SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

The objective of this study was to determine whether change in physical activity was associated with maintaining a healthy weight from childhood to adulthood. This prospective cohort study examined 1,594 young Australian adults (48.9% female) aged 27–36 years who were first examined at age 9–15 years as part of a national health and fitness survey. BMI was calculated from measured height and weight, and physical activity was self-reported at both time points; pedometers were also used at follow-up. Change in physical activity was characterized by calculating the difference between baseline and follow-up z-scores. Change scores were categorized as decreasing (large, moderate), stable, or increasing (large, moderate). Healthy weight was defined in childhood as a BMI less than international overweight cutoff points, and in adulthood as BMI<25 kg/m2. Healthy weight maintainers were healthy weight at both time points. Compared with those who demonstrated large relative decreases in physical activity, females in all other groups were 25–37% more likely to be healthy weight maintainers, although associations differed according to the physical activity measure used at follow-up and few reached statistical significance. Although younger males whose relative physical activity moderately or largely increased were 27–34% more likely to be healthy weight maintainers than those whose relative physical activity largely decreased, differences were not statistically significant. In conclusion, relatively increasing and stable physical activity from childhood to adulthood was only weakly associated with healthy weight maintenance. Examining personal, social, and environmental factors associated with healthy weight maintenance will be an important next step in understanding why some groups avoid becoming overweight.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Although being overweight or obese as a child is a strong predictor of overweight and obesity in adulthood, evidence from large prospective cohort studies consistently demonstrates that the majority of obese adults were a healthy weight in childhood (1). Increases in adiposity result from positive energy balance owing to decreases in physical activity and/or increases in energy intake over this period.

A small number of prospective cohort studies have assessed the association between physical activity patterns from childhood to adulthood and adult overweight and obesity, irrespective of baseline weight status (2,3,4,5). Findings have suggested that relative decreases in physical activity or persistent inactivity may be associated with an increased risk of overweight in adulthood. However, no studies have examined whether changes in physical activity over time are associated with maintaining a healthy weight from childhood into adulthood.

A recent systematic review of childhood predictors of adult obesity recommended that an assessment of factors that predict healthy weight maintenance is needed (6). Although some studies have examined predictors of weight maintenance irrespective of baseline weight status (7,8), only one other study has considered whether participants were healthy weight or overweight to begin with (9). This was a study following 165 white married couples for 2 years. Lean and weight-stable women reported more physical activity and lean and weight-stable men reported healthier dietary patterns than those who gained weight. However, participants who were already overweight at baseline were included in the comparison group, which does not allow for detailed examination or direct comparisons between healthy weight adults who remained healthy weight and those who became overweight with time.

Examining the behaviors and characteristics of those who maintain a healthy weight and who appear “resilient” to overweight and obesity in the current obesity-promoting environment may provide valuable lessons for obesity prevention (6,7,10). Little is known about the physical activity patterns of those who maintain a healthy weight from childhood into adulthood. This study aimed to determine whether relative changes in physical activity from childhood to adulthood were associated with maintaining a healthy weight.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Subjects

Baseline data for this study were collected as part of the 1985 Australian Schools Health and Fitness Survey (11,12), a national survey of the health and fitness of children aged 7–15. (For the purposes of this study, we refer to these participants as “children” though adolescents were included in this group.) A two-stage probability sampling technique was employed to first select schools with a probability proportional to the number of 10 year olds in primary schools and 14 year olds in secondary schools (n = 109, 90.1% response rate). The second stage of sampling involved selecting 10 children within age and sex strata within each school using simple random sampling (n = 8,498, 67.5% response rate). Approval was sought from the Director General of Education in each state and both parental and child consents were obtained.

At follow-up, the Childhood Determinants of Adult Health study located 6,840 (80.5%) of the original sample, with 5,170 of those contacted agreeing to provide follow-up data (75.6%) (13,14). A total of 2,410 young adults (between 27 and 36 years of age) attended a study clinic during 2004–2006 for physical measurements. The Childhood Determinants of Adult Health study was approved by the Southern Tasmanian Medical Research Ethics Committee.

Measurements

At baseline, BMI (kg/m2) was calculated from measured height and weight, and healthy weight was defined as a BMI less than internationally accepted age- and sex-specific cutoff points for overweight in children (15). Under supervision and in groups of four, 6,559 children aged 9–15 years self-reported the duration and frequency of walking and cycling to and from school, school sport, school physical education, and non-organized activities in the past week (data from 147 participants were excluded because of incomplete questionnaires, resulting in valid data for 6,412 participants). The sum of these activities was used as an estimate of weekly time spent in physical activity in childhood. Sociodemographic information collected included area-level socioeconomic status derived from residential postcode, smoking, parental smoking, country of birth, and language spoken at home.

