New study finds home births are SAFER for high risk women

[Jackieharris]

onlinelibrary.wiley.com/doi/10.1111/1471-0528.13283/abstract;jsessionid=82D9DA516329D2AF352315DA7E71420E.f04t03

Well this throws the cat amounts the pigeons!

I'm pro hb myself but still thought that I would probably choose hospital if I was high risk (depending on exact circs).

Of course it's not wrong to feel safer in hospital.

But then it's not unfair when someone says that it is safer to point out that the evidence doesn't suggest it is for all women.

You're not challenging people's feelings about what's right for them.

"That is my worry here. I would not want a mum with high risk complications to base her decision to home birth on the incorrect interpretation of this paper being that home birth is safer for her and the baby. "

Luckily for you, it's not something you have to worry about, as women at high risk of complications wanting to birth out of hospital tend to be spoken to at length by their consultant and midwife, and usually given a flat out refusal of NHS provision for a homebirth.

Perhaps save your concern for the very much larger group of women for whom hospital is the default option, despite the lack of discussion surrounding the additional risks posed by obstetric led birth.

For example, where have you ever heard of a low risk mother having her second baby being advised that giving birth in hospital appears to more than double her risk of ending up in HDU, or needing a blood transfusion or a GA, without improving outcomes for her baby?

I'd say, never. Nobody needs to justify hospital birth, only out of hospital birth.

Things like placenta praevia are always going to be an indication to birth in hospital.

It's the high risk labels attached to women over 40, or women with a high BMI, women who've had previous c/s, or who have mild, diet controlled gestational diabetes, or mild hypertension, that are more of a grey area.

I am sure that absolutely no-one will argue that placenta praevia, or pre-eclampsia are not high risk. There are, however, categories where different CLUs have differing opinions. Surely some of these categories are open to debate? Witness the differing dates for being classed as being overdue, and being referred for induction. Or what level of higher than desired BMI puts you into a high risk category? Or even parity - is a woman having her 5th or more child always at high risk, or should it be looked at in the totality of her general health?

Ah right the thread title and OP just referred to "high risk women".

There's a huge variety in "high risk". Part of me wonders whether home birth "high risk" cases are AS high risk as hospital ones which would explain The results. Eg I have a blood condition so high risk - but mechanics of birthing were unaffected and indeed I had a fast birth. A friend of mine goes into labour very early every time so far: this involves more risks to her/baby and she always births in hospital. I wouldn't expect - due to the risk to the baby- someone to CHOOSE hb in the first place with her condition whereas for me it might well be easier/safer.
Which might prevent a like for like comparison of risk.

"Iamrandom" couldn't agree more with you.

I cannot understand any researcher without a very strong agenda summarising the data from this study in this way. It might be also worth noting that the NICE technical team rated the Birthplace study of low or very low quality - despite it costing over 10 million of largely public money.

To be clear, table 3 clearly shows that home birth has double the risks of planned OU birth. It only becomes a higher risk in the hospital group in Table 4 where they count 'more than 48 hours in NICU' as a 'dread' outcome like stillbirth, brain damage or fracture of the baby's skull or bones!

Yes, it does not reach statistical significance (where the CI is more than 1) but it reaches 0.98 and 0.97 so within a whisker!

Lies, damn lies and statistics springs to mind! There are agendas here.

Minifingers, I totally agree women should have a choice but that choice has to be based on accurate information and this isn't.

First time that I have heard of the NICE team rating the Birthplace study as being of low quality. I wonder why they based their latest recommendations on it then?

I've read the whole paper, and all it really says is that if you are high risk and give birth in hospital baby is more likely to be admitted. iam summarises it well. I find this bit the most worrying even if not statistically significant because the abstract seems to gloss over it.

When the measure of adverse perinatal outcome was restricted to include only ‘intrapartum related mortality and morbidity’, a measure that encompassed intrapartum stillbirth, early neonatal death and specific intrapartum related morbidities, the direction of effect was reversed, with a higher proportion of adverse outcomes in planned home births, but this apparent difference in risk compared with planned OU birth was not statistically significant and confidence intervals were wide and compatible with a range of effects. Because of the small sample size it was not possible to adjust for maternal characteristics other than parity.

Anyone who thinks this paper says home births are safer in high risk women really need to read the whole paper critically, the two cohorts have very, very different characteristics.

Totally agree, RumNoRaisins.

LaVolcan, you asked why NICE based their recommendations on the Birthplace study in the latest Intrapartum Care guideline if it is rated as very low quality.

Very good question and one that needs answering. But check for yourself - page 162 -170 in the full guideline. You will see it is rated of very low quality for nearly everything it measured despite its enormous cost.

www.nice.org.uk/guidance/cg190/evidence/cg190-intrapartum-care-full-guideline3

More stats - on what evidence has the last 40 years worth of recommendations that hospital is the most appropriate setting for birth been based?

More stats - can you direct me to a good quality study involving large numbers of subjects which suggests hospitals are the safest setting for birth for either high or low risk women?

Can you show me a study linking equal or better maternal outcomes?

If this evidence doesn't exist then why is hospital the default option for everyone?

What happened to 'first do no harm'? The medicalisation of birth has risk often obstructs the normal progress of labour. It's unethical for doctors to steer women down this route if there is no clear evidence of benefit. Medicalisation should never be the default mode.

"Nonetheless, the group felt that the Birthplace study was generally a good quality observational study, because it is a very large study, it did perform adjustments for confounders, it was conducted more recently and it was based in England."

