About 20 years ago, I attended a lecture given by Andrew Stevens, a rather formidable and austere Professor of Public Health at the University of Birmingham. He was setting out his epidemiologically based approach to health care needs assessment. He and his colleagues, James Raftery, Jonathan Mant, and Sue Simpson edited a series of books on the topic that sit on the shelf behind me1. I refer to them on a regular basis. They remain both grounding and inspirational reference material.
At the heart of his lecture and books series is a deceptively simple and timeless conceptual framework that I’ve relied on ever since, and most recently in our analysis of exploring GP practice consultation rates, the gap between need and supply of consultations, and GP practice productivity.
Health service commissioning or planning had always felt like a messy, rudderless discipline to me. There were problems, and there were numbers, but no coherent way to link them. This conceptual framework seemed to offer some purchase.
The framework starts by defining three terms: need, demand, and supply, with respect to healthcare services. These must be three of the most commonly used words in healthcare planning circles, but we tend to use them loosely. For this framework at least, tighter definitions are required.
Here, ‘need’, means something quite particular, and can only be understood with respect to a specific service or intervention. A patient ‘needs’ service X or Y, if they have the potential to benefit from that service. The presence or absence of that potential is a function of the patient’s characteristics and the evidence base.
Demand is some request or outward expression of a desire for a service. It may be logged within a service’s record system, but such a record is not essential for demand to exist. It can come directly from the patient, or from someone acting on their behalf. In some cases, such as screening, immunisation, or chronic disease management, the demand may originate from the service itself.
Supply is the easiest of the three terms to define. It is the delivery of a service or intervention to a patient.
Having staked out the territory, the framework offers two key insights.
Insight 1: Need, demand, and supply, can be thought of as overlapping domains.
They can coexist within a patient’s experience, but they can also be present independently and in pairs. This is usually represented as a Venn diagram of three overlapping circles. In the central area, where need, demand, and supply occur together, a patient needs (i.e. has the ability to benefit from) a specific service or intervention, that service is requested and received. All good.
But in all of the other domains, some loss occurs. It might be that resources are wasted, or that an opportunity to improve a patient’s health is missed.
Need and demand but no supply
In these circumstances, a patient needs a consultation and attempts to make contact with their practice to arrange one, but either fails to get through, or isn’t offered one (at all, or at a time that is not sufficiently prompt or convenient). It is these circumstances that give rise to the greatest public concern and underpins much of the political debate. At worst, these circumstances lead to patient harm, when an opportunity for clinical intervention is missed. However, not all need is equal. Some might argue that GP practices should not offer a consultation to those whose needs fall below a certain threshold. Indeed, this is often described as a legitimate and clinically grounded form of rationing. But other negative consequences can follow when a patient is not offered an appointment that they feel they need. They may choose to attend their local Accident and Emergency Department instead. This is less appropriate for the patient, more costly for the NHS, and more likely to lead to a risk-averse clinical response.
Demand and supply but no need
In some cases, a patient might request and receive a consultation where no need is present. In these circumstances both the patient and the practice have failed to identify the lack of need and both the patient’s and practice’s time is wasted. Where the supply of consultations is highly constrained, then this might also lead to another patient missing the opportunity to receive a consultation. Patients are usually not medically qualified, and so should not be expected to know with confidence when their symptoms represent a need for a consultation. Making information available to patients in accessible formats and improvements in health literacy might help but will never eliminate this problem. And practices often have to make judgements about which patients to offer consultations to based on limited information, often collected by non- clinical staff. Furthermore, practices often report that patients consult when the underlying problem is not medical in nature (e.g., debt, loneliness, poor housing). Demand of this type is likely to increase when social and economic conditions deteriorate and when public services are cut back.
Need and supply but no demand
In these circumstances, a patient needs a GP consultation and receives one, but did not initiate the consultation by contacting the practice. Rather, the GP practice has arranged the appointment on the patient’s behalf. Case-finding patients in need and proactively managing their condition (particularly long-term conditions) is a comparatively new phenomenon but with the advent of the Quality and Outcomes Framework it has become a substantial component of service delivery. The losses here are minimal and relate to the costs of case- finding and proactive management. DNA’s are also likely to be slightly higher when appointments are practice rather than patient initiated. Where the practice intervention relates to some lifestyle or addiction problem, then the costs of acquisition may increase further.
Need but neither demand nor supply
Need can exist by itself, without either demand or supply, but the consequences of these circumstances are likely to similar to those when demand is present. Missed opportunities to address health needs, and presentations to other more accessible but less appropriate settings. Patients may learn not to request (demand) appointments, when previous attempts have been unsuccessful, or indeed when they hear from their families and friends or via the media that securing an appointment is likely to be challenging.
Demand but neither need nor supply
Demand can exist without need or supply. A patient might contact a practice to request a consultation, but the practice deems that no need is presents and so does not make an appointment available. The practices response to the patient may take the form of reassurance, or perhaps signposting to an alternative and more appropriate services. If managed well, the losses here are minimal. If not then the patient may experience unnecessary anxiety and in some cases escalate their demand to a more costly service.
Supply but neither need nor demand
In this scenario, a patient receives a consultation, but neither requests it nor needs it. GPs and others have raised concerns that the financial incentives associated with the Quality and Outcomes Framework, may lead to a tick-box exercise, where practice staff complete tasks (e.g. a blood pressure or cholesterol measurement) when in certain circumstances, this intervention has little clinical value.
Need, demand, and supply
The intersection between need, demand, and supply represents the ideal. A patient who needs (has the potential to benefit from) a consultation, might request (demand) and receive one (supply). But need, demand, and supply can occur independently of each other. When they do, some loss, failure or waste occurs. Most strategies seek to address the challenges of GP services by increasing the overlap between need, demand, and supply.
When a patient notices a lump in their breast, but can’t get through on the phone to book an appointment, then need and demand exist, but without supply. Unnecessary antibiotic prescribing indicates demand and supply, but in the absence of need.
Insight 2: The objective of commissioning is to increase the overlap or coincidence of the three domains.
Or, to put it another way, to reduce the instances where need, demand or supply occur independently or only in pairs. The framework offers a rubric to navigate the messy world of health service planning. If you don’t know what to do, then pick one of these outer segments, think how it might occur in the service you are interested in and what actions could be taken to shrink it. These actions should make up the core of your commissioning plan for the service.
Like all conceptual frameworks, it has its limits. It doesn’t for example incorporate the idea of cost-effectiveness. Just because a service might improve someone’s health, does not mean it is value-for-money or affordable (these ideas are covered elsewhere in the book series).
In our analysis, we used the framework to help us explore the mechanisms by which health policies seek to influence need, demand, and supply of GP practice services (see diagram below).
I commend the framework to anyone in a planning or commissioning role. It’s stood the test of time, and I’m convinced it has a role to play in helping us work through some of the seemingly intractable problems facing primary medical services. You might also be interested in Mary Dixon-Woods’ Candidacy framework and efforts to tailor this to GP services via the IMPRESS project.
Thank you, Andrew Stevens, James Raftery, Jonathan Mant, and Sue Simpson.
1 Heath care needs assessment. The epidemiologically based approach to needs assessment reviews. Edited by Stevens A, Raftery J, Mant J, and Simpson S.
