The complex interaction of causing and resulting factors of overweight/obesity

Full text


Wissenschaft & Forschung

| Original Contribution

Peer-reviewed | Manuscript received: April 5, 2012 | Revision accepted: October 3, 2012 Beleg/Autorenexemplar!

Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar.

Dies gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronische Systeme.

The complex interaction of causing and

resulting factors of overweight/obesity

Increasing the understanding of the problem and deducing requirements

for prevention strategies

Eva Hummel, Friederike Wittig, Karlsruhe; Katja Schneider, Nadine Gebhardt,

Giessen; Ingrid Hoffmann, Karlsruhe


The prevalence of overweight/obe-sity has developed into a worldwide epidemic with severe consequences [1, 2], even though numerous pre-vention projects have been and are being implemented [e. g. 3, 4]. One reason for the mediocre success may be that prevention measures do not adequately consider the complexity of the problem [5].

Against this background, the present study depicts and investigates the complex phenomenon of over-weight/obesity with the help of

Nu-triMod, a nutrition-ecological modeling technique. A qualitative cause-effect model is used to visual-ize the interactions between the causing and resulting factors of overweight/obesity described in lit-erature. Examples are presented to demonstrate additional insights into overweight/obesity and for the de-velopment of prevention measures which may be identified by the pres-entation and investigation of the complex phenomenon.

Developing the model:

the modeling technique


A qualitative cause-effect model was developed on the basis of scientific literature, with the help of the nu-trition-ecological modeling tech-nique (Nutrition-ecological Mod el-ling) (쏆Figure 1) [6, 7]. In accordance with the nutrition-ecological ap-proach[8, 9] the four dimensions of nutrition – health, society, environ-ment and economy – were consid-ered. Between March 2006 and Feb-ruary 2009, scientific literature was collected from numerous databases (e. g. Medline, Web of Science, Agri-cola, FSTA, EMBASE), using search terms such as overweight, obesity, causes and effects, together with search terms from the different


In order to demonstrate the complexity of overweight/obesity, a literature-based, qualitative cause-effect-model was developed applying nutrition-ecological mod-elling (NutriMod). By visualising many multicausal and multidimensional interre-lated causes and effects of overweight/obesity, the model illustrates two factors di-rectly causing overweight/obesity: energy balance and biological factors. All other factors influence overweight/obesity in particular via energy balance and thus in-directly via cause-effect chains. These cause-effect chains can join to feedback loops and thereby result in reinforcing vicious circles. The model increases the un-derstanding of the complex interaction of overweight/obesity by gaining know-ledge on cause-effect chains, feedback loops, multicausality and multidimension-ality. Therefore, it serves as a good basis for the development of effective preven-tion measures.

Keywords:Nutrition ecology, nutrition-ecological modelling technique, NutriMod, complexity, prevention, overweight/obesity, cause-effect model


Hummel E, Wittig F, Schneider K, Gebhardt N, Hoffmann I (2013) The complex inter-action of causing and resulting factors of overweight/obesity. Increasing the under-standing of the problem and deducing re-quirements for prevention strategies. Erna-ehrungs Umschau international 60 (1): 2–7 DOI 10.4455/eu.2013.002


-Food supply


Food availability Ser

ving size External stimuli Price Food composition Labelling Socio-economic status

Education Income Profession/ occupation

Agents of socialisation Media Peer groups Family School/ kindergarten

Social change Affluent society Fast-paced life style Ideal of beauty

Urbanization W omen’ s employment Household structures Religion Lifestyle factors Nutrition behaviour Physical activity

Media consumption Sleeping behaviour


Biological factors


other factors Genetics

Age Sex

Prenatal and infantile


Infant nutrition Birth weight Mental factors Emotions/stress

Traumatic experiences Body image Self-esteem Depression

Migration Eating disorders Prevention Quality of life Isolation Infrastructure/ neighbourhood Nutrition competences Energy balance




Gestational diabetes Environmental pollution Food production/ processing

Technological progress/ globalization



Discrimi- nation

summarized factors

causal relationship

© Authors: Katja Schneider


Wissenschaft & Forschung

| Original Contribution

mensions of nutrition (e. g. for the dimension economy: costs, advertis-ing; for the dimension health: co-morbidities, physical activity, genet-ics). System boundaries of the model were defined in such a way that the situation in western industrial coun-tries is covered.

Causes and effects of over-weight/obesity described in scientific literature are depicted as factors in the qualitative cause-effect model (쏆Figure 1). Terms in italics in this article indicate factors in the model. In the model the factors are pre-sented on a highly aggregated level: Firstly, the listed factors often repre-sent a multitude of further differen-tiated (sub)factors, which are not di-rectly visible in the model, but in-cluded implicitly. For example, the factor nutrition behaviour includes the subfactors food preferences, food choice andfood preparation. Secondly, some factors in the model are sum-marized under topics in boxes. For example, the factors media, peer groups, school/kindergartenand fam-ilyare summarized under the topic agents of socialisation. The factors were selected on the basis of their rel-evance to the subject over-weight/obesity and on the basis of scientific literature available at the time the model was built. Thus, this compilation does not claim to be complete.

