You can learn a lot from analyzing data, but after working on a time series analysis for almost a year, Nathan Tompkins will tell you that one of the most valuable lessons he learned is to think about what the data leaves out.
Tompkins is a Quality Improvement Facilitator at Providence Care in Kingston, Ontario. Providence Care operates Providence Manor, a long-term care home, Providence Care Hospital, which combines long-term mental health and psychiatry programs with physical rehabilitation, palliative care and complex care, and 18 community-based rehabilitation and mental health services across southeastern Ontario.
While attending the Masters of Science in Healthcare Quality Program at Queen’s University, Nathan conducted a three-year retrospective time-series analysis on reported falls at his workplace. The experience taught him a lot.
“My whole goal was to see if I could use the patient safety reporting system fully to do a time series analysis,” says Tompkins. “I wanted to see if [the analysis was possible] the way our system was designed.”
Tompkins experienced the typical struggles associated with reporting bias and missing information when analyzing data from incident reports. But there were also some quick lessons learned that could greatly improve the potential of their reporting system. For example, the exact time when an incident occurred was not being captured in reports, with a drop-down menu of two-hour timeframes, which made it hard to isolate fall patterns and link when falls occurred to specific activities on the units. When Providence Care recently upgraded to RL6, this was changed in the new system.
“When going through the process of doing a data analysis, you learn what all the possibilities are in the system, you learn what data is available and what you can do with it,” says Tompkins.
Conducting a time-series analysis also gave Tompkins an opportunity to step back and look at the big picture. During this research project, Tompkins used statistics to identify patterns proactively in the clinical setting. However, for healthcare organizations, it is not always feasible to devote the time and resources necessary to analyze the data from incident reports using such a comprehensive statistical approach. As such, Tompkins hopes his experience translates into improved safety for patients, clients, and residents through future opportunities to conduct focused and smaller-scale analyses on incident reporting data.
But, says Tompkins, data is nothing without context. He explains that numbers can’t account for everything – like when the number of reported falls changes over time. Is it because something has changed on the unit? Or, have reporting patterns changed for some reason?
These types of uncertainties are why Tompkins believes that combining quantitative and qualitative information is key to presenting the whole picture. “It gives us a fulsome picture of what the problem could be and what potential opportunities might exist,” he says.
And while Providence Care’s move this spring to a new facility limits the translation of Tompkins’s findings into improvements on the units, he says that his broader experience in quality improvement has taught him that supporting data with stories is one of the best ways to encourage change.
“If you’re going to change anything then you need to get to the meaning behind what it is that you’re doing,” says Tompkins. “You need everyone to understand and have a shared vision of what it is that you are trying to do and trying to improve – those stories will help you do that.”
The RL Suite can support you in achieving quality improvement in your organization. Learn more about the software or book a personalized demo.