The PDSA Cycle: How to make an improvement stick - Sekoia

The PDSA Cycle: How to make an improvement stick in a busy day

PDSA article thumbnail picture

Now you may think this is just another pseudo academic article. Well, it isn’t. This model focuses on incremental and very hands on improvements that can lift your service delivery to its next level.

Any care plan changes or adjustments are made to live up to the promises and vision of your service. Ultimately securing an attractive customer experience for your residents. But how do you know when you’re there? Sometimes a change is just a status quo.

Don’t want to read through the entire post? Here are our 3 social care examples.

It’s sensible to define successful outcomes for any care plan or activity. But making any kind of change can be a project in itself. So, what does success look like?

Some say you should look at the cold hard facts. The data. The time saved. The money saved. Others would ask people what they feel. Are they less stressed now? Do they like it better now than before the change? The best approach is often a mix of the two.

Our preferred method is the PDSA cycle. It’s a great way to clarify the desired goal and how to measure the change. It’s also a powerful tool for learning from ideas – no matter if they were successful or not.

We know you’re busy, so here’s our super-short summary of what's needed to use the PDSA cycle

Consistent data collection
Fixed structures/processes
Organisational buy-in
Patience

What is the PDSA all about then?

Like most other abbreviations the PDSA cycle sounds more complicated than it is. The cycle is divided into 4 stages.

  • Plan – the change to be tested or implemented
  • Do – carry out the test or change
  • Study – based on the measurable outcomes agreed upon before starting, collect data before and after the change. And reflect on the impact of the change and what was learned
  • Act – plan the next change cycle or full implementation.

A key part of the process is that it’s not a one-and-done project. Neither is your service. By approaching the change as a process rather than a project you’re more likely to make the transformation stick. There’s no end. It’s simply the new normal.

When is PDSA useful?

Most models, frameworks or guidelines are either too generic to be useful or too specific to work.

Likewise, the PDSA cycle is not going to be a silver bullet. They simply don’t exist. So, when is it useful? And, perhaps more importantly, when is it not?

The PDSA cycle’s best friend is SMART goals. Yes, frameworks do love their acronyms⚡️

If you have a goal that’s specific, measurable, achievable, relevant, or time-bound you can use the PDSA cycle. And there are plenty of those goals in social care, so that’s a good starting point.

If your desired outcomes are more fluffy PDSA is rather useless. Sometimes even counterproductive, since it requires tangible results to iterate on the process.

As an example you need to know what the desired outcome of a care plan is to be able to measure any improvements to this.

If your desired outcomes are more fluffy PDSA is rather useless. Sometimes even counterproductive, since it requires tangible results to iterate on the process.

Getting started with PDSA: Key questions to ask yourself

How do you get started?

Our advice is to ponder these 3 questions to get the best starting point:

1)    What are we trying to do?

  • There has to be a clear goal with measurable targets
  • The goals should be relevant and ambitious

2)    How will we know if the change is an improvement?

  • You will know by defining and measuring outcomes

An example could be if a resident meets their daily fluid target or completes their planned training exercises 3 times a week. Very tangible.

3)    What changes do we make to get improved outcomes?

  • Before starting the cycle it’s useful to map what changes can be made. Which elements can be changed to create a different outcome? Most likely, there are a lot of different things you can do. But evidence suggests that small changes are more likely to result in improvements.
  • Our advice is to pace yourself and let the iterative process of the model do its magic.

NHS’ tips for succeeding with PDSA

NHS has made some advice to make use of the PDSA cycle.

  • Plan many cycles to test ideas

PDSA’s premise is that you are not done when you’ve completed the cycle once. It’s an iterative process. If you allow the time to go through the cycle many times, you are more likely to find the best ideas and make any changes stick.

  • Test on a really small scale. For example, start with one resident and increase the numbers as you refine the ideas.

The whole process of the PDSA cycle can easily become overwhelming. Especially since it’s not a one-and-done thing. To make it more manageable, consider starting small. One resident and one area. Then expand it gradually when you’re ready.

  • Test the proposed change with people who believe in the improvement. Don’t try to convert people into accepting the change at this stage.

Change is a funny thing. People don’t like it all that much. Almost no matter how bad the current situation is. Because you make it work. And with change comes the big “what if?”. It’s uncertain and many people feel uncomfortable with that uncertainty. That’s why you shouldn’t direct your attention towards convincing them right away.

Most likely the early iterations of the cycle will expose flaws in the change. Focus on the people who are okay with uncertainty and has bought into the need for change. That’s where you’ll find success and support for the change.

  • Only install the idea when you’re confident you have considered and tested all the possible ways of achieving the change.

When you commit to a change, you put yourself out there. As leaders, as people. Make sure you choose the best method to achieve the desired outcome. But remember it’s impossible to be certain of success. There’s always an element of uncertainty. But isn’t that the beauty of it?

Bottom line. It needs to be easy to follow the cycle’s steps. If it isn’t people won’t commit to the process. If you make it difficult to follow the improvement loops the project is born to die.