So this equates to the Birthplace study of low or very low quality?

More stats - can you direct me to a good quality study involving large numbers of subjects which suggests hospitals are the safest setting for birth for either high or low risk women?

Indeed, minifingers - we have had 45 years where CLU births have been the default option for at least 90% of women. There should be no shortage of evidence to back this statement up.

'Minifingers' - I am not disagreeing with you. The recommendation of the 1971 Peel Report that all women should give birth in hospital was based on information that was as flaky as the Birthplace study.

What they do not understand is that there is no such thing as a 'best' place to give birth.

Only an individual woman can decide that. But women need unbiased evidence to make that decision.

If what is important to you is giving birth in a homely atmosphere with good support and minimum interventions, then probably an out of hospital birth is best.

If you are bothered by things like access to epidurals, not wanting to transfer and a very small increase in adverse outcomes, then probably a hospital birth is best.

There is a lot of research on the different risks for the different places of birth and women want to consider these along with the 'gut' feeling of what feels right for them.

What matters most is that women get accurate information otherwise they cannot make an informed choice.

More stats - what do you see as the primary problem with the birthplace stats? Incomplete reporting? Or is it just the inevitable lack of randomisation?

I also think it's insulting to suggest that the quality of the BP study is comparable to the paucity of the evidence in the Peel report.

"If what is important to you is giving birth in a homely atmosphere with good support and minimum interventions, then probably an out of hospital birth is best."

Except I would spell out that by 'minimum interventions' we mean less than a third as likely to have a cut to the perineum, about a half as likely to end up on an operating table, about half as likely to need a GA, less likely to end up in HDU, A LOT less likely to need a blood transfusion, a third as likely to have a forceps delivery, less likely to have an anal sphincter injury. Best be specific. Obviously this would be particularly important to someone wanting another child who should be made aware of the small increase in stillbirth in pregnancies following c/s.

"If you are bothered by things like access to epidurals, not wanting to transfer and a very small increase in adverse outcomes, then probably a hospital birth is best."

Sorry - you have said the evidence in the BP study is unreliable, yet you confidently speak of an increase in adverse outcomes? The abstract of this study in the BMJ concluded: "Women planning birth in a midwifery unit and multiparous women planning birth at home experience fewer interventions than those planning birth in an obstetric unit with no impact on perinatal outcomes".

Was also interested to note in the figures on the BMJ report of the study that there were double the percentage of vaginal breech births in the homebirth arm compared to the OU arm. I wonder what contribution this made to the adverse outcome figures?

minifingers, you have probably come across this organisation before, but if not, it's a very useful source of stats, info and support for normal birth:
www.aims.org.uk

Observational studies are low quality evidence except in exceptional circumstances, GRADE give reasonings for this.

The NICE intrapartum care guideline pages that more stats referenced earlier give reasoning for classifying each point of evidence they used from birthplace (the tables show most points being low grade).

Having said that, low grade evidence can still be very useful when it comes to writing guidance especially where it is probably the best that exists about a given subject and where it would be very difficult to carry out the sorts of trials needed for higher grade evidence. So in this example it would be pretty impossible to run a randomised controlled trial of high risk women giving birth at home versus in hospital. This is probably why NICE have used it to fairly extensively inform their guideline.

The paper referred to in this thread is another matter as secondary analysis of data throws up a host of additional quality issues and I do think that we need to separate thinking about the birthplace study in general from the data in this secondary analysis and the poor interpretation of it initially presented in this thread.

Secondary analysis, especially where outcomes are combined in order to achieve significance, is not good quality.

I would make the point again that women who are in a high risk group should not be led to believe that this data shows that it is safer for them to homebirth. They may find other information that leads them to homebirth but it would be wrong to base the decision on this data or this interpretation.

Minifingers The NICE technical team rated the Birthplace study of low or very low quality. The group that Lavolcan refers to who said the study was ok are not experts - the NICE technical team are. They are world class.

What was wrong with it? They used completely the wrong sample size. This was the main problem and for a study costing more than £10 million it is incomprehensible.

In the primary analysis they did not adjust for the obstetric group having a huge excess of women referred with problems at the onset of labour, they only partially adjusted for confounders - the obstetric group continued to have higher risk factors, they included data where less than 85% of forms were returned .......and on and on and on. Where they did adjust for both confounders and lost data, they found out of hospital birth to be consistently higher risk for first time mothers but dismissed that in their conclusions.

Finally, it was an observational study - and these are very weak for looking at safety.

Sorry, Iamrandom, our posts crossed. Agree on every point particularly the last paragraph in your post.

I want to correct my post though - I said the group LaVolcan referred to are not experts. Sorry - I meant they are not experts in epidemiology or statistics.

I have been worried by this thread but I can completely understand why the OP raised it.

I think the research community has to think a little harder about how if reports its findings in future.

I agree more stats that this is an issue that researchers and editors do need to think more carefully about.

Firstly about the data itself and how meaningful it is and secondly about how clearly it is reported. At the risk of sounding like a stuck record, well written lay summaries are vital. it just isn't reasonable to expect people to plough through methodology and data to understand what a paper means to their situation.

"they found out of hospital birth to be consistently higher risk for first time mothers but dismissed that in their conclusions."

Sorry - are you saying there was consistently higher risk for the FSM as well as for home birth?

My understanding of the study was that it was absolutely clear that the outcomes for mothers having their first baby at home was significantly poorer than for those mothers in other settings, but overall the number of poor outcomes was still low.

I don't call that 'dismissing', but you do.