Yesterday I had the opportunity to talk with the people from GE who briefed me that GE is in the process of shutting down their GE Centricity Advance product line. This was pretty big news to me since I remember just last year at HIMSS meeting with GE and hearing that for the small practice (I believe 1-10 docs) GE Centricity Advance was where they were putting all their effort. You could see the energy they had behind it. In fact, their iPad EHR app was built on top of the GE Centricity Advance solution (which is now being moved to their other EHR product lines).

You might remember that the GE Centricity Advance solution was actually created out of the purchase of MedPlexus in March 2010. At the time, MedPlexus had 100 employees out of California with the development team out of India. At the time of purchase it seemed GE’s acquisition would provide a SaaS based EHR option to the independent physician market. Plus, MedPlexus (which became GE Centricity Advance) also provided an integrated Practice Management System with the EHR.

The GE Centricity Advance website is already forwarding to the Centricity Practice Solution website and a letter was sent out to all Centricity Advance customers informing them that the product line was ceasing operation. I’ve asked for a copy of that letter and if I get it, I’ll add it to this post (or if you’re a customer that received it and doesn’t mind sharing we’d welcome it).

I was told that GE is offering Centricty Advance users a free transfer to their Centricity Practice Solution EHR software. From what they told me it seems this will include data migration, training on the new system and a license for Centricity Practice Solution. Of course, Centricity Advance was paid on a subscription model so they’ll have to continue paying the monthly fee. As with most data migrations, I don’t think we’ll know how good GE is at migrating the data from GE Centricity Advance to Centricity Practice Solution until they start to do them.

Since both Centricity Advance and Centricity Practice Solution have ONC-ATCB complete EHR certification, there shouldn’t be any problems for those that transfer to Centricity Practice Solution when it comes to EHR stimulus money. Those not wanting to move to the Centricity Practice Solution will have this as part of their decision on what to do once Centricity Advance is no longer supported. I expect there will be many in this situation since while Centricity Practice Solution is available through GE’s partners as a “SaaS” offering, I think many will want to find a true from the ground up web based SaaS EHR offering.

I asked how many providers would be effected by the end of the Centricity Advance product line, but it’s GE’s policy to not comment on those numbers.

Where does this leave GE Centricity EMR software?
GE Healthcare IT still does a couple billion dollars of business and still has three EMR software offerings:
*Centricity Practice Solution – The replacement for Centricity Advance and will be GE’s EMR offering for the 1-100 provider practices.
*Centricity EMR – Still ambulatory EMR, but for the 100+ provider practices.
*Centricity Enterprise – Acute care EMR

I’m sure that many will wonder how good the Centricity Practice Solution will do in the small practice arena. Will this basically mean that GE is no longer a player in the small 1-10 provider practices? It’s hard to say for sure, but I’ll be interested to see how the Centricity Practice Solution EHR does in this market. There must have been a reason they purchased what became Centricity Advance instead of going with Centricity Practice Solution in the first place.

On the other hand, I could see people making the argument that this is a sound strategy by GE since movements like accountable care organizations (ACO’s) and related initiatives are putting the small practice in jeopardy. We know that many hospital systems are purchasing up group practices as they prepare to become an ACO among other reasons. While we still have many small group practices, it’s worth considering how many of them will survive the changing landscape. If not many survive, then this strategy by GE could end well for them. Although, I personally believe that practice consolidation is cyclical and so I’m not ready to announce the death of small group practices yet.

Another trend that might make this a good decision on GE’s part is what I call the Smart EHR. Our current phase of EHR adoption is basically converting paper to electronic. Once doctors start requiring EHR software to do things far more advanced (see Artificial Intelligence and Genomics EHR), it will require a new kind of EHR. Maybe Centricity Advance wasn’t prepared to make this shift. We’ll see if GE’s other EHR software is ready for it.

Many have argued that EHR consolidation is inevitable. I guess I shouldn’t be surprised that part of that EHR consolidation is happening within the same EHR company. I’m sure there are more on the way as we see which EHR companies survive the meaningful use winter and come out on the other side and which EHR companies close up shop.

Update: I asked GE for some more clarification on when GE Centricity Advance would be sunset and which data they’ll be migrating as part of the data migration process. Here are their answers:
Sunset Period: We have announced that we will cease operations of Centricity Advance on June 30, 2012. The data will be available in read-only mode until December 31, 2012.

