The PHAPO working group has developed a fresh way of looking at the many social factors contributing to obesity. The causal web represents a new concept in linking a wide variety of elements that have an impact on energy input and output (see illustration). 

Although complex at first sight, this is a true reflection of the nature of the obesity problem from a public health perspective. Building the web up from right to left, column by column can aid understanding.

This new framework has been welcomed at several meetings, the first presentation taking place in August at an ICO98 satellite session - the Berzelius Symposium on Prevention of Obesity in Stockholm, where the PHAPO group held a two-hour workshop. Professor Shiriki Kumanyika, who chairs PHAPO, also presented the ‘web’ at a workshop during the annual American Public Health Association (APHA) meeting in Washington DC in November 1998 and at the WHO meeting on ‘Behavioural and Social Aspects of Preventing Obesity’ in Tokyo, Japan, in December 1998. 

The concept of the causal web was originated by Dr. Cheryl Ritenbaugh, a nutritional anthropologist and epidemiologist and PHAPO member. Dr. Ritenbaugh presented it at the Denver Initiative For Healthy Lifestyles Meeting in November 1998, the American Anthropological Association Annual Meeting in Philadelphia in December 1998, and at a CDC conference entitled "Obesity Etiology and Prevention: A (truly) comprehensive view" in Atlanta in January 1999. 

The diagram shows the factors that operate at similar social structural levels labelled across the top. The most common target for interventions to reduce obesity levels is individual behaviour (shown to the right). The large number of arrows indicate the factors bearing on individual behaviour, which challenges the notion of individual "free will" regarding food choice and energy expenditure. 

The School/Work/Home column represents the next most common focus for intervention. Each of these sites is influenced by factors further upstream. For example, although schools can be a target for changing the food supply or physical activity patterns of youth, they rarely operate independently of external bodies; funding sources, accrediting bodies, and elected school boards all have different agendas for school functioning and budgeting. The bottom line for schools is academic performance, and health issues are generally secondary. 

National policies providing adequate incentives or financial support may be necessary to bring about widespread change in schools - no matter how good a particular educational program may be. Even the provision of food within the schools encounters linkages with government food policies supporting school lunches, availability of fresh foods, and economic issues tied to the frequent situation that revenues from food sales are the only discretionary funds available to schools. 

Another example is found in urban planning. The use of stairs instead of elevators is often recommended for people who want to increase physical activity. To support this approach, efforts have been made to work with building designers to make stairs more attractive and convenient, and elevators less so. These efforts have run into myriad problems, including building codes, fire safety standards, design expectations, and architectural prerogative. Even simple solutions become complex. 

State and national level policies are linked to international markets and the level of economic development of countries. International monetary agencies and banks may encourage/require debtor countries to use more of their agricultural capacity for export rather than local food supplies, shifting the food availability and cost to all sectors. International corporations through advertising and other marketing activities greatly influence food availability and demand; foods most widely and intensively marketed are likely to have a long shelf life, a very high profit margin, and be easily transported throughout the globe. International corporations similarly influence purchasing capacity through salaries and energy expenditure through the level of labour required to earn income. 

In this web there is no single organizational locus where motivation to change behaviour can result in a decline in obesity without interference from competing interests. Coalitions with commitment to change must be formed across all levels, founded on mutual interest. This is the classic situation that requires a broad-based public health perspective.