The price of knowledge is worth knowing

We all pay a lot of money for the product of our own collective academic enterprise. With 2.5 million downloads of pdf's by Harvard University patrons from our top three publisher packages, I wondered what the costs might be. Well, thanks to Betsy Eggleston, we now have a better idea:

Elsevier package 2008 article downloads: ($.76/download)
Wiley package 2008 article downloads: ($1.52/download)
Springer package 2008 article downloads: ($2.98/download)

That is a lot of money per click but several questions pose themselves:

a) Do all libraries have a similar cost per download?

b) Is the relative cost per download similarly ordered for each of these three publishers in other libraries?

c) What is the equivalent cost for a circulated book/monograph per patron-use? Is that a fair comparator?

d) What is the equivalent cost per download for open access publications (including the author cost)?

I suspect that knowing the answers to these questions is a source of leverage and power. How can we make decisions with and on behalf of our researchers, faculty and public without knowing these answers? Should we not insist on greater transparency of the relationship of academic value and cost. If you have any additional data, feel free to enter a comment regarding this post or send me an email and I will add it to this post.

Electronic Medical Nags are Largely Ignored

First there was the machine that went ping. Then there were a lot of pings because of all the machines around patients and it became clear that most of them were not relevant (94% of alarms in the pediatric intensive care unit). Although there was some work done to attempt to reduce the noise, a lot of the monitors now have their alarms silenced. So, perhaps it is not too surprising that a recent study has shown that many of the electronic health record-driven advice, reminders, and warnings regarding medication are now being ignored. This begs the question whether the solution is primarily addressable by technology or really a matter of re-engineering the process of patient care.

(hat tip Ted Shortliffe)

Influenza in Massachusetts

aegis.4.27.2009 Originally uploaded by kohane

Time will tell whether this picture is due to a lot of worried individuals with influenza-like symptoms showing up in emergency rooms or an actual uptick in cases. Nonetheless, there is something compelling about observing, live, the distribution of cases of influenza across this state.

An equally compelling global perspective is provided by my colleague John Brownstein.

Doctors do not bill to make insurance companies smarter

Those of us who have worked with electronic healthcare data have been long aware of the limitations of billing data (aka claims data, aka administrative data) for research. They are often too coarse grained for clinical research and are inherently biased to maximize income. It is motivated by these limitations that Natural Language Processing (NLP) has become increasingly important in mining clinical records for research. What a doctor writes in her notes is much more revealing of her patient's state than what she bills for. Notwithstanding there are some significant challenges in the de-identification of textual records and in transforming these records into standardized clinical categories (e.g. SNOMED). Yet the appeal of using the clinical narrative text rather than claims data is compelling. In our work in i2b2, we have seen significant overrepresentation of diagnostic codes where a diagnostic encounter to "rule out" a disease was codified as that disease in the claims data. For example, a radiologist asked to rule out rheumatoid arthritis based on an X-ray will often classify the X-ray with a billing code corresponding to rheumatoid arthritis when perusal of the full narrative text of the radiologist's notes that there were NO findings consistent with rheumatoid arthritis.

A recent article in the Boston Globe points out additional challenges in using claims data for personal medical records. The same limitations of claims data for research appear to impinge on their utility for clinical care. My colleague John Halamka makes several useful suggestions on how to improve the use of such data, including recruiting patients themselves as collaborators in refining the categorization of their clinical records or even removing gross errors. Notwithstanding, a small number of codes are likely to be quite limiting and it may be that codifying the patient's record by using the entirety of their clinical documentation (i.e. what their care providers wrote about them) will ensure the most nuanced and most faithful representation available of what the clinician was thinking about in each clinical encounter with that patient.

Medical Justice

One of the many reasons healthcare records do not have broad adoption, is that they are burdened with a multitude of functions beyond "merely" serving as a communications vehicle among the members of the healthcare team. All these functions are then bundled into monolithic systems by single vendors. It is then typically arduous and expensive to substitute new functionality for existing functions. Perhaps one of the most onerous uses of these systems, one that is not tightly linked to quality of care, is clinical documentation for preemptive legal defense rather than for effective clinical communication and decision making. This brief article in the NY Times offers a path towards safety and rationality so that patients who are injured are justly compensated and those who are not, are not compensated. It also offers as an implicit side-effect, changes in physician behavior and use of electronic medical record systems that would be focused on improving quality of care and communications (to and from patient as well as physician).