Why "Pull" for Hospitals?


Before any of you give a knee jerk reaction about how “pull is good, push is bad” — I agree with you… pull is better than push. For manufacturing. In a factory, a “push” system pushes production and inventory along so it just piles up and queues up regardless of downstream demand. A pull system, where operation A only produces what operation B can handle can keep WIP down, as long as operations are pretty balanced in cycle time and are always available, from an uptime standpoint.

I saw this question today in the “Healthcare Management Engineers” Yahoo Group.

We are in the process of building a new Emergency/Urgent Care facility. We would like to develop a pull system where it is possible. I am wondering if any of you have used lean methodology in your Emergency Departments. Were you able to achieve a successful pull system in the ED? How did you accomplish this? What were some of the strategies and tools that were used to create a pull system?

This seems like the internal hospital consultant is looking to use a tool (pull) — it's actually more of a concept than a tool — for the sake of using a tool… because pull is good, right?

What is the business problem being solved? I've worked with hospitals that have horrible patient flow. The emergency department is “boarding” patients, which means you're stuck in the hallway. Why is that? Is it the ER's fault? No…. there aren't beds available for you to be admitted into. The beds are full, or they're closed down because there aren't enough nurses. The ER is so full that the hospital is “on diversion“, which means they aren't accepting ambulances except in severe life threatening emergencies. Bad bad bad. These are huge problems that we need to solve.

So is “pull” the answer?

With the current state…. we HAVE pull. The ER isn't “pushing” patients up to the inpatient units. They are only moving a patient when a bed is freed up (when a patient is discharged — hopefully, they're going home, not here). Space downstream (bed) opens up, bring a patient from upstream (the ER). That's pull. And that's totally broken.

I'm not saying the ER should push patients to lay in the hallways of the inpatient floor instead of the ER.

If anything, the pull should be patients pulling on resources. When a patient needs a medication, the pull system gets it there ASAP. Likewise, when a patient needs an inpatient room, the patient is “pulling” on that resource (sending a signal that it's needed), although the room isn't physically pulled to the patient.

What we need is the following:

  • Balancing of capacity (beds) with demand (patients). This means having enough rooms, enough nurses, and making sure that we're flexible enough to put patients in rooms that aren't excessively specialized (having the “wrong” kind of rooms open).
  • Standard work: we need predictable, consistent processes to make sure that patients are discharged without delay, thus freeing up beds for new patients.
  • Flow: we need to reduce delays and waiting time between process steps
  • Visual management: how does the ER know when a bed is open, so they can immediately bring a patient up?

There are a number of lean tools that I would think about using first before “pull.” Looking to use any one lean tool, rather than solving the problems at hand, brings potential for ineffectiveness. It's a bad sign when I hear anyone saying, “how do I use this tool?” instead of “how do we fix this problem??”

Update — See my discussion of this on page 33 of Lean Hospitals (via Google Books preview).

What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn – and follow me or connect with me there.

Did you like this post? Make sure you don't miss a post or podcast — Subscribe to get notified about posts via email daily or weekly.

Check out my latest book, The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation:

Get New Posts Sent To You

Select list(s):
Previous article“Assembly Line Medicine”
Next articleOverpaid Dummies Voting with their Feet at Ford
Mark Graban
Mark Graban is an internationally-recognized consultant, author, and professional speaker, and podcaster with experience in healthcare, manufacturing, and startups. Mark's new book is The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation. He is also the author of Measures of Success: React Less, Lead Better, Improve More, the Shingo Award-winning books Lean Hospitals and Healthcare Kaizen, and the anthology Practicing Lean. Mark is also a Senior Advisor to the technology company KaiNexus.


  1. Pull concept is of course good. But there are 2 other concepts of lean that have to be considered. The first one is takt. Understanding the bit or the pulse of your system. “How long does it take to service one patient?” should be our first question. The next question is “How can we attain the best, simplest flow possible. The flow approach will incover many of the wastes of the value stream and it forces us to remove those waste. Then pull could happen naturally. We still need to level load the volume so we could further optimize our system.
    Rudy Go

  2. I think an important issue to resolve, before diving into push vs. pull, is the definition of customer and product. Is the patient the customer? If so, then the patient pulls services from the hospital, and the services need to be set up as such? Is the patient the product? Then the comments made in this post are more relevant, because now the hospital is pulling or pushing the product through the process. Or maybe the physician is the customer, in which case the patient becomes a product? On a related note, defining the customer(s) in healthcare is an oft-debated and contentious issue, especially when one considers who is really paying. I don’t have a good answer to any of these issues, but I think these are questions that need to be answered first to put the goals in the right perspective.

    A follow-on comment to Rudy Go: Healthcare, and hospitals in particular, are more like high-variety low-volume manufacturing (an environment in which I work), and the concept of takt is very very difficult to apply because of the variation (even when broken down by steps, processes, or machines). I question the applicability of a large-scale takt in a hospital, given the difference, for example, between caring for a patient with pneumonia and one with multiple gun shot wounds. However, takt might be more applicable to a sub-process, for example the takt for doing a blood test or an x-ray.

  3. I couldn’t agree more, Mark. When I worked in manufacturing, we referred to the “tool trolls”, the people who just went around applying tools without any business reason. The key question is: what’s the problem? Then the proper tool for the proper reasn can be applied. Now that I’m in healthcare, it’s more important than ever.

  4. Great post, I don’t disagree with examining the root problem before applying a tool, but this case I think there are several good reasons to use a pull methodology in hospital wards.

    1.) The hospitals I have spent time in have a centralized bed coordinator, whose job it is to actually arrange the transfer of patients from the ER, or scheduled surgery patient, into an acute bed. When things back up, this person or department is under intense pressure to move patients into beds, so putting a patient into the correct type of ward is not a high priority. This results in many different types of patients in a ward, which increases the workload and stress of the nurses working the ward, which leads to sick days, which goes back to the original problem. A bit of a vicious circle.

    2.) While wards may have the option to accept or postpone an admission, they don’t typically have the option to look for a more applicable patient, because the bed coordinator will only be sending them one client. They don’t have time to look for all applicable clients for a ward, as they will be sending the other clients to other wards. While a bed coordinator can be a clinical or clerical position, it’s really the ward manager who has the best idea of which patient a ward should take next. With a push system it’s impossible to provide the wards the ability to determine which patient they can best provide for.

    So by putting in a pull system, where wards can look at the clients currently being consulted, or clients who have been accepted, they could work collectively on getting the right types of patients into the wards. With a pull system they can do this continually, because they more than anyone else, the ward staff know what happens when clients back up in ER.

    I agree on your point of having a visual of beds that are open, but I think it might be almost more important for the wards to have this system wide view than ER, as they are the ones that have to adjust when the ER backs up. If this visual could be tied into a pull system, so much the better.

  5. As we all know, the Lean principle is based on Toyota Production System (TPS) and Kanban is essential in controlling the release of materials. So, how can we use the idea of Kanban system in healthcare? I think Lean in healthcare such as a tool to reduce waste and rapid improvement projects that often do not address the primary patient flows. Therefore, most lean healthcare applications do not engage with the ‘pull’ principle. Does that mean lean has fully implemented in healthcare? Any thoughts guys

    • Yes, kanban works very well in hospitals, as a form of pull.

      See this post and also this one that each have videos about kanban.

      Lean should also address patient flow, not just materials.

      Lean is never “fully implemented” anywhere, including Toyota, as there is always waste, there are leaders who don’t always follow TPS / Lean principles, and the system should always be evolving.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.