At follow-up, BMI was calculated from measured height and weight and healthy weight was defined as a BMI <25 kg/m2. Participants completed the long version of the International Physical Activity Questionnaire, which demonstrated excellent reliability and reasonable validity in a 12-country, 14-site study (16). Participants were asked to wear a Yamax Digiwalker pedometer (model SW-200) for 7 days from which average daily steps values were calculated. Pedometer data were deemed valid if daily wear time was >8 h and a minimum 4 days of readings were collected, as recently recommended (17,18). Sociodemographic information collected included current occupation, highest level of education, smoking status, marital status, medication use, weight reduction dieting, and number of live births (females only).

Analyses

These analyses were restricted to participants who were healthy weight at baseline, who had measures of BMI and physical activity at both time points, and who were not pregnant at follow-up (n = 1,594). Healthy weight maintainers were defined as those healthy weight children who remained healthy weight as adults. “Gainers” were healthy weight children who became overweight or obese adults. At baseline, self-reported minutes of past-week physical activity was used as an indicator of weekly physical activity. At follow-up, two physical activity outcomes were used to characterize physical activity in adulthood: self-reported minutes of past-week physical activity and daily steps as measured using pedometers. Both self-reported physical activity measures were log-transformed to impose normality. To assess the direction and magnitude of change in physical activity, z-score values were calculated as the difference between individual values and mean values divided by the standard deviation within sex strata (and age strata for baseline physical activity). Baseline z-scores were then subtracted from follow-up z-scores to generate change scores, which were categorized as a large decrease (less than −1.0), moderate decrease (greater than −1.0 and less than −0.5), stable (−0.5 or more and <0.5), moderate increase (≥0.5 and <1.0), and large increase (≥1.0).

The Kruskal-Wallis equality-of-populations rank test was used to examine sex and school-level (primary or secondary) differences in the physical activity and BMI characteristics of participants at baseline and follow-up. The terms “younger” and “older” are used to differentiate participants who were primary school-aged and secondary school-aged at baseline. Log binomial regression was used to determine the relative risk and 95% confidence intervals of being a healthy weight maintainer vs. being a gainer across categories of physical activity z-score change. P values from tests for linear trends are presented. Results are presented adjusted for sociodemographic factors and stratified by sex and baseline school level.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Characteristics of the sample

Mean (s.d.) time from baseline to follow-up was 19.9 (±0.6) years (range: 18.7–21.0 years). Of the total participants, 39.9% (n = 316) of males and 65.5% (n = 525) of females were a healthy weight at both time points (healthy weight maintainers). Age, physical activity, and BMI characteristics of these participants are described in Table 1 according to whether they were in primary or secondary school during initial examination. At baseline, there was no significant difference in the physical activity levels of younger males and females, but older males reported more physical activity than older females (P < 0.01). No significant differences between males and females were evident in any of the physical activity measures at follow-up. Although older participants (males and females) reported significantly higher physical activity levels than younger participants at baseline, no significant differences were noted in follow-up physical activity between those who were younger and those who were older at baseline. Females had marginally higher BMI values than males in childhood (nonsignificant), but males had higher BMI values than females in adulthood (P < 0.01).

Table 1.  Characteristics of the sample at baseline and follow-up, by sex and baseline school level
inline image

Change in physical activity z-scores

The proportion of participants whose physical activity z-score decreased, increased, or remained stable since childhood were similar, irrespective of the physical activity measure used at follow-up (Figure 1). More than one-third of the participants were in the same category of physical activity at baseline and follow-up (stable), a third moved from a higher category in childhood to a lower category in adulthood (decreasing physical activity), and the remaining third moved from a lower category in childhood to a higher category in adulthood (increasing physical activity). No significant sex- or school-level differences existed in the proportion of participants whose physical activity was categorized as stable, increasing, or decreasing; results were therefore not presented by sex- or school-level strata.

image

Figure 1. Change in physical activity z-score from childhood (self-reported total physical activity) to adulthood (self-reported total physical activity and daily steps).

Download figure to PowerPoint

Healthy weight maintenance and change in physical activity z-score

Age (inverse), education (positive), and occupation (positive) were associated with healthy weight maintenance among males and females. Marital status among males was associated with the outcome (married less likely to be healthy weight maintainers); and among females, number of live births (three or more live births less likely to be healthy weight maintainers), weight reduction dieting (higher proportion of weight gainers reported dieting), and medication use (higher proportion of weight gainers reported medication use) were also associated with the outcome. These factors were therefore included as covariates in the respective models.

