[NB this was submitted to the BMJ as a possible blog in March 2021. They rejected it as not topical enough. But the relevance of the topic has increased a great deal in April and May as GPs have seen record workloads. So I thought i'd post it here to to remind me what I was thinking then before I write about the latest developments in the debate.]
There is a lively debate on how GPs should work once the pandemic is over. Many have adopted what NHSE calls "total triage" where patient needs are assessed before booking any appointments and as many requests as possible are handled remotely (online or by phone).
Should they stick with this approach or go back to seeing most patients face to face?
Examples of the debate are this lecture by Kath Checkland titled Who needs to see a doctor face-to-face? Or this Guardian article where Martin Marshall claims most patients prefer face to face (f-to-f) appointments.
The trigger for the debate appears to be the major drive by NHSE to get GPs to adopt non f-to-f ways to handle patient requests during the pandemic.
But the discussion of the value of f-to-f appointments risks distracting GPs from the reality they have faced and will face again. Sure, there is something special about f-to-f appointments–GPs can pick up body language cues and building a relationship with patients is easier–but this does not imply that every request needs f-to-f or that the tradeoffs achieving high-levels of f-to-f appointments are worth making.
The pre-pandemic reality for most GPs was that the majority of patients did not get a same day response to their requests and the average wait for an f-to-f was more than a week. Is that time tradeoff really worth it for the benefits of a person to person meeting? Worse, the reality of most practices was that access didn't depend on patient need but on their ability to get through a phone lottery at 8am (my GP is still like that: getting a call back on the same day is possible but required me to make 30 attempts over a 1 hour period to get through on the phone at all).
The world where the default response to every request was a same day face to face meeting has not existed for decades.
The reality is that GPs need to balance the supposed benefits of f-to-f against the tradeoffs. And not just the tradeoffs they desire, but also the ones their patients want and need.
This is where we have to test the attributes of different online triage tools.
The couple of dozen tools available are supposed to make getting and triaging patient requests easier for the patient and the GP by moving as much as possible online. The major benefit of this is to make the internal flow of work inside the practice much more efficient (but many of the available tools are quite late to that party). This frees capacity, enables the ability to triage based on need, allows more flexible responses based on need and, as a result, improves the speed of response.
Many object that moving patients online disadvantages the old and less tech literate. But this is only true if being online is compulsory. Good practices don't close down incoming phone calls or walk-ins. Good online tools allow the practice to process phone requests internally using the same tool that facilitates those requests arriving online (again, though, many do not, creating a very different experience for those using the phone).
Others object to the need for online systems to classify the patient request into simple boxes. Patients, as Checkland pointed out, often have vague amorphous requests rather than simple classifiable problems. But, again, this is a straw man that only applies to those systems that insist on asking 20 questions to try to reach a diagnosis before the GP sees the request. Not all systems work like that: some allow a simple free text request to be submitted which a human can assess and triage before deciding how to respond.
What good online triage systems offer is an easier way for patients to get their request in front of the GP. And they enable the GP to offer a faster, more flexible range of responses (including bringing the patient in to be assessed in person). If anything online tools make getting f-to-f meetings easier for the patients who need them than the traditional phone lottery.
But they also reveal some truths the fetish for f-to-f obscures. First the majority of patients don't even want f-to-f appointments if a faster useful response is the alternative. Second, patients like being offered a choice of response mode. Third, GPs can be far more effective if they adopt flexible responses as many of these take less time than f-to-f and free up time for the most needy.
The widespread nostalgia for f-to-f obscures all the tradeoffs uncovered by online tools. I'm not even sure that, if we had twice as many GPs who had the capacity to do same day f-to-fs for all, that this would be what patients preferred. But, since we live in the real world, GPs need to judge whether doing far more f-to-fs is worth the tradeoffs in time and convenience for patients. If some patients really need f-to-f assessment, how are you going to identify and prioritise them? Pre-pandemic practice often traded high proportions of f-to-f for long delays and prioritised by lottery not need.
So what should GPs do?
Don't reject online tools because they think they are inimical to good f-to-f practice. Perhaps–even if they love the benefits of their f-to-f meetings–they should choose online triage tools that make it easy for them to identify the patients who will benefit most from f-to-f.
Choose tools that ask patients what they want and don't assume everyone wants f-to-f.
Choose tools that patients like, not ones they won't use.
Monitor the results: monitor what patients want and whether they get it; Monitor their satisfaction; check that triage works effectively.
Certainly don't pine for the non-existent days when f-to-f was the perfect solution to everything with no tradeoff in time or convenience.
