“So … what exactly am I looking at here?”
If those words sound familiar, you’ve probably produced or consumed reports in a hospital operations setting. In many organizations, report packages cobbled together from several data sources are the basis of critical decision-making. They’re vital to everyday activity, yet they’re difficult to produce, maintain and deliver in a timely manner, and they often raise questions that can’t be easily answered without additional research and analysis. In other words, they can be hard to trust.
Predictions, even very accurate ones, don’t engender a lot of trust without understanding historical performance, and historical metrics are questionable without the ability to drill down to the individual patients or jobs that make up those metrics. For example, we’ve learned that the ability to explore the individual records that make up a data point will go a long way toward believing the data truly represents reality. People are naturally drawn to the “corner cases” – a few records or processes that fall well outside the norm. While those cases rarely affect average performance, the ability to access them easily is essential to root-cause analysis and stops conversations from derailing around data integrity.
Here are four tips on how to engender trust in your operational and clinical data:
- Before getting into advanced analytics, make sure stakeholders first have a good picture of historical operational performance. Whether it’s OR block utilization, inpatient discharge processing time, or ED visit breakdowns, having a solid shared understanding of historical performance sets the stage for the next level of discussion. Then, it becomes much easier to drive action based on demand forecasts, discharge predictions, nurse staffing level recommendations, OR block re-allocation recommendations, and more.
- Be open with definitions. Explain the meaning of every metric in plain English, as well as the policies that go into each calculation. There may be internal debate on whether a particular metric is a valuable one, but no one should have to question what a metric is supposed to represent.
- Make reporting drillable down to the record level. When questions inevitably arise about where a number comes from, it should be easy to dig into individual surgeries, visits and jobs to see what happened.
- Lastly, try to keep it as simple as possible. Sending out too much historical reporting at once can cause fatigue. Focus on key performance indicators that are tied to your organizational goals and give them precedence over all others. As you review, remove metrics that don’t matter to your organization and include new ones that do.
At Hospital IQ, we place huge value on data trust; making sure you trust your data is central to our mission. Gaining data trust is key to operational improvement. Historical reporting is only the beginning – it’s the first rung on a ladder that leads up to the complex analysis, forecasts of future demand and prescriptive recommendations for action that really drive a hospital’s or health system’s bottom line.