In the first article in this series I talked about changing expectations. Medicines policy is moving much closer to the centre of the political story of the NHS. The speed of new technology development is creating a public expectation the increasingly constrained resources the NHS has available are incompatible

So if expectations are changing, the obvious next question is: how do we actually influence the system in that environment? How do we induce, incentivise, support or force the behaviours and effects we want? How do we ensure that we’re using the medicines we have available for the greatest population health benefit?

Spoiler alert - it’s going to be a big change.

In this article I’ll largely be talking about levers in the context of primary care prescribing, still largely in general practice. Secondary care is a different picture, and community pharmacy isn’t (NHS) prescribing in any great volume yet. As much as I’d love to be describing medicines use across a system, where sectors and settings are irrelevant, we’re just not there yet.

For many years medicines optimisation teams have relied on a combination of local relationships and commissioned services to affect prescribing in primary care. We’ve done ‘for’ the system. 

Formularies, policies, guidelines, audits etc. All aimed at supporting clinicians to make good decisions when prescribing.

And we’ve also directly provided, commissioned, or incentivised activity and behaviours. We’ve employed medicines optimisation and medicines management teams, we’ve contracted them from other providers, we’ve paid providers directly. All to ensure that the prescriptions being written were more likely to be what we’d want, and to achieve a balance between cost, effectiveness, safety, and quality. 

But in the main it’s been transactional, and it’s been retrofitting quality into a system. We’re stewards of a system that was designed in the mid 40s, and it shows.

When I was working into the Model ICB design process led by NHS England recently, I observed that there was a broad consensus from others that good medicines optimisation ‘pays for itself’. That certainly helped to safeguard our position in the new organisations, albeit to a variable extent.

But I felt uncomfortable with it. Because I know that that view comes from the fact that we have historically corrected problems that should have never occurred. 

• Data showing that a practice inhaler usage is not optimal? Send a team in.

• Opioid usage going in the wrong direction? Incentivise (i.e. pay for) improvement.

• Too many people using non evidence based medicines? Change the formulary.

I know that analogies with other industries are often oversimplified, but bear with me on this one.

I live a few miles away from one of the largest car manufacturing plants in Europe. For much of the early industrial era, factories relied on inspection at the end of the production line to catch defects. Teams of inspectors would stand at the end of the process, identifying faulty products and sending them back to be fixed or discarded. It worked, but it was inefficient and expensive.

After the Second World War, a group of quality thinkers led by W. Edwards Deming and Joseph Juran helped Japanese manufacturers realise something profound: quality cannot be inspected into a product after it has been made. It has to be designed into the process itself. Companies like Toyota embraced that idea, redesigning production systems so that problems were prevented upstream rather than corrected downstream. The result was a revolution in manufacturing quality and efficiency that eventually reshaped industries across the world.

We have done far too much observing, inspecting and correcting the problem.

Of course I know that clinical practice and prescribing are not predictable, uniform and consistent in the way a production line is. But my point is that sending a team into a GP practice to change things that have already happened is inefficient, ineffective and too late. 

That model kind of worked when medicines optimisation teams were larger and had more resource, the interface between care sectors was less tense, and the pace of new drug development was slower.

None of those things are true anymore.

We simply no longer have the luxury of inspecting and correcting quality at the end of the process.

If medicines policy is going to function in this environment, quality, safety and value have to be designed into the system much earlier, and it has to be done in a way that understands the predictable human behaviour which makes it work, or fail. And influencing the prescription before it’s written.

The right prescribing decision needs to be the easiest one.

That’s the only way we’ll design a system to work with less friction, less antagonism and more good stuff happening for our population.

The recently published neighbourhood health framework reinforces this direction of travel. It places multidisciplinary teams, prevention, and local service design at the centre of delivery. That is not a model where medicines optimisation sits at the end of the pathway correcting variation. It is a model where prescribing behaviour is shaped much earlier, through how services, roles and systems are designed.

At some point in time, ICBs will be abolished and replaced by something else - politicians can’t help but interfere, and every new health secretary thinks they can ‘fix’ the NHS. When that happens, I’d love us not to need to design practice facing medicines optimisation teams into that new world. 

Not because I don’t believe in what we do, but because we should be working further upstream, and designing the system so that it produces the right prescribing in the first place.

That is a very different kind of influence. It requires us to be involved earlier in decisions about service design, digital systems, contracting, commissioning and operational planning. It means being prominent and vocal in discussions that are not just obviously about “medicines”.

But how do we get from here to there? 

I think there are three things we need to grasp:

1. Market management

2. Carrot and stick

3. Stepping back rather than leaning in

1. Market Management

We’ve always tended to focus time and energy on general practice. That’s where almost all of the prescribing takes place, so it makes sense. Our most recent pre-ICB structure was that of clinical commissioning group, where GPs were constituent members, and so the organisational needle was always pointing to general practice.

But we’re moving into a different world. Successive governments, of both colours, have tried to break the GP partnership model and move towards a more multidisciplinary primary care environment. Will neighbourhoods become what PCNs failed to achieve? Will the new neighbourhood health framework fundamentally change relationships? Who knows, but whether they do or don’t the direction of travel is clear.

The direction of travel is now explicit. Neighbourhoods are being positioned as the organising unit of care, with prescribing distributed across multidisciplinary teams rather than concentrated in general practice.

