UK United Kingdom

Stemming the obesity epidemic requires courageous, population-level action

Non-communicable diseases – Lennert Veerman examines the reasons for the obesity epidemic and options for controlling it. Obesity levels are on the rise the world over. This is a sign that something is…

The most promising interventions to curb obesity’s prevalence target the population as a whole rather than individuals. Tobyotter

Non-communicable diseases – Lennert Veerman examines the reasons for the obesity epidemic and options for controlling it.

Obesity levels are on the rise the world over. This is a sign that something is amiss with our diets and physical activity levels. So what causes this problem and what can be done to solve it?

It pays to first have a look at what is likely to have caused this increase in the prevalence of obesity.

Although there has been a gradual increase in the average body mass index of people in western countries over the 20th century, this increase accelerated from around 1980 onwards.

It’s important to note that it wasn’t a specific group within the population that became heavy; it seems rather as if everybody started getting a bit fatter, though some more than others.

Such a pattern suggests broad environmental changes were at work.

And energy balance calculations suggest a small mismatch between average energy intake and expenditure – in the magnitude of a can of soft drink more or 15 minutes walking per week less – is enough to explain obesity trends.

But was the driver a decrease in physical activity or an increase in consumption, or both?

Although facilitated by low levels of physical activity in many countries, a recent study published in the Lancet argues the main driving force of the obesity pandemic is an increase in consumption.

The authors show that in the United States, food availability declined until around 1970, in line with reductions in energy expenditure. But after that food availability (and presumably consumption) went up.

The 1970s was also when the food supply started to change radically. The supply of refined carbohydrates and fat increased and more food was mass prepared rather than cooked at home.

The era of easily available, cheap, tasty, highly promoted, energy-dense foods had begun.

Effective interventions

This view of the causes of the rise in obesity prevalence suggests the likely solutions lie in the area of the supply and promotion of food. And research supports that notion.

A range of interventions to reduce overweight and obesity in Australia were examined as part of the Assessing Cost-Effectiveness in Prevention (ACE Prevention) study.

Weight-reducing drugs were found not to be worth the money.

Diet and exercise interventions were found reasonably cost-effective, but they have a small overall effect on obesity in the population because the weight reduction is modest and temporary.

Bariatric surgery for the very obese was found to be cost-effective and had a sizeable impact on population health, mainly because the weight loss is large and permanent.

But many people may not like the prospect of a society in which a significant proportion of the population needs a lap band to lose weight.

The most promising interventions targeted populations, not individuals.

The first was a tax on unhealthy foods, and the second was traffic-light nutrition labelling of food products in supermarkets.

These interventions are likely to be cost-saving because weight loss prevents diseases, which then don’t need to be treated at a cost to the health-care system.

Taxation and, to a lesser extent, nutrition labelling are also likely to result in significant health gains at the population level.

And unlike the other interventions, they are likely to reach disadvantaged groups in society. This is important because these groups, such as the Indigenous or people with little education and a low income, suffer more from obesity. They are also hard to reach with health promotion interventions.

But the evidence for population-wide interventions is only moderately strong. Limited research has been done and the highest standard of evidence, the pooled results of double-blind randomised controlled trials, can never be attained.

Such trials are difficult when the intervention is taxation or regulation because they typically affect whole populations and researchers can’t hand out placebos. But by gathering evidence from different sources in logical mathematical frameworks (“models”) we know we can expect positive outcomes.

For example, if the price of bananas goes up, people respond by eating fewer bananas and more of other fruit. This kind of information can be used to estimate the effect of a “junk food tax”.

In similar research among children, a reduction in television food advertising came out as the best intervention. Again, this is a population-level intervention.

Getting real

If they really want to curb the obesity epidemic, governments will probably need to:

  • tax unhealthy foods,

  • subsidise healthy foods such as fruits and vegetables,

  • limit the availability of unhealthy foods, for example in schools, and

  • limit the advertising of unhealthy foods, especially to children.

This requires political courage. But powerful global players in the food industry are not keen on such measures and will try to prevent them.

The preparations for the current UN High-Level Meeting on Non-Communicable Diseases (NCDs) illustrate this very well.

Successful United Nations conferences culminate in the adoption of a declaration. Such declarations have a great influence on policy globally and within countries and are prepared in advance.

The draft declaration for the non-communicable diseases conference contains a paragraph proposing cost-effective measures to reduce the production and consumption of saturated fats, trans fats, sugar and salt.

But amendments to the draft text show Australia joined the United States and Canada in calling for the deletion of that paragraph.

The government denies this but it gives the appearance that in Canberra, the interests of powerful companies weigh more than the health of nation.

The 19th century pathologist and public health physician Rudolf Virchow famously said: “Medicine is a social science, and politics is nothing but medicine on a large scale.” That was perhaps overly optimistic.

Bringing down smoking rates required a political struggle for bans on advertising, smoking restrictions and taxation.

Reducing obesity rates will likely prove no different.

This is the third part of our non-communicable diseases series. To read the other instalments, follow the links here:

Join the conversation

5 Comments sorted by

  1. Troy Barry

    Mechanical Engineer

    We already have a 10% junk food tax - the GST with exemptions for unprocessed food. If there is any population data to show that this has had a good effect on obesity rates, you have a strong case. Otherwise I add my scepticism of a taxation response to my scepticism about whether it is even governments' legitimate business to concern itself with the problem at all.

  2. Lennert Veerman
    Lennert Veerman is a Friend of The Conversation.

    Senior Research Fellow, School of Population Health at University of Queensland

    Regrettably such policy changes are seldom introduced in a way that facilitates the collection of the data you ask for, Troy. The basic research design would be 'before-after comparison', which doesn't score high in terms of strength of evidence.

    Insisting on randomised controlled trial-level evidence simply means missing out on what probably are the most effective and efficient interventions to reduce obesity.

    As to the role of governments, I'd say that protecting the health of citizens is not only a legitimate business of governments, but their duty.

    1. Troy Barry

      Mechanical Engineer

      In reply to Lennert Veerman

      "Before-after comparison" might be weak and imperfect evidence, but that remains better than intuition. If the logic behind junk food taxation is strong, it should be reflected in pre- and post-GST obesity statistics. If it isn't, that isn't proof that junk food taxation is ineffective, but it means one piece of evidence we reasonably expected to find was not evident are we are less confident in the logic. Vice versa, if the obesity rates did change at the time of GST introduction that doesn't…

      Read more
  3. Dale Bloom


    Prescribed risk control measures within risk management legislation are elimination, substitution, isolation, engineering controls, administrative controls and personal protective equipment.

    Teaching the public about unhealthy foods is in the area of administrative controls, which is one of the least likely ways of reducing risk.

    Eliminating the risk by placing a ban on the sale of unhealthy foods, or requiring food companies and fast food shops to substitute unhealthy food with healthy food is the most likely way to reduce the risk of an unhealthy diet.

  4. Yolanda Newman

    Learning support coordinator

    I welcome any attempts to help deal with obesity. I have carried extra weight all my adult life and despite being reasonably intelligent, despite having lap band surgery, and despite having regular psychotherapy losing weight is as hard as ever. In terms of 'government responsibility' referred to as unnecessary by Troy above I disagree completely. Having had numerous problems in my life without the skills and resources to deal with them, I can identify hundreds of ways in which government action would have made a huge difference. Of course I believe in personal responsibility etc. but help available in terms of managing of food choices or through education programs. Having said that I am very disappointed in the government's lack of strength in the face of obvious business lobbying in the UN decision.