Making Health IT Work in a Hospital: the CMIO Should Be a Doctor

A hospital’s Chief Medical Information Office (CMIO) should be a physician, says Pam Brier, president and CEO of Maimonides Medical Center “because nobody knows a doctor’s business like a doctor.” 
As a hospital’s information technology (IT) point person, a CMIO needs to be able to persuade physicians and other health care professionals that health information technology (HIT) can help them care for patients.

It is not that Brier believes that non-physician managers can’t talk to doctors. . . After all, she herself is not an M.D. Yet she runs Maimonides, a top-ranked 700- bed teaching hospital in Brooklyn, New York.

On the other hand, Brier is not an MBA either. She has a master’s in Health Administration, which means that, unlike many hospital CEOs who went to graduate school to study business, she understands that an organization that provides health care is not a “business” in any ordinary sense of the word. A hospital is a service organization: its raison d’etre is to meet the needs of a community and its patients.

It is telling that before coming to Maimonides in 1995, Brier spent fifteen years in New York City's municipal hospital system, and  still says: “Even though I'm not working for government anymore, I still feel that I'm a public servant."

(A 2008 book that profiles Maimonides, titled Hospital: Man, Woman, Birth, Death, Infinity, Plus Bad Behavior, Money, God and Diversity on Steroids,” offers insights into what Brier means by public service. The former chair of orthopedics at the medical center told Julie Salamon, the book’s author, that he “fell out with Brier because he wanted to give priority in the waiting rooms to patients who paid out of pocket or who had full insurance: ‘People who pay for health care don't want to sit in a room with fifty people. They want to be seen in a timely manner. I think that's very reasonable.’” But Brier was not willing to make the sick queue up according to ability to pay, with the poorest at the back of the line.)  

                                Finding a Physician CMIO

For Brier, choosing a CMIO who is a doc didn’t just mean picking an IT expert who has a medical degree. She wanted someone intimately and actively involved in clinical care. At Maimonides, Dr. Steven Davidson not only serves as the hospital’s CMIO, but as the head of emergency medicine.

At the same time, Davidson is committed to Health IT. When Maimonides hired him 14 years ago, one of his conditions for coming to work at the hospital was that it implement electronic medical records (EMRs) in the emergency department.  He was in the vanguard of those who understood the potential of HIT.

Davidson quickly immersed himself in the ED’s IT requirements. “A number of us really took the lead in pressing for institutional CPOE [computerized physician order entry] implementation,” he recalls. “By doing that we were able to substantially refine what was happening in the emergency department, and we demonstrated that both from a community acceptance measure and with the financial improvements we brought to the hospital.”

As Brier points out, Davidson is not only focused on emergency medicine, but on other areas such as primary care and inpatient services as well. “So he knows a lot about care outside of the emergency room, which makes him a really good person to work on these IT issues because he is immersed in what goes on outside of his own area.”

It seems that Davidson has succeeded. The American Hospital Association has named Maimonides one of the nation’s top 100 “Most Wired and Wireless” hospitals.

        Health IT That Doesn’t Help Physicians or Patients

In a recent post on Kevin M.D., Donald Burt, M.D., the chief medical officer at PatientKeeper, a company that helps physicians integrate information from a variety of IT systems, agrees that “the CIO, who is not an M.D. can’t fully appreciate how cumbersome, distracting and unproductive a traditional hospital information system (HIS) can be. The hard truth is computerized systems that don’t fit into the physician’s workflow don’t stand a chance of being readily adopted by physicians, and that includes systems like CPOE, which are part of the ARRA-HITECH ‘meaningful use’ requirements.”

“Doctors are happy to make screen touches, mouse clicks and keyboard strokes if the application is right,” says Burt. “But too often . . . the software that CIOs often try to ‘sell’ doctors on using typically wasn’t designed with physician users in mind.”

The University of Pennsylvania’s Dr. Ross Kopple concurs: “Designers of healthcare information technology (HIT) must be exquisitely sensitive to the non-linear, context dependent, fast communication-dependent, interruption-filled, uncertain, and collaborative nature of hospital clinical practice,” writes Kopple in the Journal of Biomedical Informatics.

