Paperonium

Public health · Lancet Public Health · 2016

Study: more low-sugar drinks would prevent 1,528 diabetes cases yearly. Its own groups total 12,108.

1,528Stated total, in summary and table10,580cases left out of the stated total12,108Its eight groups, added up1,528Women 65 and over (same as stated total)5,327Men 19 to 643,289Women 19 to 641,598Men 65 and over366Boys and girls 4 to 18 (four groups)
1,528Stated total, in summaryand table10,580cases left out of the stated total12,108Its eight groups,added up1,528Women 65 and over (same asstated total)5,327Men 19 to 643,289Women 19 to 641,598Men 65 and over366Boys and girls 4 to 18 (four groups)

Diabetes cases prevented yearly by more low-sugar drinks: stated total against its eight groups

The paper says

A study modelled how the UK’s sugary-drinks tax could affect health if drink makers cut sugar, raise prices or sell more low-sugar drinks. Its opening summary says that, on hopeful assumptions, selling more low-sugar drinks would prevent 1,528 new diabetes cases a year.

Its own data show

The paper’s diabetes table also gives 1,528 as the total. But it splits the total into eight groups by sex and age, and they add up to 12,108; 1,528 is exactly one group’s figure, women 65 and over. The table’s range of likely values for the total, 4,414 to 21,785, holds 12,108, not 1,528.

Everyone in the model is in exactly one of eight sex-and-age groups, so the groups must add up to the total. For hopeful price rises they do: groups and total are both 10,861. At 12,108, more low-sugar drinks beat that, as the main text says. The summary’s 1,528, about 8 times too small, puts them far behind.

“The summary” means the paper’s abstract: the short summary printed at its top, which is the part most readers see.

Read the paper →Full technical write-up →

Health impact assessment of the UK soft drinks industry levy: a comparative risk assessment modelling study · Lancet Public Health (2016) · doi:10.1016/S2468-2667(16)30037-8