Welcome to the Vanco Numbers Plot-page. Please feel free to send me an email (by referring to this page in the subject field in order to not be filtered out by the mailbox spam filter) or to make an entry in my guestbook. Comments are well appreciated.
The two figures above should visualize the Vanco numbers. The data is based on a private communication. All unprocessed numbers (i.e. wholesales in USD and monthly and weekly prescription numbers) come from IMS Health which services are hereby greatly acknowledged.
The plot in Fig. 1 shows Vanco prescription numbers on a weekly (red, right scale, zero suppression) as well as on a monthly (blue, left scale, no zero suppression) basis (for a monthly relative growth plot click here).
Monthly data is connected with a b-spline fit line, while weekly data error bars are calculated as 3 * SQRT(value), using principles of poisson statistics in order to yield a three-sigma significance.
Please note, that adding appropriate weekly numbers does not necessarily have to yield the corresponding monthly number since the two data sets are based on different statistical samples. Nonetheless, the two sets are strongly correlated to each other because they are drawn from the same entity. The quality of the correlation is directly related to the significance of the underlying statistical sample which I personally would estimate to have improved after the change in the methodology of accounting for weekly numbers by IMS Health as of January 2007. See for comparison an old plot of the comparison between weekly and monthly prescription numbers that shows weekly numbers before the methodology change, i.e. up to and including December 2006.
Fig. 2 shows the sales of wholesalers to pharmacies in million US-dollars. Solid circles mean the total amount while open circles belong to sales on a per week basis. The first, fourth, seventh and so on (every third) point include five weeks of selling, the remaining points only four weeks. Note that the y-axis is broken in order to obtain a better resolution.
This link should open a plot showing prescription numbers and sales per week numbers (both monthly) in one image. Also, since zero-suppression is applied to both y-scales of that plot, the resolution (especially of the per-week monthly sales graph) is enhanced significantly with respect to Fig. 2.
Have a nice day, NochEinTyp
Last updated: 2007/08/07/01/45 UTC+2 |