In the last two or three years hospitals have made progress, says Burt: “A growing number of institutions are ‘teaming’ the CIO with a Chief Medical Information Officer (CMIO), an MD who offers exactly what the CIO needs: a physician deeply enmeshed in the hospital’s clinical systems who can be a credible and effective liaison and technology advocate with physicians.”

Nevertheless the doctors and nurses responding to Burt’s recent post on Kevin M.D. unanimously agreed that hospital HIT is falling far short of their needs: “It’s amazing how ass-backward the overwhelming majority of hospital software is,” says one young doctor-in-training. “The simple fact is that technology should simplify and streamline workflow, the moment it ceases to do that it  . . . ultimately compromises patient care.  Forward thinking hospitals need to realize that investing in a solid IT infrastructure vetted by a DOCTOR is the best thing they can do.  . . I’m a medical student and I know tech better than half of the jokers we call IT staff around here, but you could never pay me enough to take a ‘CMIO’ job because it would inevitably become emasculated by short sighted hospital administrators who are still stuck in the 80′s.”

It seems that in a great many cases, CIO’s are trying to coerce doctors and nurses to use a multi-million dollar Health Information System (HIS) that, as Burt puts it “the board of directors was persuaded to purchase  . . .”

Too many HIT consultants and vendors don’t have clinical experience, don’t understand a hospital’s workflow, and are primarily interested in selling the most expensive system available. The great danger is that they will manage to peddle their system to the hospital’s board with promises of great savings. A hospital board should consult with a physician-CMIO who can explain that "the key to success for hospitals is to make physicians want to use their HIS systems, rather than be forced to use the systems. Beating affiliated doctors over the head with a stick hasn’t worked,” Burt adds: “Witness the single-digit adoption rates of commercial CPOE systems over the past 40 years – and it isn’t going to work now.”

It will take time for physicians who understand HIT to find systems that actually fit an institution’s workflow in the many different settings within a hospital’s walls: the ER, the Intensive Care Unit, the nursing stations. . .

“I've noticed that the IT training for the staff and doctors before implementing the system is often very basic or irrelevant to how and what you will be using the system for,” says Andonis Terezides, a resident surgeon in the Division of Oral-Maxillofacial Surgery at Jackson Memorial Hospital in Miami. “The training never seems to cover the important factors and variables that are faced in each department and specialty. We tend to have to learn how to deal with these situations on the job in front of a waiting patient.”

Different parts of the hospital require HIT designed to meet workflow in those areas, yet departments also must be able communicate with each other through their IT. (Not long ago I wrote about a Manhattan hospital that managed to buy IT for its ER that didn’t “talk to” IT in other parts of the hospital.)

This is why I continue to believe that Washington is making a mistake as it rushes hospitals into adopting HIT—or face penalties in 2015.

I am in no way suggesting that we don’t need HIT. And some hospitals already have succeeded in installing systems that work. Maimonides appears to be one of them. But many institutions are scrambling to meet the deadline, and this is bound to lead to bad decisions.

In 2020 or 2025 many hospitals may discover that the choice they made in 2014 was less than optimal, and now they must strip out the old system, and install a new one. This would be good news for the HIT industry, but terrible news for Americans facing escalating health care costs.

As I have argued in the past, too many people are selling information technology to hospitals and doctors—and selling hard.

Moreover hospital boards should not be making the calls about the HIT systems that hospitals purchase. This is not an investment decision. It is a health delivery system decision that should be made by clinicians who understand how their hospitals deliver care in real time.  Experience tells us that when clinicians succeed in making care safer and more efficient, they save health care dollars. In the long run, they could save enough to more than pay for Healthcare IT. But not if a money-driven HIT industry calls the shots.

