How to save $40 billion in health care costs

What It Costs to Start-Up an Electronic HealthElectronic health records (EHRs) broaden access to patient data and provide the platform for pushing evidence-based decision support to clinicians at the point-of-care. This promotes optimal care for patients, reduces medical errors, optimizes the use of labor, reduces duplication of tests, and by the way, improves patient outcomes. When done in aggregate across all health providers, a team from McKinsey estimates that $40 billion of costs could be saved in the U.S. health system.

Reforming hospitals with IT investment in the McKinsey Quarterly talks about the American Reinvestment and Recovery Act’s (ARRA) $20+ billion worth of stimulus funding under the HITECH Act and estimates that 80% of existing hospital IT applications will be affected by the regulation. Hospitals will be spending about $120 billion to meet the adoption and meaningful use provisions of the Act. This equates to $80,000 to $100,000 per hospital bed. ARRA incentive payments will cover roughly 20% of this cash outlay, meaning that $60-80K won’t to covered.

But McKinsey says, “Hold on!” There are ways to recoup the spending gap between HITECH incentives and cash-out-of-the-hospitals-budget. McKinsey’s research calculates that optimizing labor, reducing adverse drug events and duplicate tests, and adopting revenue cycle management can help the average hospital save $25,000 to $44,000 per bed each year. That gets to the $40 billion in annual savings when multiplied across all hospital beds in the U.S.

In operational terms, the savings accrue through:

  • Managing inpatient beds more efficiently using equipment-scheduling software
  • Optimizing the use of clinical equipment
  • Determining optimal staffing
  • Reducing administrative waste
  • Reducing adverse drug reactions through computerized-physician-order-entry (CPOE) which cost $8,000 to $15,000 per bed each year (up to $3 million for a 200 bed hospital)
  • Managing the revenue cycle by billing unbilled services, equivalent to 0.4% of hospital services, or $4,000 per bed.

Jane’s Hot Points: The McKinsey team rightly points to three critical success factors for maximzing health IT investments that the most wired, effective hospital-adopters have learned: get critical buy-in among clinicians and hospital execs early in the HIT adoption process; ‘radically’ simplify health IT architecture; and, elegantly plan and execute.

It’s the implementation phase in health IT adoption that so often gets short-shrift. McKinsey notes that Canada’s hospital system devoted 30% of its entire budget to change management. That’s a big number, but it’s also where rubber meets road: a capital outlay of $N million is the easy part of HIT adoption. The follow-on implementation resources, both in terms of sheer dollar volume and labor/staffing, along with disruption of clinical workflow, is the hard part. But getting to meaningful use will require no small amount of implementation effort in the form of evangelism, education and training, and ongoing assistance and support.

Categories: Uncategorized

Tagged as:

39 replies »

  1. Hi Tina! Great blog. I’m like you – punster first and then plot gets in there. I see you’re from Boston. My novels are set in and around Boston and the fictitious town of Laski, Mass. Thanks for following on Twitter… will catch up with you there! Cheers!

  2. As part of my university degree, I was given a chance to volunteer for six weeks (April to June 2010) in Knysna, South Africa, with an organisation called EDGE of AFRICA. Committed to responsible travel, EDGE of AFRICA provides “volunteer and internship placements in South Africa for gap years, career breaks, university internships, school groups and corporate team building projects.” Their aim is to make a direct positive impact on the local community and environment, while allowing travellers to gain firsthand exposure to the colourful local heritage, culture and traditions.

  3. I worked for Wellpoint as an RN for 2 1/2 years. It is not HEALTH CARE it is SICK CARE that they concentrate on DENYING coverage. TOO BIG TO FAIL. Google Wellpoint subsidiaries. There are 116 all over US, one in China. None of their DRUG COMPANIES are listed here. This company has NO ETHICS. TOO many details.
    Here’s my point. Angela Braly CEO gets over 1 mil in salary and 8-9 mil in perks. This is only one company.
    I have more and I’m doing videos of information such as PPMSI that Wellpoint/Anthem/Blue Cross pays to US when we deny claims. The pay groups like Hill’s Physician’s. There are 8-10 small print pages of these groups. PERKS to all.