Data Migration: We are working with our partners and customers to figure out the best way to migrate data. We have told customers that we will migrate the following data:
a. Patient Demographics, Patient Insurance data, Fee Schedules, Appointments
b. Patient Summary
c. Patient chart
We will migrate all clinical data. We are working with our partners to determine which financial information should be automatically migrated.

One of the really interesting things I noted while attending the NextGen user group meeting had to do with the move to SaaS based EHR and other SaaS based EHR software. I partially mentioned this in the write up I did at the conference, including a tweet where I talk about how Scott Decker really pushed the idea of NextGen making the move into the SaaS based software world.

I think there’s little doubt that NextGen sees the value of SaaS based software. I think they see the convenience to doctors of not having to manage a server. Most importantly, I think they see the value of not having the healthcare data stored in EHR in silos.

One thing that Scott Decker mentioned in his keynote was improving their coding rules engine based on the feedback and experience across all of their SaaS based EHR users. I found this really intriguing since it highlighted some of the challenges and limitations of the client server EHR model that’s so prevalent in healthcare.

After hearing these comments about NextGen’s move towards more and more SaaS based software, I wondered what users at the meeting thought about the move by NextGen to SaaS EHR. The nice part of a user group meeting is I had a chance to talk to a number of them.

One company I talked to said basically, “We have 30 Citrix servers in our NextGen EHR installation. That’s a huge investment we’ve made and I don’t see us changing that any time soon.” They’ve got an interesting point. There’s a lot of money invested in training, equipment, software, and general understanding of the current client server EHR installs that NextGen employs (or is it employed?) for its large EHR customers.

It’s quite a stark contrast to consider this entrenched client server user base that is unlikely to change even if NextGen’s direction is headed towards SaaS EHR software. To be completely honest, I’m not exactly sure how this “conflict” is going to play out.

Yesterday, I created a post on EMR & EHR called The Must Have EMR Feature – An iPad Interface. that post has driven quite a bit of discussion on Twitter and Google Plus. One comment from @2charlie hit me the most though:

2charlie – Charlie Gaddy
A decent web interface wouldn’t hurt either. RT @ehrandhit: The Must Have EMR Feature – An iPad Interface dlvr.it/tYkN7

Charlie’s twitter response highlights a number of interesting ideas. The first point that every SaaS EHR company will point out is that he said a web interface. We could go into the semantics of what is “the web”, but I have little doubt that Charlie meant a browser based interface when he said web. I’ll leave the rest of the discussion of “web” EMR interfaces for another post (plus, we’ve had that discussion many times on this site).

Instead, I want to focus on his use of the word “decent.” That adjective is interesting because no one would really argue that there aren’t plenty of web EMR interfaces out there. If you look at the EHR Scope EMR Comparison site, you’ll see a huge number of web based EMR companies listed. However, when you add the word “decent” to web EMR interface, I think we could have some really interesting discussion.

At least a couple times a week I get a doctor sending me an email or posting a comment on my website saying that “all of the EMR interfaces are terrible.” I don’t necessarily agree that “all” EMR interfaces are terrible, but a lot of them do fit the description quite well. I’m sure at this point all the EMR companies are thinking about their competitors and agreeing with me.

The iPad Opportunity for EMR Interfaces
As I thought on Charlie’s comment of a “decent web interface” as compared with an iPad EMR interface, I realized that the iPad provides a unique opportunity for EMR vendors with less than stellar web interfaces. While it would be great for EMR vendors to create stellar web interfaces or improve their current web interfaces, that’s much easier said than done. Many are working on older technologies. Others have so much company culture built into their interface that it’s hard to change. Many have large user bases that will freak out at the idea of a new web interface. Etc etc etc! The point being that the culture and history of many EMR interfaces make it hard to change.

In these cases, I see the iPad as a great opportunity to start fresh with your EMR interface. Many EHR vendors could use the iPad as a way to be able to create a new interface for their EMR with all the knowledge they’ve learned over the years baked in. Doctors expect the iPad interface to be different and unique.

I’ll be interested to see which EMR companies take this opportunity and make something of it. It’s the perfect chance for EMR companies to create a paradigm shift in their EMR software without having to admit publicly the mistakes they made in their first EMR interface. Unless you happen to be from an EHR company who built the perfect EMR interface from the start. Then, this need not apply.

In my post about the common EHR implementation problem of EHR slowness, I mentioned that I’d follow up with a post on how you can avoid the EMR slowness issue altogether. It’s better to avoid than fix problems.

