Usability: Computer Systems Can Save Hospitals Money If Done Right

An article on SlashDot today references a study at Harvard that claims that electronic systems in hospitals are not in fact, saving them any money. The post on SlashDot is short and doesn’t really delve into the issues. The actual article that reviews the study in ComputerWorld does however hit the real problem right on the nose: most systems are not designed with health care practitioners in mind.

One of the quotes in the article is particularly pertinent, per the leader of the Harvard study, Dr. David Himmelstein, programmers on successful e-health systems say that: “If you need a manual, then the system doesn’t work. If you need training, the system doesn’t work.”

The missing component for a lot of these unsuccessful e-health systems is the focus on usability and user experience particularly from a clinical perspective. Beyond even gathering sufficient requirements to provide a successful user experience for doctors, nurses and other health care providers, within a hospital or other health care environment, those requirements then also need to be applied properly to the development of the e-health product. Essentially, usability needs to be considered, taken into account and applied throughout the entire product development life cycle and repeatedly re-applied over the course of the lifespan of the product during reviews.

So a part of the problem with realizing cost savings in health care through the application of information technology, is in fact, bad design on the software end and also in some cases, improper data management design that requires too heavy of a load from a hardware perspective. How many machines does an organization have to purchase just to support all of the storage needs for the vast amounts of data that are housed in your average medical record, for instance? Failover, redundancy and backup are also key considerations on the hardware side and can introduce added costs in terms of hardware to ensure 24/7 availability as well as either a ‘hot’ or ‘cold’ failover system that can be brought online to back up the main system in the event of a catastrophic failure.

In order for health care organizations to realize true cost-savings measures, the following principles and steps need to be applied to the purchase and implementation of any system:

  • Requirements. A strong requirements gathering phase so that the needs of the organization are fully understood and documented prior to the consideration of any software system. Ignore the latest buzz about any one system, ignore ‘cool’ factor purchases and concentrate on what your particular organization’s needs actually are. Don’t stint on time spent in the requirements gathering phase, understanding the business needs before making a large purchase is crucial to getting your money’s worth from your software vendor/s.
  • Planning. Once the requirements are well understood, then the purchase needs to be carefully planned before any vendors are brought in for a demonstration. Select vendors based on the requirements that are outlined and make sure that any vendor, whether providing an ‘out of the box’ system that can be customized or a vendor who will be custom-building something largely from scratch understands your requirements and includes an analysis of your organization’s workflow both administrative AND clinical and is able to customize elements of the system to match that workflow.
  • User Experience Expertise. If possible hire a UX expert or appoint a UX expert from within your organization to work with your administrators and clinicians so that their workflow is incorporated into the requirements before talking to vendors. Involve clinicians in the early stages of the project to ensure that their needs will be met by the systems. A UX expert can work with each of these different groups to capture how they work and help to coordinate with the IT end of the equation to make sure that any system under consideration will dovetail with the user experience that the users actually want and need in order to do their jobs efficiently. Bringing in User Experience early and often in health care system purchases can go a long way towards saving dollars in the long run or avoid making a bad purchase.
  • Pilot. Once all the requirements are understood and a decision to build or purchase has been made, strongly consider going into a pilot phase. If possible, get free trial versions of any vendors under consideration, or have the development team build a ‘low scale’ version of the system, then build a pilot group of testers that is a representative sample of your final end user base. Cut them loose on the system. They’ll be able to ferret out road blocks in the system quickly and provide invaluable feedback about what works, what doesn’t work and which systems stand up to the ultimate test of day to day usage. This is just like taking a care out for a test drive and no one really buys a car without taking it for a test drive. Do the same thing with your e-health systems, kick the tires, check all the gears, rev the engine up and see how the system handles at full speed. A vendor-driven demo won’t reveal flaws the way that a test drive or pilot will, though such demos are definitely useful for getting an idea of which system or systems you would like to bring in for a pilot.
  • Don’t Be Afraid To Pull The Plug. While there’s often a lot of back room politics, budget concerns and other elements involved with any system purchase, one of the biggest mistakes that gets made in almost any industry, is fear of backing out once a path has been chosen. If a product does badly in demo, if the pilot is generating a lot of negative feedback, if small problems keep cropping up, don’t be afraid to back up a step or two and re-evaluate. Bring in a different system to pilot. Re-visit some of the requirements. What are the key priorities with the system purchase? Is it meeting the needs of the end user? If not and there’s no way to meet those priorities and needs, pull the plug. Far less money will be wasted by backing out before making a final purchase, investing in customizations, documentation and training. Selecting the right system, getting it right the first time can save an organization from making a purchase that ultimately costs more and renders the organization less efficient.
  • The User Is The Ultimate Customer. At the end of the day, any efficiencies or cost-savings are going to come back to the usability of the system. Are the people who have to use it every day going to be able to use it with a minimum of training, without a manual? Are they going to be able to seamlessly work the system into their day-to-day work, using the system as a replacement for what they currently use today? Is the system going to offer improvements over what is currently in use? If the users are not satisfied at the end of the day, have not been empowered to do their work either more quickly, efficiently or accurately, then the customer’s needs haven’t been met.

Two real world examples provide a comparison of approach.

System A, implemented on the East Coast, in spite of hours of time spent by the vendor’s analysts shadowing doctors and figuring out workflow, provided a system that did not successfully take all that observation of workflow into account or apply it well to the final product.

The system in question creates backlogs of work for nurses working in specialty clinics especially, because it is not tailored for the specialty. The system apparently also experiences gradual slow-downs over the course of the day as usage peaks, grinding down to a crawl by the early afternoon hours. This likely points to some flaws in the hardware and database set up on the back end. However, more critical is the experience of the nurses using the system and the fact that they’re not able to easily enter data in a way that makes sense from a clinical perspective and have too many clicks to get through to input or read what they need to. Lock-outs and freezes are also apparently common, causing lost work and data re-entry that can take away from time spent conducting follow-up calls to patients or simply, long hours in the office after closing time.

Contrast this with a system from another part of the country that involved the clinicians at every step of the way including four prototypes and a pilot as well as a constant feedback loop via the provider outreach group and a schedule of improvements to the system to address that feedback in a relatively timely manner. While this other system also occasionally experiences hardware difficulties that can slow the system down, in terms of usability, this system is much more successful and lauded by the clinicians who use it daily and have in fact realized huge efficiencies in work, reducing turnaround time on an approval process, from 3 business days to 24 hours in some cases. Mouse-clicks and screens are organized in a logical, hierarchical fashion and the pieces presented to the end user are targeted based on who they are at log on. This allows the user interface to be relatively lightweight and the labeling on all fields is done in a way that is intuitive for the clinicians using it.

What it all adds up to, is that Himmelstein is likely right when he states that computerized systems are not currently saving the health care industry money across the board. However, this does not mean that these kinds of systems can’t save money over the long term if they are designed and implemented correctly. One of the fundamental ways to ensure that both design and implementation are correctly conducted is by espousing the principles of usability and providing intuitive systems that have a strong focus on serving the needs of the users that will use them. Without an eye to user experience design in health care systems, health care organizations will continue not to realize the cost and efficiency benefits that are possible.

Leave a Reply

Your email address will not be published. Required fields are marked *