What happens when racism is named in quality improvement?

Equity is central to how we define quality of care. However, it can often remain in the background of routine improvement work.

We often set aims, track measures, and test changes, but we don’t always pause to consider whether our improvement efforts are narrowing inequities across health and care services.

Across health and care services in the UK, disparities persist in access, experience, and outcomes for Black and racially minoritised communities. These patterns are not isolated anomalies, but reflect structural conditions, longstanding disadvantage and the ways racism operates within and beyond healthcare systems.

The Model for Improvement by Associates in Process Improvement (API) is widely used because it brings focus and discipline. Its three key framework questions should be familiar to all those working in healthcare improvement:

• What are we trying to accomplish?
• How will we know a change is an improvement?
• What change can we make that will result in improvement?

The simplicity is a strength, but the questions themselves do not prompt us to consider equity, and more specifically, racism.

In practice, aims are often broad, data are presented in aggregate, and change ideas tend to emerge from existing structures. When racism is not named, inequity can be mistaken for routine variation, allowing overall improvements to mask persistent gaps. This reflects not a limitation of the model, but of how we apply it.

Many organisations pledge and make commitments to equity, but the challenge is translating those commitments into everyday improvement work.

In 2024, the NHS Race and Health Observatory (RHO) published its 7 Anti-Racism Principles to support organisations to move from intention to action. The principles call for naming racism explicitly, strengthening race and ethnicity data, identifying bias and meaningfully involving racially minoritised communities.

Through the RHO Maternity and Neonatal Learning and Action Network, we worked with the Institute for Healthcare Improvement and ten NHS perinatal teams across England to drive improvement activities to address racial and ethnic inequities in maternity and neonatal care. Our collective learning has resulted in The Anti-Racism Model for Improvement (MFI-AR).
 

Improvement begins with clarity of purpose, but many aims still focus on overall progress without openly specifying who should benefit. 

To embed equity meaningfully, aim statements must name the racial or ethnic groups most disadvantaged and commit to reducing unjust gaps. This means acknowledging how racism has shaped access, resources and trust, and designing interventions that respond to those legacies.

Equity focused improvement starts by naming racism, examining bias in pathways and decision making, and involving those with lived experience. Clear, specific aims help align stakeholders and create accountability. Without this lens, improvement efforts risk reinforcing inequities and directing resources away from those communities most affected.

Measurement shapes where attention and resources are directed. Aggregate reporting can conceal disparities and give a misleading sense of progress. Historically, data from across the UK show people from ethnic minority backgrounds are poorly represented in healthcare, with granular data often very limited.

Disaggregating data by race and ethnicity helps expose inequities that would otherwise remain hidden and highlights the structural drivers behind them. Applying a race critical lens ensures indicators are meaningful, progress is visible to affected communities, and potential unintended harms are monitored.

Through the MFI-AR, we encourage consistent use of disaggregated data, scrutiny of incomplete ethnicity recording and integration of qualitative insight from racially minoritised communities. Progress should be judged not only by overall gains, but by whether unjust gaps narrow.

Meaningful involvement of racially minoritised staff, patients and communities can reshape how problems are framed and which interventions are chosen. This helps build trust, shift power and ensure improvements reflect lived reality. It requires removing barriers to participation and enabling shared ownership from the outset, as well as sharing learning widely to build an anti-racism culture of improvement.

Embedding anti racism in improvement therefore requires attention to power and leadership, as well as process.

Across the RHO Maternity and Neonatal Learning and Action Network, teams have begun applying the Anti-Racism Model for Improvement to redesign services with equity at the centre. Their experiences show what becomes possible when racism is named explicitly, data are used critically, and Black and racially minoritised communities are meaningfully involved from the outset. 

At East London Foundation Trust (ELFT), this meant addressing inequities in perinatal mental health access for Black African and Black Caribbean women, who were underrepresented in community services and overrepresented in inpatient care. Guided by local data, the team partnered with family hubs, strengthened cross sector learning, and embedded lived experience through service user placements and involvement in the local Perinatal Mental Health Equity Board.

Lancashire Teaching Hospitals focused on reducing disproportionately high postpartum haemorrhage (PPH) rates among Black and ethnic minority women and birthing people. A three-stage data review informed targeted interventions, including early antenatal education, multilingual resources, improved interpreter access, culturally tailored advice, and more accurate blood loss measurement. The initiative led to a sustained 3% reduction in PPH and systemwide changes such as an ethnicity-focused data dashboard, showing how embedding equity into clinical practice can drive measurable improvement. 

Embedding anti-racism into quality improvement doesn’t require a new framework or duplication; it requires using existing ones with intention: setting aims that are explicit about whose outcomes must improve, using disaggregated data to expose gaps, and working with communities authentically as partners from the start, rather than inviting them in once decisions have already been made. This requires us to examine how power, privilege, and bias shape our systems and our improvement practice, and to stop treating that work as peripheral.

Quality improvement exists to close the gap between what is and what should be, and racial inequity remains one of the widest and most persistent gaps in our system. An anti racism lens belongs at the centre of our improvement work because it helps us focus on both the historical and current conditions that shape outcomes, and not just the outcomes themselves.

As members of the Q community, I invite you to place an anti-racism lens at the forefront of all your improvement work. Approach each step with intention, stay open to frank discussions on what the evidence reveals, and be willing to adjust course as you learn. 

When we approach this work together with honesty, accountability and a shared commitment to equity, we will begin a shift to reshape the legacy of racism and challenge conditions that still allow inequities to take root and flourish.

I’m looking forward to exploring this topic further with many of you at the upcoming Q member event on 25 March. I’ll share more about the model and there’ll be a chance for you to discuss how you might be able to apply it in your own work.