The best way to approach EHR performance issues is to make them part of your EHR selection process. EHR performance issues could and should be a deal breaker for you when you’re evaluating EHR companies. How then can you identify EHR software that might have these performance issues?

Red Flag #1 – EHR Demo Slowness – Bring a red pen to your demo and every time they say something like, “It’s not usually this slow?” or “It must be slow because it’s running on my laptop.” make a BIG RED mark on your paper (or tablet if you’re advanced like that). Even one red mark should be cause for concern and investigation.

Certainly there are situations where environmental issues can cause slowness to an EHR. So, you can’t completely rule them out completely for this, but this is their demo. This is there one time to shine. If they can’t get their EHR demo running at full speed, what makes you think an EHR production environment will be much better?

You can make an extra red mark if it’s a SaaS EHR that’s providing the demo. They might say it’s just “the internet connection.” Well, guess what? Soon, that’s going to be you using that EHR and often on similar internet connections.

Of course, the message to EHR vendors is to make sure your demo runs as fast as your production system.

Red Flag #2 – Site Visit Slowness – While the demo can tell you a lot about an EHR software, it can’t necessarily tell you the speed of the EHR software. Just because the EHR is fast during the EHR demo, doesn’t mean that same EHR software will be fast in a production environment. Add this to the multitude of reasons why a site visit to a current user of that EHR is so important.

Make sure to do that site visit at one comparable in size and users to your clinic. You don’t want to look at the EHR responsiveness of a solo practice if you’re going to be a 6 provider multi clinic setup. Size matters when it comes to EHR speed.

Once on site, you can get an idea of the speed and responsiveness of the EHR software in two ways. First, observe the users of the EHR in the clinic. See if they exhibit any of the systems listed in the first section of this post. Another observation is to see how quickly they’re clicking around the EHR. If you see a lot of clicks in a row with little waiting in between clicks, that’s a great thing. If you see them click, wait, click, wait, click, click , wait. Be afraid.

The second way is to ask the EHR users. The problem with doing this is that only one response has value. If they say the EHR is slow, then you’ve gleaned some important information that’s worth checking on. If they say the EHR is fast, then you don’t necessarily know. The problem is that you don’t know what the user considers fast. What’s their frame of reference for saying it’s fast? Do they know what fast is? Have they just been using the EHR software so long that they’ve hit a rhythm that makes it feel faster than it really is? It’s a good sign if they say that it’s fast, but take it with a grain of salt.

Red Flag #3 – Use A Demo EHR System Yourself – Most EHR vendors will provide you a way to demo the product yourself. This isn’t a fool proof method to test EHR slowness, but it’s another decent test of the EHR’s responsiveness. Try it out using your internet connection and your computer hardware. Nothing like first hand experience documenting some patient visits to learn about the speed of an EHR.

EHR Speed Suggestion – Don’t Skimp on Hardware
Far too often I see a clinic skimp on the hardware requirements and regret it later. In fact, they often end up spending the money twice since they have to buy new hardware since they skimped in the beginning.

Of course, this suggestion can be taken too far as well. The computer and laptop manufacturers will try to sell you the whole kitchen and you might only need the stove and refrigerator. To put it in more practical terms, you’re going to want plenty of RAM, but do you really need the webcam, Blu-ray player, and special 100 in 1 media device?

Just because an EHR vendor says their EHR software can work on a certain hardware configuration doesn’t mean it should be used on that hardware configuration. In the middle there’s a spot between can and overkill that’s called optimal. Find that hardware configuration and you’ll be a much happier EHR user.

Conclusion
Don’t accept an EHR that’s slow. Make sure that the EHR performs at a satisfactory level. I know of nothing that frustrates a clinic more than a slow EHR.

We’re back again with our ongoing series on Common EMR Implementation Issues. Seems like readers really liked my first entry in the series about Unexpected EHR Expenses. To be quite honest, I was really happy with how that post turned out myself. It’s one of the most comprehensive and useful posts I’ve written in the 5.5+ years I’ve been writing about EMR and EHR. Hopefully we can continue that trend.

Today’s Common EMR Implementation Problem: EHR Performance Issues

I have to admit that this is a really tough problem to crack. However, it’s also incredibly common. The symptoms for this problem usually are described as, “THIS EHR IS SOOOOOO SLOW!” (This is appropriate use of ALL CAPS since they are often yelling this.) Followed by a *huff* and an angry doctor or nurse leaving their computer in a fit of rage. Other symptoms might include drumming fingers on the desk while staring blankly at the screen, lots of mouse clicks that get progressively longer and more emphatic, or the sitting back in your chair staring at the screen hoping that something will happen.

