Big Data analytics has made big strides in the past few years. But why have its adoption and use in the healthcare sector slowed to a crawl?

Is Healthcare Data Analytics Ready for Prime Time?

On October 16^th, Shahid Shah, CEO of Netspective Communications, joined me in attending the 4^th  Annual Biomedical Informatics Symposium at Georgetown University in Washington DC.

The really hot ticket was a lively discussion of the usefulness of emerging healthcare analytics technologies. The format was a Point Counterpoint-like debate of the real-life benefits and limitations of the latest data analytics technologies. Here’s a summary of the discussion.

Pro: We can use practice-based evidence generated by EMR data right now.

Dr. Nigam Shah, PhD of Stanford University walked us through the promise of healthcare data analytics based on patient data.

• Data sources. Patient medical records are used to predict and classify patient conditions and risk of their getting certain conditions. EHR as well as many other data sources such as social media and medical electronics sensor readings are used.
• How data is used. The patient feature matrix is the basic way to assemble data used in healthcare analytics. Analysts use patient profiles to determine health risk factors, look for specific data patterns and learn the health effects of specific variables.
• Benefits: Successful healthcare analytics can lead to:

• More informed decisions and actions of patients and providers.
• Promoting evidence-based practice based on predicative modeling, not hunches. or anecdotal evidence.
• Better understanding of drug use and its impact.
• Monitoring quality metrics.
• Patients getting a better understanding of the practice of medicine.

But, for some medical professionals, the picture isn’t quite so rosy.

Con: There are too many problems to implement healthcare data analytics now. `

Michael Hogarth, MD of the University of California at Davis, took a not-so-fast-buddy look at current healthcare analytics. He described very real problems, which could reduce patient safety and increase costs of medical practice.

• Data complexity. The EHR environment includes many consumers and huge volumes of data. There’s lots of data, but it’s difficult to use it.
• Data quality. EHR data quality is not very good. The lack of standards make data accuracy, completeness and access to clinical records uneven at best.
• Data access. Many users get little or no access to metadata and data sets needed to do predictive analytics.
  • Data integration. There’s a big problem linking medical records, especially free-standing data not connected to EHR. There’s a need for a tethered approach, which connects EHR and other digital information.
• Unstructured narrative data. Much important data comes as handwritten notes and other formats that computers can’t read.
• Organizational rules and best practices. Some organizations limit data access and what medical professionals can do with it.

Unfortunately, all of these problems also apply to people seeking and interpreting medical travel information.

Medical Travel: Same Data Problems No Matter Where You Go

For medical travelers and healthcare providers in the U.S. and overseas, the lack of accessible, relevant data limits patient treatment options and provider revenue.

Of the data-related problems described above, data access, integration and quality (completeness and consistency) have the biggest effect on potential medical travelers.

If you want all the gory details, you can read an earlier post, which describes how data quality and quantity limit the effectiveness of medical travel services.