This site is intended for health professionals only

Explained: Putting PHM theory into practice to improve patient care

Explained: Putting PHM theory into practice to improve patient care
By Paul Birch Associate director of applied analytics, NHS Arden & GEM CSU
30 April 2020



This is the third and final part of our series on population health management. If you haven’t already, catch up on the firstand second instalments.

Population Health Management (PHM) allows us to segment or group patients and categorise them by risk factors. The current coronavirus pandemic has prompted perhaps the first example of national segmentation in action, with the drive to identify those most at risk from Covid-19. This required special effort, however, as data sharing between healthcare sectors is still fairly new. The initial challenges have provided a very timely reminder of why safe and effective data sharing is so important.

Even without the pressure of a pandemic, the NHS is used to having to maximise its resources, which is why a better understanding of our patients’ needs is so important. This isn’t just about those that use health services – but also those that don’t.

Traditionally, commissioners have focused a lot of attention on meeting the health needs of those using secondary care, but are we doing enough to support the health needs of those not so evidently ‘in the system’? And what about those who are heavy users of primary care alone?

Increasingly, we are seeing Clinical Commissioning Groups (CCGs) and Primary Care Networks (PCNs) looking to PHM to help them understand how patients are using primary care and to predict and prevent health problems that have the biggest impact on resources.

Using the combined segmentation and stratification tools, GPs are able to understand who is attending their practice, for what reason and whether their potential outcomes are being achieved. PCNs will be able to use that data to plan appropriate interventions across their patch, with the aim of delivering more personalised care for their patients, and a reduction in overall demand on services.

Together, Herts Valleys CCG and their PCNs are using PHM to understand patient flow through primary care. NHS Arden & GEM CSU is working with them to evaluate the impact of their extended access service, to assess whether they are seeing the right people in the right locations and whether the provision is successfully reducing A&E attendances so that the services can be developed accordingly.

PHM is helping PCNs to plan ahead, too. We’re seeing an increase in demand for data that will help practices build a case for new roles such as social prescribers and determine how best to use their time. With the role’s focus on social as well as healthcare factors, integrated data sources are vital to identify patients who can benefit most from this support, such as those struggling with isolation, bereavement, mental ill health or complex social needs. This is an area where a PHM approach is a natural and necessary fit.

PHM is also enabling a more holistic approach to care, with the inclusion, where possible, of social care, mental health and housing data. We are still in the early days of combining different data sets, but the growing awareness of the impact of these factors on physical wellbeing is undoubtedly playing a role in the Government’s keenness to maintain some access to fresh air and exercise, despite the pandemic lockdown.

Where to start?

For many, the first hurdle in making greater use of PHM approaches is knowing where to start, but the PHM cycle can begin in different ways. If your starting point is an idea for a new intervention, data analysis can help you identify which patients would most likely benefit. For example, in Norfolk, PHM has been used to specifically identify patients with multi comorbidities who would be most likely to benefit from more intensive community nursing.

PHM can also be approached from the opposite direction, where you’ve identified a group of patients that are using a lot of resources. Risk stratification can help you identify suitable interventions which could improve patient health and/or reduce service demand.

We have been working with CCGs covering Coventry, Rugby and North Warwickshire where they have one of the highest rates of access to specialist child and adolescent mental health services (CAMHS) in England. We have used PHM analytics to enable commissioners to develop a better understanding of service needs and triggers and improve provision as a result.

Sometimes, it is the initial analysis of data that throws up something interesting, which leads to more detailed work and ultimately a change in how you provide a particular service. But there can be a reluctance to get started, either due to concerns about data sharing or a sense that the data isn’t strong enough.

In our experience, it’s more important to make a start than to wait for the perfect combination of data. We have been working with NHS England and NHS Improvement on a National Performance and Population Health Dashboard, which will soon be available to all health systems in England. While this is just a starting point, national data alone can help your system begin the thinking around populations – as your understanding develops and your data set improves with your own local and primary care data, you can build on your initial approach.

PHM is a constant cycle of progress and review. It can only be done by taking action. If you haven’t already, then consider making PHM one of your priorities, as soon as pandemic pressures allow.

Want news like this straight to your inbox?

Related articles