P Kopelman

Overweight and obesity cause or exacerbate a large number of health problems, both independently and in association with other diseases, and are among the most significant contributors to ill health (1–4). The intra-abdominal visceral deposition of adipose tissue, which characterizes upper body – central – obesity (assessed by waist circumference and/or waist : hip ratio) is a major contributor to the development of hypertension, elevated plasma insulin concentrations and insulin resistance, hyperglycaemia and hyperlipidaemia (metabolic syndrome). The concept of the metabolic syndrome refers to the phenomenon of risk-factor clustering – a collection of metabolic traits occurring in the same individual with the clustering presumably reflecting a unifying underlying pathophysiology that requires a holistic approach to their management (5–7). Many of the health risks associated with increasing body weight begin to appear in children and young people. Of great concern is the increasing prevalence of type 2 diabetes and associated medical complications in childhood (8–10). This early onset needs to be reflected by active management and research priorities to reduce the consequential health and economic burdens (11). Evidence from twin, adoption and family studies unequivocally indicates that inherited factors play a major role in the determination of inter-individual differences in fat mass. The identification of genetic variants influencing human fat mass is critical to obesity research. In general, mutations have been found in those rare children with extreme obesity and clear evidence for monogenic inheritance. Genetic studies in the more common forms of obesity have yet to have the same tangible success that has been seen with monogenic subtypes. Nevertheless, progress has been made in the identification of certain chromosomal loci containing genes conferring susceptibility to obesity. Body weight is determined by an interaction between genetic, environmental and psychosocial factors acting through the physiological regulation of energy intake and expenditure. Similarly, genetic, environmental and physiological factors are likely to determine the associated health risks. The genetic influences appear to operate through susceptibility genes, which may be expressed more frequently in certain ethnic groups (12). Such genes increase the risk of developing a characteristic but they are not essential for its expression or, by themselves, sufficient to explain the development of a disease. The differences in genetic susceptibility within a population are likely to determine those who are more likely to become obese and/or develop associated medical complications in any given set of environmental circumstances. Implicit to the susceptible gene hypothesis is the role of environmental factors that unmask latent tendencies to develop obesity and health risks. A challenge for the future will be to identify the genetic and environmental contributions to weight gain, the intra-abdominal distribution of fat (central obesity) and also to health problems, paying attention to those at particular risk, including certain ethnic groups, susceptible families and the socially deprived. A further challenge will be to discover whether these factors operate through weight gain, independently or in combination. Obesity is characterized by elevated fasting plasma insulin and an exaggerated insulin response to an oral glucose load: a positive correlation is observed between increasing central obesity and measures of insulin resistance. Post-hepatic insulin delivery is increased in central obesity leading to more marked peripheral insulin concentrations that, in turn, lead to peripheral insulin resistance. In both men and women, the lipolytic response to noradrenaline is more marked in abdominal than gluteal or femoral adipose tissue. Cortisol may also contribute to this enhanced lipolysis by further inhibiting the anti-lipolytic effect of insulin. Increased production of a fat-cell-specific protein called resistin by abdominal subcutaneous and omental adipose tissue is associated with decreased cellular insulin sensitivity. By contrast, there is an inverse relationship between adiponectin (an adipocyte-specific secretory protein) and adiposity and insulin sensitivity, with type 2 diabetes being characterized by low levels of adiponectin. Adiponectin is similar in structure to tumour necrosis factor-alpha (TNF-α), which paradoxically appears to be increased in abdominal adipose tissue. Increases in pro-inflammatory cytokines [interleukin 6, resistin, TNF-α and C-reactive protein (CRP)] reflect overproduction by the expanded adipose tissue mass (13,14). All of these factors contribute to the exaggerated release of free fatty acids (FFAs) from abdominal adipocytes into the portal system. FFAs have a deleterious effect on insulin uptake