Importance of retaining a national dietary guideline for sugar

A Working Group convened by the National Health and Medical Research Council's Health Advisory Committee is currently reviewing the Australian Dietary Guidelines, and reconsideration of the New Zealand dietary guidelines is also likely. Because of insufficient scientific evidence to link sugar consumption with ill health, a case has been made to abandon the guideline referring to sugar, "Eat only a moderate amount of sugar".¹

Discussion of this issue is timely as sucrose consumption appears to be increasing. For example, New Zealand national nutrition surveys suggest that, between 1988 and 1997, self-reported sucrose intake increased on average from 42 to 62 g/person per day in males and from 27 to 45 g/person per day in females.² Even more striking increases were evident in the age group 15-24 years.

Dietary studies: Much of the evidence relating diet and disease comes from epidemiological studies, both cross-sectional and prospective. The nutritional and epidemiological methods currently available are insufficiently robust to accurately assess intakes and disentangle the effects of interrelated nutrients. Prospective studies have advantages over cross-sectional surveys, but there are still problems with assessment of dietary intake (which is usually only assessed on one occasion and by methods which are seriously flawed). Much of the evidence on which diet-disease relationships is based is self-reported, and there is substantial under-reporting of foods perceived to be unhealthy (eg, those rich in sugar, fat and alcohol). In particular, people who are obese tend to under-report, and do so selectively.³

Even randomised controlled clinical trials can be difficult to interpret, an excellent example being the CARMEN study.⁴ In this multicentre trial, 398 moderately obese adults were allocated at random to a seasonal control group (no intervention), a control diet group (dietary intervention typical of the average national intake), or one of two low-fat, high-carbohydrate groups (in which the carbohydrate was derived primarily from "simple" or "complex" carbohydrate). The weight changes on the latter two diets were not significantly different statistically, and the authors therefore concluded that the nature of carbohydrate is a relatively unimportant determinant of body weight. However, closer examination of the data reveals interesting trends: body weight loss on the low fat/high simple carbohydrate diet was 0.9 kg (P < 0.05) and on the high complex carbohydrate diet 1.8 kg (P < 0.001). A similar trend was apparent for change in fat mass. With a larger sample size, these differences might have become significant.

Obesity and diabetes: The prevalence of obesity has risen sharply in Australia and New Zealand over the past 20 years. The recently released Diabesity and associated disorders in Australia 2000⁵ draws attention to some of the devastating consequences of overweight and obesity. The association between type 2 diabetes and adiposity is arguably of even greater importance than the previously well described association with cardiovascular risk factors (hypertension and dyslipidaemia). Diabetes prevalence in the Australian adult population has doubled since 1980, the increase occurring in parallel with the rising prevalence of obesity. The estimated number of Australians with diabetes has reached almost one million, of whom less than 500 000 are aware of the diagnosis. The greatly increased risk of coronary heart disease and other vascular diseases, as well as the microvascular complications of diabetes (eg, retinopathy and nephropathy), emphasise the consequences of this disease, which has now reached epidemic proportions.

This serious public health issue brings obesity to centre stage. It is caused by a complex interaction between genetic predisposition and environmental trigger factors, but the current epidemic of obesity has to be attributed to the "obesogenic environment"⁶ — the human genome has not changed in this period! The environmental facilitators of obesity are food intake and energy expenditure. While acknowledging the undoubted contribution of our increasingly sedentary lifestyle to the obesity epidemic, our discussion will focus on diet.

Diet: Much of the discourse on the role of diet in the development of obesity highlights the role of fat, due to its high energy density (kJ/g of food or beverage) and its propensity, if consumed in excess, to be deposited as adipose tissue. However, there is now evidence that fat gain is similar with overfeeding of carbohydrate or fat.⁷ Energy intake is strongly influenced by energy density. Covert manipulation of energy density (ie, providing diets of different energy density without participants' knowledge) results in sustained changes in energy intake.⁸ The increased use of low-fat products, many of which are energy dense due to their high sugar content, and sugar-containing beverages now contributes significantly to total energy intake and are examples of the means by which sugar may enhance the energy density of the diet. If sugar does contribute to excessive energy consumption and to the problems of overweight and obesity, then clearly retention of a guideline is important.

Evidence is accumulating that the form in which the sucrose is consumed is also important. A recent prospective study from Ludwig and colleagues⁹ showed a clear-cut, graded relation between the consumption of sugar-sweetened drinks and the development of obesity in children. The prevalence of obesity among children in the United States doubled between 1980 and 1994; 11% are now above the 95th reference percentile of body mass index (BMI) for age and sex. The observation that this increase paralleled the increase in sugar-sweetened soft drinks prompted Ludwig et al to enrol 548 ethnically diverse schoolchildren in four Massachusetts communities in a prospective study for 19 months. The difference in measures of obesity was related to change in consumption of sugar-sweetened drinks and other possible determinants of obesity, including physical inactivity and fat intake. For each additional serving of sugar-sweetened drinks both BMI (mean, 0.24 kg/m²) and frequency of obesity (odds ratio, 1.6) increased, after adjustment for anthropometric, demographic, dietary and lifestyle variables. Changes in diet soft drink intake were not related to obesity incidence.

Of course, an observational study does not prove causality, but it is of interest that another recently published study in an entirely different group of older individuals produced similar results. Elmslie and coworkers¹⁰ compared a group of bipolar (manic depressive) patients and matched controls; the patient group had higher rates of overweight and obesity than the controls. The bipolar patients reported a higher energy intake, the increased energy being derived almost entirely from sucrose in sweetened drinks.

