Harvard study: Computers don't save hospitals money

http://www.computerworld.com/s/article/print/9141428/Harvard_study_Computers_don_t_save_hospitals_money?taxonomyName=Hardware&taxonomyId=12

by edw519 • 17 years ago

51 46 17 years ago

computerworld.com

va_coder2• 17 years ago

I have a friend who is a hospital administrator. She recently complained about an awful Oracle system that tries to do everything and fails.

As I think about my friend with the Oracle system I wonder if the problem is technology or the companies implementing the technology?

liamk• 17 years ago

From the article it sounds like the companies are failing. Only three hospitals in the study showed marginal advantages to having software, and theirs was custom built. I'm guessing the other hospitals are more likely to have generic poorly thought out software, like your friend uses.

spamizbad• 17 years ago

One big plus of moving towards electronic medical recording keeping is you open up a whole new market for advanced software that 1) makes hospitals run more efficiently 2) provider a higher level of care for the patient and 3) flag and prevent potential errors that could result in malpractice (thus lowering insurance costs across the industry)

The software that does this isn't going to be built over night, and will likely take at least a decade to mature, but when it does the whole industry wins big.

gaius• 17 years ago

I'm always fascinated by the thinking behind these ERP behemoths. Like, does anyone at Oracle or SAP ever take a vacation, or get promoted? How? Do they use their own software to do it?

mildweed2• 17 years ago

The problem with healthcare software isn't the technology, its the people. Tons of entrenched stakeholders who have "always done it this way" on a committee full of people who have "always" done it another way. Then put more people on the committee, one or two from every department of the hospital. First you have to get the whole hospital to do their work all the same way. Once you tackle that beast, you then have to put software together that works with that single method, yet still accounts for all sorts of caveats.

maryrosecook• 17 years ago

Computers rarely save money. They usually allow people to do more in the same amount of time for the same money. In the case of hospitals, they provide a better level of care for the same money and in the same time. e.g. I was in an appointment with my cardiac consultant yesterday. He was able to pull up the results of my recent exercise test during the consultation, rather than having me wait to receive a letter with the results and his thoughts.

trapper• 17 years ago

If this were true we would see a reduction in visit times, costs per visit, patients per doctor, patients per nurse, medical errors etc. Instead we see the opposite.

The point of the research is that none of these benefits are supported by the data we have available. Individual anecdotes are the exact opposite of what we should be thinking about after reading this paper.

aplusbi• 17 years ago

That assumes that the only thing that has changed is the computerization of records. In the same time span that we've seen increased computer usage we have also seen increased red tape from insurance companies, malpractice lawsuits, etc.

I'm not saying you're wrong, just that we don't have enough information to make any claims.

CWuestefeld• 17 years ago

Hospitals focus their IT systems on addressing the reporting requirements of Medicare and Medicaid, because so much more of their income comes through those channels than any other, and the reporting requirements are so onerous that they can't be met any other way.

By way of contrast (although this is only a single anecdote), my dentist's automation is clearly much more efficient, and is very effective in streamlining his operations. Digital xrays feed directly into the database, and become immediately available for viewing on the monitor even before a conventional xray would be developed -- and are always instantly retrievable from the DB. IMHO, this is because the field of dentistry is less driven by an 800-lb gorilla, and more dependent on meeting patient needs and budgets.

aneesh2• 17 years ago

There are positive externalities from the use of EHRs, though. The whole health ecosystem benefits (including patients, other providers, insurers, etc), not just hospitals, so it's misleading to consider the hospital's bottom line as the overall impact of EHRs.

jhancock3• 17 years ago

There is no "ecosystem" that I can see since the data cannot be shared. Have you seen some examples where its providing value? I truly would like to know if I'm missing out on some great software.

I've helped a mid size doctor's office for over 10 years with their IT system. The only value it has for them is managing billing with insurers and scheduling patients. Everything else the system does (tracking lab reports, patient history) is either not practical for the doctors to use or they don't need it. The part about helping with insurers is a business they'd rather not be in but have no choice. The scheduling could be replaced with Google Calendar if not for the tie-in to the data already mandatory for the insurers. Overall the system is something they consider they have to use but a bit of a pain and provides no other value.

aneesh2• 17 years ago

I can access my dental records online. But you're right, that's probably not a common scenario.

Since you have some experience in this field, could you imagine this situation changing in the next several years, so that there is a meaningful ecosystem?

Alex3917• 17 years ago

Paul A. Strassman wrote a book about this in 1997. You can read the executive summary here:

http://www.infoeconomics.com/squandered.php

If organizations aren't following Strassman's recommendations for technology implementation, then it's not really that much of a surprise that they're still wasting money. The costs of technology are basically the same today as they were ten years ago, and it's not like human behavior has magically changed.

cakeface2• 17 years ago

I used to work for a health care information systems company called Meditech. They are one of the largest providers software and EHR to hospitals. While I was there I was amazed at how inefficient the whole process was. My daily coding work did not even involve version control! Also we were programming in a language developed in house from like 1985 or something. Its certainly easy for me to believe that a system like that would only cause more problems than it solved.

sethg2• 17 years ago

I have a friend who told similar horror stories about working for Meditech. I hope she's escaped to a better job by now.

niels_olson2• 17 years ago

I used meditech as a medical student on the wards. Very painful experience. I'm sure it was really slick 20 years ago, but it's painful now.

jordanb• 17 years ago

I have a friend who manages a hospital IT department.

He said that problem is that American private hospitals already are highly computerized, it's just that the computer systems are focused entirely billing and not at all on medical records management.

When they decide to get an EMRS, hospitals turn to their billing software vendor to provide them with something integrated with their billing systems. The vendors tack something on to the billing software without adequate requirements gathering amongst the actual nurses who will use it, so the result is a system that simply does not have the proper workflow for managing patient medical records.

While EMRSes like the Veterans Administration's VistA program exist and are well-liked by nurses and doctors, they provide no functionality to ensure that you get billed $20 for each tongue depressor used. That functionality is, of course, of primary importance to private hospital administrators. So effective EMR software gets passed over unless it integrates with hospitals' existing billing infrastructure, and the hospital billing system vendors build terrible EMR software.

jhancock3• 17 years ago

great summary. I've noticed the VA system is quite good as well and have been trying to see why other institutions don't use it. I assume the taxpayer already paid to create it, so why not open source the product and let people integrate?

j_b_f• 17 years ago

VistA has an open source release called OpenVistA that is pretty well-supported: http://sourceforge.net/projects/openvista/

jberryman• 17 years ago

I can't find a link right now, but a month or two ago there was an interview on NPR with the author of (I believe) this book:

http://www.amazon.com/Best-Care-Anywhere-Health-Better/dp/09...

In the interview he talked about how some doctors at the VA hacked together a computerized medical records system in the 70s, and how having these computerized records actually allowed the VA doctors to discover links between some dangerous drugs and disease. That's in addition to the benefits of a doctors being able to pull up 30 years of your medical history on a screen.

Wish I could remember details about the interview, but the book is probably good.

sethg2• 17 years ago

The problem "is mainly that computer systems are built for the accountants and managers and not built to help doctors, nurses and patients," the report's lead author, Dr. David Himmelstein, said...

So the software works for the only interest group in the hospital that actually has the power to spend five- and six-figure amounts on infrastructure. The system works! :-/

aik• 17 years ago

This article seems overly negative. The tone makes it sound like computers should not exist in hospitals. I wonder what their agenda is?

"The problem 'is mainly that computer systems are built for the accountants and managers and not built to help doctors, nurses and patients,'"

There are plenty of products out there specifically to help doctors and nurses, and an increasing amount are being installed in hospitals. If it is in fact true that a majority of hospitals only have billing software, I would say it is the fault of the hospital/institution in that they chose to go that route.

"Implementing e-health records nationwide would cost between $75 billion and $100 billion, Brailer said, adding that individual hospitals "will have to make sizable, potentially multi-hundred-million-dollar budget commitments." Still, he said a fully functioning national electronic health system could reduce U.S. health care costs by $200 billion to $300 billion annually by cutting down on duplicate records, reducing record-keeping errors, avoiding fraudulent claims and better coordinating health care among providers. Himmelstein called those claims "unsupported.""

The dollar amounts I don't know where come from, but the advantages of a universal health care record aren't too difficult to see. Supposedly 50+% of faulty diagnoses are due to a doctor not having the information they need (that is available elsewhere).

xiaoma• 17 years ago

"For 45 years or so, people have been claiming computers are going to save vast amounts of money and that the payoff was just around the corner," he said. "So the first thing we need to do is stop claiming things there's no evidence for. It's based on vaporware and [hasn't been] shown to exist or shown to be true."

Oh, but it has. Outside the US, a number of countries have used computers very effectively in health care. In Taiwan, my medical records are linked to my national ID number and follow me wherever I seek care. Furthermore, doctors can regularly do pull up my chart from a laptop when I meet them in their office. It's also easy to make appointments online, and they go straight to the doctor in question, without any need to go through human administrators at the hospital. It's been a huge time saver for the patients, and the doctors achieve a turn-over that would be unheard of in N. America.