Arrows between factors indicate causal relationships, showing the di-rection of the effect. For example, A

→B means that A causes a change in B. The factors and interactions shown in the model are based on re-sults of about 460 references. In an electronic version of the model, the depicted causal interactions and se-lected factors can be accessed in a hy-pertext structure, with detailed de-scriptions, explanations and a selec-tion of references (www.uni-gies obes_e.php [10]).

The complex phenomenon

of causes and effects of


Various aspects of the complex phe-nomenon of overweight/obesity will now be described in more detail and illustrated on the basis of the model.

Causing factors of overweight/obesity

Overweight/obesityis influenced by numerous causing factors. A dis-tinction must be made between di-rect and indidi-rect factors. The cause-effect model shows only two factors which directly influence overweight/ obesity: energy balance, which results from the ratio of energy intake to energy expenditure, and biological factors, which summarize genetic predisposition (geneticsin the model) and hormones/cytokines/other factors. All other factors described in litera-ture affect the development of over-weight/obesityindirectly and almost exclusively act via energy balance. For example, social changeis an indirect causing factor, which can contribute to a positive energy balance[11] and therefore to the development of over-weight/obesity[1] via changes in nu-trition behaviourand reduced physi-cal activity[11].

This example shows that over-weight/obesitycan arise from a series of interacting factors, so-called “cause-effect chains”. The model can be used to identify and retrace these cause-effect chains.

Resulting factors of overweight/obesity

Apart from numerous causing fac-tors, overweight/obesityhas many re-sulting factors. The model shows that there are both direct and, via cause-effect chains, indirect effects. For example, overweight/obesity di-rectly causes costsdue to necessary

adaptions in means of transporta-tion [12] or operating tables [13] arising from changes in body size (in the model: adaption to body size). Moreover, indirect costs arise from the diagnosis and therapy of co-morbidities associated with over-weight/obesity, such as changes in metabolism or the development of carcinoma[11].

Feedback loops

Cause-effect chains can close to form cause-effect circles, called feedback loops. These become apparent when the causal relationships between fac-tors are depicted in a cause-effect model. For example, overweight/obe-sity can directly result in an in-creased risk of suffering from depres-sion, negative body image and low self-esteem in affected individuals [14]. These mental factorscan then again lead to changes in nutrition be-haviourand physical activity, causing a consistently positive energy balance [15], which may enhance over-weight/obesity[16], thus closing the cause-effect circle. Like most feed-back loops in this model, this is a positive feedback and thereby a vi-cious circle.

As a result of feedback loops, causes can become effects and vice versa. Thus, mental factors initially are a direct resulting factor of over-weight/obesity, but, due to the de-scribed feedback loop, become an in-direct influence on its development or promotion. Consequently, in such a complex phenomenon it is not al-ways possible to distinguish cause from effect. Visualization of the re-lationships can support revealing factors which can be both cause and effect at the same time.


Aside from cause-effect chains and feedback loops, the model allows



above mentioned cause over-weight/obesity, but also by numer-ous other factors, such as social change, lifestyle factors, agents of so-cialization, etc.


In accordance with the nutrition-ecological approach, causing and re-sulting factors of overweight/obesity are assigned to the four dimensions of nutrition (health, environment, economy and society). As a result of the visualization as a cause-effect model, it becomes evident that fac-tors of different dimensions interact. This implies that cause-effect chains and feedback loops may proceed over more than a single dimension of nu-trition, and therefore are dimension transcending. For example, social change(dimension society) influences lifestyle factors(dimensions society and health), which can then lead to overweight/obesitythrough changes in energy balance. As a consequence of overweight/obesitynumerous co-morbidities(dimension health) may develop, leading to increased costs (dimension economy).

Considering the complex

phenomenon in

prevention measures

A large proportion of current pre-vention projects have the objective of changing nutrition behaviouror phys-ical activityand thus energy balance [17].

However, the cause-effect model of the complex phenomenon of over-weight/obesity shows that a multi-tude of factors influence nutrition be-haviourand physical activity. These are themselves influenced by other

factors. This means that the causes for the development of over-weight/obesity are frequently pro-ceeding to nutrition behaviour and physical activityand act via cause-effect chains. Moreover, the cause-effect model reveals feedback loops, through which effects may become causes and through which reinforc-ing effects may possibly lead to vi-cious circles.

For a prevention measure to be suc-cessful, it is therefore necessary to identify causes or possible starting points related to both direct and ob-vious causing factors, as well as in-direct preceeding causing factors along cause-effect chains, as well as to recognize and consider feedback loops [see 18].

