Dr Matthew Capehorn reviews currently available advice on obesity assessment and management, and gives a rationale for drug use

Since publication of comprehensive obesity guidelines from NICE in 2006,1 there have been few new guidelines or updates that reflect recent changes in practice and further knowledge on how best to assess and manage the obesity epidemic. Is it time for some new clinical guidelines?

Definitions of obesity

Until relatively recently, healthcare professionals still spoke in terms of ‘ideal weight’, but given variations in body composition this was always nonsense. The definition of obesity was standardised by the World Health Organization (WHO) with the well established classification of obesity based on body mass index (BMI),2 which is a measure of weight in kilograms/height in metres2. Unfortunately BMI measurement alone can be flawed; for example, in particularly muscular people (where muscle weighs more than fat). An athletic patient can have a BMI that suggests he or she is obese according to this classification, when in fact he or she may be particularly healthy, with no dangerous central/visceral fat, and may not be at significant risk of cardiometabolic conditions.

More recently, with increasing knowledge of the fat cell (previously thought of as just a store for excess glycogen, but now established as highly metabolically active in its own right), it has become possible to hypothesise on the role of the adipocytokines it releases and links between the fat cell and development of insulin resistance, inflammatory markers, dyslipidaemia, and so on. This supports the International Diabetes Federation criteria for metabolic syndrome.3

Evidence shows a strong correlation between a simple waist circumference and the dangerous central or visceral fat that sits in and around central organs (as distinct from subcutaneous fat) and increases cardiometabolic risk.4 It was not surprising that criteria for waist circumference measurements were then developed to equate with being overweight (at risk, >80 cm in women, >94 cm in men) and obese (significant risk, >88 cm in women, >102 cm in men), and these were included in the 2006 NICE guideline.1

Assessing risk of obesity

Assessing risk based on these definitions is flawed in certain circumstances. To measure waist circumference, the waist should be parallel to the floor at a height that is halfway between the lowest rib and the hip bone on exhalation. Unfortunately, there appears to be no agreement on how to measure waist circumference when the waistline has dropped and is overhanging. This overhang can be noted in patients with a BMI >35 kg/m2, which makes waist circumference recordings inconsistent in these patients. There is also debate over whether the cut off for a waist circumference representing ‘at risk’ or ‘at significant risk’ should be the same for someone who is 137 cm (4 feet 6 inches) tall, compared with someone who is 183 cm (6 feet) tall and of the same gender.

However, evidence suggests that waist circumference is a better indicator of likely cardiovascular risk than BMI,5 although it is still not ideal. Ideally we would be able to assess this visceral fat, but access to CT/DEXA scanners is limited. Biotechnology is improving, however, and there may soon be affordable tools for measuring visceral fat. In the meantime there are affordable scales that work by bio-impedence, estimating fat composition, as distinct from bone, muscle, or water, according to variations in how a mild electrical current passes through them, to give a rough guide as to whether someone has a healthy level of fat, is ‘over-fat’, or obese. However, these devices are, arguably, not entirely accurate, and results need to be interpreted in the context of body water composition at that given time, which is often ignored and can lead to misinterpretation. This is particularly noticeable in young women who retain a lot of premenstrual fluid, and is another argument against weighing too often—weight checks every 4 weeks are more than sufficient.

Definition and assessment of obesity in children

One recent change has come in how we define overweight/obesity in children. Charts are available that clearly mark the 91st/98th BMI centiles used by paediatricians, and the darkened lines that represent the equivalent International Obesity Task Force criteria (see www.healthforallchildren.co.uk). With the advent of the National Child Measurement Programme, the DH has introduced new thresholds at the 85th/95th BMI centiles, and in August 2008 announced that PCTs will be encouraged to report results to parents. Healthcare professionals will need to get used to this new criteria, and, when appropriate, waist circumference and fat composition should be checked.

Managing obesity

Over recent years, little has changed in how we manage obesity. A good history and examination is essential as always, and this should include measurement of:

  • weight
  • height
  • BMI
  • waist circumference
  • fat composition
  • blood pressure.

In adults, and appropriate children (such as obese adolescents, or even children as young as 8 years old who are considered at risk of cardiometabolic disease due to morbid obesity as defined by growth charts), blood testing should be performed to include, at the very least, thyroid and liver function tests, fasting blood sugar (including glycosylated haemaglobin [HbA1c] if the patient is known to have diabetes and this has not recently been tested), and lipids. Further blood tests may be indicated depending on the individual and other co-morbidities, and should be considered as a screen for all cardiometabolic risk.

