Sunday, July 26, 2015

The EHR and Washington

There is nothing better than a pediatrician who is an administrator, who wants to run for Governor and who in the process goes to Washington to specify how all physicians must in the future prepare their medical records. In fact in any real world one would find such a tale absurd, yet it is the truth. We now have the "meaningful use" standard, where did that come from, for the management of patient records. I thought this was stupid seven years ago, and yes even in 1992 when I first wrote of this, and still do.

Patient records are important, and part of the reason we never really got the problem solved is because it is really hard. Sometimes hard problems are hard for a reason.

At a recent AMA meeting as reported by Medpage the writer notes:

Almost immediately, physicians gave voice to the barriers to care they say are caused by electronic health records systems. Over the course of the 90-minute meeting they raised concerns over reduced productivity, the security of private patient medical records, interoperability, and government regulation. "We're removing the science from medicine," said one physician who described having to check "yes" and "no" boxes rather than being able to note subtle nuances his patients reported. "Thank God I learned to type in high school -- I never thought I'd use it," said another, explaining that she now has to make sure every employee she hires can type, regardless of the job for which they are hired.

It frankly is the arrogance of many who go to Washington who think that they know all there is to know...then again perhaps not.

Friday, July 24, 2015

Cancer Classification

There is an interesting paper from CSHL on progress on cancer classification. Linnaeus some 300 years ago came up with a classification system for various species. Aristotle was driven by his desire to classify, and ever since we have people trying their best to do that task. Patients always want to know what they have, and that is a form of classification.

We classify cancers based upon organs. We may modify it based on cell types or based on cell markers such as immunological markers. I remember back in the 60s that Leukemias were simple; acute or chronic, you died now or later. Now we have a plethora of substypes and a multiplicity of therapeutics.

But we also now genomic data. Perhaps then we should classify cancers based upon genes, not upon organs, binding proteins, or the like,

As the authors state:

Classification is an everyday instinct as well as a full-fledged scientific discipline. Throughout the history of medicine, disease classification is central to how we organize knowledge, obtain diagnosis, and assign treatment. Here we discuss the classification of cancer, the process of categorizing cancers based on their observed clinical and biological features. Traditionally, cancer nomenclature is primarily based on organ location, e.g., "lung cancer" designates a tumor originating in lung structures. Within each organ-specific major type, further subgroups can be defined based on patient age, cell type, histological grades, and sometimes molecular markers, e.g., hormonal receptor status in breast cancer, or microsatellite instability in colorectal cancer. In the past 15+ years, high-throughput technologies have generated rich new data for somatic variations in DNA, RNA, protein, or epigenomic features for many cancers. These data, representing increasingly large tumor collections, have provided not only new insights into the biological diversity of human cancers, but also exciting opportunities for discovery of new cancer subtypes.

They continue:

An ever finer classification system has many potential benefits. It is needed to capture the full spectrum of biological diversity—the "endless forms" that Darwin spoke of. It could lead to a better recognition of patient-specific disease mechanisms, and importantly, could suggest treatment options that are more accurately matched to the patient's tumor. Precision medicine, at its very foundation, relies on valid and continuously optimized disease classification that reflect the underlying mechanisms. However, a fine-grained classification system also has many potential drawbacks. The newly proposed splits may not be technically robust. Even when the finer categories are robustly supported by statistical significance and by replication, they may still lack a clear biological meaning, or have little impact on treatment options (#3 below) if it turns out that some subtypes share the same clinical endpoint, or if treatment options are limited.

Indeed, we may find it much more powerful to have a new Linnaeus type look at classification. Classifying genomically, via genes, RNA, and epigenetic factors, may help stratify and focus on therapeutics. This article raises an interesting dialog.

Monday, July 20, 2015

Windows 10?

Are you out of your mind! Next Monday Microsoft releases Windows 10. Now I have Windows 7 and it works just fine. Even have a few XP systems for lab work so who cares since they are not even networked. But, updating to Windows 10 would be akin to inserting hot peppers into my eyes. Total agony.

Would anyone allow Microsoft to change their computer? The chance that all your data, all your software, all your drivers, all your security, all your connectivity, would just vanish! A Microsoft hot line? The IRS is easier to deal with.

So for those folks happily upgrading next Monday, I remind you of Dante and the Inferno and the sign above the door to Hell, "Abandon all hope you who enter!"