by the liver and contribute to the increased hepatic gluconeogenesis and hepatic glucose release observed in central obesity. Insulin insensitivity is not confined to adipocytes because the process is accentuated by skeletal muscle insulin resistance. The relationship between insulin resistance and hypertension is well established and is underpinned by several different mechanisms. Insulin is a vasodilator and has secondary effects on sodium reabsorption by the kidney. In the setting of insulin resistance, the vasodilatory effects of insulin can be lost but sodium reabsorption appears to be preserved. Insulin also increases sympathetic nervous system activity. Hyperinsulinaemia and insulin resistance are both significant correlates of a dyslipoproteinaemic state and contribute to the characteristic alterations of plasma lipid profile associated with obesity. Dyslipidaemia progressively develops with increasing abdominal fatness and body mass index (BMI). It is first noted at a BMI of 21 kg m−2 (2), when a rise in proatheromatous, dense, small-particle low-density lipoprotein (LDL) is seen. With elevated LDL concentrations, as well as high concentrations of triglycerides, coronary heart disease risk rises. The combined effect of saturated and trans-fatty acids on plasma lipids is amplified by the lack of n-3 long chain fatty acids, which have competitive effects on prostanoid synthesis, cellular function and thrombosis. The effects of cytokines on peripheral tissues with increased intracellular lipids also lower cellular insulin sensitivity: the surge in lipids promotes proliferation of the vasa vasorum of the arterial media and apoptosis by the medial macrophages with a further release of cytokines. There is an additional complex relationship through an associated pro-inflammatory state and alterations in coagulability. A rise in blood viscosity is induced by the release of profibrinogen and plasminogen activator inhibitor 1 from adipocytes with a fall in plasminogen activator. These changes may explain the role of obesity as a promoter of intracellular inflammatory processes that result in arterial damage. Ethnic differences in CRP may explain some of the variation observed in insulin resistance across populations with comparable weight gain and associated medical complications. It is unclear whether tackling weight gain directly or treating the resulting pro-inflammatory state will have the greater health benefit. The effects of increased body fatness on cardiovascular function can be predicted. Total body oxygen consumption is increased because of an expanded lean tissue mass and metabolically active adipose tissue, and this is accompanied by an absolute increase in cardiac output. The total blood volume in obesity is increased in proportion to body weight. This increase in blood volume contributes to an increase in the left ventricular pre-load and an increase in resting cardiac output. The increased demand for cardiac output is achieved by an increase in stroke volume: an increase in stroke volume results from an increase in diastolic filling of the left ventricle. The volume expansion and increase in cardiac output lead to structural changes of the heart. The increase in left ventricular filling results in an increase in the left ventricular cavity dimension and an increase in wall stress. This thickening of the wall with dilatation results in eccentric hypertrophy. Left ventricular mass increases directly in proportion to BMI or the degree of overweight. In those subjects where systemic resistance is increased, the combination of hypertension and obesity results in an increase of ventricular wall dimensions disproportionate to the chamber radius that with time leads to concentric hypertrophy. The cardiovascular adaptation to the increased intravascular volume of obesity may not completely restore normal haemodynamic function. Marked systolic dysfunction occurs when the ventricle can no longer adapt to volume overload. Dilatation of the left ventricle cavity radius leads to a decline in ventricular contractility. A combination of systolic and diastolic dysfunction progresses to heart failure (15). An increased amount of fat in the chest wall and abdomen has an effect on the mechanical properties of the chest and leads to an alteration of respiratory excursion during inspiration and expiration, reduces lung volume and alters the pattern of ventilation to each region. Such changes are significantly exaggerated when an obese person lies down flat. The mass loading effect of fat requires an increased respiratory muscle force to overcome the excessive elastic recoil and an associated increase in the elastic work of breathing. The obesity-related changes in respiratory function are most important during sleep. Irregular respiration and occasional apnoeic episodes often occur in lean people during Rapid Eye Movement (REM) sleep, but obesity, with its influence on respiratory mechanics, increases their frequency and may result in severe hypoxia and resultant cardiac arrhythmias. Studies of obese men and women have demonstrated that the obstruction occurs in the larynx and is associated with loss of tone of the muscles controlling tongue movement. Relaxation of the genioglossus muscle allows the base of the tongue to fall back against the posterior pharyngeal wall, occluding the pharynx. This results in a temporary cessation of breathing (sleep apnoea) and associated transient fall in arterial oxygen saturation concentration – hypoxia. It is not uncommon to observe very low oxygen values during REM sleep in some obese subjects, even though their awake arterial gases are normal. Daytime somnolence may intervene, accompanied by persistent hypoxia and/or raised carbon dioxide (hypercapnia), pulmonary hypertension (superimposed on an increased circulatory volume) and right-sided cardiac failure. Such changes constitute the clinical manifestation of the obesity–hypoventilation syndrome, formerly known as the Pickwickian syndrome. A number of groups have reported an increased risk of myocardial infarction and stroke associated with sleep apnoea. Snoring is a strong risk factor for sleep-related strokes, while sleep apnoea symptoms increase the risk of cerebral infarction. Importantly, body weight, independent of several traditional risk factors, is directly related to the development of congestive cardiac failure in the Framingham Heart Study, a clinical association not widely recognized. Changes in lifestyle are major contributors to the current epidemic of overweight and obesity. Energy-dense diets rich in fats and refined sugars promote weight gain, and high sugar and salt intakes also induce an increase in blood pressure. Traditionally, low-calorie diets (800–1500 kcal d−1), which incorporate various methods for restricting intake, have been recommended for weight management. The use of low-calorie diets with a treatment period beyond 6 months has been associated with a mean weight loss of ∼8%, although with longer use (3–4.5 years), this is reduced to ∼4%. Generally, these strategies aim to provide a macronutrient composition of 30–35% energy from fat, 50% from total carbohydrate and 10–15% from protein, thereby moving individuals towards national dietary guidelines. Energy density is reduced by higher intake of vegetables and fruit, which the Dietary Approaches to Stop Hypertension (DASH) trial (16) showed also reduced blood pressure. Data from the DASH trial suggest that blood pressure can be lowered independently of weight change, especially in people with hypertension. Esposito and colleagues randomized 180 Italian men and women with the metabolic syndrome to a Mediterranean-style diet (food rich in mono- and polyunsaturated fat, fibre and a low ratio of omega-6 to omega-3 fatty acids) or to an ad lib diet, and followed them for 2 years (17). Subjects on the Mediterranean diet lost more weight but, even after accounting for this, there was a 39% reduction in the prevalence of the metabolic syndrome. The American Diabetes Prevention Programme (18) and the Finnish diabetes prevention trial (19) used a low-fat diet in combination with physical activity (at least 150 min per week) and behavioural strategies to demonstrate the feasibility of sustaining modest weight loss (3–4 kg) over time (4 years), with an accompanying 58% reduction in the risk of developing type 2 diabetes. Physical activity recommendations of 30 min of moderate activity on at least 5 days of the week are associated with improved fitness and protection from cardiovascular diseases (20). However, recent evidence highlights a requirement for 45–60 min per day to maintain lowered weight and prevent weight regain (21). Reduction in the time spent in sedentary behaviours (such as television watching) is an important strategy for increasing physical activity and energy expenditure in children and young people. Five areas for research into health risks associated with overweight and obesity have been identified. All of these areas will expand current scientific knowledge about the interrelated biological mechanisms and environmental factors that links weight gain to health risks. A priority is prevention of overweight and obesity in young people. Another priority is to test the effect of weight reduction by various dietary manipulations and increased physical activity in preventing and reducing health risks across the population. The pharmaceutical industry is actively investing in research into novel therapeutic agents for the management of established health risks and associated diseases. This will continue to require active collaboration across all spheres of academic laboratory and clinical research in the United Kingdom. No conflict of interest was declared.

