High Performance Health Care

Increasing attention is being paid to health care IT, particularly as President Obama continues to mention it as one of the key tools needed to slow the rising cost of health care. Due to this increasing interest, a progressive collection of clinicians and major IT firms have advanced the concept of Health 2.0, modeled on the emerging concept of Web 2.0. Whereas the original Web 1.0 was one way – Web sites and email, Web 2.0 is much more community- oriented with social networking, blogs and wikis.

The Health 2.0 proponents believe these and similar tools can be used to radically transform the way health care is delivered. A number of major firms (Google Health, Microsoft Health Vault, and Web MD) now allow individuals the ability to maintain their own medical records online through an Internet tool know as the Personal Health Record (PHR). The great question that will shape the future is: Who is going to maintain these medical records? These larger firms are betting that individuals will want to own and manage their own records.

As opposed to these systems, Tang and Lee make the case for the “integrated PHR” in a recent New England Journal of Medicine article. This type of system is connected directly to the providers’ own Electronic Health Record systems. It allows the patient direct access to their own data, scheduling resources and the health care team. The level of access and interaction can be set by the providers. Early patient responses are very positive. Tang and Lee characterize the integrated PHR as a “try it you’ll like it type of innovation.”

The New York Times also recently examined this phenomenon. Dr. Ashish Jha of Harvard who studies electronic health records says – “We still have a long way to go” in the widespread use of Personal Health Records.

Some interesting statistics on the use of Web 2.0 are contained in Groundswell by Li and Bernoff. Here are their categories of users (in 2007) and the percentage of Americans in each group:

• Content creators – Web sites, blogs and videos - 18%
• Critics – people who write responses to content – 25%
• Collectors – online participants who put tags on websites or RSS feeds – 10%
• Joiners – individuals who maintain profiles on Myspace, Facebook or Linked in – 25%
• Spectators – readers of Web sites, blogs and other media – 48%
• Inactives – people who have access to the Internet but do not participate in any level of social online activity as described above – 48%

Is this really happening today in our community? Recently, I was teaching health care operations in our Mini MBA for Health Care Management. The topic was using the Balanced Scorecard to implement strategy, and the students were creating strategy maps. One primary care physician in the class was creating a map for the care of diabetic patients in her practice. One of the key initiatives in the operations area was to implement e-visits for these patients. I asked her about it and she said her practice was already experimenting with e-visits and they seemed to be successful. They were beginning to use all the Health 2.0 tools to improve their connection to their patients.

The Web is transforming the delivery of health care. Once financial incentives are aligned to support this transformation, it is likely that industry interest will intensify. The competition between the stand-alone PHR and the integrated PHR may determine whether I will be getting a significant portion of my health care from Dr. Google in the future.

References:
Kaplan, Robert and David Norton. The Strategy Focused Organization. Harvard Business School Press, 2000.

Li, Charlene and Josh Bernoff. Groundswell. Harvard Business School Press, 2008.

Tang, Paul C. and Thomas H. Lee. “Your Doctor’s Office or the Internet? Two Paths to Personal Health Records.” New England Journal of Medicine. 360; 13. March 26, 2009.

Posted by Dan McLaughlin at 03:35 PM | Permalink | Comments (0)

The New York Times recently reported that “The nation spends billions of dollars a year on patients’ return visits to the hospital — many of which are readmissions that could be prevented with better follow-up care.” (This finding is based on April 2, 2009, article in the New England Journal of Medicine.)

The Wall Street Journal also reports that “MedPac, a commission that advises Congress on Medicare policy, has recommended that Medicare start a pilot program in which “bundled” payments extend beyond the first hospital stay to include, say, the first 30 days after discharge. The idea, which is also part of President Obama’s budget proposal, is that if hospitals get paid fixed rates for caring for certain conditions — and they don’t get paid more for those same conditions if patients return — hospitals will have a financial incentive to reduce the risk of readmission.”

Many hospitals are now focused on preventing hospital readmissions and one of the most basic strategies is to assure that each patient has a follow-up clinic visit scheduled.

The Mayo Clinic Data Mining Study
At the recent Minnesota Health Services Research Conference, five investigators from the Mayo Clinic presented a unique research project that used automated “text data mining” to determine whether a hospital discharge summary contained a follow-up clinic appointment.

A dataset consisting of discharge records was manually reviewed to determine whether the records contained follow-up appointment instructions. The same dataset was evaluated for the same criteria using SAS® Text Miner 3.1 software. The two assessments were compared to determine the accuracy of text mining.

Of the 6,481 discharge records reviewed, 3,576 (55.2%) were identified as containing all criteria for follow-up appointment instructions through manual review, 113 (3.2%) of which were missed through text mining. Text mining incorrectly identified 107 (3.7%) follow-up appointments that were not considered as valid through manual review. Therefore, the text mining analysis concurred with the manual review in 96.6% of the appointment findings.

The Mayo researchers concluded that text mining of medical records can accurately detect whether elements of follow-up appointment instructions are documented in hospital discharge notes. The results also suggest that text mining software can be used to identify specific appointment criteria in a large number of textual medical records, thus saving considerable resources required for manual abstraction in quality-related research and performance assessment.

The results of this research project provides a platform for Mayo to automatically track its clinic appointment follow-up after hospitalization, develop new strategies for improvement and reduce its hospital readmission rate. Today, most automated electronic medical records contain a substantial amount of information that is stored as text. Data mining provides a powerful tool to now use this data for operations improvements.

Business Intelligence
Many non-health care businesses routinely use very powerful analytical tools to operate their enterprises. These strategies are generally considered to be part of the emerging field of “business intelligence.” BI practitioners use software to copy data from operational systems into data warehouses, use powerful analytical tools on this data (data mining) and then implement operational real time decisions using automated business rules. The Mayo team used data warehousing and data mining for their study.

To understand the power of BI, consider its use by Harrah’s casino in Las Vegas. The casino tracks every customer’s gambling wins and losses in real time through the use of their customer “total rewards” magnetic cards. By using data mining tools they have determined that each customer group has a “pain point,” which is the amount a customer can lose in one day and still come back for more. Once this point is exceeded, however, the customer may not return.
Therefore, if a customer in a group (e.g. 34-year-old female, upper class zip code) loses $900, Harrah’s business rules software senses this event and sends a Harrah’s employee immediately to the customer. This “luck ambassador” then says “I see you are having a bad day. I know you like our steakhouse so I’d like to take you to dinner on us right now.” The customer’s pain has been converted into a positive experience.

Health care Business Intelligence is not yet mature. However the expansion of health information technology throughout the care delivery system is beginning to create large databases. Hopefully, BI tools will be used in the future to create innovative solutions to improve the delivery of care – just as the Mayo team nicely demonstrated.

References:

Ayres, Ian. Super Crunchers. New York: Bantam Books, 2007.

Ruud, Kari, MEd; Matthew Johnson, MPH; Carrie Schinstock, MD; Juliette Liesinger; James Naessens, ScD. “Accuracy of text mining in identifying follow-up appointment criteria from hospital discharge records.” Mayo Clinic Study.

Posted by Dan McLaughlin at 10:09 AM | Permalink | Comments (0)