There aren’t many vaccines which work after the infection has been acquired, and this Costa Rican trial of a human papillomavirus 16/18 vaccine proves no exception.

In this next study, two-thirds of the patients were still alive five years after a diagnosis of stage III colon cancer, i.e. cancer that had spread to local lymph nodes. The main purpose of the trial was to compare chemotherapy regimes, but detailed dietary questionnaires were used to distinguish between those who followed a “prudent diet”, defined as one rich in fruit, vegetables, poultry and fish, as opposed to a “Western diet” rich in other meat, fat, refined grains and dessert. There was a three-fold difference in favour of the “prudent” group, after adjustment for a wide variety of confounders. Of course these diets represented personal choices rather than randomised allocation, but it does seem reasonable to recommend these nice foods to people who people who have come through bowel cancer.

We’ve all been taught that diabetes not only increases the risk of coronary disease but also worsens outcomes after acute coronary events; but is this still true in the era of rapid intervention? The eleven trials of the TIMI series from 1997-2006 confirm that in the first year following ACS, diabetics still fare worse, but thereafter the added risk fades away.

Every now and again, somebody pops up to suggest we are measuring the wrong lipids for cardiac risk, or that we should do one measurement for men and another for women. A paper from Framingham comes to silence our groans: carry on measuring total cholesterol and HDL cholesterol and you can derive a score that is just as good as one based on fancy apolipoproteins (apo B:apo A-1) in both men and women.

High density lipoprotein cholesterol is often called “good” cholesterol because of its association with reduced cardiovascular morbidity, but attempts to raise people’s good cholesterol always seem to go wrong. One small trial, you may remember, even tried infusing the stuff into people’s coronary arteries. In the latest issue of Evidence Based Medicine I describe HDL-raising as a therapeutic cul-de-sac which ought to carry a traffic warning: this systematic review bears me out, though it is less damning in its conclusions.