Once you’ve identified that there’s a problem with EHR slowness, then begins the fun and exciting (that was written in the sarcasm font) journey to identify the real issue. The biggest challenge with identifying the slowness is that there are a multitude of places that could be the bottleneck that’s causing your slowness. Some of which you can fix, and others you have to rely on your EHR vendor to fix.

To assist you in the ugly process of improving EHR performance issues, here’s a list of possible reasons you could have a slow EHR.

EHR Slowness You’re Responsible For
Slow Computers and/or Laptops – I’ve heard of a few EHR vendors offering free iPad’s with their EHR, but for the most part, you’re responsible for buying the computers and laptops for your EHR implementation. See my “EHR Speed Suggestion – Don’t Skimp on Hardware” below for more info on buying the right hardware. Needless to say, I’ve seen many slow computers be replaced and the EHR went a lot faster.

Slow Local Internet – Your local internet (or LAN as it’s often referred) could be the cause of your EHR slowness. I could have split this point into a half dozen possible issues. Some of them might include: Bad network card, bad cabling, bad switch, bad router, bad routing configuration, bad DNS configuration, overwhelmed network, etc etc.

Of course, in most cases you’ll probably have to call your IT service provider to solve these issues. They should be able to easily test most of the above issues and prove that it works for other internet applications and so it must be some other issue causing your EHR slowness.

Slow ISP (external internet connection) – If you’re using an in house EHR server, you won’t have to worry about this as much (except for interfaces, or EHR updates). If you’re using a SaaS EHR, then this could be a major bottleneck. Good thing is that it’s easy to test your ISP speed. If you’re speed is great to other sites, but not your EHR then you can move on to another issue. If you’re speed is bad for all sites on the internet, you need to see if your ISP can make some changes to provide the speed you’ve purchases from them. Otherwise, you might just need a bigger ISP connection than you have and you’ll be able to get your EHR running much faster.

Also, be sure you don’t have employees using up all your bandwidth downloading illegal (or legal) music or videos. That can eat up your bandwidth really quickly. There’s a reason Netflix uses up 20% of bandwidth on the internet. Movie downloads/watching might be using up your internet connection as well.

Memory on Server – I see this issue most often when a clinic tries to re-provision an old server for their new EHR or when they don’t follow the suggested specs of their EHR vendor. It can also happen when you start your EHR with 1 doctor and then grow your practice to 5 doctors. More users usually requires more memory on the server. There are good tools on servers for analyzing how much memory is being used so you’ll know if this is the problem or not.

Hard Disk Space on Server – This definitely shouldn’t happen in a fresh EHR install, but often can happen over time. Servers don’t like to run out of hard disk space and can do all sorts of crazy and unexpected things if they do. Other things that cause a hard disk to run out space might be backups or large log files. I’ve also seen where the IT administrator takes a 500 GB hard drive and divides it into multiple partitions. One partition for the O/S and one partition for the data. Often they misjudge how much to give to one partition versus the other. So, the one partition runs out of space while the other one has TONS of space left.

Good planning and regular maintenance will avoid these issues.

CPU on Server – I believe this is pretty rare these days since memory is usually the bottleneck instead of CPU. However, if the EHR software isn’t written correctly, this could be an issue. Particularly on older boxes.

Complex Workstation Setup – Your IT service provider might have told you all the great benefits of a thin client setup or some sort of virtualized desktop software solution. When done right, these solutions can work fantastic and save you a LOT of money. When done wrong, they can cause you all sorts of slowness and heartache.

EHR Slowness Your EHR Vendor Must Fix
Slow Server Configuration – There are lots of ways to tweak a server to go faster with less resources. Unfortunately, most of these tweaks are likely going to have to come from your EHR vendor. In a larger hospital implementation, you might be able to work with your EHR vendor to implement some of these tweaks. In a small clinic, you’re basically at the mercy of your EHR vendor to configure the server to run fast.

Slow Server (SaaS EHR) – Yes, SaaS EHR vendor servers can go slow too. The good thing is that your EHR vendor likely has monitoring tools that are watching for any slowness so they can proactively fix it. The problem is that then you’re at their mercy to fix the slowness. Needless to say, an EHR vendor’s server support staff rarely feel the end user pain of EHR slowness. At least the pain isn’t nearly as poignant.

