Saturday, May 18, 2024

Case Study: Hospital’s Prescription: A Dose of Database Software

Datamation content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

Technology has led to countless lifesaving medical advances, such as the mechanical heart, the chip-embedded pacemaker and the CAT scan. Now, technology is catching up to an age-old health care problem — medical errors caused partly by physicians’ poor handwriting.

CareGroup Healthcare System Inc., a $1.6 billion, 12,000-employee operator of six hospitals in the Boston area, recently installed a database software system from InterSystems Corp., of Cambridge, Mass., that takes the guesswork out of pharmacy orders.

InterSystems’ Web-oriented post-relational database technology is being used to run CareGroup’s new Physician’s Order Entry system, or POE. With the POE, instead of writing orders in their own hand, doctors order drugs from any Web browser on any of the hundreds of PCs throughout the hospitals.

The system works via a seamless data connection built on InterSystems’ high-performance Cache database, its flagship product. Among the system’s benefits is helping to eliminate medical errors often caused when scribbled doctors’ orders lead to the incorrect drug or dosage being delivered to hospitalized patients.

But the POE goes far beyond the handwriting problem, leveraging database technology to pull together critical information that helps doctors make the wisest drug decisions while leaving little room for mistakes to take root, according to Dr. John Halamka, CIO of CareGroup.

For instance, he says, at the moment doctors are keying a prescription into a database, the POE links them to, among other things, that patient’s medical history. The system advises the doctors on the medically recommended drug options for the patient’s exact condition.

It also cross-checks the patient’s record, flagging any potential drug interactions, patient allergies, duplicate drug orders and other factors that could lead to medical trouble. “If you try to adjust the dose too high or too low, it wont let you,” says Halamka.

Also, the Cache database links the prescription-writing process to lab results and other patient data, such as the digital x-rays and other medical images (which are 100% digital), to help doctors with their orders. Doctors can also read fresh monthly content updates on medical and pharmacy best-practices.

Reducing Human Error

Once the order is relayed to the in-hospital pharmacy system, it activates a device that automatically opens the door to a container that dispenses the proper medicine, which is date and time stamped.

Together the structured system provides checks and balances that heretofore have been prone to human error, according to Halamka, who also serves as associate dean of Harvard Medical School and is an emergency room physician.

The system has been in use since June. “We believe we’ll remove at last 50% of all error that takes place in medicine,” Halamka says. He estimates 7 out of 100 patients in American hospitals experience a medication error. “From a patient perspective, there’s a sense of a safety net” as a result of the new technology. The doctors and nurses also feel that the new system is effective by serving to double-check their work, he adds.

Medical errors are a persistent and costly problem for the health care industry, reportedly leading to more deaths each year than motor vehicle accidents or breast cancer. Each “adverse medical event” costs a hospital an estimated $5,000, Halamka says.


It’s easy to see how it could become a problem. Under traditional hospital practices, “If a doctor wants to give you a medicine, he writes (a prescription) on a carbon paper and faxes it to the pharmacy,” Halamka says. If the handwriting is garbled, “the pharmacist might ask himself, is that 5 mg, or .5 mg? Is that in metagrams? You can imagine how dangerous it is to deliver medicine that way.”

The old process has other flaws. Once the pharmacist keys the prescription into the computer system, it’s filled by a human and delivered to a nurse for delivery to the patient. “At every handoff there’s a chance for a mistake,” Halamka says.

1 Million Patients’ Records Accessible

Enter InterSystems, which provides high-performance post-relational database software systems for Web applications to more than 4 million seats worldwide. Its flagship product is Cache, a post-relational database for e-business that helps enterprises access and inter-relate vast volumes of operational data. (InterSystems is considered the leading database software vendor in the health care sector.)

Halamka says Cache does an excellent job at helping manage a huge data load — in CareGroup’s case, 21 terabytes of data, including records on about 1 million active patients. (Its Boston-area hospitals include Beth Israel Deaconess Medical Center, Mount Auburn Hospital, New England Baptist Hospital, Deaconess-Waltham Hospital, Deaconess-Glover Hospital and Deaconess-Nashoba Hospital.)

CareGroup runs Cache on clustered Hewlett-Packard Unix servers. It uses storage technology from EMC. Its hardware is located in a secure, 13,000-square-foot data center. To protect the system, it runs backup generators and employs redundant telecommunications infrastructure to prevent the system from going down, no matter what might happen. It also uses advanced security to authenticate its users, including RSA SecurID tokens to verify users.

Halamka says the database system cost $2.5 million to develop and install; annual support and maintenance costs about $250,000 annually. It took 18 months to customize the system, including the work of 23 programmers. A team of 35 clinicians (doctors, pharmacists, nurses, social workers) worked together to ensure features that were important to them were incorporated, including current pharmaceutical updates, and continue to add new content to the system.

The POE can be accessed from nearly anywhere — from the hospital floor to doctors’ homes — using a Web browser. CareGroup maintains 8,000 PCs, or about one PC for every two hospital beds. The POE system is also wireless-ready and can be accessed via wireless laptops and handheld devices, as well as from off-site computers.

There are few other hospitals in the U.S. using such database system, so the eyes of other hospital CIOs are on them, Halamka said. In addition to health care applications, InterSystems’ Cache is used in transportation, financial services, law, telemarketing, and retail, among other industries — any business that needs high-performance, highly scalable processing applications to automate core operations.

Privately held InterSystems was founded in 1978, has 350 employees and its software is licensed to 4 million seats worldwide on 80,000 systems.

Subscribe to Data Insider

Learn the latest news and best practices about data science, big data analytics, artificial intelligence, data security, and more.

Similar articles

Get the Free Newsletter!

Subscribe to Data Insider for top news, trends & analysis

Latest Articles