Quality & Patient Safety

Ask simple questions to manage your data

Patient Safety Quality Monthly, July 17, 2008

Want to receive articles like this one in your inbox? Subscribe to Patient Safety Quality Monthly!

Ambulatory surgery centers (ASC) struggling to keep up with increasing demands to collect, analyze, and present data should first take an inventory of what information they’re already collecting and determine whether it ties into the organization’s strategy.
 
“The real secret is making sure every piece of data you collect, one, you know about, and, two, it’s directly tied into your strategy,” says Ken Rohde, author of Making Your Data Work: Tools and Templates for Effective Analysis, published by HCPro, Inc., in Marblehead, MA. “Organizations sometimes are collecting a lot more data than they know about, and that data never gets used because people don’t know that they’ve collected it.”
 
It’s critical that the data is linked to the ASC’s goals. For example, if one of the organization’s strategies is to provide exemplary patient safety, then the ASC needs to collect information that will reflect its performance in that area.
 
“If you collect the wrong data, you’re wasting your time, you’re wasting your organization’s time, and you’re frustrating everybody,” says Rohde, a senior consultant at The Greeley Company, a subsidiary of HCPro. “So you need to make sure you’re collecting the right data.”
 
Collect at the right speed
Organizations also need to ensure that they’re collecting data at the right speed. For example, when you’re driving a car, you need to monitor the gas gauge regularly. If you look at the gas gauge only every month or so, you would most certainly run out of fuel.
 
Similarly, if your process changes daily or weekly, you need to collect data on a daily or weekly basis.
 
“If your data is changing on a weekly basis, taking a quarterly or semiannual sample is misleading and probably harmful to your organization,” Rohde says. “So you need to make sure the frequency with which you’re sampling the data matches up with the speed.”
 
Ask simple questions
Another strategy is to manage your data by answering simple questions, Rohde says. One of the first questions should be: Do we stay the course, or change direction?
 
“If your data can’t help you answer that question, your data is not working for you,” Rohde says. “We have to collect information that allows us to know the status of what’s happening. That’s the ‘what’ data. This is the data that will help us know falls are under control, or, uh-oh, falls are out of control, falls are in trouble.”
 
After collecting the “what” data, ASCs need to gather the “why” data. This is the data that lets the organization know what it needs to do to get back on track.
 
Examine four questions in analysis
Organizations then move into the analysis phase. Once again, simple questions can help ASCs understand their data. Organizations should answer four key queries when analyzing their data, including questions about:
 
1. Magnitude. “The simple question here is, ‘Is the data indicating too much or too little?’ So we have 14 falls, we don’t know if that’s too many falls or too little reporting. We have to be able to answer that question,” Rohde says. “One of the key tools there is benchmarking, you have to compare to something. Either compare to what someone else does or compare it to our own historical performance. That helps us answer the simple question about magnitude.”
 
2. Direction. The question here is, “Are we getting better, or are we getting worse?”
 
3. Variability. The next question is about the variability of the data. “Is this data pretty steady, or is it bouncing all over everywhere,” Rohde says. Control charts are a good tool to help answer that question. If data have a big spike in them today, does that mean anything, or is that just a spike in the data? Be careful that the data aren’t lying to us. We would hate to change our whole organization on a blip in the data.
 
4. Rate. The last question is, “Do I have to jump on this right now, or do I have some time to think about it?” Rohde says. “How fast is the process moving? Is this something that’s going to cause a terrible disaster if we wait until next Friday, or is this something we need to work on over the next two years?”
 
Present data in an understandable way
Once organizations answer those questions, they need to be able to present the data in a way that’s understandable and meaningful to the governing body and the outside world.
 
One way to present data is with graphs, such as histograms or Pareto charts. This allows you to compare one type of data against another, but it does not look at timing.
 
A second way to present data is with time series charts. These charts don’t compare different types of data, but they assess how things have changed over time. Histogram graphs and time series charts should be used during data presentation. Make sure your data presentations answer the “simple questions.”
 
“You need to be able to compare risk, or importance, you need to be able to compare whether something is changing over time, and you need to be able to bounce back and forth between those presentations,” Rohde says.
 
The third way to present data is with a dashboard. Just like a dashboard in a car, one used for presenting data should only contain the most crucial pieces of information, not every piece of data that the organization collects.
 
“One of the key concepts in the dashboard is the same as in your car,” Rohde says. “In your car, you have the speedometer or the tachometer that have red lines on them. As soon as the needle hits those red lines, you know you need to pull over and do something different.”
 
Similarly, the data should allow you to tell your staff or your governing body, “We just hit a red line and we need to do something immediately.”
 
“That red line either says patient safety is now significantly at risk, or quality of service and satisfaction is at risk, cost is at risk, or perhaps reimbursement or accreditation are risk,” Rohde says. “We’ve reached a dangerous point.”
 
One mistake some ASCs make is to cry wolf by having too many indicators colored red. If something is 15% or 20% off the mean, that shouldn’t trigger a red flag. Red should only be used for imminent danger.
 
“You can present information 100 different ways on a dashboard, but remember the fundamental purpose of that dashboard is, first, is it safe to operate, second, what do we have to change, and, third, what are we doing successfully.”
 
Go to www.hcmarketplace.com/prod.cfm?id=5970 for more information about Rohde’s book.



Want to receive articles like this one in your inbox? Subscribe to Patient Safety Quality Monthly!

Comments

0 comments on “Ask simple questions to manage your data ”

 

    Patient Safety Monitor

  • Patient Safety Monitor

    As part of your Patient Safety Monitor membership, you'll receive Briefings on Patient Safety. In this 12-page monthly...

  • Patient Safety Monitor Alert

    This e-mail newsletter provides healthcare professionals with the latest patient safety news, while offering useful...

  • Patient Safety Quality Monthly

    Ken Rohde, Senior Consultant for The Greeley Company with over 25 years of experience in quality management. His roles in...

  • Occurrence Reporting:

    Take advantage of the information occurrence reports provide and make sustainable enhancements at your facility. Expert...

Most Popular

Related Articles