That's right, when you all find out that, oops, Microsoft had a few glitches, that Democrat speak for massive screw up, and you are now back to pencil and legal pads, you did it to yourself! Keep the XP machines, Microsoft no longer updates them, that means they are stable.

Friday, July 17, 2015


From the NASA web site, Pluton and its "moon". Interesting set of photos. But a note: In the old days we had to wear our own clothes. Today NASA seems to get everyone dressed up in matching polo shirts.

The Neo Marxists

In the Guardian is an interesting piece if you can wade through it. The author wanders endlessly and argue for the destruction of capitalism, another revolution of the proletariat.

He says:

During and right after the second world war, economists viewed information simply as a “public good”. The US government even decreed that no profit should be made out of patents, only from the production process itself. Then we began to understand intellectual property. In 1962, Kenneth Arrow, the guru of mainstream economics, said that in a free market economy the purpose of inventing things is to create intellectual property rights. He noted: “precisely to the extent that it is successful there is an underutilisation of information.” You can observe the truth of this in every e-business model ever constructed: monopolise and protect data, capture the free social data generated by user interaction, push commercial forces into areas of data production that were non-commercial before, mine the existing data for predictive value – always and everywhere ensuring nobody but the corporation can utilise the results. 

If we restate Arrow’s principle in reverse, its revolutionary implications are obvious: if a free market economy plus intellectual property leads to the “underutilisation of information”, then an economy based on the full utilisation of information cannot tolerate the free market or absolute intellectual property rights. The business models of all our modern digital giants are designed to prevent the abundance of information.

 Intellectual property rights or property rights of any time will in his mind fall away as we see the proliferation of more and free information.

He continues:

There is, alongside the world of monopolised information and surveillance created by corporations and governments, a different dynamic growing up around information: information as a social good, free at the point of use, incapable of being owned or exploited or priced. I’ve surveyed the attempts by economists and business gurus to build a framework to understand the dynamics of an economy based on abundant, socially-held information. But it was actually imagined by one 19th-century economist in the era of the telegraph and the steam engine. His name? Karl Marx.

Yes, Marx comes back again. Just like Marx, this piece is an endless rant against capitalism and its pending demise. All because of Wikipedia. After all Wikipedia did away with encyclopedias! What did away with encyclopedias was the cost and size, and frankly the useless nature of an encyclopedia.

It is always worth reading those Marxist writers and impending doom. Well back out to my garden, it is a nice summer day! 

What is Real Progress?

I read the article about Reddit in the NY Times. The article opens with:

The return of a founder to a company is a well-worn story line in Silicon Valley. In Apple’s darkest hour, Steve Jobs came back, eventually turning the computer maker into the world’s most valuable corporation. In April, Mark Pincus stepped back into the chief executive role at Zynga. And Jack Dorsey, a co-founder of Twitter, has twice returned to help right the social media company’s ship. Now Steve Huffman, who co-founded Reddit in 2005, may have one of the toughest returns of all. After being away for six years, Mr. Huffman reappeared last Friday as chief executive to pull off a turnaround of the online message board, which has grappled with a series of missteps and is embroiled in a battle to win back the confidence of its users.

Now I have dropped my Twitter and Facebook accounts. Never figured what to do with Twitter other than get into trouble with some stupid remark and Facebook became watching friends of friends making inane comments about their lives. I have looked at Facebook for "advertising" one's company but it is such a linear time sequenced medium that finding something is a problem. One can find Facebook adds but content is another thing.

Along comes Reddit. Never saw it until yesterday. It appears to be a never ending set of nasty comments. At least from a first view. This is what has become of Silicon Valley. Formerly a center for high tech, now the center for high touch. 

Then I walk around MIT and Kendall square and look up. Building after building with "smoke stacks" if HEPA filters. Each of these massive "factories" are "making" change, real change. It is what Pittsburgh was to steel a century or more ago. One measure is the parking fees. Five years ago it was $11.00. Now it is $42.00. No poor folks here! Senator Warren should try to get a parking spot...

But the point is simple. Jobs and Apple started by making something, computers. Google had a search engine, it was a breakthrough akin to a word processor. Facebook, Twitter, Reddit, Linked In et al facilitate something, but do they in any way create value? How would one measure value? In Cambridge I look at CRISPR Cas9 and see tremendous value. As in the Graduate the new thing was "plastics". in this generation it is CRISPR. It is not Twitter or even Reddit.