Key takeaways
- Developmental learning improves each link by feeding data back; design learning questions the whole chain and lays the puzzle differently.
- Improving per link runs out at some point: from 120 to 100 and perhaps 90 minutes, but no further.
- In cardiac arrest, chest compressions must start within six minutes; an ambulance arriving in the fourteenth minute meets the standard but does not help the patient.
- A network of AEDs and about 6,500 citizen responders linked to the control room raised survival in good condition from 24 to 40 percent.
- The whole programme saves about sixty lives a year in the region: a coach full of people.
Two forms of learning
The case (see part 1, part 2 and part 3) shows that the improvement team beat the standard time it had set itself: from 120 to 100 minutes, thanks to business intelligence.
Developmental learning
Many of those improvements came about developmentally, purely by feeding data back to the people responsible for a link. In this way it was decided to dispatch ambulances immediately instead of first questioning the caller extensively, and to have the ambulance crew place two needles on the way to the CT scan. Perhaps 90 minutes can still be reached by looking for improvements per link, but at some point that really ends.
Design learning
If you still want to gain a lot of time after a number of cycles of mainly developmental learning, you need to zoom out considerably. What then comes into view are possibilities to lay the whole puzzle differently: you question the entire chain you built together. Taking a different route through the hospital is an illustration. Design learning often involves considerable costs, so such a redesign must be embedded in a larger PDCA cycle to bear the investment over a longer period.
Developmental versus design learning
Successful organizations combine both forms.
Developmental learning
Design learning
Scope
One link at a time
The whole chain
Method
Feeding data back to the link
Zooming out and redesigning
Example
Dispatching at once
New route through the hospital
Costs
Low
Often considerable
PDCA cycle
Short, operational
Long, strategic
Some future scenarios
When Martin Smeekes zooms out further and looks at a redesign of ambulance care, several future paths become visible.
The drone doctor
If drone technology keeps developing, we arrive at the system innovation in which you no longer bring the patient to the hospital, but the hospital to the patient. It is already possible to bring an automated external defibrillator (AED) to a patient by drone.
EEG in the ambulance
Until now it has taken a lot of time and effort to make an EEG. A development that Smeekes worked on together with Philips is a rubber cap resembling a bathing cap. With it the EEG can be made in the ambulance while the specialist in the hospital already examines the results. The instant diagnosis determines the protocol for the ambulance at that moment: a cerebral infarction requires a different protocol than a brain haemorrhage.
The mobile operating room
New developments will probably make it possible within ten years to carry out full operations in the ambulance. We still think of the ambulance primarily as a means of transport and are stuck in the idea that you must go to a hospital to operate. Regard the ambulance as a mobile operating room and other possibilities appear. This is a deep investment that typically fits a strategic PDCA cycle with a lead time of, say, three years. And when such an innovation sees the light, many terms to which improvement professionals have committed themselves for years can go in the bin: there is no reception at the operating room at all.
A disruptive technology trend
In the future the patient can stay at home for simple procedures. Equipment is brought in, a nurse is present and the specialist operates remotely via robotics. Take defibrillators: once operated only by doctors, they now hang in all public spaces and can be used by relative laymen. They are no longer an asset of doctors in specialized medical environments, but a commodity in the public space. Such technologies have a major impact on existing chains: new chains or networks arise in which the ambulance takes on a fundamentally different role. Anyone who identifies too much with current equipment and its KPIs gets no incentive from the system to change anything. Again a plea to start with passion.
Datacratic working at its best: the citizen AED initiative
After all the improvements in ambulance transport, Smeekes kept looking for more. How can we develop a model that is not based on deploying more ambulances? The basic question remained: how do you organize things so that the patient is helped in time and residual damage is limited?
Six-minute zones
The response time of fifteen minutes is a standard for urgent calls in general. But in cardiac arrest the six-minute zones matter: chest compressions must start within six minutes to give the patient a good chance. If an ambulance always arrives in the fourteenth minute, it meets the national standard, but the patient gains nothing. After eight minutes it still makes sense to intervene; if the heart is only restarted after ten minutes, the chance of a good outcome is much smaller.
The new improvement goal became to start chest compressions within six minutes, everywhere and always. Together with Ruud Koster of the AMC and the Twente region, which has a comparable geography, the “citizen AED” was developed. From 2009 AEDs were hung in all kinds of places, forming a dense network in all municipalities. That network is worth nothing if nobody can operate it, so an informal organization of about 6,500 citizen responders was linked to the control room. When a call comes in about someone who has collapsed and is breathing strangely, an ambulance and the police are sent, and at the same time automated messages go to citizens in the network who happen to be nearby and can resuscitate. With this form of acute crowdsourcing great success was achieved: these people are on the scene before the ambulance.
The datacratic PDCA of the citizen AED
- Plan and passion. It starts with the thought: if we keep doing the same, we will not save this group. Deploying laymen who have followed half a day of AED training is the plan.
- Do and data. Citizens operate the AED and are directed from the control room, creating a far more fine-grained grid in the region.
- Check and conscious control. Research by the AMC shows that heart patients receive chest compressions sooner and that their chances of survival have increased enormously. Before the citizen AED, 24 percent of people survived a cardiac arrest caused by ventricular fibrillation in good to reasonably good condition; that rose to 40 percent.
- Act and autonomy. All Dutch municipalities adopted the approach over the years. The Dutch Heart Foundation ran large campaigns: learn to resuscitate and sign up, then you are linked to such a network.
How Diamond Decisions combines both forms of learning
Diamond Decisions helps organizations improve per link and redesign the whole. In business modeling we question the chain itself, with data analytics we feed results back to each link, and with decision makers we determine when improving is no longer enough and redesign is needed.
Epilogue
In four parts this case showed step by step how the time between the first call and the placement of a stent was shortened. By continuing to learn smartly and looking differently at available data, human lives are saved: about sixty people a year. That is a coach full of people of flesh and blood. Sixty families who would otherwise have lost a grandfather, grandmother, father or mother.
Talk to the author
Daan van BeekDiamond Decisions
Discuss your question with Daan
Tell us which challenge in this story you recognize. In a no-obligation conversation we translate the lessons to your organization.
Schedule an introduction →Diamond Decisions
Our approach in 4 steps
From inspiration and business model to data analysis and better decisions: this is how we approach it for you too.
Explore our approach →Frequently asked questions
01
What is the difference between developmental and design learning?
Developmental learning improves existing links step by step by feeding data back to the people responsible. Design learning zooms out, questions the whole chain and redesigns it. The first is cheap and fast, the second requires investment but yields larger leaps.
02
What is a six-minute zone?
An area in which help can start within six minutes after a cardiac arrest. Chest compressions and defibrillation must start within about six minutes to give the patient a good chance of surviving without serious damage.
03
What did the citizen AED network achieve?
According to research by the AMC, survival in good to reasonably good condition after cardiac arrest caused by ventricular fibrillation rose from 24 to 40 percent. The approach with AEDs and citizen responders linked to the control room was adopted by all Dutch municipalities.
04
When should an SME redesign instead of improve?
When improvements per step yield less and less while the goal is still far away. Then ask whether the process itself is still right: can steps be skipped, moved to the customer or supplier, or replaced by technology?
Further reading
About the author
About Diamond Decisions
Diamond Decisions helps organizations become better by enabling better decisions. We don’t start with data, dashboards or AI, but with the decision that needs to improve, and work back from there to KPIs, data and technology. We do that in four steps: inspiration and alignment, business modeling, analytics and data platform and data-driven decision makers. That way BI and AI don’t remain an IT project, but deliver value every day.


