Sunshine Payments Wabi-Sabi

Unlike the Western World’s cultural devotion to “perfect beauty”, in Japanese art and culture the aesthetic of “Wabi-Sabi” – beauty that is “imperfect, impermanent, and incomplete” is much admired. As beauty is in the eye of the beholder, perhaps Open Payments is thus a delight for Japanese data analysts!

Open Payments had a difficult birth – the initial release in September 2014 contained payments for just the last five months of 2013 and was maimed by problems correctly identifying recipient healthcare providers, leaving 2/3rds of the over $3 Billion in payments as “de-identified”, making the dataset pretty useless for solid analysis.

The second release of Open Payments, on June 30th 2015, corrected that very ugly flaw, for both the revised 2013 and new 2014 data, boasting that all the payments were now “matched with total confidence to a particular covered recipient”. But there are plenty of lessor flaws that can trip up the unwary analyst.

The second release of Open Payments, on June 30th 2015, corrected that very ugly flaw, for both the revised 2013 and new 2014 data, boasting that all the payments were now “matched with total confidence to a particular covered recipient”. But there are plenty of lessor flaws that can trip up the unwary analyst.

Uncommon Teaching Hospital Names, Chopped or Not

Payments to around 1,200 teaching hospitals are in the dataset. CMS chose to use the “PECOS” name of the hospital that had been originally registered with Medicare. Unfortunately these names can be quite different from the well-known hospital names we know today. For instance, you may be looking for payments to the famous Memorial Sloan Kettering Cancer Center. But its PECOS name is Memorial Hospital for Cancer & Allied Diseases. And that’s the name you will have to search for in the 2013 data. But to make it harder still, for the 2014 data CMS decided to truncate the names down to 36 characters (and upper case them), so you need to change your search to “MEMORIAL HOSPITAL FOR CANCER AND ALL” to find Sloan Kettering in the 2014 data.

Unique IDs, With a Catch

But, you may reasonably argue, CMS keeps a unique ID for each teaching hospital, right? Argh, yes and no! In their infinite wisdom they decided to change all these “unique” IDs for the 2014 data, so Sloan Kettering has ID 102 in the 2013 data, but ID 1265 in the 2014 data!

Non-covered Entities Munch on the Research Dollars

For research, as well as payments to physicians and teaching hospitals, payments to other “Non-covered Entities” are collected. These are typically hospitals, including, strangely, teaching hospitals. The thing is these mysterious Non-covered entities, in pac-man-esque fashion, are munching the vast bulk of research dollars:

But Where’s the Beauty?

Despite its pretty darn ugly flaws, Open Payments can produce very interesting results, if analyzed carefully to avoid the booby traps. Here are some of the types of valuable analysis that can be performed:

• By drug: Identify the most highly paid physicians in certain activities for a target list of drugs, categorized into speaking, consulting, and research engagements.
• By competitors: Analyze your competition – see how your competitors compare in their marketing budgets, and where and on whom they spend their money.
• By KOLs: Analyze the work of your own KOLs, find who else is funding them, and how you stack up against them.
• By Institutions: Discover which institutions are receiving major research funding in your areas of interest, and from which of your competitors.
• By Partners: Find new partners for your business plan, and analyze current partners, by identifying companies making complementary products who are spending in your target areas.
• Trend Analysis: Find trends in your area of interest – who are the rising stars, and which types of work is becoming more valuable, and for which drugs.

How Do I Do These Tricky Analyses?

Snowfish offers a white paper, Healthcare Big Data: CMS Open Payments, covering the Open Payments database and its possibilities in more detail, and also have a real example of an analysis in Excel. The example shows the 30 most highly paid psychiatrists and neurologists, where we calculate the total payments made to each physician in each of a number of categories, and show not just the companies making the payments but also the associated drugs for the payments.

If you are interested in learning more about Snowfish’s industry-leading approach to healthcare data discovery and mining, and how we can help with your custom Open Payments analyses, please feel free to reach out to us.  This is the time to capitalize on this fascinating new opportunity.

Martin Snowden is Director or Technology at Snowfish. He is an expert at database integration and analysis. Snowfish integrates, clinical, analytic, and business insights for life sciences companies. We have worked with nearly three dozen companies for over a decade and leveraging big data to help increase a company’s competitive advantage. Please reach out to us by going to snowfish.net/contact-us/.