Non-adherence to five common conditions, asthma, type 2 diabetes, high cholesterol and coronary heart disease, hypertension, and schizophrenia, is estimated to cost the NHS over £930 million a year, with around £500 million of that judged recoverable through better adherence. That is before the separate £300 million lost annually to wasted medicines. The case this post makes is that closing that gap depends on medication adherence outbound calls delivered through a specialist operating model, rather than added to the workload of teams who are already at capacity.
If you run a community pharmacy group, a pharma patient services function, or a PCN pharmacist team delivering structured medication reviews, you already know the adherence problem exists. You do not need it explained. What is less commonly mapped is why the follow-up that addresses it is structurally hard to deliver at scale inside the settings currently responsible for it, and why that points toward a different operating model rather than more effort from the same people.
There is a specific structural reason the follow-up gets dropped, and it is worth being precise about it before getting to the buyers.
The adherence problem is well understood. The follow-up is the part that fails.
Start with the scale, because it frames everything else. Between 30 and 50% of medicines prescribed for long-term conditions are not taken as recommended. The World Health Organization’s long-standing estimate is that around half of patients on long-term medication globally do not take it as intended. This is not a fringe behaviour. It is closer to the default, and it is most acute in the period immediately after a new medicine is started.
The clinical and economic consequences follow directly. When a patient does not take a medicine as prescribed, the condition is undertreated, which generates downstream cost in further appointments, additional tests, escalated therapy, and sometimes admission. A prescriber who assumes adherence may increase the dose or add a second agent when the first appears not to work, compounding both cost and risk. The £930 million figure is the visible tip of that. The fuller cost, in worsened outcomes and avoidable contact across the system, is larger and harder to count.
None of this is news to the people who manage medicines. The reason the gap persists is not a lack of awareness. It is that the intervention which closes it, structured, timely, two-way follow-up in the weeks after a medicine is started, is operationally fragile in the settings where it currently sits. A community pharmacist running a busy dispensary cannot reliably make a structured follow-up call to every patient who started a new medicine two weeks ago, because the dispensary does not pause to let them. A PCN pharmacist running structured medication reviews has finite clinical sessions, and proactive follow-up competes with everything else in those sessions. The follow-up is not skipped because anyone undervalues it. It is skipped because it is the work with no protected time around it.
That is the structural point. Adherence follow-up is high-value, time-sensitive, and the first thing to fall off when capacity is tight. Which is exactly the profile of work that benefits from a dedicated operating model rather than being squeezed into one already running at its limit.
What the New Medicine Service got right, and where it strains
The New Medicine Service is the clearest existing proof that structured adherence follow-up works. It is built on a sound clinical model. After a patient is prescribed a new medicine for a defined long-term condition, the NMS provides a structured intervention consultation roughly one to two weeks later, followed by a further follow-up consultation a week or two after that. The three-stage shape, engagement at the point of prescription, intervention around the first fortnight, and follow-up across the first month, maps onto the window when most patients either embed a new medicine into their routine or quietly stop taking it.
The evidence backs the design. The pragmatic randomised controlled trial of the NMS found it increased the proportion of patients adhering to their new medicine by around 10 percentage points compared with normal practice, with the odds of adherence improved by a factor of roughly 1.67. The longer economic modelling suggested an 89% probability of cost-effectiveness at the standard £20,000 per quality-adjusted life year threshold. The structure is right. The timing is right. The clinical logic is right.
Where it strains is delivery. The NMS depends on the community pharmacy team finding the time to make and hold those consultations alongside dispensing, and the real-world completion of the follow-up stage is where the model is most brittle. The intervention is sound, but the operational base it runs on is the same dispensary that is already absorbing prescription volume, supply pressure, and walk-in demand. The clinical model is not the bottleneck. The capacity to execute it consistently is.
This is the gap a specialist operating model is built to fill. Not to replace the pharmacist’s clinical judgment, which is the part that has to stay human, but to carry the structured, repeatable, high-volume contact that surrounds it. Before getting to how that works, it is worth being clear about who has the budget and the motive to buy it, because that differs sharply across the market.