Of course, a chorus of calls from EHR users to the EHR support line will help them understand better and fix the slowness. One call about your in house server doesn’t resonate quite as loud.

Slow or Overwhelmed Data Center Connection – Data Center internet connections are generally quite robust and built with a lot of redundancy. However, since data centers usually host many many different systems, they can also get overwhelmed. Sometimes through spikes of traffic, but more often through other nefarious attacks on the systems in the data center. Often, it’s not even your EHR software that’s causing the issue, but it might suffer the consequence. Not very common, but possible.

A little more common could be an EHR vendor that’s growing so rapidly that they can’t keep up with the demand for their EHR software. Other times the EHR vendor just did a poor job planning to expand their EHR data center services.

Poor EHR Code – Not all code is created equal. Some programmers are good at creating code that will execute quickly, but most are not. Fixing speed issues aren’t trivial. Particularly if you have a large code base that’s been created over a long period of time.

Poor EHR Design – The design of an EHR software often determines how fast it work. Designing for speed from the beginning is crucial. Otherwise, a poorly structured EHR can almost never be made fast.

Related to this is EHR software built on old technology. To use a car analogy, you can only make a pinto go so fast without gutting the engine. Too many EHR vendors are built on engines that can only go so fast. They can keep squeezing a bit more speed out of the engine, but eventually you have no other speed benefits because of the legacy technology limitations.

I’m sure there are other possible bottlenecks. Let me know of any I missed in the comments and I’ll add them to the list.

EHR Performance Finger Pointing
Another big problem with the complex list above is that it often leads to a bunch of finger pointing. Yes, sometimes it will feel like you’re back in Kindergarten again. Your EHR vendor will point the finger at your IT setup. Your IT service provider will point the finger at the EHR vendor. Then, the EHR vendor will point the finger at the hardware vendor. You’ll never be able to talk to a person at the hardware vendor and so you’ll have to use other tricks to prove it’s not them.

Needless to say the finger pointing can get really tiring really quick. Not to mention it can be very expensive as you spend money proving to your EHR vendor that it really is their problem and not your setup.

I’ll follow up this post with another on how to avoid EHR Performance Issues during the EHR selection process. I’ll link to that post once it’s up.

Side Note: This post was much longer than expected. I guess I did have a lot to say about this issue.

I think the debate over a SaaS based EMR versus a Client Server EMR is never going to end. Maybe we should just have a peace treaty and decide that whoever has a SaaS EMR is going to love the SaaS model and the benefits and features of a hosted EMR solution. The client server EMR people are going to love their in house “doctor controlled” EMR software with its inherent features and benefits.

What inspired this post? A few old threads popped up on my stats page. First, is a SaaS EMR versus Client Server EMR poll I did back in June of 2009 about which type of EMR setup people prefer. Here’s the results (as of this posting):
Client Server EMR (Client Install) – 35 Votes
Client Server EMR (Web based) – 28 Votes
Hosted Web based EMR (SaaS/ASP) – 84 Votes
Huh? – 3 Votes
Doesn’t Really Matter – 7 votes

That’s good enough as a tie for me. Probably reflects the chasm we have in EHR and EMR companies. There’s plenty of each to go around.

The above poll also led me to this post about the myth that a SaaS EHR is required to show meaningful use. I forgot that some EMR companies (or likely their sales people) were spreading these crazy myths about meaningful use.

I knew that my previous post about the cost to update an EMR would bring out the people who like to back the SaaS EMR model versus those who like to back the Client Server EMR. As I’ve said before, it’s one of the most heated debates you can have in the EMR space.

I realized in the comments of that post why it’s such a heated topic. It’s because once an EMR software chooses to go down one path or the other, it’s nearly impossible to be able to switch paths. Why? Cause if you do choose to switch you basically have to just code a new application all over. Basically, the switching costs are enormous. So, only a few software companies (let alone EMR software companies) ever change from one to the other.

Considering the high switching costs, that basically means that an EMR vendor that is SaaS based has a strong vested interest in the benefits and upside of the SaaS model of software development. The same is true for Client Server EMR software and client server EMR companies looking at the benefits and upside of the client server model of software development.

This entrenching around a software development methodology (for which they can’t change) is what makes discussing each model so interesting. Each party dutifully makes the most of whichever software development methodology they’ve been given.

