Attempting from day one to work in a completely different way with an entire patient population is fraught with danger. Therefore, it is essential to identify defined groups to target for particular attention.
A range of predictive models that combine healthcare, social care, prescribing and other data are increasingly used but this technology needs to be complemented with changes in professional practice. For example, getting professionals to better share their knowledge about individual patients.
Across the developed world the same ratios seem to emerge from segmentation of different patient groups. Typically, over 40 percent of the combined health and social care cost in a system is driven by the top 5 percent of the population, with the top 0.5 percent often using as much as 10 percent of the total. It is tempting to focus on applying population stratification approaches to these very highest users; the numbers are small and the resources are significant.
However, such people are typically very ill. Their particular conditions, co-morbidities or the stage of their disease may make the cost of their care unavoidably high. Rather than attempting stratified accountable care approaches for the most complex patients first, being able to identify people most at risk of being in this group in the future is often a more fruitful approach.
Having the analytics and methodology to identify these people early and act quickly is a key step. In developing a deep understanding of the needs of the patient group in question and how these will change over time, is significant to all that follows.
The types of analytics required to do this will have to identify the diagnostics that drive cost and risk as well as match the specific skill sets of healthcare professionals that will be needed to deliver coordinated care for this particular patient group.
When it comes to targeting important patient groups, the growing problem in every jurisdiction is not as simple as the number of patients with chronic conditions, it is the number of people with multiple co-morbidities. For example, in Scotland a mere 14 percent of those that have diabetes, have that condition on its own. This means that the analysis necessary to underpin truly coordinated care needs to dive into the relationships between the different conditions and give insights into the interventions that will have an impact on each of the conditions.
Most chronic diseases, for example, improving the patient’s nutrition and exercise will have a big impact on outcome and will do so across, rather than within, conditions. Identifying shared characteristics of high-cost patients is not just a matter of clinical indicators.
You may define your population by:
Whichever approach you decide is best, the important thing is to make a start, even if you begin small and build up the capabilities over time.