Much of the conversation about the relationship of analytics and healthcare focuses on the potential of big data
to change the healthcare industry in the years to come. However, in reality analytics can play a role for providers and hospitals right now as a tool for assessing their adoption of an EHR system, their performance as part of an accountable care organization (ACO) or patient centered medical home (PCMH), or their readiness to connect to a health information exchange
“In almost all cases, when you first go to a provider and say, ‘Here’s what our data is telling us about you,’ nine times out of ten the provider says, ‘But no, that’s not what I do. I deliver much better care,’” says Greg Chittim, Director of Analytics, Quality Improvement, and HIE at Arcadia Solutions.
With healthcare reforms shifting their focus from pay for service to pay for performance, they require that their participants (i.e., providers) report how effectively they are meeting key measures for quality care (e.g., clinical quality measures
, meaningful use
). The disparity between perceived and reported performance, therefore, could place a provider, practice, or hospital in jeopardy of falling short of important thresholds. In light of what’s at stake, providers require tools to identify and correct potential gaps in capturing data appropriately.
The simplest and most direct solution to prevent this type of outcome is to apply analytics from the get-go, beginning with the EHR system:
In an ideal case, you actually bring analytics live as soon as you bring your EHR live. You do them in conjunction because analytics is one of the best ways to measure and very quickly course-correct on the adoption of electronic health records. So it very quickly allows you to instrument your go-live process to understand who’s using the EHR, who’s not, what problems might they be having, do we need to roll out some additional training, do we need to build some additional templates, do we need to have a broader program . . . things like that.
More simply, analytics can help ensure that data quality accurately reflect care quality.
Producing quality data enables quality exchange
Low-quality data are generally the result of inconsistencies in how measures are recorded and in what format. Through a performance improvement program, providers are able to locate where the discrepancies or gaps reside. “Often, consistency in data means getting things into a structured format, but it’s balancing that getting things into a structured format with making it work within a provider’s office,” Chittim explains. “Quality of data is often one of the easiest fixes to make in terms of EHR adoption and usage because it is often just an issue of giving providers the right training and the right information at the right time.”
The quality of data is crucial not only to the providers themselves but also to any organizations that rely on this information in an actionable way (e.g., HIEs). To prove useful, analytics solutions need to first get the data in order in the provider’s eyes. “The easiest way for any kind of aggregation program like this to fail — whether it is a quality improvement reporting and measurement system or a private or statewide HIE — is for the providers to not trust the data,” continues Chittim, “And as soon as they don’t trust the data, it’s very hard to recover from that.” If providers refuse to address data quality at the EHR level, they could undermine the efforts of HIEs and prevent their patients from ultimately benefitting from the sharing of information.
“Content is one of the underrepresented issues facing both private and public HIEs in that HIEs are valuable because they bring data together and allow you to do powerful, scalable things with them. But HIEs make this assumption that data coming into them are high quality,” argues Chittim. “For HIEs to be valuable they need to be getting data from EHRs where that quality of data matches the quality of care that the providers deliver.”
While analytics is expected to take hold in future phase of healthcare reform, it could perhaps prove just as useful presently in ensuring high-equality data down the road.
Related White Papers: