Weight / Fat loss

Some highlights from the paper for you. With a bit of translation

There may be social desirability bias and random error (forgetting) in the recall of fruit and vegetable consumption.

This is a sign that the authors understand their limits, and is nice to read. Social desirability bias is where when given the chart to fill in the participants want to look better than they are, so slightly lie. Recall (or the forgetting part) coupled with this could have significant effects. It is important to note that it does seem that they have made every attempt to reduce that bias as is practical.

This study has found a strong association, but not necessarily a causal relationship.

In other words: Yes there is a link. No we don't know if eating fruit and veg actually reduce your risk of cancer directly, or through associated factors (continued below).

There are additional unmeasured confoun- ders not included in the analyses, including other aspects of diet*.

Plus social factors, location of living, etc. etc. etc. There are tonnes of confounders (translation: things that could give a false impression of a relationship) here.

It is important to note that by not adjusting for total energy intake, our results support the hypothesis that eating fruit and vegetables is associated with decreased mortality not simply through displacing other foods from the diet.

This terrible sentence suggests to me that 'you need a balanced diet', which has been suggested for just about forever. But like so much research in this field, it basically allows itself open for anyone to justify anything they want from it.