One of the difficulties in trying to compare different bodies doing the same thing is that we don’t know what good is. What is a good standard of timekeeping for trains? The answer is rather different in Switzerland from Britain. Theirs aren’t faster, just much closer to the stated time, which happens to be what passengers want.
Here’s another clue from sport. Have you noticed that the more elite the event, the closer to each other are the competitors’ times? In the Tour de France, a few minutes separate the winners from the tail end, over weeks of racing. But in your local Santa Fun Run, someone is over the line in 16 minutes, while others are strolling in after an hour and a half. Many spheres tell a similar story. Large supermarkets are barely distinguishable, despite all their efforts to persuade you of their unique offers. They can’t stay in business if they are far out of line. Modern thermal power stations around the world are all 40% efficient, plus or minus about 4%: above that range is not technically feasible, below it they can’t compete.
It’s unusual for Britons never to have flown, but a generation ago it was the preserve of the few. Aeroengines are one of the technologies which have made such low cost travel possible, and in a previous life I had a very small part in making them. The other day I caught up with an old friend who stayed in the field. He tells me that engines are getting still more efficient, by a fabulous 0.5% per year. His research might be responsible for 0.5% in about ten years time. Needless to say, all large civil engines have very similar efficiencies.
What happens, then, when we look at an area not subject to the same pressures? Say, how long it takes to be seen, treated and discharged from an English A&E? About the same number of players as in the Tour de France, with roughly the same technology and funding in proportion available to each one. The range is a factor of 3 between the fastest at 64 minutes median and the slowest at 184 minutes. This tells us not only that there is high variation, but also that the sector as a whole performs poorly. Worse, when we look at performance over time, there is no 0.5% per year to cheer us, they are actually getting slower, by about 5 minutes per year, or 4% of the average 125 minutes. I can already hear the dear things calling for more money – no, that’s not the answer. One hospital has cut its median time by 40% in the last year, and suffered a 25% rise in demand in the same period, with no special treatment. Better quality and higher performance cost less. That’s why you can all afford to fly.
But what about general practice? We regularly hear that satisfaction is terrifically high, with ratings above 80% for almost everything. Every one’s a winner and all shall have prizes. How does this relate to the variation we have discussed? A satisfaction survey can’t give over 100%, and it turns out that very few are below 70%, almost all crowd around 85% +/- 5% “very or fairly satisfied”, regardless of the question. When it comes to productivity, we don’t have a simple measure. like they do for those engines. What is revealing is that when a practice changes thinking from supply led (slots) to demand led (patients and their needs), and changes operation so GPs respond rapidly by phone, the new system makes them about 20% more efficient – within weeks. Compared to 0.5% per year that is staggering potential for change. How do we know? This is what they tell us (I could go on):
- Didcot practice saved from a £170k PMS funding cut. List now up 7%
- London practice is meeting all demand on the day, list up 10%
- Bradford salaried practice meets all demand with sessions down from 21 to 17
- Wokingham practice GPs now having to spend their duty hours on CPD, can’t fill slots after 5, and go home on time to their families. Huh.
- Liverpool practice chooses to turn their 20% saving into an easier paced day
- Edinburgh Deep End practice meeting all demand, at last access is not an issue
- Swansea practice, stress has gone, patients happy, what’s a DNA?
Back to our theme, what is the variation between practices in average waiting time to see a GP? We’ve measured the range from one day to over 10 days. ‘Nuff said.