  4. I think that does post is helpful in a way that they try find ways in on how to shell more money to spend on more important things and that techonology helpful in achieving it.

  5. I think that does post is helpful in a way that they try find ways in on how to shell more money to spend on more important things.

  6. ” with different branches and checks and balances and bills of rights and independent judiciary,”
    Had, Executive branch and their appointees write laws every day without ever going through congress. EPA is regualting emissions based on their own directives.
    Abortion, gay marriage, and many other laws have been defacto written by the judicial branch.
    Our lkegistlative branch is a joke that passes meerly the name of bills then allows Executibve and Judicial to fill in the blanks.
    What’s new is the degree to which we have allowed our protections from mob ruled to be weakended or eliminated. When else in our history have we ever had so many people dependent on the federal government for basic living?

  7. “A democracy is nothing more than mob rule, where fifty-one percent of the people may take away the rights of the other forty-nine.”
    Thomas Jefferson
    Nothing new in this thought, Nate. We have been debating the Tyranny of Majorities since the very beginning. It was a great concern for those writing the Constitution as well. James Madison’s hope was that the sheer size and diversity of the new country will ensure “…that no common interest or passion will be likely to unite a majority of the whole number in an unjust pursuit.”
    This is why we have such a complex system with different branches and checks and balances and bills of rights and independent judiciary, etc. Not to mention that we don’t have a pure democracy.
    I believe Mr. Madison was correct.

  8. when 51% get together to take advantage of the remainging 49% that is not what this country was built on. We are no longer a nation of equal rights, instead we are in a race to see who can be first to form a collition of 51% and enslave the rest. I seem to remember equality being mentioned more then once at founding.

  9. We haven’t lost anything. The fight was for keeping these decisions with the people on this side of the ocean and it was a complete success. The nature of taxation and the use of the proceeds was always debated, and we are continuing to do so today, and we will probably continue to do so for the foreseeable future.
    You have your opinions and I have mine. This is what elections are for. Sometimes your views prevail and sometimes mine do. This is exactly the process they fought for and it’s working just fine whether Bush is in the White House or Obama is Commander in Chief.

  10. we know the country has gone down the wrong path when people refuse to share equally in responsibility, when they say it is OK to tax him 50% but I don’t want to pay any taxes then we have alredy lost what our founders fought for.

  11. being truly free is having 50-70% of your personal production confiscated to be redistributed to others who refuse to work, Margalit you surly have a different perspective on our history then I do.

  12. They were fighting for representation, not for a tax free existence. The more advanced society becomes, the more dependent we become on technology and the higher the standard of living is, therefore more “stuff” becomes necessary for minimum existence and the ability to “pursue happiness”, select representation and be truly free.
    “fruits of labor from the minority of the nation that is productive”
    The producing class is not a minority. Those who do no real labor, but certainly enjoy other people’s fruits are a minority. And in my book, they ought to be happy with a bit less unearned fruit.

  13. “We, as a society, will pool our resources”
    What exactly does this apply to? So far it appears food and shelter are benefits guaranteed at birth. We have since added utilities, healthcare, education.
    I remember us fighting a war so we were entitled to the pursuit of these I don’t recall ever decising as a nation to guarantee these to everyone that gets across our border and doesn’t want to earn them. How much farther do you want to take this? Once you forceably confiscate the labor and fruits of labor from the minority of the nation that is productive what else are you going to claim people are entitled to? Does everyone deserve an electric car? What about Cable TV so they can be enformed? Internet is already in the works are a guaranteed right. I would think you could learn the slipperly slope of socialism and communism watching the past 100 years of them failing and killing tens of millions in the process without having to actually go through it. Apparently not.

