Prior artistry of patents and patients

I have written about patents over the years and their often dubious role in healthcare. A recent panel discussion at the Lown Conference introduced me to Dean Baker a Senior Economist at Center for Economic and Policy Research.

Dean Baker, Senior Economist at Center for Economic and Policy Research, pointed out that we need significant changes in our insurance providers before handing them the reins to a “Medicare Advantage For All” system – at the least changing them to non-profit insurers like in other countries. He also proposed more radical changes to bring down prices, like abolishing patent monopolies for new drugs.
“The way we pay for drugs now, it’s like paying firefighters to come to your house when your house is on fire,” said Baker.--Lown Institute A think tank advocating a radically better and uniquely American health system that overturns high-cost, low-value care

Full disclosure I have worked with patents numerous times as a scientific/medical writer and most recently as a healthcare analyst. The topic is certainly not unknowable but the granularity may not be for everyone. I am going to highlight a few high level insights in case you are data curious and would like to be able to access a little bit more information about how the sausage is made.

First, there is quite a useful data trail when you are curious about patents. The US Patent and Trademark Office (USPTO) is freely searchable but you will need to understand a bit of the underlying reference material.

In my work I am interested in the citation indexes. Think of prior art, "any evidence that your invention is already known" as bread crumbs on the trail. Relying on the published work of Torrance and West, All Patents Great and Small: A Big Data Network Approach to Valuation I characterize the value of a patent by using "relative positions of issued United States patents embedded within a comprehensive patent citation network". 

​A company, Patentfield, allows limited searches of patents/utility models for free (up to 100 records) or you can expand access for a monthly fee. This is only one approach and as your mileage may vary, you are free to develop a unique algorithm better suited to your data question.

The first graphic demonstrates a portion of available data visualization formats generated to answer industry questions or to better design network queries seeking important nodes to better identify prior or competitive art. For example, how novel is the patent? Which patent applicants referenced prior patents in the development of their current or prior applications?

The second is hypothesis generation by searching clusters of biotech filings based on citations within the patent framework. The size of the node indicates level of importance of the patent within the network.

Why does any of this deserve your attention? All roads to patent exclusivity lead back to drug development costs. But how much of the research was already pre-existing? Are we paying for true innovation or cleverly motivated opportunists? These questions are critical for informed discussions in emerging fields like immuno-oncology where the foundational research pre-dates the media headlines by decades.

We need to be mindful of the "artistry" of wide brush strokes in discussions of value, innovation, and who is truly profiting when the patient is left holding the bill.

Want to stay in touch? Connect over on twitter or if you are so inclined, buy me a coffee to show your appreciation!