I can only speculate about the money savings, but it's overwhelmingly likely they are large. Most visits have a co-pay of about 3-10 US dollars, my one emergency room visit was about 40 and the highest tax bracket is only 20 percent. Most people are only taxed at about 7-8 percent, so it isn't simply a case of a health care system flooded with public money.

US hospitals just need good computer-based systems, or if it's a national one then a single good system.

anamax• 17 years ago

> In Taiwan, my medical records are linked to my national ID number and follow me wherever I seek care. Furthermore, doctors can regularly do pull up my chart from a laptop when I meet them in their office. It's also easy to make appointments online, and they go straight to the doctor in question, without any need to go through human administrators at the hospital. It's been a huge time saver for the patients, and the doctors achieve a turn-over that would be unheard of in N. America.

Apart from the "national ID" part, that's exactly how Kaiser works, and has for some time.

I don't know how widespread Kaiser is, but it's certainly common in California.

allenp• 17 years ago

I just wanted to emphasize that the best part about Kaiser is their interface - it is super simple to schedule and receive care and it rarely felt "slugish" to go through their processes.

va_coder2• 17 years ago

You have a single payer system in Taiwan.

The other day, in the US, I had a physical therapist run a routine check of my daughter. I had to sign 4 times and there was a total of 28 pages of documents. (Georgetown Hospital). And I had to interact with three people before I could see the therapist.

halostatue• 17 years ago

I've seen similar results here in Canada with MRIs. I suspect that the problem with Health care IT automation in the U.S. is the overall brokenness of the U.S. health systems.

jfoutz• 17 years ago

amusingly, the computerworld summary doesn't include the paper's summary:

"Finally, we believe that the computer’s potential to improve efﬁciency is unrealized because the commercial marketplace does not favor optimal products. Coding and other reimbursement-driven documentation might take precedence over efﬁciency and the encouragement of clinical parsimony. The largest computer success story has occurred at Veterans Administration hospitals where global budgets obviate the need for most billing and internal cost accounting, and minimize commercial pressures."

It seems like hospitals hyper-optimize single actions at the expense of efficiency of pipelined actions. sort of like... An addition takes a single clock, a multiplication takes a single clock, but a multiplication followed by an addition takes 100 clocks. Except tongue depressors and temperature taking instead of arithmetic; money instead of time.

I doubt you could build a more cost effective system to have a doctor walk in, take your temperature and leave.

samuel2• 17 years ago

I work in IT for my country's National Health System. I made the switch only six months ago so my domain knowledge is still very limited. Anyway, I have already seen how automation may greatly improve doctors and nurses efficiency.

For example, RIS/PACS(Radiology Information System/Picture archiving and communication system) systems may (and do) save millions euros a year only in radiographic film. I've seen figures stimating the ROI and its only 3-4 years. That without mention the advantages of having radiological data instantly accesible from anywhere, which opens the door to telediagnostics and telemedicine.

That's a very clear example. In other cases it may be harder to calculate real savings(if any), but not the improved care level.

iamelgringo• 17 years ago

This just touched a big nerve with me, so be prepared for a rant.

I've been an ER/Critical Care nurse for 15 years. I've worked/contracted at over 30 hospital, and I've worked in a number of different specialties. I recently went back to school for a CS degree, and I'm working 3 days a week as an ER nurse, and I code the other 3 nights a week on my social news site for economic and financial news, http://Newsley.com.

I get asked all the time by people at Hackers and Founders meetups as to why I'm not going in to Health Care IT. And, the short answer, is that the computerization of healthcare is wrong. I currently work at a hospital that's consistently named one of the nation's "Most Wired", and the more I work at a "Wired" hospital, the more of a luddite I become. As much as I love computers, I think that their role in Health Care should be limited to a very few places.

Why? Here's the long answer:

- Culture. When a physician graduates from med school, they take a vow to "First, do no harm." The culture of safety, is everywhere. Physicians change their practice slowly based on research that often takes years to vet. Medications take years to pass FDA approval. This is all because of the very strong culture of safety in hospitals. Now, try and meld that with an IT culture where a generation in technology is 18 months, and user's heads explode. I've seen CardioThoracic surgeons who makes $750 thousand a year cutting into beating hearts struggle for an hour trying to put orders into a new computer system . A lot of physicians trained in the last 40 years never learned to type, and really don't use computers much. It's simply not cost effective to force physicians to make the huge adjustments in their practice to fit the needs of the IT department.

