A newly published academic book chapter argues that planned home birth is ‘as safe’ for low-risk women as planned hospital birth.
The evidence we have published says otherwise.
Let me show you the difference between what some of the literature claims and what the data actually show.
What the Chapter Claims
Kirstie Coxon and Melissa Cheyney, writing in Risk and Uncertainty in Maternity Care (Springer, 2026), assert that research from the UK, USA, Netherlands, Norway, and Canada demonstrates that planned home birth is as safe as planned hospital birth for low-risk women.
They frame hospital birth as a source of ‘intervention-related iatrogenesis’ and position home birth as a rational choice to avoid overuse of lifesaving interventions.
They acknowledge that the evidence is ‘frequently contested’ but treat that contestation as ideological rather than scientific.
The framing is familiar. I have read versions of this argument for two decades. The problem is that the claim does not survive contact with the largest birth dataset in the United States.
What the CDC Natality Data Show
Using the CDC natality database -- which covers all US births, not a selected research cohort -- my colleagues and I have published findings that are difficult to reconcile with the ‘equally safe’ claim. Neonatal mortality rates for planned home births attended by certified professional midwives are substantially higher than for comparable low-risk hospital births. In our published analyses, the absolute risk differences, while small in proportional terms, represent hundreds of preventable infant deaths annually in the United States.
The studies that reach the ‘equally safe’ conclusion consistently share methodological features that inflate home birth safety: they study highly selected populations in countries with well-integrated midwifery systems (the Netherlands, the UK under certain conditions), they exclude unplanned home births from the numerator while including them in the denominator at various stages of analysis, and they use transfer rates as a proxy for outcomes rather than measuring the outcomes directly.
The United States is not the Netherlands.
We do not have a seamlessly integrated midwifery-to-hospital transfer system. We do not have uniformly credentialed home birth attendants.
When Coxon and Cheyney aggregate international evidence to make a claim that applies to American patients, they are applying a finding from a context that does not exist here.
The Framing Problem
The chapter frames the debate as one between women’s autonomy and medical overreach. I want to be precise about this, because the framing is doing significant work.
No one disputes that women have the right to choose where they give birth. They do.
What I dispute is the obligation of physicians and midwives to provide that choice without full, honest disclosure of the outcome data. Autonomy without informed consent is not autonomy -- it is abandonment. A woman told that home birth is ‘as safe’ as hospital birth based on a selective reading of the international literature has not been fully informed. She has been misled by selective citation.
There is a difference between respecting a patient’s decision after full disclosure and building an academic framework that makes the clinician’s job of full disclosure harder. This chapter does the latter.
The Rise of Freebirth
The chapter also discusses freebirth -- unattended childbirth without any medical or midwifery presence -- as a ‘response to systems failures.’ This framing deserves serious scrutiny.
Freebirth represents the highest-risk birth setting we know of. When women choose freebirth because they distrust the medical system, that distrust may be understandable, but the solution is not to validate unattended birth as a reasonable response to institutional failure. The solution is to fix the institutional failures while being honest that freebirth carries risks that no amount of empowerment language makes acceptable.
Presenting freebirth as a coherent political response to systems failure, without equally prominent discussion of its mortality data, is not academic neutrality. It is advocacy with a literature review attached.
What Good Evidence Communication Looks Like
I am not arguing that hospital birth is without risk.
It is not.
Unnecessary induction, unnecessary cesarean delivery, hospital-acquired infection -- these are real. The question is whether the risks of home birth in the US context are accurately represented when clinicians counsel patients.
Good evidence communication means presenting absolute risks, not just relative comparisons. It means specifying which country’s data applies to which clinical setting. It means acknowledging that a finding from a Scandinavian randomized cohort does not automatically transfer to a rural American home birth attended by a lay midwife.
None of that nuance is present in the Coxon and Cheyney chapter’s summary statement. ‘As safe as hospital birth’ is a headline, not a clinical conclusion.
Conclusion
I have been publishing on this question since before the CDC natality database was what it is today. The population-level data on neonatal mortality at planned home births in the United States have been consistent, peer-reviewed, and reproducible. They are not the product of obstetric ideology. They are birth certificates.
When an academic chapter funded by the critical studies literature selectively aggregates international evidence to argue that home birth is equally safe, and presents that conclusion to clinicians who will use it in counseling, it has a clinical consequence.
Babies are not research questions.
Women deserve clinicians who give them the full picture -- including the parts that do not fit the narrative.