15 thoughts on “Making Health IT Work in a Hospital: the CMIO Should Be a Doctor

  1. The hospital where I do my clinical made the same mistake you cite: the IT system the ER uses does not talk to the IT system the rest of the hospital uses. The printout that is ported over to the main system is nearly impossible to read and understand.
    The paper reproductions of documentation systems is ugly and often hard to follow. I suppose that’s great if you want to keep unfriendly eyes from reading a chart, but it defeats the purpose of communicating patient care among members of the health care team.
    The failure to consult with the staff who actually have to USE the systems that are created has always been one of my biggest gripes about healthcare documentation systems.
    I find it ironic that the best documentation system I’ve ever seen wasn’t designed for actual use, but as an educational tool for students in the lab.

  2. The only way health IT is going to work nationwide is if the federal government MANDATES ONE STANDARD.
    Otherwise, what we have is a failure to communicate. We will have 50 different IT companies making 50 different health IT “solutions,” none of which talk to the other because its in teh corporations best interest to separate into fiefdoms and compete for market share rather than collaborate and form one solution that can be used nationwide.
    Instead of pouring billions into private corporations, the solution is for CMS to say to every clinic and hospital “you’ve got 5 years to implement CPRS/VISTA”
    Its the best health IT solution we have out there, and it will solve all the probelms of communication between clinics and hospitals nationwide.
    Imagine a day when you can walk into a VA clinic in San Diego and your doctor can look up every single medicine you’ve ever been on and look at the last 20 clinic notes that were done in Maine, Florida, Ohio, and Texas. Seem like a dream? It happens every day w/ VISTA in this country.
    GE, Siemens, MediTech, and the rest of the health IT conglomerate cant come up with a product like VISTA because they are only interested in parsing market share, not in really forming a true IT solution.

  3. The HITECH act that is pushing so much of this HITalong with ARRA and last year’s health insurance legislation are perfect examples of how well meaning legislation is turned on its head by today’s croney capitalism.
    Why is it that doctors are often uninvolved in an information system that they will use every day? Why is the board pushed by the executives to decide on a system? Because governance is all so centralized. Decisions are not allowed to be made at the MD or department level by the people who know their own needs best and who have a responsibility to the patient as well.
    It is seen all to clearly in IT, but the same problem infects many hospitals.
    Ms. Brier is a hero and probably a dying breed. The statements she has made publically will be used against ger should she want to work at another system. And the system she is in now will get her someday.
    PS, the doctors don’t equal the system.

  4. Jason, Look at who donates to get our current politicians elected. GE, Google, Siemens, etc. It ain’t Vista. All have a vested interest in that never happening. Chalk your idea up as an idea that should happen, but never will because of money and power.

  5. iI’ve been calling for this since at least 1996, when it became obvious (as documented at a number of deidentified cases at ) that amateurs in medicine did not cut it in clinical IT leadership roles.
    I further add the the “C” in “CMIO” should not jsut be titular, and that the incumbent should have direct control of resources, staffing, etc. Otherwise they are an ‘internal consultant’, not a “C” level official.
    As CMIO, I had direct control of: zero budget, zero grant of authority, zero personnel.
    A short while later as group director in pharma: 50+ staff, 12+ direct reports, $13 million annual budget at my disposal, $100K grant of authority.
    Quite the contrast.

  6. “Joe Says” writes:
    Why is it that doctors are often uninvolved in an information system that they will use every day?
    I think this question is especially poignant, some 60 years into the “computer revolution.”
    There’s a meta-issue here. I state it as follows:
    “Management of HIT projects without high levels of clinician involvement is, by definition, mismanagement.”
    This is a first principle. There is nothing to debate or discuss.
    — SS

  7. Jason is entirely right. We need to simply mandate that everyone figure out how to use the same system (preferably Vista, since we’ve already paid for it and despite its detractors is quite popular overall). If we are so desperate to give money to private industry we can pay consultants to figure out how to implement and integrate it.
    I’ve written about this before here, but where I work we have __ different EMRs. 1 for inpatient orders and meds, 1 for inpatient records (handwritten notes are scanned in), 1 for imaging, 1 for patients to email me, 1 for outpatient visits, and 1 for ob/gyn. None of the systems talk to each other even though a few are even made by the same company. If I want something from one system to go to the other, I have to print it out and request it be scanned into another system. Some records can’t be printed out (supposedly due to HIPAA concerns), so I have to find a computer with a big screen, take a screen shot, crop it, print it, and scan it. It is way too easy for an important piece of information to be lost.