-Disasters. If any organization has to run in a disaster, it has to be a hospital. There aren't many disasters that can prevent me from writing a note on a paper chart in a disaster. But, every computer system that I've worked on in hospitals crashes. When that happens the whole hospital is crippled. Fault tolerant, reliable systems are essential for hospitals, and frankly, they just don't have the money to spend to build them or buy them. Paper is much more forgiving in a disaster than a huge distributed computer applcation running on 1000 Windows PC's with a datacenter back in New Jersey.

-Time. More often than not, I've seen software systems slow patient care down by a factor of 1-4x. We have to spend a lot more time looking at the computer rather than looking at my patient.

-Money. I've alluded to it above. But, when hospitals actually are running in the black financially, they tend to run a 1-2% profit margin. Software systems to run hospitals usually tend to cost tens or hundreds of millions of dollars, and a large support team to manage them. Hospitals generally don't have the money to pay top notch talent to maintain a really complex IT system.

-Bugs. Bugs are a fact of life for programmers. Bugs happen all the time. But, when a bug happens in a healthcare setting, people die. I've seen in happen, and it sickens me.

-Proprietary data formats. Every application that I've ever worked with in a hospital has been proprietary. That means proprietary data formats. A typical hospital will use dozens of different systems that all have their own proprietary data formats from automated lab systems, EKG machines, vital sign monitors, billing systems, inventory tracking systems, etc... All those systems have to be integrated, and for that data to be available across systems is really hard. (see Bugs and Money above).

After years of thinking about this, and working in this environment, I've come to the conclusion that paper is simply a much better technology for the industry. Really. Paper is cheap, portable, fault tolerant and easy to change. I don't have to wait for months for a work order to go through IT to change a database schema when I want to adjust the data that I'm gathering at Triage for instance. I can just have someone print up a copy of the new form. I'll stop now, but if you want to hear more of this rant, feel free to stop by our next meetup and buy me a beer. By the time we're done we'll both need it. As you can see, it's a hot button issue for me. :)

izend• 17 years ago

Your points are quite valid, however, I must say when I went in for my last surgery a young surgeon was standing at a computer and reading up on procedures on a medical Wiki. I would have to guess that has improved his efficiency versus trying to find the information in some books?

ippisl• 17 years ago

hospital IT systems are big , complex , require training, have bugs , generate changes , etc etc... that's basiclly the essence of your rant. probably like any computer system in any complex organization.

But

they have huge potential(in the long term) , when done properly to transform medicine , reduce costs , prevent errors , give better treatments, enable better research , etc, etc.

In essence , medical IT systems are a huge problem , with a huge payoff. should'nt we tackle the problem because it's huge ?

nradov• 17 years ago

Paper occupies space and eventually becomes expensive to store. Paper can't be easily shared with other providers outside the hospital walls. Paper can't be automatically copied to the patient's PHR. Paper can't be data mined for quality control, outcomes analysis, and clinical research; we must have those capabilities to eventually be able to improve quality of care and drive down costs.

The EMR/EHR problems with usability, reliability, and interoperability are gradually being fixed (although some vendors still lag behind). At this point there's no going back.

Retric• 17 years ago

I would counter by saying software helps save lives when used dealing with medications. Knowing that Drug A + Drug B + Drug C = death is one thing keeping track that Doctor bob does not prescribe A and B, and then doctor Todd hands out Drug C is next to impossible when using paper records, unless the interaction is very well known.

The secret is finding a balance between when something is recorded and when it’s entered in a computer somewhere. While still pricy integrating a computer whiteboard with OCR let’s you mix physical and electronic records at the same time. You can even mix pen based solutions to do the same thing with paper copies. So while current solutions are horrible there it’s far from hopeless.

PS: Hospital software needs to cover a wide range of bases from billing to lawsuits which means it’s really hard to get right. But, just because it's been sold as enterprise software does not mean it's got to be terrible.

samuel2• 17 years ago

I don't think nobody is saying that is migration is easy or risk free. Some of your complaints I think are more related to how has been done the transition in a particular facility than generally applicable.

About culture, it's true that most doctors just don't type (lot of them refuse to do it). That's why they use voice recognition software or handwrite. But the love to see lab results as soon as they are ready, or prescribing drugs trough a GUI instead of handwriting prescriptions(this saves lives).

Disasters, you are completely right. That's why its needed to develop policies for working "offline". It's not easy, but can be done.

Propietary data formats. That's why Hl7 and Dicom standards exist. They're a f*cking mess, but everyone sort of support them.

niels_olson2• 17 years ago

I'm a doc, and I want to agree, but I know we can do better:

- Culture: docs use Macs. 40%. Well above the general population. And my experience has been they have a clue. These are the people doing data analysis, writing research papers, etc.