Obesity is not new (1) but it is a growing problem. While the mean body mass index (BMI; a measure of relative weight) of the population has been steadily increasing among adults over many generations, there has been a concomitant increase in the proportion of the population who are seriously obese. The latest World Health Organization estimates are that around 1.2 billion people in the world are overweight and at least 300 million of them are obese. The UK is fortunate to have year-on-year surveys that track the increasing prevalence of obesity. In 2004, 23.6% men and 23.8% women were obese and this is predicted to increase to around 33% and 28% by 2010. Additionally, the problem is penetrating into childhood, rising from 1.8% to 6.0% in boys and 1.3–6.6% in girls aged 5–10 years between 1974 and 2002/03 (2) and estimated to reach around 19% and 22% in boys and girls respectively by 2010. This, together with a greater recognition of the burden of disease and the health costs associated with obesity (especially abdominal obesity) including increased risk of coronary heart disease, type 2 diabetes and cancer (3), has taken the problem outside the traditional health/research sector and into front-line politics as a debate over roles and responsibilities rages (4). This overview draws on 34 ‘State of the Science’ review papers commissioned as part of the UK Foresight project ‘Tackling Obesities: Future Choices’. It explores the many facets of the issue from genetics to social norms and ethics from the perspectives of national experts, with resonances on an international scale. The conclusion is inescapable. Obesity is a complex, multifaceted problem with no easy or obvious solutions. Subsequent steps in the Foresight process will seek to provide clarity to this complexity through the use of an obesity systems map and will explore the environment for action through the development of a range of future scenarios. These two strands of qualitative work will inform estimates of possible future obesity levels using a newly developed quantitative cell-based model. At the heart of the problem of excess weight is a homeostatic biological system, struggling to cope in a fast-changing world, where the pace of technological revolution outstrips human evolution. Since the identification of the ob gene in 1994, research into physiological mechanisms controlling body weight has grown exponentially (5–9) including work on inter-generational effects and impact of early-life experiences that may perpetuate obesity within families (10,11). Much has been learnt about regulatory mechanisms from animal models, with the potential to perform detailed investigations, including the use of transgenic models (7). This work continues to provide critical underpinning for the development of novel drug targets for the treatment (and perhaps in the future, prevention) of obesity and/or the modulation of related metabolic disorders (5,6,8). Intriguingly, most of the significant advances in understanding have occurred in relation to the control of food intake. Indeed, research into the metabolic aspects of energy expenditure has yielded little evidence of abnormalities (12), although this does not rule out future therapeutic interventions. But in contrast, studies of the regulation of food intake have spawned new components to the human appetite regulation system at an impressive rate. This includes work on single gene defects associated with severe obesity that have highlighted critical neural pathways, especially in relation to leptin and the melanocortin system (5). Other research has shown the importance of peripheral signals derived from adipose tissue and the gut (6). New research is considering the interactions between food and biology through studies of the neural response to food-related stimuli (9). This fusion of psychology and biology provides important insights into mechanisms underpinning food habits. However, individual strands of this basic biological research are often conducted in isolation and this limits a fuller understanding of the relative importance of different pathways, their interactions and redundancies within the system. Thus, it is still difficult to reliably predict the key points for intervention. This is further compounded by the inter-individual variability in each of these components and research into gene–environment interactions is in its infancy (5). However, this is a topical area of research, in the public and private sector, not least with respect to developments in nutri-genomics, the science underpinning the concept of personalized nutrition (13). In spite of this mechanistic complexity, obesity is ultimately a consequence of a long-term uncoupling of energy intake and expenditure. Observational studies of human behaviour have sought to elucidate the detailed components of this dysregulation but are hampered by the lack of robust, objective measures of dietary intake and physical activity in large populations (14,15). There has been disproportionate attention given to debates over the relative importance of diet or inactivity in the aetiology of obesity, given that the subtle shifts in energy balance which have occurred at a population level are below the limits of detection of current methodologies. In the case of dietary intake, long-term records are based on household food purchases while historical data on physical activity are totally lacking. However, recent technological developments in the measurement of activity levels have given rise to a raft of new studies to consider the variation within a population with respect to physical activity and subsequent risk of obesity, and this is yielding important insights. For example, showing that sedentary behaviours, especially television viewing, are critical risk factors for obesity and the importance of energy expended during routine