EHR Musical Chairs
As EHR use grows across the market, we see the replacement market starting to grow, too. Depending on whom you believe, the EHR replacement market could range from 40 to 80-plus percent of existing users. (The latter seems like a stretch, but to paraphrase John Maynard Keynes, in the long run we’re all dead, which I suppose applies to EHRs as well as people.) There is obviously a large and looming replacement market in front of us, and one characteristic that separates replacement customers from first-time buyers is that they are more seasoned (embittered?) and are usually adamant about not repeating the youthful indiscretions of their past.
Until recently, it’s been hard to find hard industry data on trends like EHR replacement. There are plenty of places to go for advice on how to manage EHR replacements, but the pickings are pretty sparse (OK, barren) when it comes to real data on the buying behavior of seasoned EHR users. It’s worth noting that the one thing the Meaningful Use program has given us is better industry data than we’ve ever had before. Jamie Stockton’s excellent analyst reports for Wells Fargo Securities offer some really fresh insights on industry dynamics using CMS data (a testament to the benefits of opendata.gov and crowd-sourcing). A professor of mine used to say that there is no Geneva Convention for data, and it’s fair to say that Jamie and his colleagues have sadistically extracted a lot of confessions from the CMS Public Use Files on MU attestation.
The Wells Fargo EHR replacement report (referenced in HIStalk) looks at the results of 6,000 physicians using over 500 EHR systems who attested to MU in 2011 or 2012 and then switched to a different vendor in 2013. These 6,000 physicians represent about 4 percent of the total physicians who attested during that time period and represent an early view of the choices being made in replacing one EHR with another.
One of the most interesting pieces of information in the report is head-to-head analyses between different EHR vendors – which systems are people running from and which systems are they running to? The data is in the report, but it takes a little work to calculate, and I’m guessing that few people have the time to drill into the data to this level. Luckily, I’ve done the work for you, and here are my conclusions:
There is growing consolidation in a smaller set of vendors.
- Physicians used over 500 vendors for MU in 2011/2012. Yet, relatively few of these vendors had net gains among physicians switching vendors, with the biggest gainers being Epic, Cerner, athenahealth, eClinicalWorks, and Aprima.
- Over 50 percent of all physicians who changed vendors switched to these five vendors.
- The number of certified EHR systems has dropped almost 60 percent from MU Stage 1 (about 3,800) to MU Stage 2 (about 1,600). (I got this from the ONC CHPL site, not from the Wells Fargo report.)
Physicians fled from small vendors.
- Over 2,000 of the physicians who switched EHRs went from a small vendor to a larger, established vendor
Everyone except Cerner lost customers to Epic.
- Cerner was the only vendor to have a net gain against Epic.
- The biggest losers to Epic were NextGen, GE, and Allscripts.
Athenahealth and eCW both lost to Epic. They fought each other to a draw, and both made significant gains against smaller vendors.
- Athenahealth and eCW each lost two physicians for every one they gained from Epic.
- Athenahealth and eCW lost as many as they gained against each other.
- Athenahealth and eCW took more customers from small vendors than any other vendors, even Epic.
Physicians left Allscripts more than any other vendor.
- Allscripts had the biggest net loss of physicians among all 500+ EHR vendors.
- Allscripts lost head-to-head against every major vendor except McKesson.
I should note that there are limitations to this data. It covers only physicians participating in Meaningful Use, and it addresses only the market for replacement EHRs, not the entire market. And, while 6,000 physicians sounds like a big number, once you slice and dice the data down to the EHR-vs-EHR level, the numbers can get pretty small and thus less statistically reliable. I don’t want to make too much of this data, but to ignore it would be to make too little of it.
Is this data consistent with broader trends we’re seeing in the market? Darned if I know, but I’ll give it a shot.
