Large-scale can help general practice weather the storm

In the run up to launching our new publication Is bigger better? Lessons for large-scale general practice, our communications team warned us that journalists would home in on our mixed findings about the impact of larger scale on quality in general practice and overlook much of the rest.

And they were spot on. This is disappointing, because the report highlights other important findings about the potential impact of larger scale general practice. Specifically, these organisations have the potential to improve the sustainability of a fragile but important sector of the NHS. They can also deliver high quality extended services, popular with patients, but do not yet have much experience of doing this at large-scale.

And our findings about quality of care came with significant caveats.

Firstly, in order to compare different large-scale organisations to each other and to traditional practices, we had to choose from nationally available quality measures that were used universally, rather than analyse data that was tailored to evaluating specific quality improvement initiatives in each case study site. Thus, the data told only part of the story about quality of care.

We also noted that the 15 quality indicators we selected (in collaboration with large-scale organisations in the Nuffield Trust's Learning Network) related to four areas of care over which GPs had variable levels of control and influence. For example, prescribing is in the control of GPs and indicators in this domain improved for most measures in most sites – in absolute terms and relative to national performance. Quality Outcomes Framework measures are also largely under practitioner control and the case study sites improved scores compared to national average in five out of six domains. However, other indicators we looked at are less reliable as measures of GP quality of care. For example, many other factors beyond GP advice (such as proximity to the A&E department; advice from relatives, by-standers calling an ambulance and more), affect A&E attendance and emergency admissions. 

The three case study organisations for which we analysed 15 different quality indicators all operated in inner cities, where patients are well recognised to report lower quality scores than in other communities. They have taken over a steady stream of poorly performing practices, often small and disorganised with poor quality care. And despite this they improved in some measures.

Incorporating new practices into the data set was methodologically challenging and it is possible that improvements in new entrants were masked by the larger number of patients in the wider organisation whose care was not changing much. However, interviewees in the case studies described the systems and processes and sheer hard work that went into ‘turning around’ an under-performing practice, with regular input from senior clinicians and managers and significant resources invested in staff training and development. So it is nonetheless surprising that this did not result in measurable improvements across the broad range of indicators that we used.

Case study sites were at the forefront of implementing national policy on access, and each had introduced new ways to improve access to services. Yet these weren’t reflected in patient satisfaction scores in our research. This could be explained by the fact that, while access is an important dimension of quality, its relationship with reported patient satisfaction is complicated. Indeed, research by Paddison and others suggests that access is only weakly associated with overall patient satisfaction (which is most strongly correlated with patients’ experience of doctor communication).

From a patient’s perspective, these caveats will offer little comfort. Many of those we interviewed particularly valued the continuity of relationship with a usual GP and with the wider practice staff and were worried that this will disappear in the future. Whilst we did not interview anybody for whom continuity had been disrupted against their will (although some had chosen to take advantage of extended access in a different clinic and were pleased to have done so) this remains a significant concern for patients. We concluded that an important future challenge for large-scale organisations will be to preserve the continuity of care and ‘expert generalism’ of traditional practice. And they will need to do this while capturing economies of scale, improving access through centralising and standardising organisational systems and using technology to support the organisation and delivery of care.

In our recommendations, we suggest that leaders of emerging organisations should identify specific and measurable quality improvement goals that are consistent with local commissioning priorities. Assessing the impact of such initiatives would require tailored local data collection, but if they can be shown to improve outcomes, then we argue that this could help to demonstrate the value of larger scale in clinical areas of local importance, and create a rationale for future investment in their work.

In the mean time, large-scale general practice organisations are likely to face some difficult dilemmas. With NHS financial pressures rising and ongoing problems with GP recruitment and retention (which we concluded could be alleviated by large-scale organisations) sustaining quality will be tough; and improving it even tougher. Given the pressures GPs are facing, the potential to sustain the sector alone may be reason enough to collaborate, even if it does not demonstrably improve quality. 

We hope that the findings of this report will encourage these organisations to closely monitor the impact of service changes on patient satisfaction and clinical outcomes. In the meantime, we are clear that scaling up general practice should not be dismissed because of the mixed findings on quality. Traditional practices will need all the help they can get to weather ongoing financial storms. Larger scale can provide some of the help.