Management should still start with looking for obvious problems, such as:

  • a lack of understanding of calories
  • the effect of a ‘night off the diet’
  • effects of alcohol
  • high calorie/low fat foods.

Lifestyle advice should include:

  • using portion control at meal times
  • changing snacks to less calorie-dense options
  • encouraging healthy balanced eating
  • the importance of regular rather than occasional exercise and its effect on energy requirements.

Many areas of the UK have obesity strategies that offer nutrition and lifestyle interventions as their first tier approach, followed by referral to specialist multidisciplinary team (MDT) weight-management centres. There, patients will receive more intensive input from psychological support, dieticians, physical health trainers, specialist nurses, GPwSIs, and, when appropriate, input from bariatric surgeons. Advice from the MDT will include prescription of pharmacotherapy as appropriate, and introduction of more radical treatments such as very low calorie diets (for short periods only under the guidance of specialist input), and surgical interventions, when previous interventions have failed. However, whenever possible, steady sustainable weight loss of less than 3 kg (7 lb) per month is more than adequate; this, in my experience, helps to prevent the problem of excess skin associated with dramatic fast weight loss, which leads to a need for subsequent apronectomy surgery (tummy tuck), a procedure that is not funded by every PCT.

Future management strategies

Possibly one big change in future management of obesity, assuming sufficient funding, will be the offer of psychological support and ‘talking therapies’, such as cognitive behavioural therapy, hypnotherapy, and neurolinguistic programming. Although the evidence for the successful use of these techniques in weight management is not yet as convincing as that for other interventions, and good quality randomised control trials are needed to assess their true role, there is no doubt that some patients prefer these approaches, and we should support anything that works.


Critics of pharmacotherapy argue that weight loss gained by this method is not worth the expense; however this is short sighted. It does not reflect the weight that the individual would have reached without intervention, pharmacological or otherwise. As morbidity and mortality appear to be directly related to weight at any given time, if a patient is at a lower weight than they would have been without the intervention, then you have succeeded in reducing their risk.

It has also been argued that weight loss with pharmacology eventually plateaus and slowly reverses. Unfortunately, this effect is seen with any intervention, whether pharmacological or conventional therapy.6 Obesity should be considered as a chronic relapsing condition, and if an individual has a predisposition to weight gain, they may always have it. After weight reduction with an intervention plateaus, if sufficient lifestyle changes have not been made, further monitoring and interventions may need to be considered.

Current licensed medications for weight loss include orlistat,1 sibutramine,1 and rimonabant.7 There is little evidence to guide selection between these drugs, but clearly a suitable rationale would be to follow the simple rules outlined in Table 1.

In my experience, appetite suppressants will probably only work in about one in three patients, and the skill is to identify appropriate patients for particular medicines. Specific patient risk factors may indicate a particular medication should be used, although all these medicines should reduce metabolic syndrome criteria with weight loss.8 The improvement in HbA1c with rimonabant over and above what would be expected by weight loss alone,9 may be an influencing factor, for example, and also raises the question as to whether we should be considering using anti-obesity drugs as second-line or third-line anti-diabetic medicines.

Table 1: Rationale for pharmacological treatment selection in patients with obesity

FactorAdvice/further considerations
Possible contraindications to pharmacotherapy

Avoid medicines known to have contraindications, for example:

  • orlistat—chronic malabsorption syndrome, cholestasis, breast feeding
  • sibutramine—uncontrolled psychiatric illness, certain cardiovascular conditions including uncontrolled hypertension, cerebrovascular disease, hyperthyroidism, pregnancy, breast feeding
  • rimonabant—severe renal impairment, breast feeding, or any previous history of psychiatric illness
Patient with high fat intake

Orlistat—this pancreatic lipase inhibitor may produce good results. Patients will need careful counselling to avoid unwanted side-effects, which appear proportional to fat intake. Patients may lose weight anyway if fat intake reduces. This may question the need for pharmacotherapy, or at least question why a patient could be unsuccessful in losing weight on this medicine

Sibutramine or rimonabant—these centrally acting drugs may still be considered, but not in dual therapy

