 235 experienced SMM and 1 in 590 were admitted to critical care. (data from Kearns et al 2024).

With an epidural

In this study, 125,024 women gave birth with an epidural.

Of those women, 527 experienced SMM and 177 were admitted to a critical care unit.

This means that, of the women who did have an epidural,

1 in 237 experienced SMM and 1 in 706 were admitted to critical care. (data from Kearns et al 2024).

What does that mean?

So it’s true that women with an epidural were less likely to be admitted to critical care than women without an epidural. But the crude rates of SMM were similar, and you can also see from this that the actual risk of being admitted to critical care is fairly low in both cases.

A bit more analysis

The authors then went on to analyse different subgroups. This means that they looked at women in particular categories, to see if any interesting things could be seen. These analyses showed that:

“Greater risk reductions in SMM were detected among women with a medical indication for epidural analgesia (0.50, 0.34 to 0.72) compared with those with no such indication (0.67, 0.43 to 1.03; P<0.001 for difference).” (Kearns et al 2024).

I have no argument with this. It’s not surprising that epidurals can benefit women who have medical conditions. That’s partly because we know that they have a number of effects on the body which include lowering blood pressure, as well as reducing pain. Epidurals are often recommended to women with high blood pressure, and they can help with that. It’s always the woman’s decision as to whether she wants one or not, of course, and the other questions I raised above are still worth considering, but there is evidence of benefit.

There’s another reason which explains why epidurals can be beneficial for women who have medical conditions or are at genuinely higher risk of having a problem. An epidural can only be given as part of a larger package of care. In other words, if you want an epidural, you also have to have a cannula, IV fluid, more monitoring, and closer medical care. That’s because, like all medical interventions, epidurals have effects and can lead to problems, such as a sudden drop in blood pressure. But the extra care that is put into place to monitor the effects of the epidural, and allow quick treatment if it’s needed, can also help if you do have a sudden problem that isn’t brought about by the epidural.

So this research suggests that, if you have a heart condition, or pre-eclampsia, or high blood pressure, you might be better off having an epidural in labour than not having an epidural, at least in relation to your chance of having severe maternal morbidity. Epidurals also seemed to be beneficial for women in preterm labour in this data set, although I don’t have time to look at this finding more deeply at the moment.

But for women who do not have medical problems? The data for this group did not show a reduction in severe maternal mortality with epidurals. So, while it is the case that epidurals do seem to make a difference in women who have medical conditions (but please read on, because there are some quality issues we need to consider), this data show no benefit for health women giving birth at term

And there are still other things to consider. An epidural may or may not be right for you. Epidurals have downsides. They can limit your mobility, have side effects, and can lead to a higher chance of certain other interventions. They can cause rare but debilitating and sometimes lifelong medical conditions. So you will probably want to weigh the decision up within the context of your personal circumstances. If you want to labour physiologically and experience the benefits of that, you may not want one. If you are experiencing a serious medical condition which might affect the outcome of your pregnancy, that might be the most important factor for you. It’s an individual decision.

I haven’t written much about the evidence on the pros and cons of epidurals more widely, but I can recommend this excellent book by Henci Goer. (That’s an affiliate link).

Is the conclusion warranted?

But does this research warrant the conclusion that more epidurals would improve maternal health for all women?

I can’t see how it does.

The most significant finding relates to the care of women with problems and, as I mentioned above, there is no benefit for women who are healthy and giving birth at term. They may cause more harm in this group. That’s because, again, epidurals can have downsides as well as benefits. The likelihood of a healthy woman experiencing a problem is low, especially if you make decisions that increase your chance of physiological labour and birth. If your chance of experiencing a problem is low, and something has downsides or aspects you don’t want, then it might not be for you.

But my concerns about the way this study is being discussed are also to do with the research methods, which aren’t the most robust.

Let’s have a deeper look.

2. This study is not an RCT

It has long been established that, if we want to find out whether or not an intervention (like an epidural) is effective at helping to increase or decrease the chance of a particular outcome (like SMM), a randomised controlled trial (RCT) is the best way to do this.

This is acknowledged by the authors, who write that,

“Although a randomised controlled trial would be ideal for confirming our results, the global prevalence of epidural analgesia during labour, its established safety, and the urgency of this research make a strong case for applying our results in clinical practice.” (Kearns et al 2024).

I absolutely agree with their first point, that an RCT would be needed in order to establish whether or not these results are accurate.

I am concerned, however, that the remainder of their paragraph ignores important factors such as women’s experiences, the effect of epidurals on the physiology of labour, and some of the wider issues relating to the medicalisation of birth. Not to mention the quality issues that I am highlighting in this blog post.

These are important considerations. It’s not enough to make recommendations purely on the basis of studies like this one. Again, we need robust research, and to look at the wider picture.

3. There are some issues with outcomes

When we want to look at the possible effect of something, we need to decide and define what outcomes we think it might affect. So in this study, the researchers were looking at the effect of epidural (or no epidural) on outcomes that fit into the category of ‘severe maternal mortality.’

But that definition includes 21 different conditions, from sepsis (severe infection) to organ failure. Some of the outcomes in this category are more severe than others. Certain outcomes are very easy for clinicians to define and agree on, while others are a bit more subjective. A couple of them are the result of a clinical decision by a doctor. These points are important because the way that things are defined can lead to bias in the results.

