'Login-Logout' - Is it Costing you Time and Efficiency in your Exam Room?

One of the issues that I hear about frequently from time-strapped physicians is the speed with which they can access patient information using their EMR. Logging in and logging out can become a frustrating exercise particularly in primary care, where patient visits may be as little as 7-10 minutes. 2 minutes logging into an EMR using 2 factor authentication (e.g. a password and a token) can be a time consuming process.

This video was produced at the Royal Alexandra Hospital in Edmonton and features Dr. Mike Donoff, a professor in the department of family medicine at the University of Alberta.

The clinic uses Sun Ray technology (from Sun Microsystems). The Sun Ray is a (thin client) display device, with applications running on a server elsewhere, and the state of the user's session being independent of the display. This enables portable sessions: a user can go from one Sun Ray to another and continue their work without closing any programs. (Wikipedia)

The concept of the portable session is one that this group of family physicians finds very appealing and offers speed advantages.

What are your thoughts about hardware? Have you had any experience using thin clients? Does the concept of the 'portable session' appeal to you in terms of the way you practice.

Click on the 'Comments' link to add your thoughts.

Comments

[Alan Brookstone on behalf of Tom Stogdale]

The following insightful comment was submitted by Tom Stogdale:

With respect to the blog "Login - Logout: Is it costing you Time & Efficiency in your Exam Room?", the subject matter is very valid, and the premise in the blog is quite accurate; depending on the individual user and/or practice's workflow, continued login and logout processes can have a degrading impact on efficiency. Additionally, the reference to Sun Ray as being a potential solution for some users is also well-founded.

However, I believe the narrow focus on the workstation (in this case, the Sun Ray thin-client workstation) is a risky proposition for users.

Exclusively focusing on the device with which the user interacts is similar to buying a home because one likes the doors and windows.

There are numerous solutions that deliver the same capability referenced in the video. Sometimes referred to as "persistent sessions," but also as "follow-me computing," among a host of other references. Regardless of what it is called, this capability is for the most part independent of the workstation on which it is delivered; it is a function of the software running on the infrastructure "behind the scenes." Some vendors are beginning to deliver this capability here in the Canadian market in a myriad of ways.

It is therefore somewhat essential that before a physician or their practice rushes in to deploy any such technology, they ought to consider all of the ramifications, costs, and risk/benefits of doing so. These include:

* Technical Compatibility - Does the complete solution "fit" with the EMR to be used, as well as other technologies and applications in use in the environment. For example, the solution might work with the EMR, but might not support any other Provincial-mandated systems. What about its impact on remote accessibility? ­Does it support flexibility in multi-factor authentication, which seems to be continually evolving in various markets in Canada (some jurisdictions require two-factor, some don't. Some require use of RSA tokens, some don't)? Does using this technology require disabling of some of the core security infrastructure in the EMR, potentially opening you up to other issues?

* The "complete" solution - Have you considered all facets of your professional and personal life in looking at the solution? For example, if you use a PDA or Smartphone that synchronizes with your desktop applications (think Palm, BlackBerry, etc.), how is this supported given that the solution is based on thin-client technology? You may not be able to install the desktop software for that device on a thin-client solution. But if the particular vendor says it can be supported, will it be supported by the EMR vendor? Will the solution allow some thin-client and some thick-client workstations simultaneously, which would overcome this? Is this supported by the solution vendor? Is it supported by your EMR vendor? How is remote access managed for say, a Smartphone if you want to look at your schedule while in transit or at the hospital? How can you access your EMR from home? ­Does this have an impact on accessing any provincial applications like the electronic health record or any of its components?

* ­Network Bandwidth - Have you considered the impact on bandwidth requirements? Some studies have indicated that use of thin-client technology may require massive amounts of bandwidth? Is the bandwidth into your environment - be it delivered by your EMR provider, the province itself (e.g., British Columbia), or directly by an ISP, sufficient for this? Various architectures that deliver the "persistent sessions" manage this differently, and thus may or may not be suitable for your environment, unless bandwidth (and by implication cost) is not an object.

* ­Support - Is the solution - the total solution including all considerations above - supported by the EMR vendor, or would you be creating a finger-pointing quagmire by implementing the technology which the EMR vendor doesn't themselves support? Is there a formal relationship, between the persistent session provider and the EMR vendor, that allows them to provide support for it or are you immersing yourself in the middle of a complex, multi-organizational support scheme?

* ­Cost - What is the cost of the total solution? Not just the workstation, but the total solution, including supportability. ­Since any solution is comprised of both front-end costs and back-end costs, looking at merely the cost of the thin-client workstation is inadequate. Furthermore, it may be inexpensive to implement, but expensive to maintain and support. If it works 99% of the time, it might not seem like a major concern unless the 1% of the time it doesn't work requires two weeks to fix. And to overcome the two-week delay, you have to spend additional money to have "spares" around. Is this factored into the calculation? ­Also, if you have to give-up certain things you are used to (e.g., synchronization of your BlackBerry or iPhone with your contact list) what is this worth?

These are but a few of the considerations anyone looking at persistent sessions should consider. No doubt, there are solutions that can work for your chosen EMR, but far too often physicians can become enamored with what appears to be a simple thing that in the end has hidden costs - both real and conceptual - which were never identified.

I highly encourage readers to consult with their EMR provider around how they can properly and efficiently deliver and support this type of capability before concluding one particular technology is right for everyone.

[Tim Janzen]

I think these are all very valid comments.
Thin clients can require a tremendous amount of network bandwidth. As well. in the demos showing Sunray capability, it is never clear how many units are acitve or accessing the servers concurrently. I suspect this could dramatically impact their logon speed. I do believe that Sunray can also provide a card reader to be attached to a PC; therefore if you need PDA, smartphone synch capability, I believe you could configure specific access points as PCs. In our clinic, physicians have PC's in their offices, but thin clients in their exam rooms.

BTW thank you for the very insightful comments.

[Alan Brookstone for Doug Girvin]

The following comment was submitted by Doug Girvin, President of Stantive Technologies Group and is posted on his behalf (Ed).

The comments regarding bandwidth, log in time etc. are actually not valid, nor are all thin clients the same. The SunRay uses about 160k in bandwidth (about the same amount of bandwidth as a high quality VoIP call), which is also able to be throttled by the server software according to available bandwidth. The bandwidth used is actually an encrypted stream of pixels only to the device. Unlike traditional Windows or Linux based thin clients, there is a small amount of firmware on the device that provides the ability to find and connect to a session through traditional LAN or VPN connections. No data is ever sent to the SunRay which is why it is a preferred solution in high security and military applications. It also (with monitor) uses about 12 watts of power vs. ~180watts for a traditional PC. Because it is not running an application or OS, it's life expectancy is about 7-10 years or approx. 3 times that of a typical PC on a evergreen contract. Finally, a typical 1RU dual CPU server can support over 100 SunRay clients concurrently and provides automatic server pooling with multiple servers for loading balancing and failover, with no additional configuration or clustering software required. I agree with the requirement to validate compatibility and application support, but the assumptions on power, bandwidth, etc above are clearly incorrect.