Three buyers, three different reasons to act
Outbound adherence follow-up is not a single market. It is three distinct buyers with different economics, different funding, and different conversion paths. The commercial case lands differently for each.
Community pharmacy groups
For a multi-site community pharmacy group, adherence follow-up is both a clinical obligation and a service line. The NMS and related advanced services are funded, which means follow-up that is completed reliably is revenue captured rather than revenue left on the table. The operational problem at group scale is consistency. Across dozens or hundreds of branches, completion of structured follow-up varies with how busy each dispensary is on any given week. A specialist outbound layer standardises that contact across the estate, so the follow-up gets made wherever the patient is, regardless of which branch is under pressure that day. The team in the branch is freed to handle the consultations that need to be face to face and the clinical conversations that genuinely require a pharmacist in the room.
Pharma patient services and patient support programmes
This is the buyer with budget already allocated. Pharma patient services teams fund branded adherence and patient support programmes precisely because non-adherence erodes the real-world effectiveness, and the commercial value, of a medicine. A patient who stops taking a product in week three does not refill, and the clinical outcome data that supports the product weakens. Structured outbound follow-up, delivered through a model that is NHS-appropriate and clinically governed, supports persistence on therapy in exactly the window where most early discontinuation happens. Because the budget exists and the commercial logic is direct, the path from a conversation to a programme is shorter here than anywhere else in the market. This is the buyer for whom the operating model is not a cost to justify but a capability to deploy against an existing line.
PCN pharmacist leads running structured medication reviews at scale
PCN pharmacists running structured medication reviews are addressing adherence at the level of patients on multiple medicines, where the stakes and the complexity are highest. The SMR itself is clinical work that has to stay with the pharmacist. What surrounds it, identifying patients due for review, making contact, confirming attendance, and following up afterward, is high-volume coordination that consumes the clinical sessions it should be protecting. An outbound layer that handles the contact around the review lets the pharmacist spend their limited sessions on the review itself. The model supports the SMR programme rather than substituting for any part of the clinical judgment inside it.
Across all three, the structure is the same. The clinical conversation stays human. The structured, repeatable contact that wraps around it, and that is currently being dropped for lack of protected time, moves to a dedicated operating model that can run it at volume.
What the specialist operating model actually looks like
The reason this work belongs in a specialist model rather than bolted onto existing teams comes down to three properties of the contact itself.
First, it is high volume and repetitive. Every patient starting a relevant new medicine generates the same sequence of contacts at the same intervals. That is precisely the profile that overwhelms a team doing it by hand between other tasks, and precisely the profile a structured outbound system handles well.
Second, it is time-sensitive in a way that does not tolerate delay. The intervention window for a new medicine is the first few weeks. A follow-up call made late, or not at all, has missed the moment where it changes the outcome. A model with dedicated capacity makes the call inside the window every time, rather than when the dispensary or the clinic happens to have a gap.
Third, it requires a real conversation, not a notification. This is where the adherence problem mirrors the appointment-reminder problem. A text reminding a patient to take a medicine confirms a message was sent. It does not surface the patient who has stopped because of a side effect, the patient who never understood the instructions, or the patient whose first language is not English and who needs the explanation in a form a templated message cannot give. A two-way voice conversation does all of that, and it escalates the patient who needs a pharmacist to a pharmacist, with the reason already captured.
There is a cohort dimension to this that matters for adherence specifically. The patients at highest risk of stopping a new medicine early are often the same patients least well served by digital reminders: older patients, patients managing several conditions at once, and patients whose first language is not English. Around 11 million people in the UK lack basic digital skills or do not use digital technology at all, and that group skews older and in poorer health. A voice conversation reaches them in a way an app notification or a text does not, and it can flex to confirm understanding, repeat instructions, or answer a question in the moment. For an adherence programme accountable for outcomes across a whole population rather than only its most digitally confident patients, that reach is not a nice-to-have. It is the difference between a programme that closes the gap and one that widens it.