  8. There is no incentive for them to communicate. They all make more money doing it that way. It’s like having a betamax, VHS, HD DVD, blu- ray, with an 8 track player in each office. Follow the money, look at the donations.

  9. This is just another example of how health care personnel are being increasingly marginalized in designing and managing our health care system of the future. The much bigger issue I see is what will happen as more and more doctors become employees of major health care systems, and thus beholden to the administrations that run these institutions and determine how resources are apportioned. These buisiness people do their jobs very well with a clear motivation to grow the corporation and to increase revenue, which run counter to what we want from our health care systems presently. Physicians have been systematically co-opted into these health care organizations, but often with little say in the management of these entities.
    Health care IT has largely been designed to do what these decision makers want, which is to allow better coding and capture of encounters and thus maximize revenue, while giving us less than well thought out clinical platforms.
    Just another example of when you turn the shop over to the bean counters, expect to get a bean counter mentality with a different definition of success (increased profit) rather than what should to valued (better health outcomes at a lower cost)

  10. Keith, Sharon, Scott, Jenga, Jason, Panacea, Joe Says, Dr. Rick
    I agree with what you are saying.
    Except– I would argue that some major health systems really are patient-centered, and doctor-centered.
    I think of Geisinger, Kaiser,as well as some academic medical centers (Maimonides, UCSF, Dartmouth,Mayo, Pegout Sound . . . At these places the CEOs and other executives are not bean-counters.
    Moreover, as we develop accountable care organizations, they will be paid in a way that rewards better outcomes and cost control.
    Those medical centers that focus on increasing revenues are going to have a hard time as both Medicare and private insurers change how they pay for care.
    If these hospitals get into financial trouble, their administrations will be replaced.
    Wow, what a crazy system-All of that scanning, photographing, cropping, printing . . . Sound like a total nightmare.
    We do need top-down direction, and it should have come from Washington.
    But I’d be wary of hiring consultants to figure out how to implement HIT. What’s really needed is an MD within the hospital (or an MD the hospital hires) who knows a great deal about IT, can observe workflow in different parts of the hospital, and put together an integrated system that suits the needs of different departments. ..
    Scott– Please do keep on writing about these problems. AS you say: “Management of HIT projects without high levels of clinician involvement is, by definition, mismanagement.”
    And thanks for the links.
    Yes, I’m afraid the lobbyists representing the companies making money on HIT made sure that rather than adapting VisTa for widespread use, the government turned to the private sector.
    What you say is so true.
    “Letting the market decide” has created chaos.
    I just can’t understand how a hospiatl CEO, CFO,or board would let a hospital buy a health IT system for the ER that doesn’t “talk” to the health IT system used in the rest of the hospital.
    I recognize that an ER has special needs, but it is possible to have an integrated system that works throughout the hospital . .
    Joe Says–
    Brier has been at Maimonides for 15 years,andbefore that she survived NYC’s public hospital system.
    I doubt that anyone is going to “get her.”
    I also don’t think she represents a dying breed, but rather a new breed of forward-looking reform-minded administrators.
    The fact she is a woman may have something to do with her very different outlook. . . . Looking ahead, I suspect that more and more women will become hospital administrators and that, in many cases, they’ll help break down the very rigid hierarchy that has stood in the way of change. . .
    Dr. Rick–
    I agree–people do need to listen to Silverstein.

  11. I disagree with you Maggie. I see no evidence that things will change, or at least change in any major way under the new leadership. Even if things change for a short while, it will change back.
    There is simply too much money in medicine and it entices the self serving psychopathic megalomaniac “leader” types. They don’t play fair.