  14. I choose what I do. Anything else is tyranny. “Society” has an agenda. Physicians have ethics. You would be amazed how few life and death decisions actually change the outcome. Avoiding futile care and helping with death and dying are of immeasurable value.

  15. “But I choose.”
    No, you don’t.
    Going to medical school for 12 years and graduating with large debt does not entitle you to choose who lives and who dies.
    We, as a society, will pool our resources and pay you as much as we can afford. If that’s not enough, you are of course free to decline, but there will always be someone else, equally qualified, willing to apply for the job, because, comparatively speaking, we pay rather well.

  16. The cheapest healthcare is to dissolve CMS and the Joint Commission, have real tort reform. and let doctors do health care. All patients will be charged and will pay the same price. They can buy insurance if they want it. If they can’t afford it, they can’t have it, unless I think they are really indigent. Then I will give them everything I can. But I choose.

  17. It does not matter. When the patients’ records vanish from the terminals in one fell swoop without warning in a 551 bed hospital, redundancy is irrelevant.-propensity
    This has to be dinasour world. Something like a Oracle RCA would fail 1 in billion times and then you can have daily exports of data from which one can always restore.
    Let me describe the Hospital IT issue a bit more succintly from skills perspective. No respectable IT guys join hospital because they don’t provide any sort of career to them. Now if only hospitals could consolidate that will put an end to every hospital having a CIO and IT executives. The flip side is that we will have another monopolistic organization in hand.
    I think we had enough of small business style physicians and hospitals. They really need to conglomerate because that small friendly neighbourly image is nothing but a mirage unless you are in that remote rural area and they also end up with severe disabilities to do anything progressive such as using software or checklists. Just a few large hospital and physician chain would automatically kill the plethora of HIT software and leave the strongest one.

  18. Corpuscle Connie: That is your name? A bit odd, do you not think?
    The commenter above regarding the study gives a link to an abstract of the study which is published in a journal that is available to subscribers. In fact the comment itself is the abstract.
    This synposis is not very meaningful. The details of the study are. The conclusions of the study may in fact be meaningless in regard to the efficacy or lack thereof of any EMR or EMR/PM system. Impossible to know without checking the details.
    In any case the correlation that the study asserts may be worthwhile investigating for any number reasons, but in and of itself does not conclude anything, certainly not what the commenter presumably intends as evidence of something.
    The nay-saying nonsense from a certain group of physicians on this issue is astonishing. inchoate’s comment is apropos.

  19. “EMR use does not imply/guarantee EBM, similar to the fact that a brand new Lexus does not guarantee seat belt use.”
    rbar, this is true, but if the vehicle has no seat belts installed, it does guarantee that seat belts will never be used.

  20. Any health care organizations that are needing to be convinced as to the value of EMR first need to sit down and list those areas where they are inefficient. If this inefficiency can then be correlated to a lack of immediate patient or statistical patient information, then there is a possibility that EMR can improve the situation. For those organizations that do not yet have a continual improvement or process improvement philosophy, they really need to do so independent of EMR because they are likely quite a ways from being as efficient as they could be.

  21. I really think that elderly should be given a health cares shouldered by government in all countries. Though I know it is not possible with some of the poor countries, but it should be prioritize in countries that have resources.

  22. Excellent point, rbar. The pediatric cardiologist discussed earlier could have just as easily said that the test was old and needed to be repeated; or, if using paper, could have asked his secretary to get the report and decided not to repeat it. I don’t think the EMR was the direct cause of his decision not to repeat the test.

  23. I perceive that an incorrect connection is made between EMR use and evidence based practice. Apart from picking up allergies (which is sthg a pharm tech could/should do when filling an inpatient med order), one can do exactly the same shitty “aggressive” wasteful medical care via EMR. EMR use does not imply/guarantee EBM, similar to the fact that a brand new Lexus does not guarantee seat belt use.

