Dr. Andy

Reflections on medicine and biology among other things

Wednesday, December 07, 2005

Uh oh

Computerized physician order entry (CPOE) is looked on as a panacea which will decrease medical error, improve efficiemcy, and improve patient safety. Only it looks like it has some major, unintended consequences, like increasing death according to an article titled "Unexpected Increased Mortality After Implementation of a Commercially Sold Computerized Physician Order Entry System: in this months Pediatrics.

The authors, from my institution, studied the outcomes of kids who were transported to Children's Hospital of Pittsburgh (CHP) for care. Transport is used to move patients who need more specialized care than the referring institution can provide. In Western Pennsylvania, CHP is the primary site for tertiary (i.e. highly specialized) pediatric care, so many smaller hospitals will transport kids. Reasons can range from need for ICU level care (most outside hospitals don't have pediatric ICUs) to requirment for specialist input. Kids who are transported range from not that sick to in extremis, with the overall level of illness being less than you might expect. To be fair, community hospitals have a real range of expertise, so some just aren't comfortable with really sick kids. In addition, it is generally better to transport before someone is in extremis.

Anyway, the findings were suprising and alarming:
Among 1942 children who were referred and admitted for specialized care during the study period, 75 died, accounting for an overall mortality rate of 3.86%. Univariate analysis revealed that mortality rate significantly increased from 2.80% (39 of 1394) before CPOE implementation to 6.57% (36 of 548) after CPOE implementation. Multivariate analysis revealed that CPOE remained independently associated with increased odds of mortality (odds ratio: 3.28; 95% confidence interval: 1.94–5.55) after adjustment for other mortality covariables.
Put in plain English, kids transported after implementation of CPOE had a more than 3 fold increased risk of death. Ouch

The authors go on to speculate on what went wrong. They conclude that doctors and nurses spent more time sitting at the computer and less taking care of patients:
The usual "chain of events" that occurred when a patient was admitted through our transport system was altered after CPOE implementation. Before implementation of CPOE, after radio contact with the transport team, the ICU fellow was allowed to order critical medications/drips, which then were prepared by the bedside ICU nurse in anticipation of patient arrival. When needed, the ICU fellow could also make arrangements for the patient to receive an emergent diagnostic imaging study before coming into the ICU. A full set of admission orders could be written and ready before patient arrival. After CPOE implementation, order entry was not allowed until after the patient had physically arrived to the hospital and been fully registered into the system, leading to potential delays in new therapies and diagnostic testing . . . .

This initial time burden seemed to change the organization of bedside care. Before CPOE implementation, physicians and nurses converged at the patient's bedside to stabilize the patient. After CPOE implementation, while 1 physician continued to direct medical management, a second physician was often needed solely to enter orders into the computer during the first 15 minutes to 1 hour if a patient arrived in extremis. Downstream from order entry, bedside nurses were no longer allowed to grab critical medications from a satellite medication dispenser located in the ICU because as part of CPOE implementation, all medications, including vasoactive agents and antibiotics, became centrally located within the pharmacy department. The priority to fill a medication order was assigned by the pharmacy department's algorithm. Furthermore, because pharmacy could not process medication orders until they had been activated, ICU nurses also spent significant amounts of time at a separate computer terminal and away from the bedside. When the pharmacist accessed the patient CPOE to process an order, the physician and the nurse were "locked out," further delaying additional order entry.

Before CPOE implementation, the physician expressed an intended order either through direct oral communication or by writing it at the patient's bedside (often reinforced with direct oral communication), with the latter giving the nurse a visual cue that a new order had been placed. The nurse had the opportunity to provide immediate feedback, which sometimes resulted in a necessary revision of that order. In addition, these face-to-face interactions often fostered discussions that were relevant to patient care and management. After CPOE implementation, because order entry and activation occurred through a computer interface, often separated by several bed spaces or separate ICU pods, the opportunities for such face-to-face physician–nurse communication were diminished.
They go on to note that they've made some changes (orders can be entered before patient arrives), but many problems still remain. A second physician still needs to sit and enter orders and there remains an unacceptable lag in getting medications started.

