How Do You Improve the Quality of EHR Data for Healthcare Analytics?

A month or so ago I wrote a post comparing healthcare big data with skinny data. I was introduced to the concept of skinny data by Encore Health Resources at HIMSS. I absolutely love the idea of skinny data that provides meaningful results. I wish we could see more of it in healthcare.

However, I was also intrigued by something else that James Kouba, HIT Strategist at Encore Health Resources, told me during our discussion at HIMSS. James has a long background in doing big data in healthcare. He told me about a number of projects he’d worked on including full enterprise data warehouses for hospitals. Then, he described the challenge he’d faced on his previous healthcare data warehouse projects: quality data.

Anyone that’s participated in a healthcare data project won’t find the concept of quality data that intriguing. However, James then proceeded to tell me that he loved doing healthcare data projects with Encore Health Resources (largely a consulting company) because they could help improve the quality of the data.

When you think about the consulting services that Encore Health Resources and other consulting companies provide, they are well positioned to improve data quality. First, they know the data because they usually helped implement the EHR or other system that’s collecting the data. Second, they know how to change the systems that are collecting the data so that they’re collecting the right data. Third, these consultants are often much better at working with the end users to ensure they’re entering the data accurately. Most of the consultants have been end users before and so they know and often have a relationship with the end users. An EHR consultant’s discussion with an end user about data is very different than a big data analyst trying to convince the end user why data matters.

I found this to be a really unique opportunity for companies like Encore Health Resources. They can bridge the gap between medical workflows and data. Plus, if you’re focused on skinny data versus big data, then you know that all of the data you’re collecting is for a meaningful purpose.

I’d love to hear other methods you use to improve the quality of the EHR data. What have you seen work? Is the garbage in leads to garbage out the key to quality data? Many of the future healthcare IT innovations are going to come from the use of healthcare data. What can we do to make sure the healthcare data is worth using?