For example, the authors only counted a postpartum haemorrhage as constituting severe maternal morbidity if the woman was admitted to critical care. Thankfully, most of the women who have a postpartum haemorrhage do not need critical care, but we know nothing about what happened to them in this study, and whether there might be anything interesting to learn in relation to whether or not they had epidurals. Given that a number of research studies show that having an epidural increases the chance of postpartum haemorrhage, this is not a minor point.

Even though data were gathered from the medical records of half a million women, this number was too small to see some outcomes. This should give you an idea of how rare they are. When outcomes are unlikely or rare, it’s quite easy for one ‘chance’ event here and there to make something look better or worse than it really is. It’s not even the case that we can call for bigger and better studies. This one was already very large. Sometimes, we need to acknowledge that there are some things that we cannot know with certainty.

4. Medical records are known to be inaccurate

This is a minor but important point. Studies like this one are based on things written in someone’s medical records. And medical records can be notoriously inaccurate.

That doesn’t mean that they should never be used, and I have cited plenty of studies based on data gathered from medical records myself.

However, it is an important point to bear in mind, especially when researchers are using such data as the basis for making significant claims and calling for changes which will affect many people.

5. The study makes unhelpful, unsupported claims about epidurals and ethnicity

I found myself somewhat confused about the claims that the authors of this study were making about ethnicity. They clearly acknowledge the important of considering ethnicity and inequality:

“The latest UK Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries report underlines the uneven distribution of maternal morbidity and mortality, with deaths in women from black ethnic groups four times higher than in women from white ethnic groups, and the mortality risk twofold higher in women from the most deprived areas compared with least deprived areas. Recent UK based studies have shown that women from ethnic minority groups and socioeconomically deprived areas are less likely to receive epidural analgesia, although the underlying reasons remain unclear.” (Kearns et al 2024).

However, I find it hard to understand how they are justifying their conclusion given the very high proportion of white women in Scotland and the study.

The problem is racism, not race

First, in 39% of cases, ethnicity wasn’t recorded. This isn’t the fault of the researchers. But it’s another illustration of how using medical records as the basis for research can mean the findings aren’t as helpful as they could be.

The researchers themselves note that, “the population of Scotland is predominantly white, [so] our results might not be generalisable to more diverse populations.” (Kearns et al 2024).

They do argue that a US-based study which included a more diverse population had similar findings. However, maternity care in the US is even more medicalised than in the UK, and many of the other issues that I am raising in this blog post apply to the similarly designed American research.  

I am concerned that this is another situation where people seem to be focusing on race then they ought to be focusing on racism.

There is no evidence that offering Black and Brown women more medical tests and interventions is beneficial. In fact, I have concerns that this will lead to more problems, rather than less.

6. There are other types of possible bias that we need to consider

There are a few other things that we need to consider with this study. A few of these are discussed in the paper itself, so I would urge you to look there if you would like more depth. For example:

• There are a number of confounding variables in this study, and the non-RCT design makes it really difficult to know whether these are affecting the results. This is another reason that we should be very careful before making recommendations based on the results of this study.
• There are, for instance, clear differences in the women who do and do not opt for epidurals. The authors did things to try to take these differences into account, but this raises other questions. It is almost impossible to separate out the effect of different maternal characteristics and the effect of having an epidural. This is where a prospective study would be of much more use, even if it wasn’t an RCT.
• Other types of anaesthesia are used when there is a major emergency, such as a severe postpartum bleeding. As the researchers discuss, this might have the effect of making epidurals look more favourable.
• Women who do not have epidurals may be having other pain-relieving medication, such as opiates, or Entonox. However, these groups and interventions aren’t being separately looked at in the research. We need to bear in mind that the ‘non epidural’ group may include women whose SMM is related to the effect of other forms of pain relief.

7. There is also an issue with epidurals and access

There is also a bigger picture about equity and access which needs to be considered.

In a nutshell, it is well understood that epidural access is inequitable.

Having an epidural is also an elective decision, both on the part of the woman, but anaesthetists and other maternity staff play a part in whether or not a woman who asks for an epidural will get one, and when.

So what’s my conclusion?

As my friend and colleague Rachel Reed pointed out when we discussed this study, this research is based on women having non-physiological birth in hospitals, and it doesn’t show a benefit for healthy women giving birth at term, so it’s of limited use for many women.

We’ve seen this before, with studies like the ARRIVE Trial. In that situation, the nuances of possible bias may not even be relevant. But I have detailed several of these issues in the hope of helping more people understand the importance of unpacking research, asking questions, and looking beyond the headlines.

These are the same things that I write about and teach people to understand in my books.

There’s one last thing I want to mention, too.

The authors of this paper, like many others, are keen to place blame for increasing morbidity on women for being older, having a higher BMI and so on. Yet, as my own research shows, it’s far more complex than this.

The reason that women who are older, larger, and/or darker-skinned have more problems in some situations has more to do with the way they are treated in systems of health care than because of anything to do with their bodies.

It’s racism, sizeism, agism and misogyny that are the real problems here.

And epidurals aren’t going to help with that.

---

---

---

Books to help you become informed and make the decisions that are right for you…

---

---