We’ve got a rapidly growing private sector, largely arising from dissatisfaction with NHS providers - 13% of all antibiotic prescriptions are now from private providers. 

Community Pharmacy is going to become a legitimate, viable and competitive alternative for primary care prescribing. Those of us in charge of designing the systems need to ensure that we’re setting this up to succeed. It won’t be easy, but there are some fantastic examples of innovative practice emerging, and these need to be supported and allowed to flourish. The long term gains are enormous.

General practice may still be the glue that holds patient records and neighbourhood health together, but they won’t be the gatekeeper any more. It’s up to us as professional leaders for medicines to ensure that prescribing quality and safety is baked into whatever structures come next.

That fundamentally changes what “market management” means for us. It is no longer about influencing one dominant provider, it’s about shaping a mixed economy of prescribing across NHS, private and community-based services.

2. Carrot and stick

The second thing we need to remember is that incentives still matter.

Healthcare professionals, like everyone else, respond to the environment around them. Expectations, incentives, workload pressures and organisational priorities all shape behaviour. And if we really try, we can largely predict the effect they will have.

Historically medicines optimisation has leaned much more heavily on the carrot than the stick. We’ve incentivised improvement, funded additional activity, and provided teams to support better prescribing. In many ways that approach reflects the culture of the NHS; it runs on relationships, trust and discretionary effort.

But the environment we’re moving into will require us to be more deliberate about how we use both carrot and stick.

That doesn’t mean abandoning any sense of a collaborative approach. Relationships remain the most powerful lever we have. If prescribers feel that medicines teams are working with them rather than policing them, change happens faster and with far less friction.

But it does mean being clearer about where the boundaries sit. And the overlap between prescriber and contractor needs to be navigated, on both sides.

The challenge is doing that without damaging the relationships that make the system work.

Because the NHS still runs on discretionary effort. Policies can set expectations, and contracts can set minimum criteria. But they don’t deliver care. That still relies on clinicians choosing to do the right thing in a system that is often under intense pressure. And the system still relies on both general practice and community pharmacy going above and beyond the defined and detailed scope of their contracts.

So the art of the carrot and the stick is not choosing one or the other. It is knowing when to use each, and doing it in a way that preserves trust. Because if we get this wrong we will create avoidable antagonism. And antagonism is the last thing a system under this much pressure needs.

3. Stepping back rather than leaning in

The third thing we need to grasp is that sometimes the most effective intervention is not to lean further into the interface between organisations, but to step back from it entirely.

Medicines optimisation teams spend an extraordinary amount of time managing the friction between parts of the system.

• Primary and secondary care.

• General practice and community pharmacy.

• National and regional expectations and local delivery.

Shared care protocols, prescribing transfers, monitoring responsibilities, pathway ownership. These conversations consume huge amounts of time and energy. Trust me, it’s been a long week for me.

And often the instinct from us is to lean in harder. Fix the problem. Write more guidance. Be the honest broker in the room. 

It’s exhausting. And it’s exacerbated by pointless power struggles.

But usually the real problem is not the behaviour of the people at the interface. The problem is that the interface itself has been designed badly. And us stepping in just covers that up,  we become a lightening rod for everyone’s complaints, concerns and frustration. And there is of course the expectation that we can just open the great big ICB chequebook and sort the problem out with money.

In the environment we are moving into, we simply will not have the capacity or resource to keep doing that. Instead, medicines leaders need to get more comfortable stepping back and asking a more fundamental question:

Should this interface exist at all?

• Should prescribing move between sectors in the way it currently does?

• Should monitoring responsibilities sit where they currently sit?

• Should a pathway rely on multiple organisations negotiating handoffs every time a patient moves through it?

• Do the contracts and commissioning systems we’ve put in place make all this harder than it should be? 

• Are we pitting parts of the NHS against each other? And if so, have we made that decision consciously?

Sometimes the most effective form of system leadership is not smoothing the interface, it’s removing it.

Designing pathways where responsibility is clearer, transitions are fewer, and clinicians are not constantly negotiating operational boundaries while trying to care for patients.

We’ve created the most complex public health system in the world, and we keep layering more complexity and more interfaces on top - neighbourhoods, accountable healthcare organisations, ICBs, OPICs, regions. And every time there’s a new layer? A new contract. A new tension point. Something else to argue over.

The interface won’t be solved by a new contract, and certainly not by the vague concept of ‘neighbourhood’. It will be solved by ignoring the contract and getting the people in the room to design the new system. 

But whenever the ICB is in the room, everyone looks to us. Either to make the difficult decision, to decide whose side we are on, or to open our wallet. And none of that ever fixes the problem. 

What would happen if we were just not in the room at all?

It would probably be much harder in the short term. But it may well reduce friction and strengthen relationships enormously over time. And in a system where relationships are stretched and resources are tightening, reducing friction may be one of the most valuable contributions medicines leaders can make.

The neighbourhood health model might change where care happens, who delivers it, and how patients move through the system. But it won’t, on its own, solve the fundamental problem medicines optimisation has been grappling with for years.

We have built a system that relies on correcting decisions after they’ve already been made, and that is becoming unaffordable, both financially and operationally. But if we don’t actively design a new system, and take chances in doing so, we’ll see medicines use getting more expensive, more variable, increasing inequalities and less likely to deliver the population health outcomes we all want to see.