Insights about multicausal inter-actionsbetween the factors are cru-cial since the influence of a preven-tion measure can counteract other influences within the complex phe-nomenon. Thus, both the factor used as a starting point for the pre-vention measure as well as inde-pendent of this, other factors along the according cause-effect chain may be affected by other factors. If these influences are neglected they may at-tenuate or prevent the success of the measure. For example, a measure to increase schoolchildren’s motivation to eat more fruit and vegetables (in the model: food preferencesas a sub-factor of nutrition behaviour) would not be successful if it neglected the existing opinions and trends in the peer group. Or, the motivation may

tween a multitude of factors in the dimensions health, environment, so-ciety and economy, along the complete food supply chain. In order to develop integrative solution approaches for complex nutrition-associated problems, multidimensionality and complexity in the field of nutrition are investigated and considered [23].

NutriMod(Nutrition-ecological modelling) is a modelling technique used to develop a qualitative cause-effect model on the basis of scientific liter-ature and expert knowledge. According to the nutrition-ecological ap-proach [8] the method is characterized by including the four dimensions of nutrition (health, society, environment and economy) along the com-plete food supply chain, so that all relevant aspects of a complex theme are considered [6].

Interdisciplinaritymeans that scientists from at least two disciplines co-operate concerning common objectives and results [19]. The intention is to combine their disciplinary perception and knowledge to an integrated, not additive result. True interdisciplinary research defines and solves its problems independently of disciplines [20].

Transdisciplinaritycombines interdisciplinarity with the participation of non-scientific actors [e.g. 19, 21]. Participation means that actors affected by the problem and having an own perspective concerning the problem to be solved (e.g. government, economics, civil society) are actively in-volved in shaping the research process. The aim of transdisciplinarity is to contribute to solve so-called “real world problems” [22].


Wissenschaft & Forschung

| Original Contribution

be enhanced, but along the cause-ef-fect chain no change in food selection (subfactor of nutrition behaviour) is induced, because fresh fruit and veg-etables may not be available at work or in school (in the model: availabil-ity) or the price is too high. For prevention of overweight/obe-sity it is therefore necessary to con-sider the multicausalities apparent through the model.

As the phenomenon is multidimen-sionalthe conception of prevention measures requires involving differ-ent scidiffer-entific disciplines and fields of practice via interdisciplinaryand

transdisciplinaryapproaches. The model can support the identification of the relevant actors by visualizing different factors in single dimensions [7, 8].


If prevention measures are to be suc-cessful, it is important to take into account the complexity of the phe-nomenon of overweight/obesity. For this, it is necessary to enhance the understanding for the multidimen-sional interaction of causing and re-sulting factors and to recognize cause-effect chains, feedback loops and multicausality. Accordingly, the nutrition-ecological modelling tech-nique NutriMod was used to visual-ize the direct and indirect causing and resulting factors of over-weight/obesity with their complex and multidimensional interactions in a qualitative cause-effect model. This visualization can enhance the understanding of the problem and thus provide a basis for the concep-tion and implementaconcep-tion of effective prevention measures. The model makes it apparent that in addition to two direct causing factors numerous indirect factors along cause-effect chains and feedback loops contribute

to the development and promotion of overweight/obesity. Thus, the model can support to identify deci-sive causes, which may have been neglected or which root in vicious circles. These may then provide pos-sible starting points for prevention measures. The model also reveals which factors might interfere the success of a measure, if their multi-causal interactions are not recog-nized and considered. The model also portrays the multidimensionality of overweight/obesity, allowing to identify relevant disciplines and fields of practice for inter- or transdiscipli-nary approaches, which can be in-corporated into the conception of prevention measures.

The present qualitative cause-effect model thus provides a good basis to allow for the complexity of the phe-nomenon of overweight/obesity and to develop appropriate prevention measures.


We wish to thank Katrin KELBCHfor

support in literature search and evaluation for the preparation of the final version of the model. We also wish to thank Birte MENG, Lina



HALISCH, Ina HOTTMANN, Ulrike KAM -POWSKY, Miriam KURZ, Anne-Katrin


Anna NEFF, Tobias RÖSCH, Kathrin


par-ticularly Leonie KNORPPand Annett

WEISERfor helping to develop an

earlier version of the model.


