HIStalk Quotes
Who's Responsible for Poor Clinical Systems Outcomes, Vendors or Customers?
Most of us agree that the hypothetical benefits of clinical information systems (including CPOE) have rarely been realized and documented. Hospitals keep buying, implementing, upgrading, and replacing, spending millions of dollars and stressing out staff, yet few examples exist of anything more than the reduction in minor errors that would not have harmed patients anyway. Nothing I've seen suggests improved productivity, reduced errors, decreased malpractice claims or premiums, or higher provider satisfaction.
Would other industries keep spending despite these pathetic results? Does anyone understand how few products exist in the clinical system marketplace, how old they are (either in design or technology,) and how few customers have installed them? Is it clear to the lay public that most hospitals are too little and too unprofitable to afford the big-name systems being installed by the big-name hospitals that aren't making much of a difference anyway? We have 5,000 hospitals and the best penetration of any given mainline product is, what, maybe 300 hospitals?
I absolutely believe that today's clinical applications leave a lot to be desired. They are often poorly designed, patched instead of rewritten, kicked out the door too early due to unreasonable commitments and disincentivizing contracts that encourage both vendor and customer to emphasize deadlines over quality.
Still, here's why I blame hospitals for the predictably unimpressive results being delivered by all of this clinical system churning.
First and foremost, hospitals are buying the stuff's that's out there, warts and all. They aren't doing their homework to find known flaws. They're whipping out the checkbook without regard to whether there's really anything suitable to buy. They'll kick out their current Product A in a tizzy to get Product B, while hospitals just like theirs boot Product B and announce high hopes for Product A. They buy on marginal "features," believe the claims of slick sales guys wearing $1,000 suits and working for their fifth HIT company, and believe their organization will have better results than all the high-profile failures before them.
Here's where hospitals are most guilty, however: they pay lip service to quality, change management, process redesign, and support of the clinician. Once the check's been signed, the IT types take over and immediately get mired in ugly details, forging an often-contentious vendor relationship and pushing clinicians mercilessly to just get the damned thing up and running so they can high-five and move on to the next IT project. All of the high-falutin' ideals go right out the window when the going gets tough, which is usually about five minutes after the contract has been signed.
All of us joke about how "you've seen one hospital, you've seen one hospital." Why is that? Why are we proud that no best practices are followed, no standards are applied, and no apparent rhyme or reason exists to explain why every hospital insists on doing things the same illogical way they always have? How many ways can you schedule a med, label a lab specimen, book surgery time, or document nursing care? Five thousand, apparently. No wonder it's such a pain to train even an experienced new hospital hire -- whatever they've learned is probably useless in another organization.
How can off-the-shelf software automate this mess and still remain a viable solution for enough customers to make the vendor profitable? Answer: it can't. That's why the handful of major clinical systems have been hacked to death: to find a large enough customer base, they had to be Frankensteined into an Everyman system that works for no one. There's plenty to dislike in every clinical system whether your hospital is 50 beds or 5,000, mostly because it was written to be sellable to hospitals from 5,000 beds to 50. If the auto industry was like the clinical system one, we'd have only ten cars to pick from, all of them competing for the exact same customers. Would you like a Hemi engine or a hybrid? Sporty or luxury model? Left-hand drive or right? Huge trunk or convertible? We've got them all, right here in this same model that unmotivated, marginally trained mechanics will adapt to your needs before sending you flying down the Interstate.
Every time I look at upcoming enhancements from one of the clinical systems vendors at my hospital, probably 75% of what they're developing doesn't interest me in the slightest. Canadianization. Interface changes that offer no new functionality. Database tweaking to improve poor system performance. Integration with other vendor products that we don't use. Reporting for a particular state or organization that we don't care about. Bug fixes for problems we've never had, for functions we've never used. The downside of trying to serve a heterogenous audience is that they all want something different, and plenty of software is ruined trying to keep everyone happy. Maybe every vendor should be like Epic Systems: don't be so greedy to take on customers that you know will fail; instead, target your product to those who will succeed and who meet your sweet spot.
Most of all, hospitals are really bad at truly changing processes. They are outstanding at making emotional, breast-beating pronouncements about they'll be different than the 100% of customers who've preceded them in trying and failing to shoehorn in new software without the hard work of changing people's jobs. There's always an excuse afterward: a relcacitrant VP, an indifferent nursing executive, a lack of physician support, newly discovered gaps in software functionality, impending JCAHO visits, a financial crisis, a shortage of licensed staff or fear of offending them, unionization, or changing priorities. Once the fun of going on site visits and getting attention from the vendor's suited minions is over, it's back to business as usual. I've been on the inside several times for this kind of enthusiasm lifecycle and it never varies. All of the high ideals go right out the window because it's hard, thankless work that no VP in his or her right mind and with career ambitions will champion.
Experiences like those at Children's Pittsburgh are unfortunately common, I suspect. They can knock Cerner all they want, but they must be a pretty stupid consumer if they bought software with all those problems or listened to incompetent consultants (despite no complaints from other hospitals using the same products and consultants.) Any wake-up call is as much as a warning to hospitals as it is to vendors. You alone are responsible for the tools you purchase and how they affect patient care. Buy carefully and implement even more carefully. In fact, if you're not serious about implementing change, just save the money and use it to hire more nurses and build more parking garages.
Poetic! Amen! When I was on the consulting side, I so quickly became
disenchanted with the product we were selling because I knew these
"truths":
I spend much of my day in healthcare IT. However, I still work in
non-healthcare IT departments for a large portion of my consulting work and
I have to say that the environment you describe sounds very similar to the
government IT landscape. Now, it' not an excuse but hopital CIOs shouldn't
feel alone in their "one off" type of non-repeatable (or hard to repeat)
solutions for healthcare because the same problems exist in other sectors
like government.
Have really enjoyed discussions on this and on Matthew’s blog on this
topic…I have voiced many of these points in the past in the workplace..yet
it seems that the IT Pied Piper is just leading so many hospitals down the
road towards a “me too” and “the ultimate solution” for all the ills of the
industry…most recently this has led me to the impossible position as
follows…with a background of 20 years in IT, clinical, healthcare admin and
business process improvement, I have been less than enthusiastic about a
current initiative to replace my hospital’s CIS and eventual conversion to
a EMR…for one thing the IT department is understaffed and under achieving,
this smells like a failure in the process; for another, the choice to bring
in a consultant for the RFI and RFP sessions is a monumental waste of
resources, over 100K for a small community hospital; and for another, the
underwhelming quality of the products and their demos makes it a choose the
“least of the worst” multimillion dollar decision. I have tried to talk
about “how about we look at what’s wrong in our current processes first?”
but the steamroller to “automate” healthcare is too fast and furious for
that voice to be heard. I am certainly a proponent of technology, but I am
afraid we are heading to a situation where we simply change from bad
processes to automated bad processes, not considering the staff and
organization culture and competencies in the selection process.
Let us not forget the multiple purposes for which these systems were
designed - enabling the provision of care, documenting the care provided,
billing for the care, tracking the work flow components of the care, and
gathering various types of data to be sold and combined and crunched into
best practices (someday). The systems are not designed to serve a single
master well, and so serve many masters poorly.
Who's Responsible for Poor Clinical Systems Outcomes, Vendors or Customers?
