In The Media – Calories on Restaurant Menus
An article in The Sunday Times (16/7/23) gives results of a small survey that compares the calories stated for an item on the restaurant menu with values obtained in a laboratory.
20 dishes were assessed and 10 were found to have a significant difference (greater than 20%) between menu and lab calories. Of these, five were high and five were low.
Trading Standards Officers allow a value to be 20% out from the stated value, so if the calories stated on the menu were 500, the lab result ought to be between 400 and 600.
It should be stated that the lab in this case used bomb calorimetry for its testing. In this method, the food is incinerated and the heat produced is measured. This method is not actually legal in the UK for determining calories in foods. The restaurants will use the formula:
9 x fat + 4 x protein + 4 x carbohydrate + 2 x fibre + 7 x alcohol
So discrepancies would be expected.
At first sight, allowing a 20% difference may seem quite generous. Usually in the food factory, where they have a lot of control, this accuracy can be achieved. But it’s much more difficult in a restaurant. Most manufactured products have stated weights which need to be correct. The chef in a busy kitchen just doesn’t work like that.
However, there are many factors involved. First the lab analysis is not as precise as you might think. Particularly, it can be hard to take a representative sample, even after maceration.
There also needs to be some allowance for variation from one item to the next. For example, the thickness of the pizza dough or the size and fat content of a steak can make a big difference.
In the restaurant, the serving size can vary a lot for various reasons. And cooking methods, temperature and length of cook can vary.
It’s our opinion that in a real-life situation, it is being unrealistic to have all calorie values on a menu to be within 20% of the analysed value for a specific dish on a specific day.
Restaurants normally use a nutrition calculator like NutriCalc to obtain their calorie values. It might seem that sending a dish off to a laboratory could provide better accuracy, but that misses the point that that dish is just a snapshot in time which could be untypical.
Even after doing a reality check on what you can expect for the accuracy of calorie values in restaurants, some of the discrepancies do look alarming. One dish was found to have three time as many calories (727) as stated on the menu (236). You would have to look at operator error in this case. Perhaps a ‘per 100g’ value was used instead of a ‘per serving’ value.
In order to get accurate results from nutrition software, it’s essential to use good data, enter the recipe in full together with all quantities used and check the work (important).
The Sunday Times article implies that putting calories on menus is not worth doing and they may be right. We certainly feel that in order for diners to get full information, they should also get to see the calories in the alcoholic drinks, which are currently exempt. That large glass of Pinot Grigio provides 225 calories. A Mars bar gives you 228!