Population Health in Primary Care: Stop Treating It Like a Buzzword

For years, “population health management” has been thrown around as if saying it enough times will magically transform care. PHM strategies, dashboards and shiny tools are now a standard feature of national policy documents, ICS plans and conference agendas. Yet the reality on the ground, particularly in primary care, is far less polished.

If we’re going to be honest, the NHS has developed a habit of talking about population health as if it’s a methodology we already know how to use well. The truth is more uncomfortable: most systems still don’t have the analytic capability, Primary Care representation, or cross-sector trust required to do PHM properly. And until we fix that, the gap between ambition and delivery will keep widening.

Let’s start with the obvious: Primary Care analytic capability is nowhere near where it needs to be

General Practice holds the richest, most granular, most up-to-date view of the population. It’s the dataset everyone wants to use but rarely understands. Paradoxically, GP data is the least invested-in segment of the analytics workforce.

Most Primary Care “analytics” functions are actually bits of operational reporting, QOF trackers, or templates dressed up as insight. In many systems, the people analysing primary care data have never worked a day in General Practice. They’re often brilliant analysts but they come from acute, public health or CSU backgrounds. They don’t understand the realities of access, coding culture, appointment book behaviours, variation in digital maturity, or the nuance of how data actually gets created in practices.

The result? A huge amount of noise, misinterpretation and distrust.

I’ve lost count of the number of times a GP has pushed back on a system data pack, not because the practice is being defensive, but because the analysis just doesn’t reflect how care is actually delivered.

We cannot meaningfully do PHM if the people analysing the data don’t understand the environment it comes from.

Data sharing isn’t “blocked by GPs” - it’s blocked by poor relationships and system misunderstanding

Another uncomfortable truth: system partners often put GP data into the “too hard” box far too quickly.

I’ve seen areas try to deliver Population Health Management using only secondary care data because it’s simply easier to get. Easier to extract. Easier to interpret. Easier to analyse with the staff they currently have.

But here’s the problem:

If you’re only using hospital data, you’re not doing PHM.

You’re doing activity analysis, not population insight.

Primary care data isn’t hard to access because GPs are obstructive. It’s hard because:

• The legal landscape is genuinely complex
• Decision-making sits with hundreds of independent contractors
• Trust in the wider system is fragile
• The historic ask has often been extractive, not collaborative
• Many systems still use outdated technical approaches that simply don’t work in 2025

When GP data sharing works, it’s because relationships and purpose have been built, not because someone wrote another DPIA.

PHM is not a buzzword - it’s a three-step discipline

Strip it back and PHM is simple:

1. Understand your population – segmentation, risk stratification, unmet need
2. Put in an intervention – at scale or for a small cohort
3. Evidence the impact – did it work, for whom, and why?

I’ve seen PHM done brilliantly. In one area, we targeted a tiny cohort of high-risk patients with a personalised intervention that fundamentally reduced avoidable escalation. The cohort was small but the impact was big, because it was meaningful and intentional.

I’ve also seen the other extreme: millions invested in a PHM tool with no capability, no capacity and no workforce to act on the insights. A dashboard is not a PHM strategy. Having access to data is not the same as being able to do something with it.

The best PHM happens when you stop doing it to primary care and start doing it with primary care

We must stop treating PHM as something done by public health to the system or imposed by an analytics team that sits miles away from frontline reality.

Some of the most successful PHM work I’ve led has involved setting up local system intelligence functions with representation from:

• General Practice
• Local Authorities
• Public Health
• ICBs
• CSU teams
• Acute and Community Trusts
• Mental Health providers
• Hospices

When these groups come together around a shared purpose, the tone changes completely. Insight becomes co-owned. Data becomes a shared asset, not a political tool. People start to advocate for each other inside their own organisations. That’s when PHM stops being theoretical and starts becoming system behaviour.

The NHS doesn’t need more tools, it needs more trust.

We’re getting health inequalities wrong

The system’s current approach to health inequalities is heavily skewed towards patients we already know about: people who are registered, coded, visible in the dataset.

But the real inequalities sit with those who don’t appear in our data at all, because they don’t access care. They don’t book appointments. They don’t respond to recall. They don’t hit our “high-risk” lists because they were never on our radar in the first place.

PHM dashboards can identify variation, but they can’t find the people who don’t show up in the data. That requires:

• Community partnerships
• Outreach models
• Non-traditional data sources
• Lived experience insight
• Proper engagement with wider Primary Care

If our PHM work only focuses on known patients, we will continue to miss the people who need us most.

We must stop treating Community Pharmacy, Dentistry and Optometry as optional extras

The wider primary care sector has been an afterthought in PHM conversations for far too long.

Community Pharmacy, Dentistry and Optometry see thousands of people who never touch general practice. They hold data we rarely use. They spot problems nobody else picks up.

But the issues are familiar:

• No routine access to their data
• Inconsistent quality
• Lack of integration
• Limited system confidence in how to use it
• Tools not designed for their workflows

If we are serious about early identification, prevention and real PHM, these sectors must become fully embedded partners, not satellites.

We need to build a Primary Care analytics workforce that actually comes from Primary Care

One of the biggest missed opportunities in the NHS is the absence of analysts based in Primary Care, understanding Primary Care, and trusted by Primary Care.

Every ICS should have Primary Care analysts embedded in practices and PCNs just as confidently as they have analysts embedded in acute trusts. Without that, we will continue to misinterpret GP data and mis-design interventions.

This is not a skills problem. It’s a strategic choice problem. We’ve chosen not to invest in Primary Care analytical capability. The result is predictable: PHM becomes secondary care led by default.

If we’re serious about integrated care, we need analysts who sit across organisational boundaries and build professional relationships with peers in trusts, local authorities, mental health, community services and hospices. Without those relationships, PHM stays siloed and superficial.

So what needs to change?

A few practical, honest steps:

1. Stop treating GP data as a technical extraction problem
   a. It’s a relationship and trust problem. Solve the right problem.
2. Build Primary Care analytics as a distinct profession
   b. Recruit analysts who understand General Practice or train them intensively with GP exposure.
3. Co-design PHM with Primary Care, not around it
   c. If GPs aren’t involved in shaping segmentation, risk models and interventions, you’re building an academic exercise.
4. Put equal weight on wider Primary Care
   d. Community Pharmacy, Dentistry and Optometry must be part of the PHM data ecosystem.
5. Focus on people who don’t access care at all
   e. Otherwise, we’re measuring comfort rather than inequity.
6. Develop local system intelligence functions
   f. Bring partners into one shared space. Insight is a team sport.
   
   

Final thought

The NHS can keep talking about PHM as if it’s something we’ve already mastered. Or we can be honest: we’ve built the tools, but not the capability.

The real opportunity now is to stop chasing dashboards and start building relationships, skills and insight where they actually matter, close to patients, inside primary care, and across the full system.

That’s where PHM becomes something more than a buzzword. That’s where it becomes the way we deliver care.

If you’re interested in exploring these themes further, Conor will be joining us at the HETT Leaders’ Summit for a fireside chat on Population Health, Health Inequalities and Prevention Strategies. The session will delve into practical lessons from across the system, how real-time data and population segmentation can support earlier intervention, and what it takes to truly connect primary, community and secondary care. We’ll also look at where digital tools, virtual care models and prevention strategies are making a measurable difference and where opportunities are still being missed.

Join us at HETT Leaders' Summit, taking place on 12th February at Royal Armouries, Leeds.