Blogs

Grappling with IT Challenges in Healthcare

Several obstacles may be encountered when applying technology to help improve healthcare systems

Owner, Forest Rim Technology

If there is one universal sentiment, it is that everyone wishes for good healthcare. And healthcare is undeniably a large industry, consuming a significant part of the US gross national product (GNP). So there are many reasons—all of them good ones—to pay attention to anything that can improve healthcare. But for all the good intentions, performing information processing in healthcare has been problematic, and there are many challenges when processing information related to healthcare.

 

Technical decisions and doctors

In the world of healthcare, doctors are at the top of the pecking order, as they should be. In terms of training and responsibility, this ranking is appropriate and proper. However, when it’s time to make technical decisions—especially strategic decisions—a doctor is not necessarily the right person to make them. Just as doctors have trained all their lives to administer proper healthcare, so technicians have studied to become experts in managing technology. Because medical doctors probably are not qualified to make technical decisions, especially strategic decisions, when they do engage in making technical decisions, those decisions may not be the best ones for their organizations, their colleagues, and their patients.

It’s common within medical organizations to make decisions based on consensus by a committee. While this approach may work well for some classes of medical decisions, when it becomes necessary to drive the vision and technology decisions for the medical organization, making decisions by committee often does not lead to the best outcome. There are lots of reasons why making decisions by committee may not be a good idea. Committees tend to be slow, they require a consensus, and most of all they do not necessarily have a single vision. They often have multiple visions. There are many occasions when the only way to succeed with a technology is to have a singular focus to drive the decision making.

 

Focus on details

It’s generally understood that doctors concentrate on details. When a doctor prescribes medication, the type of medication and the dosage must be considered with the specifics of a patient in mind. Doctors, therefore, have a natural propensity to focus on detail, but that focus can derail the progress that needs to be made in other areas. For example, during a database design meeting for a medical data warehouse, the entire day was spent discussing the definition and nuances of two words that needed to be placed in the data warehouse. Understanding those two words was surely important. But having hundreds of thousands of other medical terms that needed to go into the data warehouse created a big problem. At the rate the data warehouse design was proceeding—two words at a time—it would have taken eons to complete the design.

Consider electronic medical records (EMRs). An EMR is not a communication tool. It is a documentation tool that is used primarily for legal and reimbursement purposes. The software design—current state—is to replicate the paper chart. And except for during a patient’s hospital stay or a visit to a clinic, the purpose of the paper chart is to retain a historical document. By its nature, the design is flawed. It provides data to physicians for care, but physicians and nurses do not communicate “care requests” in the EMR. Nurses and doctors use the EMR to document the care they have already given. The only real repository of communication in the EMR is the ordering action. Doctors use the order function to request—communicate—clinical tests. The flaw in the design of the EMR is that its architecture has started with replication of a paper legal record, not the actions of care that doctors and nurses follow in everyday patient treatment.

 

Textual data

But perhaps the largest constraint to success in information systems for the healthcare community is the notion that healthcare notes need to be in the form of narrative text. Doctors and nurses need to communicate with each other, and it is through healthcare records that most of this communication takes place. Having the information on the healthcare record in the form of narrative text is natural. For clinical purposes and the care of a single patient, narrative text format is what needs to happen.

But the problem is that as long as text is in a narrative form, the computer cannot be used for analysis. If there are 100,000 computerized records that must be analyzed, computer analysis using narrative text is required. The narrative text needs to be put into the form of a database for research analysis to take place. Narrative is good for the treatment of a single patient, but not for the analysis of 100,000 patients.

These obstacles are among several that IT professionals in healthcare face. Please share any thoughts or questions in the comments.