Food and diet

Are packed lunches less healthy than school dinners?

"Packed lunches worse for kids than school dinners," Sky News reports, while Mail Online reports they are as unhealthy as they were a decade ago.

In 2006, researchers looked at the contents of lunchboxes from around 1,000 children aged 8 to 9 years old from primary schools across England. In 2016 they repeated this assessment, with a sample of around 300 children.

Researchers found that over the decade between studies, there had been very little change in the contents, with ham sandwiches on white bread and crisps still being most common. There was some positive change with a decline in sweets, chocolate and sugary drinks by about 10 to 14%. However, the vegetable content of lunchboxes remained very low.

The main difficulty is in knowing how representative this second very small sample of schoolchildren is of the UK in general; particularly given study participation (in terms of schools who agreed to take part in the study) was so poor in 2016 compared with the previous decade.

As more than half of primary schoolchildren have a packed lunch, another problem is the lack of any official guidance covering school lunchboxes, unlike those that exist for school dinners.

Nevertheless, the study highlights the need for government and education authorities to understand the reasons why healthy food choices are not being made and to support everyone in achieving healthier lifestyles for themselves and their children.

Read more advice about healthy school lunch box recipes.

Where did the story come from?

The study was conducted by researchers from the University of Leeds with funding received from the university and partially by Unilever (GB). The study is published in the peer reviewed medical journal BMJ Open, and is freely available to access online.

Mail Online coverage is fairly representative of the findings, while BBC News gives helpful information on what a healthy lunchbox could look like. The headline of the Sky News coverage, however, is misleading, giving the impression that the content of lunchboxes has been directly compared with school dinners, which this study did not do.

What kind of research was this?

This was a cross-sectional study comparing 2 surveys on packed lunches conducted in 2006 and again in 2016 to see how things had changed over the decade. Such studies are useful for looking at patterns and trends, but they cannot explain the reasons behind any observations.

What did the research involve?

In 2006 researchers had conducted a packed-lunch survey of children aged 8 to 9 years old attending 76 state primary schools across England. In 2016 researchers contacted these schools again and asked them to participate in another survey. With only 12 agreeing to take part again, they also contacted another 75 schools, receiving replies from 8 schools. This gave a total of 20 schools.

Assessments took place between June and July 2016, the same as they had been in 2006. This involved researchers attending the schools and working through a questionnaire with children aged 8 to 9 (school year 4). They reviewed and weighed the content of lunchboxes to see which foods did and did not meet food standards as set out in the School Food Plan. The plan was published in 2013 and sets standards for foods provided in schools, along with other actions to help schools change the way children eat in schools and how they learn about food.

Foods that met the School Food Plan standards included:

  • protein-rich foods
  • low-fat starchy foods
  • dairy foods
  • fruits and vegetables
  • water, milk, pure fruit juice

Foods that did not meet the standards included:

  • sweetened drinks or diet soft drinks
  • confectionery (such as chocolate and sweets)
  • savoury snacks (for example, crisps)

Some sweet treats, such oat-based flapjacks or cereal bars, did meet school standards.

What were the basic results?

Of the 20 schools participating, 2 schools subsequently dropped out, leaving a final sample of 18 schools (323 children) assessed in 2016 compared with 76 schools (1,148 children) in 2006.

The most common foods were roughly compatible in both years, including white bread in around 2 in 3 lunchboxes, ham, cheese or chicken fillings in around half of lunchboxes, crisps or corn snacks in half, and confectionary in up to 2 in 3 lunchboxes.

There were some positive changes though, with confectionery, cakes and biscuits reducing by around 10% in 2016 and sweetened drinks by 14%. However, vegetables remained low and were the least common food in boxes. Additionally, 1 in 4 children in 2016 were not bringing a drink from home. They may be drinking school water, but this was not known.

The proportion of children's packed lunches that met all School Food Plan standards (provision of 5 healthy food groups with none of 3 unhealthy groups) increased from 1.1% in 2006 to only 1.6% in 2016. However, the proportion containing none of the 3 unhealthy foods (even if they did not have all the healthy ones) did increase from 9% to 16%. The proportion containing all 3 of these unhealthy foods also decreased from 27% to 21%.

However, the level of some nutrients had also decreased between the years, including a 35% decline in vitamin C content and an 8% decline in vitamin A and zinc.

How did the researchers interpret the results?

The researchers conclude: "Packed lunches remain low quality with few meeting [nutritional] standards set for school meals [as laid out in School Foods Standards]. Provision of sugars has reduced due to reductions in provision and portion size of sugary drinks and packaged sweet foods, however, provision of some nutrients has worsened."


Overweight and obesity is a global health problem with the latest data showing that up to 30% of primary school children in England are overweight or obese. This survey study suggests that although there are some signs of improvement, with rates of sugary drinks and snacks decreasing slightly, there is still a long way to go in improving the health and nutritional content of children's lunchboxes.

However, the key limitation of this research is the low participation rate and that we cannot be sure this sample is nationally representative. The 18 schools participating in 2016 represent only 12% of those invited to take part. The participation rate was also notably poor compared with a decade ago. The number of children taking part in 2016 is less than a third of the number in 2006. Overall, we have no idea how typical these 323 children – or the 18 schools they are from – are of the national average.

This is also an assessment of food on a single day in the summer. This may be typical of the rest of the year (for these children) but we cannot be sure of that.

Other points to be aware of:

  • We cannot assume children in Year 4 are representative of other primary or infant school years.
  • This is a sample of English state schools only; other areas of the UK or fee-paying schools may be different.
  • We do not have any information about school dinners to compare with, for example knowing whether children's school dinner choices (or options) are healthier or not.

Lastly, and most importantly, we do not know the reason why this sample of packed lunches in 2016 contains such a low proportion of healthy foods. For example, it could be that awareness and education around healthy eating is not reaching most people or certain groups in particular, such as those from deprived areas. Or it could be financial reasons, for example, school dinners not being subsidised or some people thinking it's too expensive to prepare healthy food.

Understanding these possible reasons is probably the key area for national and local government and education authorities to address to try to make changes to ensure that all groups in society are able to make healthy food choices for themselves and their children.

NHS Attribution