- Disasters: the military uses computers in disasters all the time. As a prior Navy line officer with some disaster response experience (http://nielsolson.us/Haversian/science/social_science/disast...), I think I can safely say computers can be hugely leveraged in disasters. In fact, the military's medical record system is deployable: you can clone a segment of the database and ship it with the unit into theater, then merge the changes back in.

- Time: thus far, I agree with you. The Navy's surgery program in San Diego is actually on probation for being over on work hours and under on patients. The AHLTA system is fantastically inefficient (apparently the implemented the whole thing without understanding HL-7 messaging). I experience this every day. The paper charts are much faster. And our ER scans there notes. That would be fine by me. Run OCR on the scans.

-Money: barring some major paradigm shift affecting the relationship between doctors and hospital administrators, I suspect this will be the last thing to be saved.

-Bugs: happen. We rely on computers in healthcare all the time in real time, from your vitals in the ER to the Da Vinci machine.

-Proprietary data formats: there are growing standards: -- http://code.google.com/apis/health/ccrg_reference.html -- http://medical.nema.org/

Overall, I really hope this problem gets cracked. Soon. Because if you've tried to go back through a hard films x-ray room that's three blocks away and locked at 2 am because you really want to know WTF is going on with this patient, you realize paper is crazy.

dsteinweg• 17 years ago

I'm curious, could you provide any details around the death caused by a software bug?

cakeface2• 17 years ago

http://en.wikipedia.org/wiki/Therac-25

smiler• 17 years ago

Of course, your perspective on the money is about hospitals and profit making. Over here in the UK with a nationalised health service we of course have a much different view ;)

We are trying to implement a centralised patients records system, but alas it's not going too well. I am sure we will get there. I am sure most of these problems stem from contracts being awarded to large consultancies (Accenture etc) and from my experience, don't seem to have any actual decent development talent but merely people who can project manage, sit on conference calls and create pretty architecture diagrams. Small and nimble teams are the way forward.

"I've seen CardioThoracic surgeons who makes $750 thousand a year cutting into beating hearts struggle for an hour trying to put orders into a new computer system . A lot of physicians trained in the last 40 years never learned to type, and really don't use computers much"

Please! They should learn to type. Learn to join the real world. Why molly-coddle a few old surgeons who are on $750k / year? They'll be retired soon anyway and the next generation will takeover. They should either learn or get out

aho• 17 years ago

I was surprised to learn how just little computerization there is in the ER. Each medical device is basically a standalone unit, with its own input and display. There is no central computer system that automatically alerts doctors if the devices connected to a patient are configured in a potentially dangerous way. For example, apparently it is surprisingly common for surgery teams to use a ventilator, switch to a heart-lung bypass machine, but then forget to turn the ventilator back on once the heart-lung bypass machine is off. This can cause permanent brain damage if the patient doesn't get enough oxygen. There should probably be some system that warns the doctors if both devices are turned off. Unfortunately there is no standard communication protocol for these devices, which makes it impossible to build such a system. Here is a presentation on the state of affairs if anyone is interested: http://www.mdpnp.org/uploads/Capitol_Hill_NSF_CPS_MD_PnP_9Ju...

ecq• 17 years ago

Harvard study: Computers don't save hospitals money

They save lives..

raheemm• 17 years ago

I would attribute this failure to the vendors/developers/IT experts in the healthcare sector. IT works extremely well in many sectors that need speed, accuracy and reliability such as finance, transportation, manufacturing, etc. So why can't IT provide the same level of service in healthcare?

mildweed2• 17 years ago

Its not just about the money. Its about increased quality of healthcare.

I think one of the important things to note is that the true benefits of healthcare reform are going to come with IT in the form of payment/reimbursement reform legislation and a restructuring of Health Insurance. As this article also notes, greater public benefit of IT comes from the data exchange pieces which are being put in place, big news this week with Kaiser sharing data with the VHA, I expect this to be a trend that snowballs more information into the NHIN.

Other pieces of Healthcare IT that will benefit outcomes and burden on the system that this article doesn’t mention is Telemedicine for greater collaboration around providers, that is a sample among other emerging technologies that it will take government funding to really skyrocket.

Taking a massive IT project inside the four walls of a Hospital can be tricky since you have to account for some soft metrics like time saved, reduced LOS’s, and other non direct financial rewards. The other piece that does not have widespread adoption is the analytics piece, mining the data and being able to use for process and efficiency improvement is HUGE, doubt that was taken into account either.

teyc• 17 years ago

Toyota's kanban was done with cards and physical cues. Computers aren't as flexible.

Harvard study: Computers don't save hospitals money

Related Stories