Wolfson Institute of Preventive Medicine - Barts and The London

Blood pressure trial meta-analysis audit trail:

User tips

How to use the site

  • The site consists of five top-level figures (as in the BMJ paper) linked to 55 figures at the level below.
  • Click any linked summary square on a top-level figure to see the subsidiary figure showing the contributing studies.
  • Not every summary square links to a subsidiary figure: in general, those that do not are composites from a number of subcategories which do lead to subsidiary figures.
  • In the subsidiary figures, click a linked summary square or some linked text to see a reference, or
  • Click a summary diamond or back to get back to the top-level figure you just left.
  • In the reference, click the publication name - if linked - to see the full text or in some cases the abstract. Please note that the availability of this content may depend on what subscriptions you have or your institution has.
  • Some totals are less than the sum of the individual categories, because some trials include more than one category.
  • Click home at any time to return to the mini-site's home page,
    or back to return to the previous page, or use your browser's back function.

Usability and PDFs

  • The site relies largely on PDFs to provide easily scaleable graphics.
  • It has been tested on a variety of browsers. It works well on standard browsers with a recent Adobe Reader plugin configured to open PDFs in the current browser window. In this case, it should be possible to navigate easily in and out of the PDFs as if they were standard web pages.
  • If you have problems seeing the pages as intended, you may be able to alter your browser/plugin configuration to help. If this proves impossible please contact us (the website manager's email address is linked at the bottom of every standard web page) and we will endeavour to help or to provide a workaround.
  • You can use the Adobe plugin's view controls to zoom in or out or fit the page to the screen.
  • Most objects on most pages are links; a few (some totals, lines with no studies, etc) are not. The cursor will change to a pointing finger when it is over a link. If it remains unchanged then the item is not linked.
  • Some browsers are sometimes quite slow in the transition between PDF and HTML pages. For example, Chrome and Safari seem somewhat worse in this respect, and Firefox and IE a little better, but we do not know what influences this.

Links to references

  • Most rows in the sub-figures just go to a single destination reference. You can click the square or the reference name or number and any of them will take you there.
  • Some rows have links to two or more destinations. This is difficult to represent on these pages. What we have done is this:
    • If the row specifies a pair of references next to each other or a larger contiguous block., the link goes to the first one. The reader needs simply to look down to see the next reference, or the next two, etc.
    • If the references are not contiguous, then most of the line still links to the first reference, but the second one is underlined in green and clicking it will take you straight there.