The Rest of the Story

Yesterday, while looking for something unrelated, I came across an article that discussed a genetic test that could predict the occurrence of Stevens-Johnson syndrome (SJS) in persons being treated with a prescription drug, carbamazepine (Tegretol).  This is not a subject of general interest, really, since carbamazepine (CBZ) is being used less often now, and the condition occurs only 8 times in 1,000,000 patient-years of exposure. 

It got me to think about the economics of the testing.  Genetic testing tends to be very expensive.  I wondered if it ever would be feasible to do such testing routinely, before starting treatment with a medication.  How much would it cost to prevent one case of Stevens-Johnson syndrome?   What is interesting here is not so much the specific case of CBZ and the risk of SJS.  Rather, what if we someday have a panel of genetic tests that could be used to predict both positive and negative responses to medications in individual patients?  That would be great, but it would cost an awful lot.  This kind of thing is one of the reasons that the cost of medical care keeps going up, and obviously, we have to draw the line somewhere.   And why is it that these things always seem to drive costs up?  In the computer industry, innovations always seem to drive costs down.  Why can't that happen in medicine? 

Actually, it does happen.  It happens a lot.  It's just that the moment you free up some money by cutting costs somewhere, a new demand eats up all you just saved, and more.  For example, the cost of treating psychosis has been reduced greatly through the development of atypical antipsychotic medications.  Many persons who used to have no hope of living outside of a hospital now are able to live in the community.  Some of them have fairly normal lives.  In fact, it is possible that you have spoke with some of them today, without realizing it.  These drugs are expensive, typically costing several hundred dollars per month.  Compare that to several hundred dollars per day for a hospital, and it's obvious that the savings are considerable. 

Back to the comparison with the computer industry.  In the 80's, the purchase of a computer was a major capital investment.  Now, a computer is a commodity, not unlike a television set.   That happens with prescription drugs, too.  Today's multibillion dollar blockbuster is tomorrow's aspirin.  In the computer industry, there is a lingering question: will it ever happen, that computing power is so cheap, that there will be no point in developing a new personal computer that is faster than last year's model?  If so, a lot of people at Intel will be looking for work.  And consumers won't be buying computers at CompUSA.  They'll pick them up at K-Mart or Meijer, at least here in Michigan.  Other places have their own megastore chains.

"Honey, could you get a loaf of bread and some tofu on the way home? And while you're there, how 'bout getting one of those 10 gigahertz, quad-CPU computers, with 16 gigabytes of RAM?  They're on sale today, and Betty has to edit some digital video for school tomorrow."

This raises an interesting, unanswerable question: will there ever come a time when the vast majority of medical practice can be conducted without resorting to a branded medication? 

In psychiatry, it does seem feasible.  I don't think it is likely that we will get to the point that we will use no branded products, but I easily could imagine a day in which 90% of the prescriptions are for generic products.  Similar situations could arise in other specialties, such as gastroenterology, or even cardiology.  It probably will not happen for infectious disease or oncology, at least in this millennium.  Still, it is entirely possible that we may see a day when the annual cost of prescription medication starts going down, at least on a per capita basis. 

Will that lower the cost of medical care overall.  No.  We'll just spend the money on other things, like heart-lung transplants.