Of course, from the clinical perspective it’s sometimes hard to cut through all this discussion and get good information on the real pros and cons of each model.

In that vein, I’m looking for a couple EMR and HIPAA readers that would be interested in making the case for one or the other. All you’d need to do is create a guest blog post on the pros and cons of your preferred method. If needed, you’d also be welcome to do a response post to the other method’s post as well.

If this interests you, leave a comment or let me know on my Contact Us page. I think this could be really interesting.

On a different note, it looks like I’m going to be attending the AAFP conference in Denver next week. Is anyone else planning to be there? Anything I should know about the conference to get the most out of it?

Miguel sent me the following email about local server EHR and Web Based (SaaS) EHR:

A lot of vendors in Puerto Rico are selling their local server application over the web application. In fact, to my view, they have very weak arguments when selling Local Server vs Web based application.

Can you direct me where to get additional information regarding the comparison of the two? Do you have an estimate, from the 100% physicians that are using EMR in US, what is the proportion of physicians using local server? What would you recommend?

This is a tricky question and the question that really divides many EMR vendors into their various camps. The tricky part is that both camps are right in their assertions. So, there is no clear winner. From my perspective you can make the case for either solution.

However, in certain situations one type of EMR might win over another. For example, if you’re in a place where your internet connectivity is not reliable, then you probably should go with an in house EMR instead of a web based EMR. Many doctors who don’t have formal IT support avoid an in house server and go with a web based hosted EMR to avoid the lack of IT support of the in house server.

I’ve written quite a few times about SaaS EMR and so a scroll through my previous posts will provide insight on a number of other topics including this post discussing the SaaS vs Client Server EHR. I should take the info and add it to this EMR and EHR wiki page. Maybe someone else can help with that too.

I don’t think anyone has an idea of the percentage of user who use a local server vs a web based EMR. I did do this EMR poll back in June, 2009 that showed a split decision between SaaS EMR and the 2 different style of client server EMR.

Finally, here’s the section from my EMR selection e-Book (which everyone should buy) that talks about the SaaS (web based) EMR vs. the Client Server (local server) EMR:

SaaS (hosted/web based) EMR versus Client Server (in house) EMR

This is one of the most heated questions you can ask EMR vendors when considering an EMR.  For an EMR vendor, choosing one or the other becomes like a religion.  My personal belief is that either model is reasonable.  Certainly the SaaS EMR people are correct that web based systems are the major trend in technology and that EVERYTHING is going web based.  However, it is also true that there are some things you can do with a client server EMR that still aren’t as effective with a web based system (ie. complex document workflow).  Some EMR vendors are combining the two models by having an in house server that is web based.  Others are putting their client server EMR in a data center also so they get the advantages of a SaaS EMR while still having some of the client server benefits.  For those that do not know the differences in SaaS versus client server, here’s a high level summary of the advantages and challenges of each model.

SaaS (hosted/web based) EMR

A SaaS EMR is one that is hosted by the EMR company (or partner of the EMR company).  Access to the EMR is done through a standard web browser. (Note: Client Server EMR can be hosted by the company and accessed using terminal server software as well, but that isn’t usually considered a SaaS EMR for purposes of this description.)  The biggest advantage to a SaaS EMR is a clinic doesn’t have to pay for the server and associated IT help to support a server in the office (ie. server room, tech support, redundant network, UPS, backups, etc).  SaaS EMR vendors reasonably argue that most clinics in house IT support cannot provide reliable and redundant server support the way a SaaS EMR can provide.  Part of this is due to the lack of expertise of in house IT support (or lack of in house IT support altogether) and the other part is due to lack of funds to build a reliable and redundant server environment.  Another advantage of SaaS EMR is that since they are web based they are available anywhere you have an internet connection.  When a SaaS EMR updates its software, you will automatically get the latest and greatest features of the software.  This can be a good and a bad thing depending on whether the latest updates were well tested and if they included features that would help your office.  Since a SaaS EMR uses a standard internet web browser, you will not need to spend time installing special software on each computer in your office.  This is even more beneficial when your SaaS EMR does an upgrade to the software.

