Why didn't Willett's study show a bigger impact on cancer from these healthy foods? There could be a number of reasons.
One possibility raised in an editorial accompanying the study is that the food questionnaires used to gauge people's diets may not be entirely accurate. In this study, as in many others, participants were asked how often they had eaten particular foods over the past year; if their recollections were flawed, the study results may be, too.
If that's the case, then it's possible that the protective effect on cardiovascular disease is even greater than the study showed, and that there actually is an effect on cancer that the study couldn't find.
The time frame of the study may also have disguised an effect of fruits and veggies on cancer risk. Because cancer can take decades to develop, it may simply take longer follow-up to find a benefit.
Or, Calle said, it may be that what people ate more recently has more of an impact on heart disease, while diet at a younger age has more of an impact on cancer. The Harvard researchers only tracked what participants ate during the course of the study, not during earlier periods of life.