1. The industry is slowly starting to consolidate on a smaller number of vendors. The financial injection of Meaningful Use Stage 1 put the market on steroids for a while, with lots of new EHR companies forming overnight to meet surging demand, but the inevitable industry shakeout now seems well underway. It’s unclear whether this is mostly a supply-side phenomenon (i.e., failures of startup vendors whose flaws were masked by MU dollars) or a demand-side phenomenon (i.e., physicians choosing to move to a smaller set of vendors, or hospital consolidation driving physicians to their enterprise vendor). Regardless, technology life cycles in every industry follow similar patterns. There is lots of product innovation at the outset, but the market eventually settles on a smaller set of dominant models that narrows market choices but enables more industry standardization. Process innovation starts to dominate over product innovation as users feel confident about investing in new organization processes that embed these new technologies. Which brings me to point #2.
2. There is growing separation between the enterprise EHR segment and the retail, small-practice EHR segment. With growing consolidation in ambulatory healthcare delivery from hospital acquisition of physician practices, as well as ACO formation, EHRs are increasingly being run by hospitals and ACO-like organizations who are trying to engineer fundamental process innovation for accountable care and patient-centric care. Are these still mostly just buzzwords? Yes, definitely. But an EHR vendor who doesn’t have management, products, and services to meet the needs of a complex array of enterprise users – such as IT, care and quality management, revenue cycle, security and compliance, and C-suite – doesn’t stand a chance in the enterprise space and may as well just focus on the market for small, independent physician practices. I got to peek over someone’s shoulder at a recent KLAS report on EHRs in the large practice space (11-75 physicians) that rated products not only according to what clinicians think, but also according to what IT and C-suite personnel think. More often than not, vendors that did well with clinical users did not do so well with IT and C-suite personnel, and vice versa. Athenahealth, Epic, and Greenway stood out as the only vendors who scored solid cross-enterprise ratings. Increasingly, EHR purchasers are in enterprises, and they value staff maturity and accountability, project management, product stability, and user support at least as much as nice user interfaces. In addition, clinicians within these enterprises seem to be gaining more respect for their IT and administrative counterparts as they become aware that those nice user interfaces are really ugly when they’re slow or unreliable.
3. Epic and Cerner seem to be moving ahead of the pack in reaching down into the enterprise ambulatory segment, while athenahealth and eCW seem to be battling it out in reaching up into the enterprise ambulatory segment. It’s unclear whether it’s better to start with a hospital product and step it down into the large ambulatory space, or start with a small office product and scale it up into the large ambulatory space. Deeper integration with hospitals than just lab results delivery is a fundamental requirement in this space, so the larger hospital vendors have a leg up there, as well as in their experience working with professional IT and administrative staff. The ambulatory vendors have the hearts of many clinicians though, which is important, but as noted above, will only go so far if they can’t scale their management and technology to work in higher-level settings.
4. If you’re an enterprise purchaser, your choices are mostly limited to a relatively small set of mature vendors. There are hundreds of EHR systems out there, but how many of them are backed by mature and robust implementation processes, deep technical bench-strength, strong support teams, and administrative features to centrally manage multiple practice EHRs? That takes more than a few Ruby programmers and a year’s supply of Mountain Dew. In looking to buy or replace an EHR, put a lot of emphasis on testing the vendor staff’s depth, experience, maturity, and professionalism. They will support you in the face of the unexpected, which in healthcare these days is just around the next bend.
5. If you’re a small-practice purchaser, you have more choices, but buyer beware. The small practice space is where a lot of product innovation will still take place for some time to come, but that comes at a price – lots of small vendors who are incredibly creative on the technology side don’t have the scale or backing to provide a lot of development and user support. That may be fine, because as a small practice your needs are easier to support because you don’t need complex interoperability, or population health, or IT administrative control, or high-powered analytics. You still want a vendor who will come running when you’re down and who will be around in the years to come, so don’t get too dazzled by shiny screens with no real people behind them.
It’s amazing what conclusions one can draw from a little data and a lot of artistic license! As more data is released by CMS, I’ll be sure to steal more insights from Wells Fargo Securities’ data analyses and share them with all of you.
Micky Tripathi is president and CEO of the Massachusetts eHealth Collaborative. The views expressed are his own.