According to this comparison of Canadian data for the withdrawal of dietary needs, underestimations can help to explain the supposed fall in sugar consumption

This recent report by Health Canada on sugar consumption was an interesting reading.

Not so much about sugar, and yes, the report says we eat too much of it (I will come back to it), but rather from the point of view of nutritional recall. It seems we are getting worse.

According to their determination, when comparing the calorie totals of the dietary recall data with those predicted by the respondent's age, the BMI (measured, not automatically reported), sex and activity levels, if the self-reported intake was less than 70% of what you predicted, ranked as a sub-reporter, and if it was over 142% of those predicted, you ranked as over-reporter.

Overall, compared to 2004, people were much more likely to underestimate and are less likely to report excessively. In adults, inadequate journalists increased by 22% (from 28.2% to 34.5%), in children aged 9-18 by 59% (from 16.5% to 26.3%) and in younger children more than doubled (from 6.7% 14.1%). At the same time, excessive reports decreased by 45% in adults (from 13.6% to 7.4%), in children aged 9-18 years by 41% (from 22% to 12.8%) and in young children by 35 % (from 27.6% 18%).

So how does this affect sugar consumption data?

Well you may have read that consumption of food sugar is decreasing. And maybe it is (the nutritional recall data is full of error). But these figures suggest that while consumption is shifting, with more added sugars than food and fewer drinks, the total amounts, if only reasonable journalists are concerned, have remained virtually the same and have risen slightly in children.

As to what is happening, could it be that with the constant debate about the dangers of unhealthy diets people are less likely to want to reveal what they eat? Either way, it is further evidence that we need a better way to monitor dietary intake and a possible confusion for those who say sugar consumption is decreasing.