The major disadvantages of a SaaS EMR are: internet connection dependence, EMR data not stored on site, and reliance on your EMR vendor.  Access to a SaaS EMR is completely dependent on a clinic’s internet connection.  Since the SaaS EMR is stored offsite in the vendor’s data center, any loss of internet connectivity means the clinic is without an EMR.  The solution to this is to have redundant internet connections (where possible), but also often means an increased cost for your internet connection.  Cellular broadband cards have helped to lower the cost of clinics having a redundant internet connection in many places.  Many rural locations with poor internet connectivity should probably avoid using a SaaS EMR.  Many clinics are also leery of SaaS EMR because the patient data in their EMR is stored in the vendor’s data center instead of on site.  Some SaaS EMR vendors will provide a backup copy of your data which you can store locally, but this is not very common and cannot usually be done at regular intervals.  SaaS EMR vendors argue that there’s no need to store a copy of your data locally since the server where your data is stored uses enterprise level backup to avoid any loss of data.  Ensuring these backups are completed appropriately and your SaaS EMR server is always available means you as a clinic are relying on your EMR vendor’s expertise in setting up those processes and configurations.

Client Server (in house) EMR

As would be expected, the advantages and disadvantages of an in house EMR mirror those of a SaaS EMR.  In house EMR software is traditionally done through a client install on a computer which accesses a server stored in the clinic.  Since the server is stored on site, you are no longer dependent on your outside internet connection.  Access to the EMR is done through your more reliable local network.  This also means that all the data from your EMR is stored in your office.  Many people would argue that client server EMR software is faster and can do more than web based software.  Web based software is making major strides in this regard, but there are still some features of an EMR that are better implemented by a client server EMR.

The biggest challenge associated with an in house client server EMR is that it requires a certain amount of local IT expertise to support your local server.  Many EMR vendors will assist your local IT support, but they still usually require some local IT support.  The quality of your local IT support matters regardless of which EMR you choose, but is more important with a client server EMR.

Another challenge with an in house EMR is that you are the one required to make the backups.  Some people consider this a pro since then you can be sure that the backups are done regularly and properly.  However, most people would argue that this is a problem with an in house server.  The reason for this is that too often making sure the backups are done and done correctly is forgotten or not done at all.  This is very common since backups aren’t appreciated until some major disaster happens and it’s too late.  Some local IT companies will partner with you in this effort and this can help solve this problem.

One of the most irritating parts of a client server EMR is the need to install the client software on each computer.  Certainly this is less of an issue the smaller your clinic, but it still can be a pain to manage.  Remember that this is not just a onetime event.  When your EMR software gets upgraded (usually 2-3 times a year), you will need to make the rounds to upgrade the software.  Certainly many EMR vendors have automated the upgrade process to some degree.  You can also often automate this process using active directory.  However, this upgrade process does create just one more area for something to go wrong with your EMR or require special IT support.  The good part is that this means that you can do the upgrades on your own timetable.

Hybrid Model

Some EMR vendors do a mix of the two options above.  They might have a server stored on site, but still have an EMR that uses web based technologies.  This still means you need the in house IT server support, but means that you don’t have to rely on your external internet connection to access the server.  It does however, usually mean that you can access your EMR from anywhere with an internet connection.  It also means that you can use a standard web browser to access your EMR instead of having to install a client on each computer to access your EMR.

This is not meant to be a comprehensive list comparing SaaS EMR with client server EMR.  Instead it’s meant as an overview of the major differences between the two types of EMR setup, but should give you enough information to choose which option will work best for your office.

I’ve come across a number of websites and people who’ve made the assertion that with the short time frames for meaningful use, a SaaS EHR is the only option to be able to meet the meaningful use requirements in a timely manner. Let’s see if I can do my part to clarify this idea which isn’t completely accurate.

First, there is still plenty of time for a clinic to implement an EMR of any type and get EMR stimulus money. At some point this might change, but at this point we are still far enough out that time is not an issue. Although, I’ll admit that it would be helpful if CMS and HHS would finally get some EHR software certified and provide some practical meaningful use details. Of course, these details shouldn’t be stopping doctors from evaluating and planning for their EMR implementation.

Second, it is worth acknowledging that in general a clinical practice can implement an EMR faster if it’s a SaaS EMR and not a client server EMR. The time for the server to be shipped to your office alone just takes time not to mention getting an IT person or your EMR vendor to install the server in your office. However, if you need more computers and a laptop to be able to use your SaaS EMR, you’re going to be waiting for computers to arrive anyway. Generally though, SaaS EMR is faster to implement than client server.

