When was the last time you looked at a hospital bill, or one sent directly from your physician? The reason I ask is I have been spending some time trying to develop a clear enough picture of Accountable Care Organizations (ACOs) to describe them easily. After all, ACOs are not something you can touch and see. You cannot just walk up to the third floor of Our Lady of Perpetual Billing and be shown an ACO.
I think it is extremely important to understand what an ACO is before trying to build one. Much of the difficulty in building an ACO has to do with the fact that something, in fact a great many somethings, will have to change for an ACO to function. The question then becomes, change from what to what?
Back to the hospital bill. Scan through the list of charges, and then press the F5 key to let me know when you are ready. Now, highlight all the line items that charge you for the care you received … I found the same thing; there are not any. Volume versus value. Caring for you versus doing stuff to you. The bill of charges under today’s business model is a blow-by-blow description of what was done to you; x-rays, medicines given, IVs, etc.
Hospital billing is not unlike a hotel bill; it is just longer and you do not earn frequent illness points. Embedded in your hospital bill are charges for food and cable television, just like you had been staying at the Four Seasons.
So, as we move from volume to value, how will that impact healthcare information technology, assuming anyone remains standing after Meaningful Use and ICD-10? I keep preaching about how the hospital’s business model must change in order to understand what will be required of IT. To do so, let us compare two very different business models and their operations, both of which are in the same industry.
Hyundai and Bentley. Volume to value. Just-in-time manufacturing versus don’t-rush-me manufacturing. Nobody would argue with the fact that the information systems and business processes needed to run Hyundai’s business are very different from those of Bentley.
I watched a show on how Bentleys are built. A team of people is assigned to each car. Depending on the car’s options, some people roll off the team and others are added, but the team “owns” the car from start to finish, and each subsequent person inspects the work of the prior person.
At Hyundai, it is not apparent that anyone “owns” the car. People have line responsibility; they own a piece of a process. I could be the “left lug nut guy,” having absolutely no responsibility for the rest of the car.
I think this is the degree of change an ACO will require in order to be effective. We will have to change from being lug nut specialists to becoming care owners. This then brings us back to the question of what IT systems will be needed to charge for and manage care.
Unlike moving from ICD-9 to ICD-10, there is no mapping model to guide the change from today’s business model to an ACO model. Three IVs and one MRI do not translate well to 4.5 Accountable Care Units (ACUs) which are then billed at whatever happens to be the going rate.
Today’s systems calculate charges based on what was done to you — $86 million gazillion for the MRI. If requested, nobody in finance or information technology will be able to vivisect the bowels of SAP or Lawson and show you where the information is that records how much the MRI procedure costs. Few can explain how the business processes and information systems that support today’s lug-nut charging model can support and report how the hospital manages its business. Nobody even pretends to explain how effective those same processes and systems are at reporting the quality of care delivered.
The ACO model will require processes and systems that capture, allocate, and report costs. The ACO model will also require processes and systems that can aggregate people and procedures into ACUs and relate patient costs against those ACUs.
We do not have those systems. Since current hospital systems are incapable of really managing today’s business requirements, we should not adapt them to the ACO model.