I sat in on the conference call this afternoon on the new HIMSS Leadership Survey for 2007. The survey this year was based on a larger sample (360 respondents) and had some interesting findings. Of course most all the focus of the conference call was on EMR adoption. After about 1% adoption annually between 2002 and 2004, the current adoption rate has grown to a very respectable 17% - hey, EMRs may actually happen. I can only sigh to think of all the hospitals replacing all their spot vital signs monitors or other needless crazy things soaking up their budgets. Don't get me wrong - medical device data should be electronically acquired and integrated into EMRS - getting good workflow and not spending an arm and a leg is what's hard. The survey reports that 69% of hospitals are either installing or already have a fully functional EMR.

Good news for many of my clients: the current priorities for IT (for the next 12 months) are "reduce medical errors/promote patient safety" (54%, up from 50% last year), "replace/upgrade clinical systems" (a big increase from 29% in 2006 to 48%), and of course the seemingly ubiquitous "implement an EMR" (almost unchanged at 48% this year). "Process/workflow redesign" also got a 5% bump up to 35% in this year's survey. That process/workflow redesign would get separate mention is interesting, given that almost any new automation requires changing the way you do things. Of course hospital folks (being just like the rest of us) are notorious for resisting change, so it's encouraging to see this get special recognition.

Along with the emphasis on reducing medical errors and improving patient safety IT priorities, the top issue facing health care was reported as "improving the quality of care" - a whopping 69%, compared to 36% in 2006. Second was "patient satisfaction" at 55%, almost unchanged from last year's 51%. A new entry on the list this year is "Medicare cutbacks" at 52%. When you troll the Health Affairs blog, or visit The Healthcare Blog, they're all about transforming health care. The elephant in the living room is the fact that the feds have been transforming health care since the introduction of DRGs in the 1980s and their continued reductions in reimbursement rates. Nothing transforms an industry like falling revenue, well except for the big mill steel industry... oh, and airlines after deregulation, hmm, the record industry... and how about those US car manufacturers? Not surprisingly, "demand for capital" almost doubled from 18% to 31% this year - it seems the list of potential capital expenditures with an attractive ROI just keeps getting longer. In fact, under Most Significant Barriers to Implementing IT was "lack of financial support", up slightly over last year to 20%.

Under Most Important Applications (next 2 years) both CPOE and EMRs dropped several points to 47% each. "Clinical information systems" jumped from 17% in 2006 to 46%. The 15 point drop for "bar coded meds management" (from 58% to 43% this year) elicited a question from one of the callers. With point of care meds admin applications at a market penetration of around 18%, why does the importance seem to be dropping. The short answer was they don't know, followed by supposition that not enough budget chasing too many projects was to blame. It was noted that bar coding for patient identification and other applications are seeing continued adoption in hospitals. In fact, Technology Adoption (next 2 years) is lead by "bar code technology", raising from 69% to 74% this year. I think "bar coded meds management" dropped because the market is fractured between systems from HIT vendors who focus on pills and infusion pump vendors who focus on the pump. Whats needed is a system that encompasses all meds administration (that doesn't cost twice as much as today's systems).

One of two new entries under Most Important Applications was "point of care data collection" at 29%. This group must be made up of hospitals that tried to get nurses to type medical device data into their EMR, or they know that over half of all point of care diagnostics test results never make it into the patient's chart (not to mention nurse's complaints about docking stations and workflow with point of care diagnostics connectivity). The other new entry under this category was "evidence based medicine at point of care", also at 29%.

Some point of care computing devices were mentioned under Technology Adoption. "Tablet computers" and "handheld PDAs" (aren't all PDAs handheld?) made the list at 62% and 59% respectively. Interestingly there is no mention of either computers on wheels (COWs) or wireless phones. There are two predominate use models at the point of care for computing devices, occasional and constant use. Charting, meds admin, vital signs collection are all examples of occasional use. Most hospitals employ COWs or prepositioned computers at or near the point of care for these apps. I would guess most all tablet computers would be mounted on a COW - they're certainly too heavy to carry around. Some vendors use PDAs for their occasional use applications. The nurse-carried use model supports applications that require continual accessibility, things like surveillance, alarm notification, voice communications, other event/message management, and worklists. (Yes, wireless phones are a point of care computing device.) These mostly nascent applications require immediate and highly reliable response from the caregiver, and thus must be with them at all times. So I'm surprised COWs and wireless phones didn't make the list - it's not like they're not being bought by a lot of hospitals.

Finally, the survey ends with Satisfaction with Vendor Performance, which was decidedly down. The "very satisfied" category evaporated from a not very high 8% last year to zero. "Satisfied" fell 10 points to 60%. The largest growing category was "dissatisfied" which went from 4% last year to 14%, followed by "neutral" which increased 8 points to 26%. Like meds admin, this trend also elicited questions from the media. HIMSS chair, Buddy Hickman, attributed this falling from grace raising customer expectations, and a notable gap between vendor marketing and their ability to deliver.

The industry is going through a transformation that is challenging both vendors and buyers. On the technical side, the emergence of middleware solutions and what I call the "enginification" of applications is causing waves and lots of new areas are seeing automation for the first time. The single vendor solution model is braking down, both because the level of automation in hospitals is starting to exceed the ability of any "single application" to address it (not to mention the ability of a vendor to cobble one together through acquisitions), and IT architectures like SOA and web services are making middleware a natural solution for enterprise-wide services (think Emergin and event management). Likewise it seems that every new application has an engine for this and an engine for that - rules engines, messaging engines, interface engines, positioning engines - geez, they're everywhere. All these engines must be tailored and/or configured on a per enterprise basis - a significant undertaking. The promise of these new technologies (the marketing) is tremendous, but the lack of experience with them (on both sides of the table) leaves the delivery and utilization of those promises wanting. We'll get through it.