Business intelligence in healthcare: Special report

In this section, you'll learn the data management challenges hospitals and healthcare groups face and the actions some organizations have taken to improve their data mangement programs and processes. You'll also find expert advice and best practices and key data management definitions and terminology for the healthcare industry.

Business intelligence in healthcare: Special report
Business intelligence in financial services: Special report
Business intelligence in retail: Special report

Healthcare business intelligence (BI) news and case studies

It's nearly impossible to turn on your TV or pick up your local newspaper without reading about the healthcare industry. It's a hot button issue all across the U.S. and while Washington debates on the best way to reform the nation's healthcare system, healthcare groups and organizations are doing what they can to provide better and more accurate care for patients with the help of data mining, BI software upgrades and new data analysis strategies.

BI software helps hospitals fight swine flu With millions of students returning to school this month, public health officials are bracing themselves for a resurgence of the H1N1 virus, also known as swine flu. But, in addition to flu vaccines, health officials in New York and New Jersey have another weapon to fight the virus: BI software. Emergency Medical Associates (EMA), a consortium of doctors that contract to staff and run 21 New York and New Jersey hospital emergency departments, is tapping BI software from SAP Business Objects to monitor patient data to detect upticks in symptoms like cough and fever that could portend an onslaught of swine flu cases. Learn how EMA plans on using Business Objects and if hospitals will (and can) follow in their footsteps.

BI issues and challenges in healthcare
BI operations can be a challenge for any company, but when it comes to the healthcare industry there's an added layer of complexity to muddle up the process. Healthcare organizations, such as hospitals, outpatient clinics and HMOs, collect and analyze not only patient financial information but also sensitive data about a patient that's governed by strict privacy rules.

The Health Insurance Portability and Accountability Act (HIPAA) is a prime example of the delicate nature of healthcare data and how businesses deal with the data. Since HIPAA requires extra security and administration of the security, healthcare groups have to think carefully about how they structure their security models, so only qualified individuals can look at patient private information. It's It is possible to be successful with business intelligence in healthcare, according to expert Elizabeth Pappius, but can be challenging for some businesses.

Some healthcare organizations may struggle when it comes to structuring the BI team and determining where it fits within the organization. Pappius said she's seen the BI team organized in several ways, including sitting under the chief medical officer (CMO), chief financial officer (CFO), chief operating officer (COO) and IT at healthcare organizations. Each structure has its advantages and disadvantages, Pappius said, but in the end, the best place for the BI team to be situated is with the group that'll give the team the support, attention and resources it needs to be successful and with a group with an executive-level manager to champion the BI team's cause.

Since HIPAA requires extra security and administration of the security, healthcare groups have to think carefully about how they structure their security models...

Data quality and data consistency for electronic medical records is another issue that can plague healthcare groups, as many systems are configured differently. And though they create for better patient care by eliminating redundant testing, lost test results and medication errors, the data coming from those systems that analysts use for BI tasks are rarely useful unless there's a lot of transformation of the data, Pappius said. And, many organizations have to combine data from different sources.

The end result: huge data consistency issues.

A lot of reporting and analysis in healthcare groups is conducted ad hoc, but unfortunately there aren't many BI vendors or products in the market that are a great fit with the type of reporting and work medical BI analysts need to do. There's nothing wrong with a lot of the work being ad hoc, Pappius said, calling it the "nature of the beast." What healthcare companies need to do, she said, is leverage only the features in BI products they need and no more. She also said some businesses might need more than one tool to meet their ad hoc and standard reporting needs.

Improving BI in healthcare companies
Although optimizing BI can be a difficult for healthcare groups, the good news is that some organizations have been successful in this effort and have seen a significant payoff from business intelligence. University Health Network (UHN) did just that when it attempted to improve its BI capabilities by signing on with BI heavyweight Cognos and, in the process, secured favorable contract terms for its partner hospitals.

Prior to the BI initiative, UHN used a homegrown data warehouse developed in an ad hoc fashion. The major issue was that it caused three nearby teaching hospitals to lack important information held by UHN's finance and human resources departments. At the time, UHN used corporate scorecards to monitor everything from patient satisfaction to infection rates; however, the data came from several unconnected sources, meaning long, tedious hours of data entry that resulted in a lot of errors and a demoralized staff. UHN knew changes needed to be made and began the vendor and software evaluation process ultimately choosing Cognos.

When the implementation is complete, UHN's finance and human resources data will be incorporated into a "refurbished" data warehouse, its corporate scorecards will gather data from consistent sources with built-in accountability mechanisms and non-management employees will have a wider array of operational-level, self-serving reports to choose from.

Improving care and lowering costs
Before implementing BI software, Jefferson Medical Center didn't have a good way to monitor employee productivity. The hospital believed it could improve care and lower costs if it could measure productivity more effectively, according to one hospital employee. The hospital faced some unique challenges, however, including receiving more than 48,000 patients in ER with a larger number being low-income patients and constantly trying to invest in new healthcare technology and improve its services.

Although optimizing BI can be a difficult for healthcare groups, the good news is that some organizations have been successful in this effort...

The medical center also wanted to track employee productivity but the paper reports handed to managers were often two months old and some reports were based on incorrect data. To counteract this, Jefferson Medical Center turned to WebFocus BI for help. The system pulls data from its patient accounting software, payroll software and other internal systems. It's also integrated with third-party sources, such as U.S. benchmarking data, enabling the hospital to see how it's doing relative to national averages. If the hospital finds disparities between internal metrics and national averages, it studies its processes and decides if and how changes should be made.

Data mining in the healthcare industry
According to business intelligence expert William McKnight, there are numerous uses of data mining in the healthcare and related industries (i.e., biotechnology, pharmaceuticals and healthcare insurance) that provide a lot of advantages to businesses. Mastering some data mining techniques, McKnight says, can mean more and faster analysis.

And data mining is widely (and successfully) used in the healthcare industry; for example, for mining claims to determine the best providers and procedures for particular conditions, determining diagnostic aids for certain procedures and protein analysis for drug development.

Enterprise master patient index (EMPI): Healthcare's MDM
As organizations continue to mature their data management programs many are looking towards master data management (MDM) as a way to ensure they have the highest quality data available at their fingertips -- and the medical and healthcare industry is no exception.

For healthcare professionals, an MDM program often comes in the form of enterprise master patient index (EMPI), a form of customer data integration (CDI) specific to the healthcare industry. Some organizations are using EMPI to fuse patient identities across systems and then use that data to feed even stronger data to the data warehouses for better analytics.

And according to experts, MDM in the healthcare industry can lead to better screening, diagnosis and treatment decisions and allows hospitals to personalize patient care.

• Read the next section -- Business intelligence in financial services
• Want more healthcare BI and data management news? Check out our vertical industry resource center

Business intelligence in healthcare: Special report
Business intelligence in financial services: Special report
Business intelligence in retail: Special report