As the health care bill is being debated in the US Senate, it's a good time for us going into/already in the health field to think about the next steps in US health care. I'd like to talk about 2 topics in this post: evidence-based medicine (EBM) and electronic medical records (EMR).
-----
Evidence-based medicine (EBM)
What is EBM? My personal definition of EBM is the application of science to the practice of clinical medicine. Sounds intuitive and obvious enough, and it sounds like all of medicine does/should operate this way, right? Apparently not.
To elaborate, EBM is the utilization of epidemiological studies (the study of disease patterns) and biostatistics to help physicians make more informed decisions when deciding on treatment or between two treatments, whether they be drugs, surgical intervention, or whatnot. Also, EBM allows for the ability to understand and critique the literature of medical studies in order for each physician to come to his/her own conclusions on the studies.
Arguably, a class on EBM is one of the most important courses a med student will take. It is also a class many med students loathe to take for various reasons - from their dislike of statistics, to the dry nature of the material, to the impersonal and "cold" methodology of analyzing data from many "anonymized" patients. This is perhaps best highlighted with the recent brouhaha in the news over changes in screening guidelines for breast cancer and Pap smears:
These revisions were generated in part by looking at and analyzing the current epidemiological data on those cancers. There is indeed a culture clash highlighted with these revisions - between those who dislike EBM and those who champion it. In my opinion, it would be wise to withhold final judgment on the new guidelines until one has read the study itself and decided if it's applicable to his/her practice.
Now, of course with all epidemiological studies the data is on pooling together a population (or populations) rather than focusing on a given individual. So while there is no "average patient," it's probably a good starting point. The guidelines don't apply to women with an above average risk of breast cancer or cervical cancer, i.e. a strong family history of breast cancer or testing positive for a BRCA1/2 allele variant.
Medicine is moving more and more towards EBM, as it should be. The more data we have, the more scientific knowledge we possess, the better we can develop guidelines and starting points when treating patients. But it should be noted that individuals are not a population, nor does a population represent an individual. EBM is a tool, and a very valuable one at that, but it shouldn't be the sole tool for a clinician.
-----
Electronic Medical Records (EMR)
There has been much excitement and distress over implementing EMR systems across hospitals in the US. Certainly EMRs have their advantages, but also less know are their current disadvantages. Suffice to say, EMR systems in the US are still nascent and may not quite ready for prime time.
The advantages of an EMR system are manifold. EMR reduces the amount of paperwork (save some trees!) which could aid in decreasing the possibility of paperwork becoming lost in the process. EMR has the advantage of being more easily organizable, such as creating tabs for "Medications," "Allergies," "Surgical History," "Family History," etc. Ultimately, EMR systems will be searchable (i.e. search for "allergy to penicillin") and more readily transferable between hospitals. So if I moved from Michigan to Wisconsin, a doctor in Wisconsin would be able to pull up my health records via EMR from Michigan once I've authorized it, rather than asking the hospital/clinic in Michigan for my health records to be faxed/sent over.
However, there are currently many obstacles to the final vision of EMR systems. My friend, Alicia, a genetic counseling student, went to their conference recently. And at their conference, there was a representative from a company that creates an EMR system. The representative painted a rather pessimistic outlook on EMRs.
One of the primary obstacles is that every hospital that uses EMRs use a different EMR system. Therefore these EMR systems aren't compatible with each other. So if a doctor in Wisconsin were to request my health records by EMR, his system has to first be compatible with the EMR system used in that hospital/clinic in Michigan. There's currently no communication between EMR systems and there appears to be little (if any) communication between the companies that design these systems - logically so, as they're competitors.
Another obstacle is the transition from paper to EMR. Someone has to either manually enter all the data into the EMR or scan the paperwork into the system. Both require considerable time and money. Paperwork can be easily buried, lost, or forgotten in the transition to EMR, potentially with devastating outcomes.
Lastly, there is some difficulty in making EMRs searchable. The cumulative medical/health data of a single individual from birth to death is enormous. EMRs would require a massive amount of storing capacity, and to design a search function that can find just what you want within that massive pile of data is currently out of reach. And a physician doesn't have the time (nor the patience probably) to look through a person's entire EMR to find that one relevant nugget of info.
So does this all mean that EMRs are doomed to fail? Not necessarily. There are obstacles, yes, but not insurmountable. If there was an incentive (perhaps from the federal government) for companies designing EMRs to communicate and collaborate, it's possible that a basic package that's compatible with all systems could be achieved. More specialized functions and widgets could be added on that need not necessarily be compatible with other EMR systems. There is also potentially a market niche in developing compatibility software that allows data from one EMR system to be accessed in another. The transition from paper to EMR is unavoidable, and it's just something that will have to take time and be toughed through (unless someone comes up with an ingenious method). And I've full confidence in the progress of technology to one day make EMRs searchable. I mean, look how far computers have advanced within the last 20 years!! But for the near future, making tabs for different clusters of data is probably the next best thing.
A nation-wide EMR system is currently a dream, and probably not something achievable within the next 20 years. But if there's a will there's a way, and being an optimist, I could see the potential and possibility of a nation-wide EMR system being in full use within the next 5-10 years. Where there's a will there's a way.
It's interesting you mention this, because I think there are medical practices that are still based on conventions or legislation set by Congress, that don't have any basis in evidence at all. For one, measuring blood pressure at regular checkups was supposedly started when a patient at a clinic suggested it. And now everyone does it, but considering how transient bp is, is there really any meaning to that measurement? Hmm.
Speaking of EMR, I think of my professor last year mentioning the example of Singapore as a success story for EMRs. According to what he said, the citizens are completely comfortable with releasing their medical records to health care. Of course, Singapore is highly educated, rich, geographically small, and their norms support implementation of EMR. Their example doesn't shed too much light on the United States experience, unfortunately, but it's fun to look at anyway ;)
Original Press Release in 2004:
http://www.moh.gov.sg/mohcorp/pressreleases.aspx?id=1366
Flyer:
http://www.moh.gov.sg/mohcorp/uploadedFiles/News/Press_Releases/2004/EMRb_Flyer_Eng03path.pdf
Hmm, interesting stuff. I didn't know that bit about blood pressure at regular checkups, though I think there is some meaning to measuring it.
As for Singapore, even if their experience is completely different than ours in the US, it's still worth a look just to see how well it works for them. So yeah, fun to look at indeed. :-)