When physical activity patterns were assessed using self-reported physical activity at follow-up (self-reported at baseline), older females with moderate decreases, stable, or increasing physical activity z-scores were 26–37% more likely to be healthy weight maintainers than older females whose relative change in physical activity showed a large decrease (Table 2). However, only relatively stable physical activity was significantly associated with an increased likelihood of healthy weight maintenance (PR 1.37, 95% confidence interval 1.07–1.76). A similar trend was observed for older males (22–28% more likely to be healthy weight maintainers), but as for females, the differences were not statistically significant.

Table 2.  Associations between healthy weight maintenance and change in relative physical activity (self-reported total minutes at follow-up), by sex and baseline school level
inline image

When physical activity patterns were assessed using objectively measured daily steps at follow-up (and self-reported physical activity at baseline), younger females with smaller decreases, stable, or increasing physical activity z-scores were 22–32% more likely to be healthy weight maintainers than those whose physical activity showed a large decrease (Table 3), although few associations reached statistical significance. In addition, younger males and older females whose physical activity demonstrated the largest increase in z-score were 34 and 25% more likely to be healthy weight maintainers, respectively, than those whose physical activity z-score showed a large decrease (nonsignificant in males).

Table 3.  Associations between healthy weight maintenance and change in relative physical activity (pedometer-determined daily steps at follow-up), by sex and baseline school level
inline image

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

The aim of this study was to examine whether relative change in physical activity from childhood to adulthood was associated with maintaining a healthy weight in a prospective, population-based cohort of young Australians. Compared with those who had large deceases in physical activity z-scores, females in all other categories of change were more likely to be healthy weight at both time points. However, relative risk estimates were small, and few reached statistical significance. In males, no significant associations were observed between change in physical activity z-score and healthy weight maintenance from childhood to adulthood.

Although observational studies examining physical activity patterns from childhood to adulthood as a predictor of healthy weight maintenance are rare, some studies have assessed physical activity patterns from childhood to adulthood as a predictor of adult overweight and obesity, irrespective of baseline weight status. As well as having different outcomes, these studies have used different measures and classifications of physical activity which makes direct comparisons difficult. Evidence from these types of studies however does suggest that persistent inactivity and decreasing physical activity levels are associated with a significantly increased likelihood of overweight and obesity, compared with those who remain active or increase their activity levels (2,3,4).

In this study, the likelihood of healthy weight maintenance was more strongly associated with physical activity z-scores in females than in males. Similar findings have been observed in studies examining physical activity patterns and development of overweight and obesity (2,4). It has been postulated that the sex differences observed may be related to differences in body composition; because females have a higher percentage of fat mass than men, definitions of weight status based on BMI may be a better marker of adipose tissue in women than men. BMI does not differentiate between muscle mass and fat mass and misclassification of some athletic males as overweight may have occurred (4). This seems a likely explanation for these findings.

In females, the likelihood of being a healthy weight maintainer was generally similar in those whose physical activity z-scores increased over time and in those whose physical activity z-scores remained stable. This is an interesting finding because the stable group includes (but is not restricted to) those whose physical activity z-score may have been persistently low. Two Finnish studies have found that while decreasing physical activity was predictive of overweight and obesity, being persistently inactive was not associated with a higher risk of obesity. Although this finding seems unexpected, it is plausible that individuals whose physical activity is relatively constant, irrespective of the overall level, are better able to maintain energy balance and a healthy weight than those whose physical activity declines over time and who may need to reduce their dietary intake to maintain energy balance. Alternatively, Yang and colleagues proposed that individuals who decrease their physical activity tend to be those who adopt a range of poorer lifestyle and dietary behaviors that lead to an increased risk of overweight and obesity (4).

Because of the different measures used at baseline and follow-up, change categories were created from z-scores, calculated to standardize physical activity levels. It should therefore be noted that these categories represent relative and not absolute changes in total physical activity. It is possible and likely that decreases in physical activity occurred across the entire sample (19). Participants with positive physical activity z-scores (categorized in this study as having “increased” physical activity) may actually have decreased their physical activity levels in absolute terms, but these decreases were lower compared with others in the sample.

This study had some limitations. The measurement of children's physical activity by questionnaire is challenging (20). While no reliability or validity information was available for the childhood physical activity measures, children of this age group have been shown to report their physical activity with reasonable validity and good reliability (21). Children's responses in this study showed trends consistent with other research in similarly aged children; boys reported more physical activity than girls, physical activity increased with age in boys, while girls' physical activity increased until the age of 13 but then declined (22,23). