Responding to Covid-19: Reshaping local care models

This is the fourth article in a series of articles on issues arising from, and responses to, Covid-19 in the UK and beyond. In this, I look at how local care models are being reshaped in the delivery of health services.

The cases presented here were discussed as part of data analytics ‘huddle’. The session was chaired by Harry Evans from NHS England and Improvement. Presenters included:

• Dr Dan Alton – GP, CCIO and Clinical Lead, Berkshire West CCG
• Steve Laitner – NAPC, GP and Public Health Specialist
• Mark England – Health Navigator
• Dr Farzana Hussain – GP, Newham
• Scott Durairaj – Co-Chair South East Covid-19 Disparity Advisory Panel

My interest in data comes from an understanding of its value – for individual and public good. Working alongside John Kellas, we’ve supported local health and care analysts in Bristol and developed an open knowledge base, Advancing Analytics and AI in health and care. Work is ongoing with a training event in July.

And at the moment, change is happening fast. For example, in health we’ve seen a rapid shift to virtual consultations. And there’s never been a greater need to link local authority data with health and social care records.

Population health management during Covid-19

Lockdown in the UK commenced on 24 March 2020.

In Wokingham, with a population of 160k, the most urgent challenge was putting in place mechanisms to support vulnerable populations. This cohort is made up of a wider group than those advised to ‘shield’.

The first challenge was how to identify these individuals. In Wokingham, teams brought together data from:

• the shielded list;
• those assisted in bin collection, sheltered housing, and vulnerable persons list from the local authority; and,
• Individual Patient Activity (IPA) to define further vulnerable person cohorts which could include factors such as hospital admissions.

This required pooling data from the NHS, the local authority, and the Third Sector. It was collected and shared under emergency powers with the appropriate permissions. It will be deleted going forward as per regulatory requirements.

The second challenge was to develop a system so that vulnerable individuals have one point of contact. In Wokingham, they put together a single team whose responsibility it is to work across and coordinate health, care, and voluntary sector responses. This is managed via weekly meetings.

The third challenge is to rapidly address issues as they emerge and iterate solutions. Important here is individual and group needs. For example, some new parents were struggling without the usual support structures in place including family and friends, and health visitors. In responding to this, a virtual group consultation was introduced.

Segmentation, streaming & new models of care(ing)

Another important factor in considered Population Health Management is to use a segmentation approach to develop approaches at both cohort and individual levels in how care is provided.  This is about understanding the needs to different populations based on factors including age and the complexity of needs. This is not a one size fits all approach.

The key in developing such an approach is driven by the need to provide wrap around models of care for individuals. This is important to address given the complexity of needs.

Predicting and preventing unplanned care

In Vale of York, 1% of the most vulnerable patients utilise 53% of hospital care. Many of these patients are transient over time. The challenge here is that simply looking at activity trends over time doesn’t provide the insight needed. Instead, there’s a need to identify individuals at the right time to put appropriate interventions in place.

Using an AI powered patient identification system, proactive health coaching has been introduced. Health coaching is used to empower patients and ensure that care is coordinated across the system. This has led to a marked decrease in health costs associated with many of these patients, across non-elective and elective care, and through a decrease in the need for A&E attendances. Most importantly, health coaching has lead to a reduction in mortality rates.

Using practice data in response to Covid-19

Dr Farzana Hussein, a GP in Newham, noticed as a result of Covid-19, child immunisation rates were dropping sharply. The practice list has about 5,000 people. On a ‘normal’ pre-lockdown week, there would be about 12 immunisations. Within two weeks, this had dropped to only 3.

Not only was it clear that parents had concerns about bringing babies into the surgery, but there was a workforce issues that needed to be addressed. Eighty percent of the surgery staff are Black, Asian, Minority Ethnic (BAME). Given the disproportionate impact of Covid-19 on BAME groups, it was important to reduce face to face time with patents.

Dr Hussein reviewed her appointment books and considered how to improve immunisation rates while also safeguarding the surgery’s staff. A staff member suggested a ‘drive-through’ clinic for immunisations.

Generally an immunisation consultation would last for 10 mins. They reduced the time to 2 mins with a phone conversation while the parents waited in the ‘car’ with the baby. Newham is not unique in terms of many people not owning cars. For the purposes of the ‘drive through’ buggies and prams were classed as vehicles, albeit there was garden in front of the practice that could be used for this purpose.

And they have made good progress with increasing immunisations to 8 per week. While not up to the level it was pre-Covid, and there is more to be done, the practice continues to learn and adapt.

Addressing requirements of the BAME workforce and communities

BAME communities are disproportionately impacted by Covid-19. They are affected in terms of the proportion of infections, the severity of the virus and the rate of deaths. And this is an urgent issue for health and care staff with for example BAME groups constituting about 70% of the care workforce in London.

The South East Covid-19 Disparity Advisory Panel developed a work package to look at different staff challenges specifically related to risk assessments for BAME staff. Data is critical for understanding the proportion of staff who may be at higher risk of infection and the complications that come from it, where they are working, how many may be off sick due to Covid-19, and how many people have been tested. It is also about understanding people’s unique circumstances – how people live, how many are in a household and so on. But this isn’t happening across the system.

And this also needs to be addressed at a community level. At present, the messaging around Covid-19 is blunt. Some local councillors and Members of Parliament are helping nuanced the messaging to fit community needs but there’s further urgent work to be done.

This is an urgent issue for health and care staff, requiring Boards to understand their workforce and address specific needs. This is not an equalities issue – this is about the health of the workforce and the measures that organisations must put in place to ensure workplaces are safe for all staff.

Conclusion

A theme that runs through these case studies is the importance of place based data, bringing together data sets from a range of organisations to address the current emergency at the local level. Systems and processes have been rapidly develop to make this happen. What would have taken months, is now taking days.

Going forward, it is essential that this improved coordination of systems, and data, continues. This is about supporting the most vulnerable, as well as the wider population.