Tuesday October 25 2011
The results do not show that fizzy drinks cause violent behaviour
“Teens who down more than five cans of soft fizzy drinks a week are more likely to be violent or carry a weapon,” reported the Daily Mirror. It said that researchers believe the “sugar or caffeine content in carbonated, non-diet drinks could be to blame – although they admit there may also be other factors involved.”
Many newspapers covered this study of 1,878 US high school students. Researchers surveyed the teenagers on how many non-diet soft drinks they drank and their violent behaviour. Those who drank five or more cans of non-diet soft drinks a week were about 9 to 15% more likely to say they had been violent towards others in the past 30 days, or to have carried a weapon in the past year.
Despite the level of news coverage this study received, the results do not show that fizzy drinks cause violent behaviour. This is because the findings are from a single survey that assessed soft drink consumption and violence at the same time. As such, we cannot be sure which came first and therefore whether one could have contributed to causing the other.
It is important that we put these results into context. The participants may not represent all teenagers. The study was based in schools and so may have not included the manifestly most violent teenagers who may have been excluded from school, or those who had been incarcerated. It also excluded children at private schools.
The violence in this study also ranged from pushing someone to threatening them with a weapon, and no indication is given of how severe the average level of violence was.
The causes of violence are complex and unlikely to be simply due to the consumption of fizzy drinks.
Where did the story come from?
The study was carried out by researchers from the University of Vermont and Harvard School of Public Health in the US. It was funded by the US Centers for Disease Control and Prevention. The study was published in the peer-reviewed medical journal Injury Prevention.
This story is covered in several newspapers. Though the reports included quotes from experts highlighting some of the limitations to this study, it could have been made clearer that the findings cannot tell us whether soft drinks cause violent behaviour.
What kind of research was this?
This was a cross-sectional study assessing whether there is a link between soft drink consumption and violence among adolescents in the US. The researchers say some people think that diet, including level of sugar consumption, may be linked to antisocial behaviour. They say one theory that might explain such an association is that people who consume a lot of sugary drinks may do so because they have low blood sugar levels, which have been linked to irritability and violent behaviour.
This type of study assesses two factors at the same time, and does not tell us which came first. This means that it cannot prove that one factor caused the other.
What did the research involve?
The researchers surveyed a sample of 1,878 public high school students from Boston in the US. They asked them how often they drank non-diet soft drinks in the past week, and whether they had carried a weapon or engaged in physical violence with a peer group member. They then analysed the results to see if those who drank more sugary drinks were more likely to have engaged in violence.
The survey included students in grades 9-12, who would be aged about 14 to 18 years. Religious and private schools were not included in the survey, nor were schools where students were transitioning back to school following incarceration, or schools for disabled children. Of the eligible schools, 71% participated and about four classrooms were selected at random from each school, with one class sampled for each grade. Of the 2,725 eligible students, 69% participated and filled in the study questionnaire.
Students were asked how many non-diet soft drink cans (12oz or 355ml) they had drunk in the past week (a 20oz [591ml] bottle was counted as two cans). Based on their answers they were grouped into those who drank up to four cans in the past week and those who drank five or more. The students also answered questions about whether they had been violent towards other adolescents, another child in their family or someone they were dating in the past 30 days. Violence was defined as:
- physical fight
- kicking or choking a person
- attacking or threatening a person with a weapon
They were also asked if they had carried a gun or knife anywhere in the past year.
In their analyses, the researchers compared violent behaviours between those who drank soft drinks more often and those who drank them less often. These analyses took into account factors that were also assessed in the questionnaire and could influence their results, including age, sex, race, body mass index (BMI), typical sleep patterns, tobacco use, alcohol use and having family dinners.
What were the basic results?
The researchers found that 29.8% of the participating adolescents reported drinking more than five cans of non-diet soft drinks each week, and 70.2% drank less than this. Adolescents who drank more than five cans a week were more likely to have used tobacco or alcohol in the past 30 days.
Overall, 30.8% reported carrying a gun or knife in the past year. In the last 30 days, 44.4% reported being violent to a peer, 19.5% being violent in a dating relationship, and 31.6% being violent to a child in their family.
Adolescents who drank more than ﬁve cans of soft drinks in a week were signiﬁcantly more likely to:
- have carried a weapon: 40.3% compared with 26.8% who drank four cans or less a week
- have been violent with peers: 56.7% compared with 39.1% who drank four cans or less a week
- have been violent with children in their family: 42.0% compared with 27.2% who drank four cans or less a week
- have been violent with dates: 26.2% compared with 16.2% who drank four cans or less a week.
The link between soft drink consumption and these measures remained even after taking into account factors such as age, sex and race, which could influence results.
The researchers found consumption of high quantities of soft drinks was associated with a 9 to 15% higher likelihood of engaging in violent behaviour or carrying a weapon. The link between high soft drink consumption and violence was similar to the links between violence and tobacco or alcohol use, which were associated with a 6 to 20% higher likelihood of engaging in violent behaviour. The link between high soft drink consumption and carrying a weapon (9% increase) was weaker than the link between tobacco or alcohol use and carrying a weapon (15 to 26% increase).
How did the researchers interpret the results?
The researchers conclude that there was a strong link between soft drinks and violence. They say that this 'may be a direct cause-and-effect relationship, perhaps due to the sugar or caffeine content of soft drinks, or there may be other factors, unaccounted for in our analyses, that cause both high soft drink consumption and aggression'.
This study has found a link between soft drink consumption and violent behaviour. However, there are a number of limitations to this study that need to be considered when interpreting its findings:
- The main limitation is that it was cross sectional. This means that it cannot establish which factor came first: soft drink consumption or violence, and therefore cannot say whether one might have contributed to the other.
- The study took into account some factors that could be contributing to an association between violence and soft drink consumption, but there may be other factors having an effect. For example, it did not take into account the adolescents’ socioeconomic status, which seems likely to be contributing to this relationship.
- The study had to rely on the teenagers’ reports of their own soft drink consumption and violent behaviour, and there may be some inaccuracies, particularly with regard to violent behaviour.
- The violence assessed in the study ranged from pushing someone to threatening them with a weapon. This is a wide range and the study does not split this into different levels of violence, meaning that we do not know how severe this violence was.
- The study included adolescents who were attending public school in the US and willing to complete a questionnaire. These teenagers may not be representative of all teenagers. In particular, the most violent teenagers are likely to have been excluded from school or to have been in correctional facilities. The study also excluded children attending private school who are likely to have different socioeconomic profile.
Explanations for links found in research might not always be causal. The researchers mention that low blood sugar may plausibly be linked to both aggressive behaviour and consumption of sugary drinks. This study raises questions about such unmeasured variables rather than providing answers.
The causes of violence are complex, and unlikely to be simply due to the consumption of fizzy drinks.