Read more and download NHS’ PDSA template 👇

https://www.england.nhs.uk/wp-content/uploads/2021/03/qsir-plan-do-study-act.pdf

You can’t study without information. Just like you can’t revise for a final without books. You need data you can trust to make informed decisions.

But so what? How can you use the model in social care?

It’s all well and good to know a process for improvement. But if it’s unrealistic or otherwise unattainable in a real-world context it’s all for nought. Our favourite thing about this method is its versatility. To prove it, we’ve come up with some of the most diverse examples we could think of. (Full disclosure: it’s Monday when we write this so we might not be at our creative peak)

Example #1: Wound Care

Scenario: You are a nursing home and want to create care plans for the people you support. More specifically for wound care. It’ll probably involve many people. Care staff and nursing staff and potentially a GP. There are a lot of moving parts and people need to work together effectively.

Using PDSA: First, you gather experts who’ll create hypotheses about the resident and summarise them in a care plan. When an injury or a cut occurs, you find a way to document it and track it. How did it happen and why? And treat the wound. After you find a way to measure it. With wounds, you typically take a picture of the treatment etc. and maybe even continuous measurements.

Staff need an easy way to access the needed information (the care plan). Promptly and at the point of care. Likewise, they need an easy way to offload documentation in a system that can accommodate it. Stuff gets lost in translation in the journey between the point of care and the computers in the back office. And if this process is not consistent? Then the PDSA cycle is practically useless since the consistency of the data will be off.

The great challenge in this scenario is to collect data in a consistent and structured way. Information is only useful when it’s provided in the correct context. Otherwise, it’s just clutter.

Example #2: Loneliness

Scenario: You are focusing on reducing the feeling of loneliness amongst your residents. It’s a difficult subject to address. Feelings are subjective. They’re tricky to track through data, right?

Well not necessarily. You can break down the resident’s actions into indicators of social connection. And find a suitable way to combat loneliness.

Using PDSA: You might decide that a good measure of loneliness is how many times a resident seeks social contact on their own. You set a goal that a resident should seek contact 2 times a week. You track this weekly. The PDSA cycle will be useful in determining if the methods used to achieve the goal are fruitful.

The first iteration of the cycle might be looking at pictures and doing reminiscence exercises with the resident. After a week or two, you go back and look at how many times the resident sought contact.

But how do you make sure people gather the same data? What if staff define “seeking social interaction” in a different way? Everyone needs to be on the same page, and that’s easier said than done.

Our solution?

Use templates to make sure data is structured, agreed upon, and easy to record.

Example #3: Medicine Administration

Scenario: You want to improve the way you handle medicine. Reducing the amount of forgotten medicine and wrong medicine administered. Where do you start?

Using PDSA: Medicine is a difficult thing to address. It’s something that’s heavily scrutinised – and for good reason. Residents requiring medicine have it described in detail in their care plans. It’s not a lack of information that’s the issue. All staff members do their best to keep up with medicine administration. It’s not a lack of care.

It’s a lack of time. And lack of access to pertinent information.

The tools you can use to make information more accessible could be to introduce tablets with information on them. If you move information to the point of care, the likelihood of it being useful increases significantly.

After, you follow the number of errors that happens. Are they going up or down? Remember to account for natural variance in the numbers. And if that change didn’t work? Iterate and try again. It’s not easy but it’s worth it.

What do all these examples have in common?

They all collect data in a disciplined and structured way. You can’t study without information. Just like you can’t revise for a final without books. You need data you can trust to make informed decisions.

The project keeps failing. What now?

You’ve followed the PDSA cycle. Clarified what you are trying to achieve, how to measure it, and people want to back the change. But still, the PDSA cycle keeps on coming up short. Something’s not working. It’s a failure.

Or is it?

The one area where the PDSA cycle struggles is if you want to introduce big changes quickly.

Otherwise, there’s always something you can take away from each iteration of the cycle. Maybe the change was too much too soon? Or not specific enough?

By using the PDSA cycle, you are following the scientific method. The method that has been used to find virtually all breakthroughs through history. The method will succeed in the end. The secret?

It’s based on hypotheses and data-driven in its evaluation. That’s also why the length of each PDSA cycle will vary. From change to change and organisation to organisation. It all stands and fall with the data you are able to pick up.

Remember the famous quote by Thomas Edison:

“I have not failed. I’ve just found 10,000 ways that won’t work.”

Conclusion

The fun doesn’t stop here. Variants of PDSA have been used for the longest time and there’s one change on the horizon that has the potential to forever change it.

Artificial intelligence (AI).

It’s a buzzword alright. But for good reason. And it’s not some far away pipedream either. It’s happening right now. The foundation of the PDSA cycle and any AI involvement is structured data. The next logical step would be to integrate the two.

Right now, we are pilot-testing AI in a nursing home in Surrey. Imagine receiving notice that certain residents’ quality of life is in decline. Or various health conditions are deteriorating. All based on objective data. Simply there to complement the expertise of the care professionals. Read more about it here: Predictive Care Planning

The future is closer than you think.