Giving a small service change a fair test

An idea for improving a service deserves a clear question and a chance to be examined in practice. Before changing a routine, a manager can ask what the proposal is expected to improve, what would count as useful evidence and who might experience its effects. Those questions give a team something specific to learn. They also create a basis for deciding whether the change should be adjusted, tried again or adopted more widely.

Ask a question the test can answer

The Institute for Healthcare Improvement describes testing a change through planning, trying it, studying the results and acting on what is learned. NHS England similarly places learning, data and collaboration with people delivering and receiving services at the centre of improvement. Although these resources concern healthcare, their emphasis on structured enquiry offers a useful starting point for thinking about everyday administrative changes.

Start by describing the difficulty in the existing process. Identify where it occurs, who encounters it and what information supports that understanding. Talk to the people doing the work before settling on a solution. Their explanation may reveal a missing instruction, an unnecessary handover or a dependency that the proposed change needs to address.

Then write a prediction that can be examined. Specify the result you expect and the reason you expect it. Choose an observation or measure that relates directly to that prediction. Agree what would make the result disappointing as well as encouraging. This gives the review a reference point that was set before the outcome was known.

Keep the first attempt bounded

Decide the scope of the initial test: the part of the process involved, the people taking part and the period over which observations will be gathered. Check the necessary permissions, information access and existing requirements. Everyone involved should understand what is changing and what arrangements continue to apply.

Give the test a named owner and a review date. Explain how difficulties should be raised and agree when the test should be paused or stopped. Make sure the people taking part can return to an appropriate working arrangement if the proposed change proves unhelpful. These practical preparations deserve attention even when the idea appears straightforward.

Keep the question manageable. If several parts of a process change together, record that clearly and be cautious about attributing the result to any single element. Leave enough time to notice how the revised arrangement is actually used, including any informal work people do to make it function.

Look for the effort behind the result

IHI’s guidance on measurement asks teams to consider whether a change in one part of a system creates problems elsewhere. Collect information that helps the team judge the whole task. Alongside the intended benefit, consider corrections, duplicated effort, requests for help and work transferred to someone else. A quicker step can still leave additional work elsewhere. Ask the people using the process what became easier and what required extra attention.

Use clear definitions when comparing observations. Check that the work being compared is sufficiently similar and record relevant differences in circumstances. An early result can be encouraging while still leaving questions about how the change will work with different people, a busier period or less immediate support. Describe those limits when sharing findings.

Make the next decision explicit

At the review, bring the prediction, observations and people’s feedback together. Explain what the team learned and what remains uncertain. Agree whether to adapt the idea, test it under different conditions, retain the existing arrangement or prepare for wider use. Record the reasoning so that someone outside the original discussion can understand the decision.

If the change is going to continue, identify the instructions, resources and communication needed to support it. Decide who will check that the expected benefit persists and when the team should reconsider the arrangement. Give the people affected a clear explanation of what has been agreed and how they can raise further difficulties.

A useful first step is a short note prepared with the people closest to the work. Set out the problem, the proposed change, the prediction, the observations to collect and the date for reviewing them. Keep the note accessible during the test and return to it when deciding what happens next. That turns an improvement idea into a shared piece of learning.

Sources and further reading

• Institute for Healthcare Improvement: Model for Improvement, Testing Changes
• NHS England: What is improvement?
• Institute for Healthcare Improvement: Model for Improvement, Establishing Measures