Where do breast cancer research dollars go? Three data visualizations of 9k grants and $4.8bn

The US, thanks to a (still relatively) healthy economy, high level of education, entrepreneurial spirit, willingness to invest public dollars (though this is decreasing alarmingly) and open attitude to bringing the smartest people in the world here, has long been a leader in R&D. This includes the field of biomedical research, and specifically the war against cancer. It can still be argued that given the “casualties” in this war, we don’t invest enough. I thought it might be interesting to take a snapshot of the state of funding for breast cancer research.

I collected a database containing 9,491 grants totaling $4.8bn with funding start dates from 2009 to early 2013, by searching the SciVal Funding data base with the term “breast cancer”. That represents over a billion dollar investment per year! By way of context, the annual budget of the National Cancer Institute is around $5bn.

The grants go mostly to US institutions, but also some in other countries. Funding is from public and private sources, mostly US-based. It does not include clinical trials, funding for which is not captured in such public databases.

Because of the lag between grants being awarded and being represented in such databases, the number of grants and dollar amounts in 2012 (and 2013 which is only partially represented anyway) appear lower than previous years, but this is not evidence of a decline in funding. I have therefore not presented the data by year in these visualizations. Instead, consider it a snapshot of recent funding on this disease.

Funding by State

In the first visualization you can see funding by state (click here for an interactive version of the visualization). Overall funding follows state population, although that isn’t necessarily expected.

Research funding is allocated mostly in a competitive process, adjudicated by peer review. Therefore, it follows population in the same way that centers of academic excellence and research follow population: big universities and research institutes are often found in major cities. Looking through the lens of breast cancer research, we can see that in general funding follows population. Some states are above the line, and these are known for their large research centers in Boston, New York and cities in California and Pennsylvania.

You can do sate-to-state comparisons by selecting them from the drop down menu. Mouse over things to get more info in the tool tips. You can zoom the maps and restore them by clicking on the pin icon.

Breast Cancer Funding by State

Funding by Institution

In the second visualization I have used institution as the focus (click here for an interactive version of the visualization). Note, countries outside the US are also included and for these the map will not work.

A major caveat here is that institutions are listed in the grant database in many different ways, which is why the map is useful. I geo-encoded the institutions using Mechanical Turk for all the different ways they were listed. The difference in listing comes from the way that the institution name is entered into the grant forms. The best way to look at these data is therefore to narrow in one a state or two, and see the distribution of funds in the tree diagram below (the thing that looks like a mosaic).

This visualization gives you a pretty quick view of where the majority of the breast cancer research in your area of the country is currently going on.

You can again do state-to-state comparisons by selecting them from the pull down menu. For example, comparing California and Texas shows that the top four institutions are, in order, UCSF, MD Anderson, Baylor College of Medicine and Stanford (again, remember these are breast cancer numbers only). Houston has about $240M and San Francisco $162M, but if you add in Oakland and Berkley you get to the same approximate number.

Breast Cancer Funding by Institution

Funding by Source with grant details

Lastly, I have shown the funding by source (click here for an interactive version of the visualization). In the database I collected the vast majority of the funds come from the National Institutes of Health, which will come as no surprise to most. The bubbles represent funding allocated to specific states, so that by selecting a given sponsor you can see where they direct their research. Of course some sponsors are state specific, as their names usually reveal.

Below, in the table, you can then see some details of the grants, including the sponsor, institution, PI, title, year it started and dollar amount for that year. If you look at the National Institutes of Health (NIH) you will see the multi-year grants represented as separate entries for each year that funds were given.

This visualization allows you to compare the impact of different agencies in terms of research dollars. For example, want to know how the Cancer Prevention and Research Institute compares to the Susan G. Komen for the Cure? Select both from the drop down menu, and visually compare the size of the bubbles. (Note, in this case the colors are the same – just click on one in the color legend to highlight.) It looks like CPRIT funds over 3x as much as Komen in Texas, but as Komen is active nationwide, it dwarfs CPRIT by 3x or so at that level. Of course Komen is focused on breast cancer, and CPRIT funds research on all cancers. Remember to “mouse over” the bubbles to see more details.

Breast Cancer Funding by Source

I offer these views to help inform discussions of where breast cancer research should be heading.

As an aside, there were only 5 hits for the search for “male breast cancer” in the grants database, and none of these grants were focused exclusively on this form of the disease.

6 thoughts on “Where do breast cancer research dollars go? Three data visualizations of 9k grants and $4.8bn

  1. Oliver, this is very interesting. What got you thinking about this question? Also, Tableau looks kind of nifty. Did you learn how to use it in your current administrative position? I had you in my mind as more of a microscope, petri dish, or rodent kind of guy and this data presentation format looks more suited to displaying demographic or epidemiological data.

  2. Elizabeth – thanks. I used to be more of a lab guy, but in 2010 started doing some administrative work, and now am actually giving up the lab – my last grant runs out in a few months, and the chance to get it refunded at present is small. In the meantime I have been learning about data visualization, and we have started using Tableau and other tools to look at academic activity to guide decision making, so you are spot on. There is a free version of Tableau that can be used to make public visualizations based on simple spread sheet data – Tableau adds maps and all sorts of other goodies pretty readily.

    Anyway – I am always wondering where the money is going, and how effective it is. This latter part is very hard. But as I connect more with the advocacy community I understand the desire of survivors to be involve in funding decisions. I think an important first step is to get a feel for what is being done. And so I thought I could make a contribution to this discussion. So I wonder how useful these kinds of things are.

    Helen- thanks for the feedback. Most of the work was done via Mechanical Turk – an Amazon site where you can pay people a few cents to do a simple thing – in this case look up the city, state and ZIP of an institution.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s