Energy from drinks (regardless of whether it is from sugar, fat or alcohol) adds to total energy intake, and does not displace energy from other forms.¹¹ Furthermore, compensation at subsequent meals for energy consumed in the form of liquid (drinks) appears to be less complete than for energy consumed in solid form (food) (ie, people overconsume more easily when excess energy is in the form of energy-containing beverages).¹²

While these new data suggest an obesity-promoting effect of sugar-containing beverages, it may also be relevant to recall data published some 30 years ago.¹³ Middle-aged men were asked to replace, as far as possible, sucrose with foods rich in starch to maintain energy balance. Despite regular advice and encouragement from a dietitian they were unable to maintain energy balance and lost weight, presumably because of the greater satiety-promoting qualities of the starchy foods. While not providing direct evidence for sugar as an aetiological factor, these observations do suggest that recommending a reduction in sugar may be a potentially useful public health measure in countries where obesity and its comorbidities have reached epidemic proportions.

Implications for Indigenous populations: Our discussion has particular significance for Australian and New Zealand Indigenous populations, who have very high rates of lifestyle-related chronic diseases occurring at much younger ages than the non-Indigenous population.^14,15 Indigenous people often have poor-quality diets, high in sugar and fat, and depleted in fruit and vegetables. For example, Lee and coworkers¹⁶ analysed the food supply at six remote Aboriginal communities in the Northern Territory using the "store turnover" method, and found a very high consumption of sugar per se and in soft drinks. The diets had high levels of animal fat (mainly from poor-quality meat) and very low levels of fruit and vegetables (ie, energy-dense and nutrient-poor). Most of the sucrose was consumed in liquid form as sugar in tea and in carbonated beverages.

Retention of a dietary guideline for sugar: The evidence we have assembled here leads us to strongly advocate the retention of a dietary guideline for sugar in Australia and New Zealand. In fact, given the marked rise in consumption of carbonated beverages in Australia over the past 30 years (47.3-114.4 L/person per year between 1968-69 and 1996-97), we also advocate an addition: "Consume only moderate amounts of sugars and foods and beverages containing added sugar". This is in line with the recently revised US dietary guidelines,¹⁷ which include the recommendation "Choose beverages and foods that limit your intake of sugars". There is also a strong population health rationale for mandatory labelling of foods with clear information, including sugar, fat, and total energy content.

References

1. Williams P. Sugar: is there a need for a dietary guideline in Australia? Aust J Nutr Diet 2001; 58: 26-31.
2. New Zealand food: New Zealand people. Key results of the 1997 National Nutrition Survey. LINZ Activity and Health Unit, University of Otago for the Ministry of Health. Dunedin, NZ: University of Otago, 1999.
3. Heitmann BL, Lissner L. Dietary underreporting by obese individuals - is it specific or non-specific? BMJ 1995; 311: 986-989.
4. Saris WHM, Astrup A, Prentice AM, et al. Randomized controlled trial of changes in dietary carbohydrate/fat ratio and simple vs complex carbohydrates on body weight and blood lipids: the CARMEN study. Int J Obesity 2000; 24: 1310-1318.
5. Dunstan D, Zimmet P, Welborn T, et al, on behalf of the AusDiab Steering Group. Diabesity and associated disorders in Australia: the accelerating epidemic. Report of the Australian Diabetes, Obesity and Lifestyle Study. Melbourne: International Diabetes Institute, 2001.
6. Egger G, Swinburn B. An "ecological" approach to the obesity pandemic. BMJ 1997; 315: 477-480.
7. McDevitt RM, Poppitt SD, Murgatroyd PR, Prentice AM. Macronutrient disposal during controlled overfeeding with glucose, fructose, sucrose or fat in lean and obese women. Am J Clin Nutr 2000; 72: 369-377.
8. Stubbs RJ, Johnstone AM, O'Reilly LM, et al. The effect of covertly manipulating the energy density of mixed diets on ad libitum food intake in "pseudo free-living" humans. Int J Obesity 1998; 22: 980-987.
9. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective observational analysis. Lancet 2001; 357: 505-508.
10. Elsmlie J, Mann JI, Silverstone JT, et al. Determinants of overweight and obesity in patients with bipolar disorder. J Clin Psychiatry 2001; 62. In press.
11. Poppitt SD, Prentice AM. Energy density and its role in the control of food intake: evidence from metabolic and community studies. Appetite 1996; 26: 153-174.
12. Mattes RD. Dietary compensation by humans for supplemental energy provided as ethanol or carbohydrates in fluids. Physiol Behav 1996; 59: 179-187.
13. Mann JI, Truswell AS, Hendricks D, Manning EB. Effects on serum lipids in normal men of reducing dietary sucrose or starch for five months. Lancet 1970; 1: 870-872.
14. Daniel M, Rowley KG, McDermott R, et al. Diabetes incidence in an Australian Aboriginal population: eight year follow up study. Diabetes Care 1999; 22: 1993-1998.
15. Simmons D, Harry T, Gatland B. Prevalence of known diabetes in different ethnic groups in inner urban South Auckland. N Z Med J 1999; 112: 316-319.
16. Lee AJ, O'Dea K, Mathews JD. Apparent dietary intake in remote Aboriginal communities. Aust J Public Health 1994; 18: 190-197.
17. Nutrition and your health: dietary guidelines for Americans. 5th edition, 2000. Home and Garden Bulletin No 232. Washington, DC: Department of Agriculture, Department of Health and Human Services; 2000.

Authors' details

University of Otago, Dunedin, New Zealand.
J I Mann, PhD, DM, FRACP, Professor in Human Nutrition and Medicine.

Reprints will not be available from the authors.
Correspondence: Professor Kerin O'Dea, Menzies School of Health Research, PO Box 41096, Casuarina, NT 0811.
kerinATmenzies.edu.au

Make a comment