daily activities as a contributor to overall energy expenditure, which has clear implications for transport policies and other aspects of the built environment (15,16). In modern societies, overt exercise accounts for a very small proportion of total energy expenditure and is likely to play only a minor role in preventing obesity, although positive benefits on wider disease risk should not be ignored (12). Measuring dietary intake outside the laboratory remains problematical, but by combining data from experimental, observational and controlled intervention studies, a number of specific dietary risk factors have been identified including a high energy density, characterized by a diet high in fat and low in fibre and the consumption of sugar-rich drinks, each compounded by large habitual portion sizes. These provide promising targets for intervention and are consistent with other strategies for the prevention of chronic disease (14). This offers potential opportunities for interventions by the food industry through reformulation of existing products and innovation with healthier options, tailored to meet the nutritional needs of a largely sedentary population (17). Research into the physiological systems controlling body weight is vital but it is increasingly recognized that biology operates within a social and cultural context (18–20). The choice of lifestyle for the population at large, or individuals, is neither a pure product of genetics nor freewill, but a melting pot, heated and stirred by the influence of the wider environment. Understanding these wider determinants of food intake and physical activity is a key area of research, spanning across areas as diverse as family dynamics, school policies, urban design and media influences (16,18,21,22). Most of this work is presently conducted at a population level and more investigation at an individual level is warranted to explain the inter-individual variability in obesity rather than overarching population trends (18). A focus on socioeconomic and ethnic disparities will help to bridge the widening health inequalities (23). However, population-based interventions must be sensitive to the need to promote a healthy body weight, without excessive thinness, especially among individuals predisposed to eating disorders (24). At present, the evidence base for the prevention of obesity is heavily biased towards causes rather than solutions. Data from intervention studies are few in number and limited in scope (15,25). Most are confined to controlled research settings. Few interventions have been overtly successful in reducing the prevalence of obesity and even the most promising have not been widely replicated (25). Areas identified for future development include multidisciplinary approaches (25), overcoming ambivalence to change (19), establishing new social norms, and, critically in the case of children, parental engagement (18). This is likely to require greater involvement of research partners outside traditional academic/health professional arenas. The need for large-scale studies involving diverse teams over long periods of time also challenges existing funding mechanisms and data-gathering processes (4,26). In contrast to interventions for the prevention of obesity, there have been some recent advances in treatment development. Treatment is hampered by the physiological mechanisms which favours weight maintenance. However, long-term studies of intense individual interventions show that sustained changes in diet and physical activity can lead to modest but sustained reductions in body weight, of approximately 5%, over many years, with concomitant improvements in comorbidities, especially the risk of type 2 diabetes and overall reductions in the cost of health care (27). Pharmacological interventions produce significantly greater weight loss while the treatment is continued (28). Advances in bariatric surgery means that this is now a well-recognized and effective intervention for obesity, with weight losses of 10–30% and substantial reductions in new cases of diabetes and cardiovascular mortality. However, although this is a large weight reduction for an individual, the impact on the population mean weight is small, because the number of individuals for whom surgery is appropriate is small. At present, only a small proportion of people who are obese receive optimal care because of a lack of resources (including trained staff) and clinical management practices which prioritize the treatment of comorbidities over weight loss. There is a need to identify individuals most likely to benefit from treatment. The BMI is a useful measure for population surveillance but has limited sensitivity at an individual level. Instead, it needs to be combined with information on body fat, other risk factors including family history, to make improved risk assessments. Future treatment-based research needs to focus on improving the success of behavioural interventions and developing new, more effective drug therapies. While there are established public–private mechanisms for investment in the latter, behavioural interventions have, historically, been poorly supported. This will not be reversed quickly given the paucity of trained researchers in this area. Importantly, treatment should not be divorced from prevention. The maintenance of weight loss and the prevention of weight regain is a critical yet under-researched component of treatment (27). In the short term, significant progress could be made by the implementation of existing knowledge. Epidemiological analyses of the prevalence of obesity have provided a catalyst for research. Some countries have the unenviable reputation of an even greater proportion of obese people than the UK. Indeed, the prevalence of obesity in the UK has been steadily tracking that of the USA with a lag of about 10 years. Obesity