Does prevention save money? __ Yes __ No

DocWPatientOr…it’s complicated.

The New York Times today published a story titled, “No, Giving More People Health Insurance Doesn’t Save Money.” A piece of the argument is, as the author Margo Sanger-Katz puts it, “Almost all preventive health care costs more than it saves.”

What do you think? What’s the evidence? Leave aside, for the moment, the “big duh” fact that at least in the long term saving people’s lives in any way will cost more, because we are all going to die of something, and will use a lot of healthcare on the way. Leave aside as well the other “big duh” argument: It may cost money, but that money is worth it to save lives and relieve suffering. Leave that argument aside as well. The question here is: Does getting people more preventive care actually lower healthcare costs for whoever is paying them?

My thoughts? #1: No consultant worth his or her salt trying to help people who are actually running healthcare systems would take such a blanket, simple answer as a steering guide. Many people running healthcare systems across the country are seriously trying to drop real costs, and how you do that through preventive care is a live, complex and difficult conversation all across healthcare.

#2 thought: It depends. It needs analysis. It depends on which preventive tests, screens, and prescriptions you’re talking about, and how it is decided whom to help with them. Sanger-Katz’ article only shows that we cannot assume that every preventive screening or test saves money and/or is worth the money spent. Mammograms, for instance, show no benefit (no extra tumors caught, no lives saved) over breast exams alone (Canadian Breast Cancer Study, n=89,000 over 25 years).

This is true of many preventive items, including the annual checkup — it’s hard to show a true benefit from them. So yes, if you assume that every preventive test, screen, or prescription is worth it, and then you give more people access to those, you’ll end up spending more money. Equally important, the assumption is that you screen everyone, and you do it the most expensive way, like giving older people regular colonoscopies as a test for colon cancer. There are far less expensive ways to pre-screen people for that. This one assumption alone costs an estimated $10 billion per year in the U.S..

The problem is that these assumptions mean giving a lot of medical care, much of it not even effective, to people who are well. There are reasonable ways to narrow the focus of expensive, personal, procedural preventive care and maintenance to the 5% or so who really need it. Find that 5%, give them extra care, and you will save money.

Productivity? In Healthcare?

by Joe Flower

Obamacare is built on the assumption that healthcare can be more productive, that we can squeeze more health per dollar out to the system that is built to give it to us. Practically everything I write is based on the same idea — big time. I believe we could do healthcare better for half the money we pour into it now.

There is a widely-cited theory that this is fundamentally impossible, popularized by William Baumol, a New York University economist, in a 2012 book, The Cost Disease: Why Computers Get Cheaper and Health Care Doesn’t. Baumol trades on the idea that healthcare is mostly the individual labor of highly trained professionals (doctors, nurses, and technicians) whose labor cannot simply be baked into machines and software. So we can’t expect healthcare to become any more productive, especially as healthcare keeps getting more complex.

We can’t both be right. What’s the daylight between these two radically different points of view?

I believe that the Baumol argument assumes many things that are simply not true. These include:

  • We are using doctors and other personnel at their highest and best use (when in fact we waste masses of clinician time on documentation and other processes that do not add value at all, let alone value that only they could add)
  • The goal against which productivity should be measured is provision of healthcare processes, such as how efficiently one can do a gall bladder removal or an uncomplicated birth (as against, say, improvement in health of patients and patient populations)
  • There is not much wasted motion within those processes (when in fact there is a great deal wasted, as anyone who has applied lean manufacturing principles to healthcare processes has discovered)
  • There is little that a doctor or nurse does that could be supplanted by a machine, or helped by automation in a way that would make it more efficient (a laughably bald assumption being disproven every day in every sector of healthcare).

Most importantly, the Baumol argument assumes that everything we do in healthcare is necessary and beneficial (when in fact at least a third of all we do is waste, unnecessary, not helpful and often actually harmful). What is the productivity of doing a procedure more efficiently, if it is a procedure you actually shouldn’t be doing at all? How do you measure the productivity increase of dropping an unnecessary or wasteful procedure (such as routine colonoscopies as mass screening for colon cancer)?

The flip side is: The Baumol argument takes no cognizance of the many procedures that could be supplanted by less expensive treatment paths, such as medical management for simple back pain in the place of complex back fusion surgeries.

The search for greater productivity in healthcare will lead us somewhat toward streamlined processes, a fair amount to automation, and massively to simply not doing what does not need to be done.