Equipped with a range of tools and techniques, an analyst can support better decision making, and ultimately drive better health outcomes… But how do we achieve this?
As part of the Health Economics Unit’s collaboration with the Strategy Unit and Nottingham and Nottinghamshire Integrated Care Board (ICB) to increase knowledge and understanding we developed a series of training courses; the second course of this training focused on helping analysts to drive better decision making. Fraser Battye and Steven Wyatt, of the Strategy Unit, share some of the insights from these training sessions.
An introduction to problem structuring
In a nutshell, problem structuring is all about making sure the question that needs to be answered by a piece of analysis is clear and well understood. The first element of this training was led by Steven Wyatt, Head of Research and Policy at the Strategy Unit, focusing on the concept of problem structuring. Steven suggests that problem structuring is an essential skill, but one that is often overlooked in analyst training programmes.
“Unfortunately, well-formulated requests for analysis are unlikely to fall into our laps”, says Steven. “And so, problem structuring is a vital step to take before beginning any data analysis task. Analysts should not take a request on face value, but instead spend time testing, probing and pushing back to fully understand the question in hand before making decisions about how to tackle it.”
Decision-makers have a clear view on the problems they face, but their requests for support are constrained by their limited knowledge of analytical methods. This leads to simplistic analyses that do not make full use of analysts’ capabilities.
“How many times does an analyst spend a significant amount of time delivering a piece of work, only to see that nothing is done with the analysis they provide?” asks Steven. “Often this is because the question being asked could never address the underlying issue. Spending more time up front, discussing and exploring the problem context increases the chances that an analysis will be useful.”
Of course, this isn’t always comfortable, and analysts may have to fight their way through some politically charged environments, and juniors to the profession may find it difficult to push back against more senior positions. Often, we’re faced with a lot of different stakeholders with different views and requirements, all giving an opinion on the work needing to be done. This can be challenging.
“I’d love to be able to give a proven formula”, says Steven. “But problem structuring is more of a craft than a science and, despite it really being quite intuitive, it can take time to master. That said, it is worth investing time in as the rewards are considerable – greater job satisfaction, less tedious repetition, and more impact.”
Making better decisions
Even when strong analysis answers the question at hand, it doesn’t always lead to quality decision making. But we can do something about this. We each make thousands of decisions every day – choosing our clothes or deciding whether to have another coffee at 3pm, for example. Some of these decisions we make instinctively, and some will warrant a little extra thought or research (Is it going to warm up in the afternoon? How tired am I feeling?) but it’s not often that we consciously run a decision-making process in our minds.
With this in mind, the Strategy Unit’s Fraser Battye, looked at what better quality decision making looks like. Having spent many years understanding decision making theory and practice, Fraser has put much of his time into helping people in the NHS learn how to make better quality decisions. The session focused on encouraging critical thinking about difficult problems and highlighting that there are different ways to assess the options available to us when considering big questions like changing how we invest in certain services, reshaping care pathways and adjusting organisational structures, for example.
“The basic principle”, says Fraser, “Is that if people are helped to think more critically, they will make better decisions which lead to better outcomes”. So why don’t we always take the time to think critically about problems and make better decisions? Ultimately, good decision making is hard because it’s uncomfortable and, despite us making so many decisions every day, it can be quite difficult to learn and perfect.
Fraser introduced the group to an idea shared in Daniel Kahneman’s Thinking, Fast and Slow that suggests we have two internal decision-making systems. System one is automatic, fast and often unconscious. It’s always available and requires little energy or attention but can be prone to biases and errors. System two is a more controlled, critical way of thinking. Engaging system two, however, needs more work. It’s cognitively expensive in that it takes a lot of energy, so our brains try to avoid using it. As a result, we are unlikely to employ enough of system two in our decision making unless we make a conscious effort to engage it.
Learning a new skill such as this takes time. Psychologist and economist Kahneman suggests that “…acquisition of skills requires a regular environment, an adequate opportunity to practice, and rapid and unequivocal feedback…”. However, especially in health and social care, it’s rare to find ourselves in this kind of environment. Research with leaders in the sector has highlighted many reasons why good decision making is difficult to achieve. The historical, political and other restraints they perceive to be in the way, lead them towards maintaining the status quo and making quick decisions over and over again rather than taking active decisions to do things differently.
So, what role can analysts play in improving decision making practices? Firstly, there is a neat link to the first part of this blog where we talked about problem structuring. If analysts can work to improve the questions being asked, they can ensure that the insight they provide is more likely to fuel good decision making. Secondly, and perhaps more uncomfortably, if we can be involved in early discussions about issues as well as in the decision-making phase, we have an opportunity to provide constructive challenge and test unfounded theories, backed up by our data analysis. Ultimately, despite the uncomfortable nature of good decision making, it can make such an enormous difference to the outcomes we get with relatively little additional effort.