In the past I've defended CPOE, but it is hard to do so after this study. Of course, many of the problems can be potentially overcome (e.g. by preentered "sets" of orders that often go together) and the centralization of pharmacy is really a separate issue, but I still find this article very troubling.

10 Comments:

At 2:48 PM, Blogger Allen said...

To me, the truly troubling thing is that we put up with this "You have to do what the computer says" mindset.

I like the things computers can do for me, but we're a LONG way from having all the bugs worked out with integrating technology and medicine. We're sentient beings and over trained professionals, not drooling automatons who need to do a 'stimulus-response' loop at a computer. Every CPOE system needs to have 'things are too important to sit at a keyboard' bypass.

My own hospital has this problem with the radiology PACS syste. I used ot be able to get a stat portable CXR in about 3 minutes. Now if I can get it in 10 I'm lucky, because the developer system won't take a "John Doe" for evergencies, so care waits for Admissions to Input a Name and Number into the network, then the Clerks can order the Xray, then Xray can develop the cassett they've been holding in their hands while we scowl.

Ultimately, we need to approach these systems with the mindset that the computer system needs to work for the people and not the other way around.

 
At 2:49 PM, Blogger Allen said...

I really should preview before I post. I can spell, but cannot type.

 
At 10:34 AM, Anonymous DML said...

I agree a lot with what Allen wrote and I see things like this all the time as well. I led a team that developed a web-based lab order/resulting system over the last four years. The system is functional and works reasonably well, especially compared to some third party systems that our hospital has purchased and has in production. I suppport an ED system that is maddeningly error-ridden.

However, even our home-grown system needs a lot of babysitting and we had an issue with a stat order from a remote hospital a few nights ago. We got it resolved within the system, but the key thing that we do at the hospital where I work is having a backup or bypass for essentially every CPOE type of system. If we hadn't gotten the issue resolved, the specimen would have arrived at our lab with a paper requisition, and the result would have been faxed to the ordering hospital. Our system provided the result just a little bit faster than the paper req/fax-back would have, but we have to have that back-up system.

I think part of what Andy talks about goes back to a huge problem that we are nowhere close to resolving. If those remote patients coming to his hospital had a medical record in a standard format that could arrive ahead of the patient, along with any recent lab orders and results, diagnoses, etc., and those things fed automatically into Andy's system, things might be better. Medication interactions and medication allergies are a key area of concern in many cases.

But, despite the best efforts of organizations to standardize data (HL7's CDA/CRS, for example), we are far from having a true data interchange nationally, or even on local levels. There are a few regional healthcare organizations that have solved this reasonably well, but not many.

The fact that we have so many big players out there with vested financial interests in various formats, systems, technologies, and data exchange formats means that we are probably a long way from solving that problem. I am currently working on some statewide initiatives in my small northeastern state to see if we can standardize just a small subset of patient data for application to chronic care disease management, primarily diabetes and then CAD/CHD next. Even in a small state with fewer than 20 hospitals, the data formats are so varied and complex that we are stuck in the arguing phase.

I have spent years working on data normalization efforts for genetics research and healthcare and the progress is painstakingly slow.

I know Andy is a free-market type, but this is one that I don't see being solved without government intervention or the consolidation of a number of large healthcare software companies or organizations into a smaller number of powerful entities. There is no "Microsoft" in the market at present that can get most everyone to look at things their way.

Maybe before I retire in 20 or 25 years, we'll see some real progress in this area and realize some real benefits from all of the investment being made.

 
At 3:41 PM, Blogger Aggravated DocSurg said...

What a fascinating and well timed study that certainly falls into the category of "the great gaping maw of unintended consequences." We have, over a period of many years, developed and honed systems which work pretty well in the delivery of timely care for acutely ill patients -- and each institution has done so in its own way.

