According to a survey commissioned by Sage Growth Partners, over 50% of leaders are underwhelmed by population health management technology as they say their current population solution doesn’t meet their needs. This has highlighted a pressing need for effective analytics and data to be embedded in population health solutions, particularly amidst the coronavirus pandemic.

In this interview for Healthcare IT News (HITN), Professor Mahmood Adil talks about the current state of population health data and actionable ways to embed it into the healthcare system.  

This interview has been edited for length and clarity.

HITN: Can you start off by defining the term population health management to our readers?

Adil: If we use the NHS (UK) as an example, in 1948 most of the health challenges were acute challenges and it was episodic health so most of the time you get admitted into hospital either for surgical or acute medical causes. Fast forward to 2020, over 25% of our population have chronic diseases. That’s almost one fourth of the population. Population health management is understanding of population health needs through big data and providing evidence-based data-driven interventions to achieve the desired health outcomes for subsets of population with similar needs rather than just for the individuals. For example, managing chronic disease in the elderly population. With the resources we have got at hand, we have no option except to go for population health management and that means you need to use the data to do the risk profile of the population and then have a population-based intervention plan to improve outcomes.

HITN: Some studies have revealed that healthcare leaders want analytics and care coordination functionalities embedded in their population health management solutions. What are the actionable ways to improve this going forward and to ensure effective population data management?

Adil: Population health management has a few dimensions and so when we talk about actionable ways we need to understand those dimensions very clearly. One dimension is the need to think about the big data because it means that if we want to look at the population we need to understand what sort of data we can collect and analyse. The other dimension is the implementation of AI and machine learning to change it into applied intelligence. Over the years we have collected a lot of data and a large variety of it. If anyone wants to look at actionable ways to improve things, we need to look at what sort of capacity and capability we have on machine learning and AI to provide applied intelligence to professionals, policy-makers, patients and the public at large.

The other factor is big data. AI can give you the diagnoses on what sort of problem the population is having and what their healthcare needs are. It can’t give you the solution. The solution needs to come from the leaders and that solution must encompass some digital solutions.

Nowadays, mobile phones can be used as digital tools for intervention. We need to look very innovatively at how best we can use the other digital technologies in order to improve the outcomes of the population health analysis.

Last but not least, we need to work upstream by looking at prevention. Most of the time, if a patient has already become a patient, to me, it is too late. On the contrary, if you work on population health management we can identify which part of the population is high risk. For example, childhood obesity, as there are over 400 million obese children in the world. We can also look at how best to work with them in advance so that they do not develop certain diseases such as diabetes later in life, and if they do get it, they should have the tools to manage it properly and avoid any further complications as much as possible. 

HITN: How can data sharing evolve population health management?

Adil: There are three major groups of factors that can improve health outcomes. Firstly, having a good health service, like many other countries in Europe and America. However, health services can only play 25% in the role to improve health outcomes. People often think that if a country has the best hospitals then that country’s population health is better than the others, but that is not true. A good example is America. They have many of the best hospitals in the world but the country’s overall population health outcomes are not good.

The other big factor is lifestyle, meaning levels of smoking, drinking and exercise within a population. If we are unable to deal with lifestyle problems, even with the best of the best hospitals are unable to improve health outcomes. Broadly speaking, lifestyle makes up 30%, health services make up 25% and broader determinants of health make up 40% to improving health outcomes. These are factors we ignored in the past but we can’t ignore in the future.

Broader determinants are key such as income, employment, education, housing and crime. People often don’t see that they are directly related to health. An example is COVID-19. If you look at the outcomes of the pandemic, patients are admitted to the same hospitals, given the best care but the outcome of the patient from deprived areas have much higher levels of mortality. We need to take into consideration health services, lifestyle and broader determinants of health at the same time. This is the reason we should start collecting data on all three.

Another aspect is the secondary use of data and sharing data on lifestyle factors. It is important to encourage people to allow sharing their health data for research and innovation purposes.

HITN: In what ways is predictive analysis being used, particularly when looking at societal issues?

Adil: One good practical example is we have a predictive analytic tool in Scotland called the Scottish Patient at Risk of Readmission and Admission (SPARRA). The population of Scotland is 5.3 million and we are collecting a wide variety of health data about every person living in Scotland. SPARRA utilise that data to do predictive analytics on everyone in Scotland, which shows the chances of the person getting admitted into hospital next year. This predictive analytics information is provided to all the general practitioners, they look at your risk score, and then they contact high risk patients in advance so they can proactively manage them in the community.

HITN: Scotland is said to have some of the best health service data available anywhere in the world, with few other countries having information which combines high-quality data, consistency, and national coverage. When looking at other countries, what improvements do you think can be made? What are the lessons we can learn?

Adil: We need three types of data to improve population health management: lifestyle, social determinants and health services. In Scotland, we have data on all three and it is very comprehensive. The question we need to be asking is, is it important that we need data or applied health intelligence? Changing data into applied health intelligence is the key to success – and this is what we are good at in Scotland.

The first lesson other countries could learn is the ways to move from volume to value. If you want to create value you need to improve the health outcomes of the population. Health means physical, social and mental wellbeing of the population. If you want to achieve that goal you can’t just keep collecting more data but must change into applied health intelligence.

Many countries are data rich and intelligence poor. We collect a lot of data but are unable to change to value proposition and the way we do that is by being very clear of the problem we need to solve. Applied intelligence is dependent on what question or problem you’re trying to solve. This is a cycle you have to be very clear about. What are your health population management problems? What sort of data are you able to collect? How best can you use that data to apply intelligence? How you share it with the key stakeholder and policy-makers in order to improve the outcomes?

HITN: What kind of innovation and developments can we look forward to in this area? What are your hopes for the future?

Adil: I would say the future is bright. I always say that the time has come for us to move from stethoscope to data scope. We collect a lot of data, however, we are unable to use it to improve population health management. Data scope helps to give insight and diagnoses the issue and helps to improve outcome at population level. By using machine learning and AI we should be able to say how best a particular population can get managed so that their overall outcomes get improved.

Drug companies are developing personalised medicines that are very specific or relevant to individuals. By combining genomic, clinical, lifestyle data and broader determinants datasets we could create ‘precision population management’ leading to predictive, personalised and proactive prevention – that would be a great innovation. An element of such innovation is already happening in the UAE and other countries.

The other innovation that I feel is important is the development of data literacy among health professionals and doctors. You will find that in a world where everything is data, particularly in the health sector as shown during COVID-19, unfortunately, our workforce data literacy is very poor in terms of how they are able to access, understand and utilise data. We need to enhance our professionals’ understanding of data and health informatics so that when we give them the digital tools they should be able to use them effectively to improve individuals as well as population health outcomes.

Developing these and other new innovations is my mission. The HIMSS conference is going to be an excellent platform to share my ideas, motivate others and co-create digital solutions to improve population health management all across the globe.

Thank you for your time. More information about HIMSS & Health 2.0 European Digital Event (7-11 September 2020) can be found here.