We are soon to publish our work about developing a system for describing and relating same-day emergency care (SDEC) implementations, i.e. an SDEC classification system. But what makes a good-quality classification system, and how will it help us reduce avoidable admission?
The technical term is a "taxonomy" but, no, it has nothing to do with tax. A taxonomy is a way of setting out how similar things relate to each other. They are ways of classifying or categorising things.
For example, we classify living things by setting animals apart from plants and, under the umbrella of "animals", setting birds apart from monkeys. Another example is a library, where you might find the fiction section set apart from the non-fiction section and, within the fiction section, you might find crime thrillers set apart from romance novels.
Some ways of classifying things are hierarchical, like the taxonomy of life. Other classification systems use 'facets', i.e. different ways of looking at something. We say more about these, below.
Some ways of classifying things are hierarchical, like Carl Linneaus' taxonomy of life. For example, humans are primates, which are mammals, which are animals.
To find humans using a hierarchical classification system, you would say:
I'm looking for an animal...
...that is a mammal...
...that is a primate...
...that is a human.
Going from broad categories to narrows ones might be good way to store information but it isn't the natural way we look for information. An alternative, called a 'facetted' classification system, is often preferred (see 'Adding filters').
Another way of classifying things is by using 'facets', i.e. different ways of looking at something. For example, when you shop online, you might use the filters to specify the kind of thing you are looking for: something small, something green, something not too expensive. Applied together, the filters help you find what you are looking for.
Using our example of finding humans, we might say:
I'm looking for an animal...(which includes slugs, giraffes, and humans)
...with two legs...(which still includes birds, kangaroos, and humans)
...that doesn't have fur... (which still includes birds and humans)
...that doesn't fly.
The facets allow you to search based on what matters to you in the moment rather than searching through the hierarchy that someone else decided upon. The down-side of these facetted classification systems is that you will still have to sift through the filtered set of results, just like when you are shopping online.
We described the kinds of classification in terms of how they help you to find things. But the real purpose of a classification system is to help see how similar things relate to each other, in an insightful way. When you look at things through a good-quality classification system, you see them in a whole new light. Eureka!
For example, classifying the parts of plants shows us that tomatoes are a fruit rather than a vegetable. This is despite supermarkets selling them in the vegetable aisle. And, when thinking about making dinner, classifying ingredients as salt, fat, acid, and heat empowers you to make substitutes and create new dishes.
So, how will a classification system for SDECs help? Well, we are developing a classification system for SDECs because we want to make sense of all the variety that we are seeing. At the moment, it's a bit messy when you look at all the ways that SDECs have been setup and run, across the country. It is hard for SDECs to reduce avoidable admissions when they don't know what is working and not working for similar SDECs. We need to know what differences make the difference between avoiding and not a avoiding admissions.
Keep in mind that an SDEC classification system is not blueprint or design guide for setting up and running SDECs. Instead, it intends to be a way seeing SDECs that helps various kinds of people to use, run, monitor, and evaluate them. We've written out some examples, below.
Classifying SDECs can help patients understand the types of SDECs, similar to how they benefit from knowing the different types of emergency department, i.e. the Minor Injuries Unit is only for minor injuries and during opening hours while the A&E is for more serious issues and never closes.
Classifying SDECs can help SDEC doctors and nurses by showing them which SDECs are most like their SDEC. The doctors and nurses can then talk to these similar SDECs to hopefully share best-practice. Of course, the doctors and nurses can also use the taxonomy to see which SDECs are quite different. They might want to talk to these other SDECs to see if there might be a better way of doing things.
Classifying SDECs can help hospital managers by showing them the different kinds of SDECs they might have in their hospital. The managers might never have realised that their SDECs are setup and operate quite differently from one another, which might explain why it is difficult to manage them all in the same way, for example.
Classifying SDECs can help academic researchers to know which SDECs are similar and which are not. This helps researchers to know if they should put SDECs in the same group for their study, and to know what factors should be taken into account when doing their analyses.
Classifying SDECs can help government policy makers by showing them the variety of SDECs that have come about as a result of their policy and guidance.
This project is about studying SDECs and their ability to reduce avoidable admissions to hospital. The more we know about what SDECs are, the the better we can operate, monitor, and evaluate them to find out what works and what doesn't.
As part of our first work-package, we surveyed hospitals in England to find out how SDECs are set up and run. The hospitals' described a variety of SDEC implementations and understandings of what SDECs are. Our classification system is an attempt to make sense of the reported variety in an insightful rather than descriptive way.
We are soon to publish details of our classification system and how we developed it. You can stay informed by signing up to our quarterly bulletin to receive our updates.