Anyone who has received some training on QI methodologies and the model for improvement in particular or who has participated in a QI project will have come across lots of data, both quantitative and qualitative. But do we ever stop to think why we put so much emphasis on collecting, analysing and interpreting data when we are ‘doing’ quality improvement? This article sets out to examine why data means so much and why a QI project relies upon data when we are ‘measuring for improvement’.
William Edwards Deming (1900–1993) was an American engineer, statistician, professor, author, lecturer, and management consultant who is often regarded as the ‘grandfather’ of modern quality improvement thinking. His practices were and remain widely influential in the motor industry and beyond. This often-quoted wisdom sums up much of why data is important; basing our improvement on facts (data) prevents us from falling into the trap of following an opinion or hunch about what is the best course of action.
Essentially data can serve a number of purposes when you are doing work to make improvement happen: tracking and confirming that improvement is taking place, helping to tell a story and aiding learning. Let’s have a closer look at each of them in turn?
First and foremost, you collect data and plot it out over time to answer the second question in the model for improvement: How will you know that a change is an improvement? Without the data, you revert to having an opinion about whether change is causing any good effect on your service.
The second importance of data is that it can help you to tell a story, especially if you make the data visual and therefore accessible to an audience. A chart showing that data has gone up or down easily gets across whether a project is having impact or not. Always remember to define and tell your audience which direction is the ‘direction of goodness’! Data on a story board (as seen above) also helps you to engage more people in your work; the data invites you to have a discussion about it, which can often lead to staff, service users or carers joining a QI project team and new perspectives on your improvement work.
Finally, data plays a key role in helping you learn in two different senses. Gathering data about your system or service before you get in to a QI project is a crucial part of exploring a problem. Pareto charts and audit/survey results can all provide you with information to narrow down what you need to improve. Indeed, you need data to tell you what the baseline output of your system is and hence helps you decide how much you need or want to improve; it answers the ‘by how much?’ question when you are trying to agree a SMART aim statement.
And the other aspect of helping you learn relates to passing on your learning to others. Data that visualises your improvement helps others to quickly learn about your successes. A QI poster such as this example from Specialist CAMHS Services in Milton Keynes above demonstrates how the data supports the learning from this project.
So, we have seen why data is vitally important in QI, but there is one caveat on all the above and that is to ensure that your data gathering is in proportion to your needs. We often say “as much as you need, but as little as you dare”. Quality Improvement does not have to be an exact science (although there is science behind it!) and so collecting too much data (making an industry out of data) or having a perfect view of your system’s performance can seriously distract you from testing changes to make improvement happen. It is always best to ask yourself “What do we need to know?” before you go about gathering data, so you know why you are collecting it.
Contact Information
For support or further information on any aspect of improvement work in CNWL, please contact the QI Team in the Improvement Academy at: