Intended for healthcare professionals


The state of care in England’s maternity services

BMJ 2023; 383 doi: (Published 20 November 2023) Cite this as: BMJ 2023;383:p2700
  1. Mary Dixon-Woods, director1,
  2. Zenab Barry, maternity advocate2,
  3. James McGowan, clinical research associate1,
  4. Graham Martin, director of research1
  1. 1THIS Institute, Cambridge, UK
  2. 2Public representative, Cambridge, UK
  1. Correspondence to: M Dixon-Woods director{at}

Time for a fresh approach to regulation and improvement

The latest Care Quality Commission (CQC) report on the state of care in England is far from an encouraging read.1 Although the healthcare system is under serious strain, maternity services are among the areas identified as especially challenged. The problems identified in maternity care, while shocking, come as no surprise. The sector is seeing repeated high profile organisational failures and soaring clinical negligence claims, together with grim evidence of ongoing variation in outcomes, culture, and workforce challenges and inequities linked to socioeconomic status and ethnicity.234

The cybernetic model of regulation offers a potentially useful way of understanding where effort needs to be directed.5 In this model, regulation comprises three interlinked elements: standard setting, monitoring, and mechanisms to secure improvement; absence of (or flaws in) any of the three elements makes failure more likely. Standard setting and monitoring, though imperfect, are the stronger elements of England’s current health and social care regulatory system. But the third component—making improvement—is weaker.

The CQC has wide ranging powers of enforcement under civil or criminal law where needed, but responsibility for improving challenged organisations lies elsewhere, and evidence for the effectiveness of the current improvement mechanisms remains limited. Of the 41 NHS hospitals under the “special measures and challenged providers” regime6 between 2013 and 2018, only six were rated good by mid-2018, with six re-entering the regime.7

Unproved improvement interventions

Maternity services enter NHS England’s maternity safety support programme if the CQC rates them as “requires improvement” or “inadequate” in the “well led” or “safe” inspection domains. Typically, the support includes improvement advisers providing mentoring, advice, guidance, and promoting quality improvement methods. However, no evaluation of the programme’s effectiveness has been published. Some trusts have been in the programme for several years, and some trusts that exited the programme have re-entered.8 An earlier observational study using routine data reported little evidence of improvement in lower performing maternity units on inspection ratings.9

Of course, the maternity support programme is not the only improvement effort targeting maternity services. Recent years have seen no shortage of recommendations and initiatives seeking to stimulate, incentivise, advise, or support services to get better. The large volume of initiatives and recommendations, and the range of bodies issuing them, is itself a problem. “Priority thickets”10 may result, where units become overwhelmed, distracted, and confused, paralysing their ability to respond and often lacking the resource and expertise to convert recommendations into effective local solutions.

The volume problem is compounded by common weaknesses in existing improvement efforts.11 Many are not evidence based, do not include maternity service users or staff in their design, do not align with human factors and ergonomics strategies, lack support for implementation, and are uncosted. Perhaps most damningly, few are formally evaluated, so the evidence base for supporting improvement remains seriously underdeveloped.12

A further threat to the effective functioning of the cybernetic model of regulation is that a poor CQC rating—even when a fair judgment—can make things worse rather than better. Affected healthcare staff may experience emotional distress, including feeling ashamed, demoralised, and stigmatised,13 which may act as a barrier to improvement. And staff recruitment and retention, which are essential to improvement, may be more difficult when an organisation is reputationally damaged. In this sense, one element of the regulatory system (monitoring) is potentially destabilising another (improvement).

Despite the scale of current improvement efforts, it remains unclear whether they are tackling the right problems in the right way at the right level. Some issues in maternity services are deeply structural, including workforce shortages, poor quality estates and facilities, and inadequate technological infrastructure. These require specific action backed by policy support and investment. But a disturbingly recurrent factor blamed for problems in quality and safety of care is culture. Simply exhorting colleagues to behave better towards each other and those they care for has had limited effect. More evidence based strategies for improving culture, including professionalism, civility, respectful communication, and anti-racist behaviours are much needed.14

Improving maternity care is a priority for families who use maternity services and staff who work in them. An evaluation of the CQC’s maternity inspection programme is under way, but a hard look at how “improvement” is currently being done and where efforts can add most value is now overdue. It will require rationalising and streamlining of current approaches, meaningful co-design in partnership with families and staff, and building an evidence base through rigorous evaluation.


  • Competing interests: The BMJ has judged that there are no disqualifying financial ties to commercial companies. The authors declare the following other interests: GM and MD-W are conducting an evaluation of the national maternity inspection programme, commissioned by the Care Quality Commission. Further details of The BMJ policy on financial interests are here:

  • Provenance and peer review: Commissioned; not externally peer reviewed.