Get PACE removed from clinical guidelines – sign the petition!
Sasha invites you to sign a new petition to the HHS to protect patients against the PACE trial, CBT and GET ...
Discuss the article on the Forums.

What is wrong with this graph?

Discussion in 'Action Alerts and Advocacy' started by Snow Leopard, Jul 10, 2012.

  1. Snow Leopard

    Snow Leopard Hibernating

    Messages:
    3,834
    Likes:
    7,934
    South Australia
    cfs spending USA.jpg

    What is wrong with this graph? (Hint, the NIH has a mandate to consider disease burden when allocating funding)

    No prizes for guessing which point corresponds with CFS. Keep in mind this is a logarithmic graph, so dots that are some distance away from the mean are a magnitude of order (over 10x) away from what would be expected.
     
    sue la-la, Simon, Tito and 7 others like this.
  2. biophile

    biophile Places I'd rather be.

    Messages:
    1,520
    Likes:
    16,491
    Outliers should be ignored because they are inconvenient.</sarcasm>
     
  3. SOC

    SOC

    Messages:
    7,803
    Likes:
    16,334
    I wanna know what those outliers with a high spending to burden ratio are, especially the one on the far left that has the lowest burden but more funding than a half-dozen higher burden illnesses. Just what makes those illnesses important enough to the NIH that they spend a disproportionate amount of money on them?
     
  4. user9876

    user9876 Senior Member

    Messages:
    1,740
    Likes:
    10,274
    In the UK cancer gets a disproptionate amount of the treatment budget. Not sure about research,
     
  5. L'engle

    L'engle moogle

    Messages:
    1,513
    Likes:
    2,187
    Canada
    OK, obviously the spending and burden are way out of proportion (especially given the magnitude of the scales), but is there something else I'm missing? I feel kinda stupid right now. :D
     
  6. I can be worse. In the Netherlands, a similar graph would put ME infinitely far below the horizontal axis.
     
    sue la-la and Tito like this.
  7. biophile

    biophile Places I'd rather be.

    Messages:
    1,520
    Likes:
    16,491
    Would be interesting to see how much money is being spent on psychobabble.
     
  8. *GG*

    *GG* Senior Member

    Messages:
    5,547
    Likes:
    3,495
    Concord, NH
    Which one is CFS? Do you have a link to the source, so I can try to understand this graph? What is DALYs?

    GG
     
  9. Snow Leopard

    Snow Leopard Hibernating

    Messages:
    3,834
    Likes:
    7,934
    South Australia
    Hepatitis-B

    The point in the top-centre is HIV-AIDS and the point to the right of that is infectious diseases (which is skewed due to the high HIV-AIDS funding).

    Keep in mind there are a whole bunch of diseases that were not included in the 2004 WHO burden of disease study. CFS was not included either, but I calculated the DALY data from the 2003 Australian study, accounting for the USA prevalence (I used 0.42% from the Chicago study).

    The biggest ratio between US burden and funding is not shown on the graph - Malaria, since the US disease burden was less than 1 DALY per 100,000. But this disease has global impact. Still, some diseases like HIV/AIDS are still overfunded when considering disease burden on a global level. (see: http://www.plosone.org/article/info:doi/10.1371/journal.pone.0016837)

    The other 'low' outlier is Migrane...
     
    sue la-la and SOC like this.
  10. Snow Leopard

    Snow Leopard Hibernating

    Messages:
    3,834
    Likes:
    7,934
    South Australia
    DALYs are a measure of disease burden developed by the WHO that accounts for both years lived with disability and years lost due to premature mortality. http://www.who.int/evidence/bod/

    CFS is the one at the bottom.

    The data for the graph is sourced from here:
    http://www.who.int/healthinfo/global_burden_disease/estimates_country/en/index.html
    The CFS data, adjusted for US prevalence was sourced from the Australian figure and assumes similar epidemiology otherwise from an adjustment for US prevalence based on a US population based study (the Chicago study).
    Australian figures: http://www.aihw.gov.au/publication-detail/?id=6442467990
    Chicago study: http://www.ncbi.nlm.nih.gov/pubmed/10527290


    Also note that when adjusting the prevalence between 0.2-1%, it remains the most distinct outlier.
     
    sue la-la, beaker, mezombie and 3 others like this.
  11. medfeb

    medfeb Senior Member

    Messages:
    407
    Likes:
    1,417
    This is excellent, Snow Leopard. Thank you
    Is Annex table 10 the right table to look at for the DALY for CFS or is it some other table?
     
  12. justinreilly

    justinreilly Senior Member

    Messages:
    2,496
    Likes:
    1,212
    NYC (& RI)
    this is great! SL, is it possible to somehow label each disease on the graph? This is a perfect graphic for us to use! pls pm me if you can. I plan to put together a brief "white paper" for legislators, etc and this again is perfect.
     
    beaker and WillowJ like this.
  13. biophile

    biophile Places I'd rather be.

    Messages:
    1,520
    Likes:
    16,491
    Tito, MeSci, beaker and 3 others like this.
  14. *GG*

    *GG* Senior Member

    Messages:
    5,547
    Likes:
    3,495
    Concord, NH
  15. Graham

    Graham Senior Moment

    Messages:
    1,174
    Likes:
    10,779
    Sussex, UK
    But the problem is that the scales are logarithmic and so the image bears no connection to the reality, unless you are of a very mathematical persuasion. That is why, when I did the video on funding both for the PACE analysis and for Cort's recent blog, I used the area of bubbles to show the differences (we aren't very good at perceiving volumes, especially when rendered in 2D).

    I'd like to see this graph rendered more like those from information is beautiful.
     
    sue la-la likes this.
  16. Graham

    Graham Senior Moment

    Messages:
    1,174
    Likes:
    10,779
    Sussex, UK
    I've been thinking (always dangerous). How about considering the amount spent divided by DALY as a measure? I've done a spreadsheet of the results from that table: top spending is tuberculosis, at $15m per DALY, then AIDS at 4.98, sexually transmitted diseases at 4.06, then a large clump working down from 1.55 to 0.13 (ischemic heart disease), followed by injuries (0.08), depression (0.07) and chronic obstructive pulmonary disorder at 0.04. But I don't have the figures for ME, so can't do that one.

    What puzzles me is that the spending figures don't quite match the table from NIH (http://report.nih.gov/categorical_spending.aspx)
     

See more popular forum discussions.

Share This Page