levels in the USA now exceed 30%. This gives a glimpse into the future if we fail to act now. Ongoing analysis is essential for surveillance and monitoring and there is scope to refine these procedures to enhance the usefulness to researchers and policymakers through rigorous evaluation (26). Further analysis of these datasets can also inform future projections to underpin the provision of public services, including health and pensions and assess the economic implications (29). It will also assist in the development of an overarching strategy to tackle obesity (4). Our understanding of the risks of chronic disease associated with obesity suggests that even 100% success in preventing new cases of obesity among young children will leave a substantial burden of ill-health and associated costs for years to come (3). These evidence reviews suggest that long-term strategies (spanning several generations) encompassing both prevention and treatment for adults and children and which also take account of the needs of groups at different lifestages and with diverse circumstances would be most likely to be effective at tackling obesity. However, just as obesity develops slowly, within both individuals and populations, so too will it take time to establish new habits and build new structures to support a healthy diet and enhanced physical activity. Interventions will only be effective if they are designed to have in-built sustainability. One of the important lessons from efforts to reduce smoking and alcohol is the lag time between research and action, whether among individuals, in terms of behaviour change or at a population level in terms of public health interventions. Controls on smoking provide an interesting case study with respect to the acceptability of different types of interventions (30,31). Over the last 50 years, policymakers have moved from the basic provision of information and advice, through the facilitation of healthier options (e.g. through use of nicotine replacement), active discouragement of the unhealthy behaviour (e.g. taxation, advertising restrictions) and onto regulatory action (e.g. bans on smoking in public places). Action on diet (including alcohol) and activity is far behind, social marketing techniques will contribute to the raising of awareness, the influencing of public opinion and resetting of social norms (32). Campaigns in relation to alcohol have successfully reset social norms with respect to drinking and driving. However, this has not been accompanied by a decrease in overall alcohol intake (31). Here, as with other diet and alcohol interventions, most observers accept the need for such campaigns to coexist with parallel environmental interventions to support and facilitate behaviour change. However, it is important not to be over-reliant on a single approach (33). These ‘State of Science’ reviews have clearly demonstrated that evidence for specific actions to tackle obesity is neither complete nor perfect and the obesity debate is fuelled by differing interpretations of the scientific literature. Meanwhile, the burgeoning prevalence of obesity, especially in children, has resulted in sustained pressure to act and to act quickly. The resulting tensions illustrate just one of the ethical challenges which arises in any strategy to tackle obesity (34). Neither is it an easy position for scientists who seek the ‘best evidence possible’, nor for policymakers who fear interventions which do not guarantee success. Greater interactions between scientists and policymakers will help to ease these tensions and maximize the potential to develop and execute effective interventions in a virtuous circle of continuous improvement which combines scientific development, policy implementation and joint evaluation.

OBJECTIVE: To understand lay beliefs and attitudes, religious teachings, and professional perceptions in relation to diabetes prevention in the Bangladeshi community. DESIGN: Qualitative study (focus groups and semistructured interviews). SETTING: Tower Hamlets, a socioeconomically deprived London borough, United Kingdom. PARTICIPANTS: Bangladeshi people without diabetes (phase 1), religious leaders and Islamic scholars (phase 2), and health professionals (phase 3). METHODS: 17 focus groups were run using purposive sampling in three sequential phases. Thematic analysis was used iteratively to achieve progressive focusing and to develop theory. To explore tensions in preliminary data fictional vignettes were created, which were discussed by participants in subsequent phases. The PEN-3 multilevel theoretical framework was used to inform data analysis and synthesis. RESULTS: Most lay participants accepted the concept of diabetes prevention and were more knowledgeable than expected. Practical and structural barriers to a healthy lifestyle were commonly reported. There was a strong desire to comply with cultural norms, particularly those relating to modesty. Religious leaders provided considerable support from Islamic teachings for messages about diabetes prevention. Some clinicians incorrectly perceived Bangladeshis to be poorly informed and fatalistic, although they also expressed concerns about their own limited cultural understanding. CONCLUSION: Contrary to the views of health professionals and earlier research, poor knowledge was not the main barrier to healthy lifestyle choices. The norms and expectations of Islam offer many opportunities for supporting diabetes prevention. Interventions designed for the white population, however, need adaptation before they will be meaningful to many Bangladeshis. Religion may have an important part to play in supporting health promotion in this community. The potential for collaborative working between health educators and religious leaders should be explored further and the low cultural understanding of health professionals addressed.