Patient has willpower to reduce food intake but struggles with hunger cravings

Sibutramine or rimonabant—these centrally acting drugs may be unsuccessful if patient can ignore/override natural signals of ‘feeling full’. A drug trial will still be useful, but reassessment of basic principles of weight loss and talking therapies may be more appropriate. If food intake can be reduced but patient struggles with hunger, a centrally acting drug may work well

Orlistat—may be considered

Patient with specific

Any weight loss will reduce risk from metabolic syndrome factors8

Rimonabant—studies suggest a reduction in HbA1c greater than that expected from weight loss alone,9 which may be of additional benefit to a patient with uncontrolled overweight or obese type 2 diabetes

Choosing a medicine acceptable to patient Many patients are knowledgeable about the anti-obesity medicines available. They should be informed of available safe options, counselled about possible side-effects, and involved in deciding which pharmacotherapy they wish to try, if any. Compliance and patient control is a key to success
Effectiveness of current medicine

Orlistat—if patient fails to lose weight, this suggests calorie intake exceeding energy requirements, and consultation should focus on reassessment of basic principles, talking therapies, or consideration of one of the centrally acting medicines

If a patient on sibutramine or rimonabant fails to lose weight, reassess the basics, alternative therapies, or consider prescribing orlistat

HbA1c=glycosylated haemaglobin

NICE guidance

The recent Technology Appraisal 144 from NICE on the newest of the anti-obesity drugs—the selective cannabinoid receptor antagonist rimonabant—is significant for the number of changes it incorporates.7 This appraisal appears to have adopted the more pragmatic and realistic target of 5% weight loss at 6 months. Treatment with rimonabant may continue beyond 6 months provided at least 5% of the starting weight has been lost. However, rimonabant use should cease if the patient returns to his or her starting weight. There is overwhelming evidence that anti-obesity medicines help to prevent weight regain.10-12

Previous NICE guidance and BNF licensed indications concentrate on 5% weight loss at 3 months. Most practitioners have aimed for 10 kg weight loss at 6 months but this has often been unachievable. The NICE clinical guideline recommends a weight reduction goal of 5–10% in 6–9 months,1 and the more recent technology appraisal7 emphasises the change in accepting 5% at 6 months as being a more realistic achievement in practice. Despite the sometimes overambitious personal goals of the patient, the aim for the healthcare professional remains at helping him or her to achieve 10 kg weight loss for its well established health benefits (see Table 2).8 In practice, however, this can be difficult to achieve and, if weight is regained, GPs are advised to consider stopping pharmacotherapy, despite any significant previous success.13

A more pragmatic approach was always needed and I feel the new NICE guidance on rimonabant shows us the more realistic way to go.7 This approach could easily be adapted to apply to orlistat and sibutramine but presumably would require a change in their product license to fit this model.

Table 2: Benefits of a 10 kg weight reduction8

Mortality 20–25% reduction in total mortality
30–40% reduction in diabetes-related deaths
40–50% reduction in cancer-related deaths

Reduced risk of developing diabetes by >50%
30–50% reduction in fasting glucose
15% reduction in HbA1c


10% reduction in total cholesterol
15% reduction in LDL
30% reduction in triglycerides
8% increase in HDL

Blood pressure

10 mmHg reduction in systolic BP
20 mmHg reduction in diastolic BP

HbA1c=glycosylated haemaglobin; LDL=low density lipoprotein; HDL=high density lipoprotein; BP=blood pressure
Adapted from Jung R. Obesity as a disease. Br Med Bull 1997; 53 (2): 307–321, with permission

Role of surgery

When pharmacology fails to achieve weight loss, many areas of the UK are still not meeting the current NICE guidance on when bariatric surgery should be considered and used. All the following criteria should be fulfilled before surgery is considered:1

  • BMI ?40 kg/m2 or 35–40 kg/m2 with co-morbidities such as type 2 diabetes or high blood pressure
  • appropriate non-surgical treatments have not achieved or maintained clinically beneficial weight loss for 6 months
  • the patient is being treated in a specialist obesity service and is sufficiently fit to undergo surgery and anaesthesia.