  24. propensity wrote:

    Patients are not airplanes.

    your rejoinder could not more perfectly render your cluelessness had I written it myself.
    here’s a simplified entry point for your long, slow climb to understanding:
    patient : your practice as air passenger : airplane
    you should also read What’s Keeping Us So Busy in Primary Care? A Snapshot from One Practice (NEJM April 2010), not so much for its findings as for the authors’ practically giddy enthusiasm over producing what to any other competently managed business is (and has been for a long time) a regular, and vital, periodic report on its operations.
    Because, “dr” propensity, insofar as your professional accomplishment, it doesn’t matter to anyone if you manage to validate yourself to yourself; it matters that you can do so for everyone else.

  25. Dearest Wendell,
    That link worked here. Fix your technology and catch up with the world.
    HITECH is wasting time and money. It is an experiment, using unwitting patients as guinea pigs.
    Doctors are advised not to participate and rebel against these deceptions and threats. Complain about how you waste your time, complain about the dangers to your patients, complain when your patients are neglected by the clicking nurses.

  26. Although implementing the solutions looks challenging, the tremendous benefits accrue as a result is rewarding. There must be a beginning and not everyone like changes. Nevertheless, determination from the top management would see the rapid roll out of the healthcare automation throughout the nation.

  27. HITechxpence: This is one of many studies. The details on this study are not available at that URL. In particular the datasets used have to be checked carefully. The details on systems used, implementation, etc. have to be understand to determine why the reported results are what they are.
    Based on the methodology described: “Marginal effects estimated using fixed effects (within-hospital) OLS regression.” for the study, gross datasets are used to establish a correlation, but any cause for the correlation is unknown. God here is in the details.
    “if propensity and Techner ran the FAA they would yank all that dangerous computing gear out of airplanes & have pilots land those suckers by the seat of their pants.”
    Why these two commenters appear repeatedly as nay-sayers is unknowable from their repeated commentary.
    inchoate’s quotation above is accurate based on their commentary. Blind rejection of technology that neither appears to understand at all.

  28. Savings NOT:
    Health Serv Res. 2010 Aug;45(4):941-62. Epub 2010 Apr 9.
    Electronic medical records, nurse staffing, and nurse-sensitive patient outcomes: evidence from California hospitals, 1998-2007.
    Furukawa MF, Raghu TS, Shao BB.
    School of Health Management and Policy, W. P. Carey School of Business, Arizona State University, PO Box 874506, Tempe, AZ 85287-4506, USA. michael.furukawa@asu.edu
    OBJECTIVE: To estimate the effects of electronic medical records (EMR) implementation on medical-surgical acute unit costs, length of stay, nurse staffing levels, nursing skill mix, nurse cost per hour, and nurse-sensitive patient outcomes.
    DATA SOURCES: Data on EMR implementation came from the 1998-2007 HIMSS Analytics Databases. Data on nurse staffing and patient outcomes came from the 1998-2007 Annual Financial Disclosure Reports and Patient Discharge Databases of the California Office of Statewide Health Planning and Development (OSHPD).
    METHODS: Longitudinal analysis of an unbalanced panel of 326 short-term, general acute care hospitals in California. Marginal effects estimated using fixed effects (within-hospital) OLS regression.
    PRINCIPAL FINDINGS: EMR implementation was associated with 6-10 percent higher cost per discharge in medical-surgical acute units. EMR stage 2 increased registered nurse hours per patient day by 15-26 percent and reduced licensed vocational nurse cost per hour by 2-4 percent. EMR stage 3 was associated with 3-4 percent lower rates of in-hospital mortality for conditions.
    CONCLUSIONS: Our results suggest that advanced EMR applications may increase hospital costs and nurse staffing levels, as well as increase complications and decrease mortality for some conditions. Contrary to expectation, we found no support for the proposition that EMR reduced length of stay or decreased the demand for nurses.
    PMID: 20403065 [PubMed – in process]

  29. if propensity and Techner ran the FAA they would yank all that dangerous computing gear out of airplanes & have pilots land those suckers by the seat of their pants.