That is the operating model: structured outbound voice contact at the right intervals, holding a real conversation, capturing what it finds, and routing the patients who need clinical input to the clinician with full context attached. The clinician’s time is spent on the patients who need it. The routine contact that was never going to get made consistently by hand gets made.
Jackie is built for exactly this kind of structured outbound work. It runs the contact at volume, holds the conversation, records the outcome, and escalates to the pharmacist or the patient services team when a call needs a person. The point is not to remove the human from adherence support. It is that the human element, the clinical conversation, only happens reliably if the high-volume contact around it is carried by a model with the capacity to carry it. You can see how the outbound side runs in more detail on the Jackie outbound page, and how the same structured-conversation approach performs on inbound volume in the Park Street deployment.
Anchoring to the standards: CG76, NG5, and medicines optimisation
The clinical case sits on established guidance, which matters for any buyer who has to defend the model internally. NICE CG76 on medicines adherence frames non-adherence as a problem of supporting patients to make informed decisions and take medicines effectively, not a problem of patient compliance to be enforced. That framing is important. The outbound model is a support mechanism, an extra point of contact where a patient can raise a concern, ask a question, or flag a side effect, not a surveillance tool. NICE NG5 on medicines optimisation places structured, patient-centred review at the heart of safe and effective prescribing, which is precisely the activity the operating model is designed to scale.
Building the model on these standards is what makes it NHS-appropriate rather than a generic call-centre overlay. The contact is governed, structured, and oriented to supporting the patient’s own decision-making, in line with how the guidance defines good practice. For a pharma patient services team or a pharmacy group, that governance is not a detail. It is the difference between a programme that survives clinical and information-governance scrutiny and one that does not.
Frequently asked questions
How much does medication non-adherence cost the NHS?
Non-adherence across five common conditions, asthma, type 2 diabetes, high cholesterol and coronary heart disease, hypertension, and schizophrenia, is estimated at over £930 million a year, with around £500 million considered recoverable through improved adherence. Separately, wasted medicines cost the NHS roughly £300 million annually. The wider cost in worsened outcomes and additional contact is larger still.
Does the New Medicine Service actually improve adherence?
The randomised controlled trial of the NMS found it increased adherence to a new medicine by around 10 percentage points compared with normal practice, with strong evidence of cost-effectiveness. The clinical model is sound. The challenge is delivering the structured follow-up consistently inside busy community pharmacy settings, which is where a dedicated outbound operating model adds capacity.
Why use outbound calls rather than text reminders for adherence?
A text confirms a message was sent. It does not surface the patient who has stopped because of a side effect, never understood the instructions, or needs an explanation in another language. A two-way voice conversation does, and it routes the patient who needs a pharmacist to a pharmacist with the reason captured. Voice contact also reaches older and digitally excluded patients more reliably than text.
Who buys outbound adherence follow-up?
Three buyer types. Community pharmacy groups, where follow-up is funded and consistency across branches is the operational challenge. Pharma patient services teams, who already fund branded adherence programmes and have allocated budget. And PCN pharmacist leads running structured medication reviews at scale, who need the coordination around the review handled so clinical sessions are spent on the review itself.
Does an outbound model replace the pharmacist?
No. The clinical conversation stays with the pharmacist. The model carries the high-volume, time-sensitive, repetitive contact that wraps around the clinical work and that is currently dropped for lack of protected time. The pharmacist’s sessions are spent on the patients who need clinical judgment, with the routine contact handled in parallel.
See how the outbound model runs
The adherence gap is well understood, the clinical model that closes it is proven, and the part that fails, consistent structured follow-up at volume, is exactly what a specialist outbound operating model is built to carry. If you run a pharmacy group, a patient services function, or a PCN pharmacist team, a 20-minute demo shows how the follow-up calls run, how escalation to your clinicians works, and how the model maps onto NMS, SMR, or a branded adherence programme. The four-week pilot runs on your existing setup so you can see the completion rates on your own patient cohort before committing to anything.