HFMA 2018: population health management can deliver rapid results

Caz Sayer (pictured), GP and chair of London GP federation Haverstock Healthcare, said the suggestion that these changes took 10 years to deliver was ‘rubbish’. ‘you can achieve [a good] level of impact in a couple of years. But I cannot emphasise enough that to achieve value you need to take two things into account,’ she said.

‘You cannot do this without valued partnerships. This is about everyone getting behind this approach and being signed up to it. And you can’t achieve it unless you have shared values. So you have to tap into the things that matter to the people who are delivering and receiving services, which is the improvement in outcomes and then the opportunity to reinvest where you have achieved those savings.’

Dr Sayer, former chair of Camden Clinical Commissioning Group, described the CCG’s approach to population health management, which has led to some dramatic improvements in service performance. Starting in 2011/12, the commissioner segmented its population using a simplified version of the Bridges to Health model. Six segments were identified from those members of the population who had never accessed secondary care, through patients with a long-term conditions (LTCs) but who were stable and up to the most complex patients with LTCs and a high potential to use secondary care.

The core to population health management was understanding that ‘achieving best value is different for each segment’, Dr Sayer told the workshop.  So, for healthy members of the population, the focus needs to be on keeping them healthy and, if they get ill, getting them back to normal function as soon as possible.

For people with an LTC, the health system should be identifying these conditions early and putting support mechanisms in place to ensure their LTCs impact as little as possible. And at the most complex end of the segmentation, the focus should be on maintaining quality of life or managing end of life in the most appropriate way.

‘The response of the system has to be different depending on the segment,’ she said. Further analysis of the data in Camden also revealed that deprivation had a greater impact as complexity increased. ‘Having one size fits all to address different groups makes no sense, yet that is what health systems do,’ she added.

In Camden, a different approach involved getting patients from different segments to identify outcomes that were important to them and then to involve them in co-designing services.

Changes within services for the frail elderly involving greater use of multi-disciplinary teams led to the achievement of patient-defined outcomes. For example, the multidisciplinary approach managed to keep 72% of patients spending the same or more time at home (compared with the previous period) despite a ‘normal’ expectation that patients would deteriorate. There was also an 18% reduction in emergency bed days, as well as corresponding savings.

Dr Sayer also highlighted major successes with the diabetes service and services aimed at improving the well-being of 16-24 year olds in the borough. In the latter case, a Minding the Gap service provided holistic support (with a hub and outreach components) that aimed to provide help that went well beyond healthcare. An independent evaluation suggested the service delivered a return of £3.40 for every £1 invested, although the savings were spread across multiple areas including health, employment, the criminal justice system and benefits.

She argued that segmentation offered the right way forward, but said that a tool on its own was not enough – health systems also needed to act on the data. ‘You won’t achieve change unless you can change clinical and professional behaviours,’ she said. ‘But using these tools can help to get them thinking.’

Members of the HFMA Healthcare Costing for Value Institute can read more about the Camden approach in a case study A population approach to value-based healthcare