Now, every part of the delivery system is having a completely different delivery method being forced upon them, at the same time, which will inevitably lead to all sorts of unexpected problems. Everyone who has worked with computers understands how this works --- lots of bugs and beta programs to work through --- but these "bugs" will most certainly result in problems with delivering care in the acute setting.

I, too, believe in a more market-oriented system, which would allow the natural progression of such systems to occur at its own pace, rather than in a governmentally-imposed manner. And I would invite Microsoft, and whoever else would be interested, to devise CPOE systems without the fear of monopoly suits from the feds.

 
At 8:25 PM, Anonymous Anonymous said...

I am not computer-savvy. Neither am I generally a hospitalist. I will never forget being unexpectedly rounding in the hospital the week that the computerized system (now dismantled) was implemented. The doc on call the night before (also unexpectedly - the hospitalist was sick) had thought she'd ordered PRN nitrates for a cardiac patient. She hadn't. So, of course, the patient had active chest pain & I couldn't get onto the damn computer to get her any meds and the nurses were not permitted to take a verbal from a doc physically in the hospital.

Finally, some kind nurse took pity on the patient & actually practiced nursing & we got the patient through the crisis with flying colors.

I didn't mind entering the orders in the computer. I didn't mind the concept of an electronic system & it might not have been so bad were I more adept. But the system, now scrapped, was not Internist friendly and was totally unable to cope with an emergency. A system which works fine with routine post-op orders does NOT lend itself to an emergent & fluid situation. Wiser heads than mine willl need to figure out how to satisfy all. It will not be easy. And I fear there will be loss of life in the process.

 
At 2:47 PM, Anonymous Anonymous said...

Every study that has come out on CPOE so far has indicated that it initially increases errors. It also increases physician work load (as much as 316% in one study). These problems can be overcome but it will take time, effort, and every possible mistake will be made at least once and patients will face the consequences of this. Worse yet is the fact that every hospital has to re-invent the wheel because of the mish-mash of computer systems that need to talk to one another. You can't just buy Microsoft CPOE 1.5 and stick it in the CD drive. The order/medication system at most hospitals has been in place for decades and has steadily been "tweaked" all that time to make it work. It is the height of hubris to think that someone could sit down and whip up a new system that works better with no incidents or unintended consequences. This is yet another poorly thought out unfunded mandate by the federal government. The same is true of the changes in residency programs in my opinion.

BladeDoc

 
At 10:48 AM, Anonymous pixelrn said...

So glad to be aware of this study. Thanks for posting.

We have recently implemented a POE system on my unit and the frustrations involved with this have been difficult to validate and quantify. When confronted with some of the frustrations involved the reaction from the powers that be are often no more than "it will take some time to get use to."

It's difficult to be objective but nurses seems to spent a significantly greater amount of time charting. Currently we toggle back and forth between two systems (one for charting and one for physician orders) and these systems do not communicate with one another.

Other problems encountered:
--Computerized messages to pharmacy that somehow get "lost."
--having to wait for the physician to enter the orders for a newly admitted patient
--the ungodly mess that occurs on the MAR when orders from a transferred patient are d/c'd.
--the "stat" orders that go uncommunicated.

It's one thing to be "frustrated" with these growing pains. It's quite another when patient safety/mortality is affected. I wish someone would do a study like this on my unit.

 
At 2:08 PM, Blogger M Omori MD said...

Granted that this article raises some good issues about Leapfrog's CPOE system, such as everyone is locked out of the system as the pharmacist take off orders from the system, but no one has mentioned the CENTRALIZATION of the pharmacy once this system was implemented at U Pitt. This implies that ALL DRUGS were mixed my the pharmacy and there were no cache of drugs available on the Units, which is just plain stupid. You can't 100% blame leapfrog for the lack of a fail safe system.

 
At 6:55 AM, Anonymous Anonymous said...

Refer to Diabetes for
useful information

 
At 3:17 PM, Anonymous Peregrin said...

I think everybody should glance at this.
Sudbury hotels | sexy pumps | sexiest women

 

Post a Comment

<< Home