Of course, this doesn’t mean that you can’t quickly implement a client server based EMR. For example, I implemented a local doctors office in a week from when the server arrived. It was an incredibly fast implementation. Other than ordering time (which they had to order workstations also), it was as fast as any SaaS EMR implementation. So, it’s certainly possible. You just better make sure you have the right IT people supporting your implementation.

My point in this post is that it’s mistaken to say that SaaS EMR is the only option that’s fast enough to implement in time for meaningful use. Many of the client server EMR companies out there have really streamlined the process for installing a server in a clinic. Although, this is not true for all of them. So, it’s a question worthy of asking any EMR company if you’re looking at compacted time lines.

At least for now, it’s a mistake to rule out a great client server EMR just based on the meaningful use time line. We’ll leave the other arguments for ruling out a client server EMR in favor of a SaaS EMR for another post.

Many of you will remember my post making the case for in house EMR backups versus SaaS EMR backups. It was my first swing at my favorite part of blogging. I personally call it blog sparring. Basically, two (or more) bloggers discussing various viewpoints about a certain issue (I welcome other bloggers to join in).

Well, Lyndsey from Nuesoft brought the President and CEO, Massoud Aibakhsh, in on the fun in a post they called, “Continuing the Discussion…Data Backups: Leave it to the Experts.”

I appreciate Massoud’s response and he does make some interesting points about what is possible when a SaaS EMR vendor does the backups correctly. There’s no doubt that a SaaS EMR provider has more resources available to do a more robust backup, disaster recovery and business continuity plan than a small doctors office with a single server. His points about possible HIPAA breaches are also worthy of serious consideration. However, that kind of avoids a discussion of the points I made about relying on your SaaS EMR vendor to do the backup.

Nuesoft, why don’t you offer your end users a nice single click download of all their patient data which a doctors office could store in a nice HIPAA secured place in their office weekly/monthly or some other reasonable amount of time? Then, you’ll have the best of both worlds for a doctor’s office. They have your enterprise level backup, availability and load balancing and they’ll have a local copy of their data which helps them sleep better at night knowing it’s safely stored away in their office.

Of course, I’m not really trying to single out Nuesoft. I don’t know ANY SaaS EMR vendor that provides this service and that’s really unfortunate. Who’s going to be the first EMR software company to step up and provide this kind of support to the doctors?

If you’re really brave, you’ll even provide it in a format that they can extract the data themselves should they so desire (say if your company gets bought by someone else). I actually believe I heard of one EMR vendor that does this (client server though). They provide all of the data from their EMR in a nice exportable XML file which could easily be maninipulated.

I’m sure many in the room reading this will say, but what about vendor lock in? Why would I as an EMR vendor make it easy for my users to export their data out of my system? If I do that, they might *gasp* leave for another EMR vendor.

I of course asked the above mentioned EMR vendor about this problem. Their answer was a confident, “there are some that might leave because it’s easy to leave, but so far people have no reason to leave our EMR because they like it so much.” Kind of an interesting concept. Make an EMR that people love so much that they have no reason to change even though the door is open.

Granted, I’m not naive enough to think that some won’t leave. I’ve heard many a horror story of doctors leaving an EMR (for good reason or not) and then realizing that the grass wasn’t greener on the other EMR vendor side. Many of these doctors end up heading right back to their old EMR vendor.

One day a SaaS EMR vendor is going to revolutionize the backup process for their end users and start providing this level of data backup to their users. I know I’d be impressed with a SaaS EMR vendor that had that much faith in their product that they’ll give you a regular export (backup) of all the clinic’s data.

Now, back to some other comments from the Nuesoft post. In it, you asked me, “Do you seriously think banks use some of the services you mentioned to back up financial data?”

Well of course banks don’t use the services I mention to back up their financial data. Although, find me a bank that has 1 banker and 6 employees and I’ll show you a bank that uses the services I mention to backup their data.

At the end of the day. Let me liberally use a quote from the movie Shrek:

Shrek: Backups are like onions.
Donkey: They stink?
Shrek: Yes. no.
Donkey: Oh, they make you cry.
Shrek: No.
Donkey: Oh, you leave them out in the sun, they get all brown, start sproutin little white hairs.
Shrek: NO. Layers. Onions have layers.

Yep, backups are all about layers. The more layers of backup you have, the happier you’ll be. I know there’s been a number of times in my IT career that I’ve had to go to my 2nd, 3rd and 4th options to recover all the data from various backups I’d done. The more well designed layers of backup you have, the happier you’ll be if (when?) disaster hits.