Many countries are considering the implementation of front-of-pack nutrition labels as a strategy to address high and increasing levels of overweight and obesity. A growing body of work demonstrates the superiority of labels that use colour and/or provide a summary indicator of product healthiness to enhance comprehension. However, previous studies have been confounded in determining the relative effectiveness of these two attributes by comparing labels that also differ in other ways. The present study tested labels that varied only on use of colour and/or reliance on a summary indicator across an international sample to provide unique insights into the relative importance of these attributes.

Participants were randomised to see one of four variations of the Health Star Rating label that differed on the basis of use of colour and sole provision of a summary indicator.

Australia, Canada, China, India, New Zealand, the UK and the USA.

Adults (n 7545) in seven countries were exposed to online choice tasks requiring them to select a preferred breakfast cereal and then nominate the healthiest cereal.

Overall, the coloured versions, and particularly the one with just a summary indicator, outperformed the monochrome version that included nutrient-specific information. However, there were some differences by country, with results from Canada and China indicating superior outcomes for monochrome labels and those providing nutrient-specific information.

The results highlight the importance of colour, but suggest that the introduction of front-of-pack nutrition labels should be preceded by country-specific formative testing to identify potential differences in outcomes.

To evaluate the feasibility and potential effectiveness of a lifestyle intervention (diet and physical activity) among women with history of gestational diabetes mellitus (GDM), delivered by trained facilitators.

Fifty-six normoglycaemic or prediabetic women with prior GDM were recruited at mean of 17 months postpartum. Socio-demographic, medical and anthropometric data were collected. Six sessions on lifestyle modification were delivered in groups (total four groups, with 12–15 women in each group). Pre and post intervention (6 months) weight, body mass index (BMI), waist circumference, 75 g oral glucose tolerance test, blood pressure (BP) and lipid parameters were compared.

The intervention was feasible, with 80% of women attending four or more sessions. Post-intervention analyses showed a significant mean reduction of 1.8 kg in weight, 0.6 kg/m2 in BMI and 2 cm in waist circumference. There was also a significant drop of 0.3 mmol/L in fasting plasma glucose, 0.9 mmol/L in 2 h post glucose load value of plasma glucose, 3.6 mmHg in systolic BP, and 0.15 mmol/L in triglyceride levels. Changes in total cholesterol, low-density lipoprotein-cholesterol, high-density lipoprotein-cholesterol and diastolic BP were non-significant.

This study showed feasibility of the lifestyle intervention delivered in group sessions to women with prior gestational diabetes.