Bariatric surgery should be considered as first-line treatment if the patient has a BMI >50 kg/m2 and surgery is appropriate.1 However, some PCTs will not fund bariatric surgery at all, despite this being a cost-effective option,1 especially when it may prevent and often affect remission of type 2 diabetes in obese patients.14

If we cannot convince those that hold the purse strings that obesity surgery is cost-effective, then how do we ration it? At present, we have anecdotal situations in some PCTs where patients are eating more in order to meet their local criteria for surgery. There are, arguably, grounds for rationing surgery for those patients with an excessively high BMI, perhaps BMI >60 kg/m2, to enable more people with BMIs of 35–50 kg/m2 to receive treatment. This might also encourage some lifestyle changes and weight loss in the very morbidly obese, prior to surgery, and inevitably increase their chances of survival and sustainable change post-operatively. These lifestyle changes could be initiated in specialist centres, and, perhaps, be combined with the short-term use of endoscopic gastric balloons, which although only licensed for 6 months, can produce results in patients prior to surgery for gastric band or bypass.15

Need for updated guidelines

Clearly there is a need for a fresh look at how we manage our increasingly obese nation. The National Obesity Forum (NOF) is currently reviewing available clinical guidelines for the management of obesity and now has Regional Obesity Groups for networking and sharing practice in most areas of the country. More details can be found on the NOF website (nationalobesityforum.org.uk). Any new clinical guideline should encourage the use of all available assessment measures/tools in order to gain a more reliable result. Any single measurement should merely be used as an indicator for further investigation. If all these criteria suggest overweight/obesity then we clearly have a duty of care to this patient to provide appropriate advice and management.

There will always be those who suggest that we should not ‘medicalise’ obesity, as it is a social problem, which it is, but healthcare professionals treat social problems every day. At present GP practices collect QOF points for treating the consequences of obesity, rather than treating the cause.

We need to tackle the obesity epidemic by provision of sufficient resources, and by sharing best practice, in order to develop local, regional, and national strategies that can reverse current trends.


  1. National Institute for Health and Care Excellence. Obesity: guidance on the prevention, identification, assessment and management of overweight and obesity in adults and children. Clinical Guideline 43. London: NICE, 2006.
  2. World Health Organisation. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Teach Rep Ser 2000; 894: i–xii, 1–253.
  3. International Diabetes Federation. A new worldwide definition of the metabolic syndrome. International Diabetes Federation Conference Berlin 14 April 2005. www.idf.org/home/index.cfm?unode=32EF2063–B966–468F–928C–A5682A4E3910
  4. Despres J-P, Lemieux I, Prud’homme D. Treatment of obesity: need to focus on high risk abdominally obese patients. BMJ 2001; 322: 716–720.
  5. Nanchahal K, Morris J, Sullivan L, Wilson P. Coronary heart disease risk in men and the epidemic of overweight and obesity. Int J Obes (Lond) 2005; 29 (3); 317–323.
  6. Wadden T. Treatment of obesity by moderate and severe caloric restriction. Results of clinical research trials. Ann Intern Med 1993; 119 (7, pt 2): 688–693.
  7. National Institute for Health and Care Excellence. Rimonabant for the treatment of overweight and obese patients. Technology Appraisal 144. London: NICE, 2008.
  8. Jung R. Obesity as a disease. Br Med Bull 1997; 53 (2): 307–321.
  9. Scheen A, Finer N, Hollander P et al and the RIO–Diabetes Study Group. Efficacy and tolerability of rimonabant in overweight or obese patients with type 2 diabetes: a randomised controlled study. Lancet 2006: 368 (9548): 1660–1672.
  10. Pi–Sunyer F, Aronne L, Heshmati H et al and RIO–North America Study Group. Effect of rimonabant, a cannabinoid-1 receptor blocker, on weight and cardiometabolic risk factors in overweight or obese patients: RIO–North America: a randomized controlled trial. JAMA 2006; 295 (7): 761–775.
  11. James W, Astrup A, Finer N et al. Effect of sibutramine on weight maintenance after weight loss: a randomised trial. STORM Study Group. Sibutramine Trial of Obesity Reduction and Maintenance. Lancet 2000: 35 (9248); 2119–2125.
  12. Torgerson J, Hauptman J, Boldrin M, Sjostrom L. Xenical in the prevention of diabetes in obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients. Diabetes Care 2004: 27 (1); 155–161.
  13. British National Formulary. BNF 55. London: Royal Pharmaceutical Society, 2008.
  14. Dixon J, O’Brien P, Playfair J et al. Adjustable gastric banding and conventional therapy for type 2 diabetes: a randomized controlled trial. JAMA 2008; 299 (3): 316–323.
  15. Fernandes M, Atallah A, Soares B et al. Intragastric balloon for obesity. Cochrane Database Syst Rev 2007: Jan 24, issue 1. CD004931.G