NHS 4-Hour ER Targets: What Happens When Hitting the Number Matters More Than Helping the Patient

TL;DR: When leaders impose arbitrary targets without improving the underlying system, people will find ways to hit the number without helping the patient. The problem isn't the people — it's the design.

I'm here in the UK for work, and the BBC morning news show had a story about what they call “Accident & Emergency” here in the UK. There's a lot of discussion, generally, in the news about improving true service quality instead of relying on targets.

The BBC reported years ago on the challenges the National Health Service (NHS) faced with emergency department (“A&E” in the U.K.) targets. On the surface, the four-hour requirement–that patients be seen within four hours of arrival–looked like a success.

Before the target, about 23% of patients waited longer than four hours. Within two years, that number dropped to just 5%. A clear improvement, right?

Not exactly.

The Problem: Gaming Instead of Improving

To meet the target, hospitals found creative ways to “comply” without necessarily helping patients. For example:

• Patients were admitted to a ward at the 3-hour-58-minute mark–whether or not admission was appropriate.
• Ambulances sometimes idled outside A&E so patients technically hadn't “arrived” until staff were ready to receive them.
• Extra staff were pulled into the emergency department during reporting periods, while surgeries elsewhere in the hospital were canceled.

These moves hit the target but created other risks, costs, and delays. In short: a system optimized for the metric, not the mission.

Rational Behavior, Broken System

I don't blame frontline staff for this. Their behavior was rational. When leaders impose arbitrary targets, people will do what's necessary to avoid punishment. It's predictable human behavior.

The real problem is the system. As Deming reminded us, “A bad system will beat a good person every time.”

As Brian Joiner wrote in Fourth Generation Management, there are three things that can happen when people are pressured to hit a target without having effective improvement support: distort the numbers, distort the system, or improve the system. The first two are dysfunctional — and they're almost always easier than the third.

The NHS examples fit Joiner's prediction perfectly. Ambulances idling outside? That's distorting the system. Admitting patients at 3:58, whether it was clinically appropriate or not? Same thing.

What Leaders Should Ask

The BBC story–and follow-up research from the London School of Economics–make an important point: targets can bring some benefit, but we have to ask if the benefits outweigh the costs. And if we see gaming, how do we redesign performance systems to discourage it?

There's a difference between an arbitrary target and what Donald Wheeler calls a “law of nature” — a target with a scientific basis. A 60-minute door-to-needle time for stroke patients has clinical evidence behind it. A four-hour A&E target imposed from above, without a method for achieving it, is arbitrary. Arbitrary targets don't come with a method for improvement. They just come with pressure.

From sales quotas to customer-satisfaction scores, we've all seen examples where pressure to “hit the number” leads to distorted behaviors.

More posts on “Gaming the Numbers.”

The Better Alternative

Instead of piling on more targets, leaders could start by studying the system that produces the current results. What does the data actually show over time? Is the system predictable? If so, improvement requires changing the system — not demanding better numbers from the people inside it. That means involving frontline staff in diagnosing problems, reducing the fear that drives gaming, and measuring progress through real patient outcomes rather than activity metrics that are easy to manipulate.

When hitting the target becomes more important than helping the patient, we've optimized for the metric instead of the mission. And as Joiner's framework suggests, that's the predictable result of pressure without method. The question worth sitting with: what targets in your organization might be producing the same pattern?