1. WHO (Weltgesundheitsorganisation Regio-nalbüro für Europa) (Hg). Die Herausfor-derung Adipositas und Strategien zu ihrer Bekämpfung in der Europäischen Region der WHO. Zusammenfassung. Kopenhagen (2007). URL: E89858G.pdf?language=German Zugriff 16.03.12

2. WHO (World Health Organization) (Hg). Obesity: preventing and managing the global epidemic. Report of a WHO consultation. WHO technical report no. 894. Genf (2000) 3. Doak CM, Visscher TL, Renders CM, Seidell JC (2006) The prevention of overweight and obesity in children and adolescents: a review of interventions and programmes. Obes Rev 7: 111–136

4. Summerbell CD, Waters E, Edmunds LD et al. (2005) Interventions for preventing obe-sity in children. Art No: CD001871. Cochrane Database Syst Rev (3): 1–78 5. Hilbert A, Ried J, Schneider D et al. (2007)

Primäre Prävention der Adipositas bei Erwachsenen. Eine interdisziplinäre Analyse. Herz 32: 542–552 M.Sc. Eva Hummel1, 2 Phone: +49 (0)721-66 25 268 Fax: +49 (0)721-6625 552 M.Sc. Friederike Wittig2 Dr. Katja Schneider1 Dr. Nadine Gebhardt1

Prof. Dr. Ingrid Hoffmann2

1Arbeitsgruppe Ernährungsökologie Institut für Ernährungswissenschaft Justus-Liebig-Universität Gießen Wilhelmstraße 20, 35392 Gießen 2Institut für Ernährungsverhalten

Max Rubner-Institut, Bundesforschungsinstitut für Ernährung und Lebensmittel

Haid-und-Neu-Straße 9, 76131 Karlsruhe verhalten.html

Conflict of Interest

The authors declare no conflict of interest ac-cording to the guidelines of the International Committee of Medical Journal Editors.


7. Schneider K, Hoffmann I (2011) Potentials of qualitative modelling of complex health issues. Am J Health Behav 35: 557–567

8. Schneider K, Hoffmann I (2011) Nutrition Ecology – a concept for systemic nutri-tion research and integrative problem solving. Ecol Food Nutr 50: 1–17 9. Schneider K, Hoffmann I. Ernährungsökologie: ein integrativer Ansatz für

Ernährungsforschung und -praxis. In: Ploeger A, Hirschfelder G, Schoenberger G (Hg). Die Zukunft auf dem Tisch. Analysen, Trends und Perspektiven der Ernährung von morgen. VS Verlag, Wiesbaden (2011), S. 389–404

10. Schneider K, Wittig F, Mertens E, Hoffmann I. Übergewicht/Adipositas: komplexes Zusammenspiel von Einflussfaktoren und Auswirkungen. Hyperlinkmodell. Gießen (2009) URL: Zugriff 16.03.12

11. Munsch S (2002) Epidemiologie der Adipositas. Verhaltenstherapie 12: 278–287 12. Hauschild SW, Wirtz S, Köster N, Nölken W (2005) Der adipöse Patient im

Ret-tungsdienst. Notfall Rettungsmed 8: 207–215

13. Schröder T, Nolte M, Kox WJ, Spies C (2001) Anästhesie bei extremer Adipositas. Herz 26: 222–228

14. Greenberg I, Perna F, Kaplan M, Sullivan MA (2005) Behavioral and psychologi-cal factors in the assessment and treatment of obesity surgery patients. Obes Res 13: 244–249

15. Kielmann R, Herpertz S (2001) Psychologische Faktoren in der Entstehung und Be-handlung der Adipositas. Herz 26: 185–193

16. Holub M, Götz M (2003) Ursachen und Folgen der Adipositas im Kindes- und Ju-gendalter. Monatsschr Kinderheilkd 151: 227–236

17. Müller MJ (2009) How are we going to turn the obesity prevention experience? Obes Rev 11: 101–104

18. Hummel E. Komplexe Probleme: Herausforderungen für Problemlösungen. In: Hoff-mann I, Schneider K, LeitzHoff-mann C (Hg). Ernährungsökologie. Komplexen Heraus-forderungen integrativ begegnen. oekom, München (2011), S. 170–176 19. Defila R, Di Giulio A, Scheuermann M. Forschungsverbundmanagement.

Hand-buch für die Gestaltung inter- und transdisziplinärer Projekte. vdf Hochschulver-lag, Zürich (2006)

20. Mittelstraß J. Die Häuser des Wissens. Wissenschaftstheoretische Studien. Suhrkamp, Frankfurt/Main (1998)

21. Klein JT, Grossenbacher-Mansuy W, Häberli R et al. (Hg). Transdisciplinarity: Joint problem solving among science, technology, and society. Birkhäuser, Basel (2001)

22. Pohl C, Hirsch Hadorn G. Gestaltungsprinzipien für die transdisziplinäre Forschung. Ein Beitrag des td-net. oekom, München (2006)

23. Schneider K. Konzeptionelle und methodische Entwicklungen zur Bearbeitung kom-plexer ernährungsassoziierter Fragestellungen im Rahmen ernährungs-ökologischer Forschung und Lehre. Dissertation, VVB Laufersweiler Verlag, Gießen (2012)