Dashboards are great tools for visualizing data. Large amounts of data that might be tedious to analyze can be quickly summarized and displayed as charts or a single indicator. However, tables are better for the comparison of individual items. In tables, indicators can assist visualization and improve understanding of the data in the table.
So, how can you use these indicators? And more specifically, how can they help pinpoint Length of Stay (LOS) opportunities for hospitals and health systems? Let’s look at the two examples mentioned above: standalone indicators and indicators that are part of a table.
There are many types of indicators in DivePort, Dimensional Insight’s dashboard software. They are all simple to use and easy to configure. You can see a list of all these indicators in Dimensional Insight’s online help.
It’s great to have a lot of options from which to choose. However, having many options to choose from presents the challenge of deciding on the most appropriate indicator to communicate the data you display. To do that, here are some tips.
- Use indicators to enhance context via visualization, like using the color red to indicate an undesirable number.
- Choose indicators that are intuitive and obvious. Users should be able to interpret the data quickly rather than spending a lot of time attempting to understand the data.
- Indicators should drive the users towards the meaningful insight of the data to make actionable decisions.
- Leverage QuickViews that act upon your indicators for diving capability and more interactivity.
The KPI or metric used for this article is Length of Stay (LOS), an essential metric in the healthcare world. Figure 1 shows a collection of different types of standalone indicators for LOS. At the top of the image, there are text indicators that provide numbers. The bottom of the image shows visual indicators which offer additional aspects to the KPI, like color, shape, and position that add meaning.
These indicators highlight different aspects of the KPI LOS. For instance, the first visual indicator compares the Actual versus the Expected value of LOS. The types of visual indicators from left to right are: Fuel Gauge, Fill Gauge, Circular Gauge, and Vertical Bullet. The visual indicator on the leftmost and the rightmost displays the value of Length of Stay Actual versus Expected. However, they use different approaches to presenting that data.
You’ll see a black line in the yellow in the fuel gauge, which is the expected target value for the Length of Stay. But because our actual Length of Stay is higher than expected, the pointer shows that the value is in the red color zone.
The second gauge is a Fill Gauge and displays the ratio of actual versus expected Length of Stay. The target is ‘1.0’. But because the actual value is higher than the target, the fill is colored red. If the actual Length of Stay were less than or equal to the target, the fill would be colored green.
The circular gauge shows that there is a 27% opportunity for improvement in terms of reducing the length of stay.
All the indicators are easy to configure. Users have a lot of control over their appearance and can choose different colors for the targets. Users can also edit the target value and use markers or Measure Factory standard measures to define the values in these indicators.
So this is the first approach, where you can use standalone indicators.
Now, let us look at a second indicator type, which adds meaning to numbers in a table.
In Figure 2, we have implemented the second approach. We have incorporated indicators in our tabular presentation. We have used DivePort’s measures portlet, but you can also use the matrix portlet to define these indicators. This table is divided into three sections to compare Volumes, Utilization, and Opportunity. You can select different values from the QuickViews: “Facility,” “Discharge year,” or “Service Line,” and you’ll see that all the numbers in the table change. For example, if we choose the Service Line “Endocrinology,” the table will display Endocrinology Volume and Endocrinology Opportunity. The best part about this page is that you can start at a high level and dive deeper for more information. You can see data for a KPI such as Length of Stay at an organization level. And then, you can dive deeper to find out if there’s an opportunity in any service line or any providers where you can reduce the Length of Stay.
The indicators in this table are easy and straightforward to configure, and there are many alternative indicator styles that you can incorporate.
Figure 3 utilizes all of these indicators. There are standalone indicators, text indicators, and visual indicators. There is also a table at the bottom of the image, which has indicators right in the table. So if we take a look at the example, we have selected all Service Lines. Looking at the “Attending Provider Service,” this view shows that Internal Medicine has the highest opportunity to improve by reducing the length of stay. The opportunity percent total is 55.3, and the most significant opportunity. The green circle in the racetrack indicator under “Opportunity” emphasizes that this is the most significant opportunity. These indicators provide information about the actual and expected LOS value. If the actual is higher than expected, the slope indicator points upwards and is colored red to indicate an undesirable value.
Figure 4 shows that we can dive further down on internal medicine by Attending Provider to determine if any specific provider shows a need for improvement. Here, the first provider shown has an 8.4% of the length of stay days. And there’s 9.5% of opportunity.
Figure 5 shows that if we want to dive further down, we can check the type of diagnoses that make up these values and see if there’s any room for improvement. We can dive on the provider’s name by Principal Dx Clinical Class to figure out the diagnoses with the greatest need for improvement. So from this dive, we can see the two top diagnoses, Diabetes and Congestive Heart Failure. They both are chronic conditions.
Figure 6 shows that we can dive further down by an account ID on congestive heart failure to find out if there are any specific account IDs with a high level of opportunity for improvement. Here, you can see the top three account IDs out of 16, which have the highest opportunity for improvement.
This is how you can start with the high-level Length of Stay information and then dive deeper into the numbers to find if there’s an opportunity for improvement or not.
To summarize how we achieved this analysis, we used standalone indicators and indicators that are part of a table. Like we saw in the dashboard example, while creating a dashboard page, you have to keep the user’s perspective in mind to create intuitive and actionable dashboard pages with effectively communicated information. Do not add so much information that you confuse the user. Also, make the dashboard pages interactive so that users can dive into the numbers. Users can slice and dice and get the information that they seek. Proactively diving into the data for the information they need can drive a user towards a meaningful insight into the data. And finally, indicators are an excellent opportunity to improve visual appeal for dashboard pages.
For more information, you can visit Dimensional Insight’s online help.