I was involved in a study that tried to simulate what impact fruit and vegetable price subsidies would have in the United States. We found that you could save 10,000 lives, due to prevention of strokes and heart attacks alone, with a relatively small 1% subsidy across the board on all fruits and vegetables. The idea was that you would only be inducing small changes in behaviour, but those changes over time would reduce risk.
It is difficult to look at what other jurisdictions have done and the impact on fruit and vegetable consumption, because we actually know very little about what people eat. When we try to find out what people eat, we end up doing things like dietary recall studies. They let us get at individuals or households, but are subject to what people think the interviewer wants to hear, or how well they can remember what they ate in the last 24 hours.
When we start looking at broader scales, we rely on things that agricultural economists call disappearance data, where we impute what people must have eaten based on what was produced, exported, or known to have been destroyed.
It's kind of hard to get at these population-level effects of what would happen, because we don't always know very well what does happen. Certainly anything the members of this body could do to encourage our data-collecting agencies to keep an eye out for that in their existing things would be very welcome.