Saturday, January 12, 2013

Quality and Integrity of Information in Healthcare

Quality and Integrity of information is only as good as its solution documentation and resource accountability. The best way to measure info quality is to detail (Zachman style) a failure of the system, figure out accountability and share the understanding of what went wrong and how to fix it. For example, in a healthcare system, when the registration system goes down, presumably it fails over to a hot swap database which is current with account numbering. If the account numbers are not in synch the risk is high for duplicate encounter numbers to be given to patients.

The day of the failure and duplicates goes by, patients are admitted. But, the next day when their bracelets are scanned for medication some patient’s account numbers are rejected because the patient names don’t match. The doctor or nurse looks up a CT Scan by the account number and sees another patient with a different name in the search result. An admin scans a consent form, enters the account number and scans in the form under a different name.

So, what happened here? Why were duplicate account numbers permitted to enter the Electronic Medical Record system in the first place? Well, this is apparently, okay according to the information architecture -- when merging two patients this could be a scenario. What? ECM solutions with systems like Documentum would call this a relation, but would never “merge”, thus erasing the history (and integrity) of information.

After discharge the patient goes home, and the hospital forms are scanned and indexed. When the QA person looks at the actual scanned form, she sees a different patient name. How will this get fixed? The Hospital Information Management department needs to process these patients through coding so the information can get sent to the insurance company and the hospital can get paid for services.

First, what happened? Was the registration faulty failover to blame? It introduced the duplicates and team did not even know it. The EMR team did not know what happened. The integration team did know about it, but didn’t sound the alarms. This issue was exposed by the nurses with medications and admins trying to scan in the front line.

Second, what was done to fix it? Amid quiet desperation, a slow resolution was worked on. No accountability, no consequences. The issue with healthcare information management is that it is patched together by longtime friends who watch each other’s back – this is not isolated to healthcare, obviously. Personal knowledge over sharing is still the preferred mode of operation in most IT shops.

Knowledge Sharing

Documentation, sharing knowledge, is the first hurdle to get over. It is not easy to revamp inventories of applications, what they do and why. Departments like to hold onto the applications that they “own” even if efficiencies and consolidations at the enterprise level makes sense. So one way to force it is the point to an industry regulation and say, “we must do this to comply with the law”, or risk heavy fines. Or, the security and data integrity risks are too great and HIPPA regulations will nail us.


Accountability, having consequences, is paramount to achieving the minimum quality standards of information needed to function without careless mistakes. For example, if the backup and restore process were tested and signed off on by directors do you think the duplicates would have happened? Accountability helps morale of the team(s) in IT as it exposes what happened, who did it, why they did it, and should offer up the best ways to fix the issues. If the director or manager was negligent, figure out why and fix it. If the architect did not follow protocol, call it out for everyone to learn from. This is not a playground here, this is people’s lives, more serious than systems like hedge fund transaction services and yet financial and pharmaceutical systems run more smoothly and with higher quality.


Quality, doing the right thing, is not achievable without knowledge sharing and accountability. Of course, the strategy and goals filtered down through a strong program management structure is vital as well, but I’m focusing just on a few concepts and an outcome that healthcare IT desperately needs to get right, every time.


There is a gap between personal integrity of workers in a hospital and information integrity of the patient's medical record. Hospitals have to balance the costs of both types of integrity. However, it's time to respect and do no harm to information as well as a person's health. It's time to innovate how we 

No comments: