The current Babylon GP at Hand model will not work for everyone

NHS England have made a commitment to make digital primary care available to everyone. But if the future of primary care is digital, then it will need to be a different model to that currently delivered through Babylon GP at hand. How can the approach be adapted to meet the needs of a broader population?

The recent independent evaluation of the impact of Babylon GP at hand (BGPaH) showed signs that BGPaH is providing good quality care to currently registered patients – generally younger and healthier people from more affluent areas. Notably, few of their patients had long-term conditions, and those who did have these conditions reported feeling confident managing them.

While the findings are positive, they were for a population that is very unrepresentative of those using general practice. There are 14.2 million people with 2+ long-term conditions in England, who account for over half of all GP visits nationally. Rates of visits are particularly high amongst those who are overwhelmed with managing their long-term conditions, which account for nearly one-quarter of the total.

The NHS faces a huge challenge to provide good care for patients with complex needs, and the question is still open as to whether, and how, technology can help. In this blog, we reflect on the evaluation and its potential implications.

Good patient experience

It's clearly excellent news that patients using Babylon GP at hand were positive about the quality of care they received. 93% said they had confidence and trust in their healthcare professional, and 87% thought that their healthcare professional was 'good' at treating them with care and concern, with 90% feeling listened to and 88% reporting being given enough time. A majority stated that the quality of care was better than that experienced at their previous practice.

Unfortunately, there is an absence of objective data about patient outcomes, which is a problem because patient experience does not correlate perfectly with clinical outcomes, but a high proportion of patients felt that their needs were mostly met, in their last appointment (90%).

A note of caution is needed because only 6.4% of individuals surveyed responded to it. While the authors of the evaluation say this was 'sufficient for analysis purposes', surely response bias is a significant concern.

No clear differences in use of healthcare services

One concern about Babylon GP at Hand is that it might lead to increases in demand for hospital care. The evaluation concludes that BGPaH patients were 'not using more secondary care resource after registration than similar patients'. Our own view is that the evaluation was inconclusive as to whether or not the model led to higher rates of secondary care use.

The chief limitation relates to the inability to control for fundamental differences between individuals who self-select for BGPaH and patients who remained at other practices, partly caused because of gaps in the primary care data, and partly by the inherent difficulty of measuring the factors that might have motivated patients to join BGPaH. As a result, it is not possible to interpret, or be certain about the extent to which differences in service use can be attributed to the BGPaH service, rather than to other factors.

The rate of deregistration, and reasons for this, requires further investigation – especially given that of those patients surveyed who had, or said they intended to, stopped using BGPaH, more than half (51%) cited unhappiness with quality of care.

Positive experience amongst workforce

One of the most striking and encouraging findings was that GPs working for BGPaH report satisfaction in terms of work-life balance and opportunities for support, development and monitoring. There were concerns that the working arrangements may limit 'the opportunity for young clinicians to develop their generalist skills', particularly given the unrepresentative nature of the registered patients.

Still, the evaluation raises useful learnings about the aspects of the model that that attract GPs to work for BGPaH, which will contribute to an ongoing debate surrounding workforce policy and planning supported by recent changes to the GP contract.

What about the wider applicability of the model?

While there are really positive findings from the evaluation, there are reasons to doubt that the model as it currently operates will work for many patients with complex needs:

• Continuity of care with the same GP has been shown to produce better health outcomes as well as lower rates of unplanned hospital admissions. It might be crucial for older people with multiple health problems. Following the evaluation, it remains unclear whether a model like BGPaH can provide continuity of care. Many of its patients actively chose access over continuity of care, and were happy with the choice they have made. Yet other patients might feel differently.
• Patients registered with BGPaH may have longer journeys for face-to-face appointments than other patients. The evaluation showed that most patients were aware of this fact when they registered, but on the whole had not considered what they would do if they were too unwell to travel for a face-to-face appointment.
• While internet penetration continues to rise, there are significant gaps, for example Ofcom found that 35% of those aged 65-74 are not using the internet, and the 2019 Consumer Digital Index states that 11.9 million people (22%) do not have essential digital skills. There are substantial portions of the population who could not use a digital-first offer.

Conclusions

We are not suggesting that the model for general practice needs to be the same for everyone. But if digital technologies are going to transform health care in the way that has been claimed, then they will need to work for a broader population than has so far participated in BGPaH.

While the evaluation provides useful evidence, it remains an open question whether technology can help with the most urgent challenges faced by the NHS and its patients.

This article was originally published in the HSJ on 5 June 2019.

Geraldine Clarke (@GeraldineCTHF) is a Senior Data Analyst, Josh Keith (@joshkeith) is a Senior Fellow, and Adam Steventon (@ASteventonTHF) is Director of Data Analytics at the Health Foundation.