A&E divergence schemes, what does the evidence tell us

If only people wouldn’t go to A&E and we had better alternatives. What does the evidence actually tell us
Nb the below is contentious

Happy to argue it out with you?
Investment in GP will prevent A&E attendances. This hypothesis is back in the headline with the Whittaker study in PLOS.

this is the peer review paper from the greater Manchester CHLARC project. It’s a great paper.

The Basu editorial sums it up nicely. Basically good paper, there’s almost certainly an effect but mind the absolute impact in population terms and be careful about assuming a net save.
Key section here – 

“Nevertheless, Whittaker and colleagues observed a 26.4% reduction in patient-initiated referrals to emergency departments for “minor” problems among patients with increased access to primary care, as compared to their counterparts in the control group.

Headlines and Subtler Lessons

For health services researchers, the most informative results may be found in the authors’ careful process evaluation. Providing enhanced primary care access was costly and produced a heavy workload. An average of 35 additional hours of appointments per week were made available per practice in the treatment group, resulting in ~33,000 additional primary care appointments booked at primary care practices and US$4.8 million in costs. By contrast, about 11,000 emergency department visits were averted, which would have cost ~US$1.1 million. Hence, expanding access to primary care did not result in a cost savings. The intervention may still be cost-effective, which requires longer-term data on health outcomes.

In an era in which provider satisfaction has been added to the oft-cited “triple aim” of improving health, improving quality of care, and reducing costs, a concern posed by the results is whether narrowly focusing on emergency department utilization has superseded the quest to increase access to primary care in a manner that can be well-maintained into the future despite a heavy workload and time investment for primary care teams. While expanding access to primary care has become a major focus of many international efforts (such as the effort to achieve “patient centered medical homes” in the US or broader universal primary care access internationally), expanding access by including evening or weekend hours comes at a price, requiring additional information on the long-term impact of these expanded hours. In some regions, frequent users or “super-users” of emergency departments require far more services than simply providing evening or weekend hours to avert unnecessary emergency department utilization.”

It’s worth saying this point was in the original GM CHLARC paper but was buried on p453 of the report, and funnily enough didn’t make it into the newspaper and GP mags headlines… Wonder why

Both the Nuffield and King’s Fund refute GP availability, GP contract, 111 and minor ailments having much to do with A&E pressures. There’s a briefing in parliament library that says much the same. 

A&E stats released over the last 10 years or so, state type 1 ED attendance ‘remains stable’ – the increase, as usual, is in walk-in a&e types.

So here’s where I think I am evidence wise

1. Hospitals are mostly pretty efficient places (happy to be argued down there). Hospitals are most often a lowest cost option. Attendances are not very expensive, so hard to imagine it could ever be less costly. I’m fairly sceptical, on the basis of the evidence at least, about out-of-hospital strategies for problems that have a genuine need and currently go to A&E.

2. More GP are a good thing (see the Deep End paper) for all sorts of reasons. Arguably the only thing we should be investing in the NHS is GP (and wider out of hospital care etc). 

3. More GPs doing to divert people with minor ailments away from ED, but at a cost that isn’t worth it (this Whittaker  study)

4. Pharmacy minor ailment schemes are going to do same. The eval is far less robust, on face of it seem like a good thing, but perhaps some signs (esp from states, see the health affairs Ashwood paper) that they stimulate net demand (the folk who would otherwise have toughed it out at home) and are net cost not net save

5. 111 etc seems complete waste of money (maybe that’s just my prejudice and I don’t have the evidence off hand – happy to tAke the challenge there.

6. GP in front of house ED also has mixed at v v best evaluation and may not be net save, certainly won’t divert people from ED but may lead to admit from ed reduction. So- evidence please. My recollection is that the now famous Purdy evidence review concluded GP in ED might be effective but unlikely to be cost effective. 

7 Small number of practices offering extended hours? Ditto – love to see the evidence. I’m not overwhelmed with it. I’ve heard underwhelming things on the basis of the first wave evals wrt to a&e.

8 Closing all 3rd lanes on the motorway eg WIC? Again, evidence please?
So all up there isn’t much that seems effective at ed diversion, maybe we should stop trying to disrupt the path of least resistance flow wise with continuation of non evidence based schemes and things where there’s evidence of ineffectiveness  and focus in stuff we do have evidence they work?

Where to go with it all

Obviously it’s difficult, we’ve been told GPs must have longer/7 day services – therefore it will happen regardless of evidence. Maybe it’s a question of what is least worst/damage limitation option?

Don’t get me wrong, I’m not arguing against investment in General Practice -far from it, see my blog in the Deep End where is the NHS going wrong blog. I’m saying we need to be guided by what the evidence does tell us when we make investments and expect certain things to happen.
But everyone wants to do it. It didn’t work first time here – so we are going to do it again … sigh
Any thoughts on that, happy for challenge!
Thanks to the few folk that commented and added thoughts. I’ve not named them so as to protect the innocent.


Whittaker Associations between Extending Access to Primary Care and Emergency Department Visits: A Difference-In-Differences Analysis 

Where is NHS going wrong. Arguments FOR investment in GP.

Basu. Reduced Emergency Department Utilization after Increased Access to Primary Care



Quality Watch. Focus on: A&E attendances. Why are patients waiting longer? 

What’s going on in A&E? The key questions answered | The King’s Fund

Retail Clinic Visits For Low-Acuity Conditions Increase Utilization And Spending






  1. What evidence of embedding community nurses in A and e to reduce admissions for elderly. We been doing this for one month not 24/7 but 34 patients away in 31 days back to community. Our nurse rapid response service. Lots soft benefits directing people to intermediate care etc


  2. Thanks Greg, very informative and thought provoking.

    I think the path of least resistance is a very good point. It got me thinking about the patient pathway to A&E for those where treatment could be provided outside of the hospital. I can see 2 broad groups:
    1. The here and now: go straight to ED and don’t pass Go (could be 73.5% of attendees (Cowling 2014 paper found 26.5% of attendances followed unsuccessful attempts to access a GP))
    2. The failed attempts: go to the ED if no GP appointment is available today or within the next week

    The small/negligible impact of some policies may be because there’s an unrealised assumption that things like extended hours will change behaviours of group 1. Changing the distribution of these groups is likely to be extremely difficult and require the resources to ensure patients aren’t let down and revert back to their old ways. We would need to refer back to the evidence of supply-side approaches to see how this reversion is avoided – which leads us back at square 1 and your main point, so my thoughts here are unhelpful…

    Anyway, thanks again for the comments, very much appreciated.

    Cowling TE, Harris MJ, Watt HC, et al. Access to general practice and visits to accident and emergency (ED) departments in England: cross-sectional analysis of a national patient survey. Br J Gen Pract 2014; 64(624):e434–9 doi: 10.3399/bjgp14X680533


    • One of the points that folk like John Appelby (and others) consistently make is that minors have grown fastest and majors basically grown with population growth.

      Given that demand side interventions don’t seem awful successful, maybe we should just close down supply. roemers law?
      I’ll be stood right behind you in the Overview and Scrutiny a Committee when you are explaining that one!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s