  30. “In IT world redundancy equals safety.” _Vikram.
    It does not matter. When the patients’ records vanish from the terminals in one fell swoop without warning in a 551 bed hospital, redundancy is irrelevant.
    The only redundancies with which I am familiar are the veneers from CIOs and hospital administrators who layer it on thick with vendor articulated statements: “…but patient care was not impacted”.
    I have experienced the sudden breakdown first hand and that is always a lie. Patients die and are injured from the delays in treatment. Trial lawyers, take note.

  31. “Managing inpatient beds more efficiently using equipment-scheduling software
    Optimizing the use of clinical equipment
    Determining optimal staffing
    Reducing administrative waste
    Reducing adverse drug reactions through computerized-physician-order-entry (CPOE) which cost $8,000 to $15,000 per bed each year (up to $3 million for a 200 bed hospital)
    Managing the revenue cycle by billing unbilled services, equivalent to 0.4% of hospital services, or $4,000 per bed.”
    If we’re really designing EHRs to do all these management chores (aside from drug interactions), no wonder they’re unusable by clinicians.

  32. In IT world redundancy equals safety. By skimming staff, equipment to just in time they take away capability to handle surge in patients or sudden depletion if staff such as during storm. Can’t argue with adverse drug events management, reduce waste and revenue cycle management.

  33. This report does not consider the horrific mistakes that are caused by the HIT CPOE, and falls of patients due to the neglect while the nurses are clicking away.
    There are few if any reports that are not sponsored by the HIT industry that show benefit of HIT. One stroke due to neglect costs $.5 million with a rehab chaser and SNF care.
    There many reports on the FDA MAUDE database depicting death, injury, risk, CPOE breakdown, CPOE defects, misidentification, inappropriate radiation, dnager causing lack of usability, etc. The FDA states this is the tip of the iceberg.
    DAVID BLUMENTHAL blows them off, depreciates them, ignores them, etc calls them names: anecdotes.
    Hey guys, send more of these anecdotes of death, injury, falls, strokes, bleeds, overdoses from CPOE to the FDA.
    Use Form 3500A at MedWatch. The FDA wants them.

  34. Intelligent and methodical implementation, most importantly commitment and adequate time on the part of users, is the most important single factor in realizing any benefit from EMR/PM application installation. Without that, no system will perform very well and the cost will be exorbitant.
    This assumes implementation of good software, of which there is plenty available.
    I marvel every time I see the break-down of presumed costs of software acquisition and implementation.
    There is no reason why the costs have to be anywhere near as high as the accompanying exhibit shows. As I usually note, existing computer equipment and/or used equipment where additional or replacement equipment is needed and which is available for 1/10 of the cost of new equipment is almost always adequate to run any new software installed.
    In addition any EMR/PM system that either (1) is itself FOSS – e.g. VistA for hospitals/hospital systems, PatientOS for hospitals or ambulatory facilities, openEMR for smaller physicians practices or (2) may have a licensing cost itself but is built primarily with FOSS (e.g. eClinicalWorks written in Java and using the MySQL RDBMS) offers software that is either completely free of any licensing cost or obviates a significant part of the licensing cost by being based on FOSS.
    Outside “consultant” services, if the services are managed intelligently by the implementing entity, do not have to anywhere near as high as the number included in the exhibit.
    In other words the figures in the exhibit are grossly inflated if a buyer spends a modest amount of time in researching the options available.

  35. Interesting post. We published a research study on electronic record keeping by Himmelstein et al in our January 2010 issue. The Harvard research team concluded the following: “As currently implemented, hospital computing might modestly improve process measures of quality but does not reduce administrative or overall costs.”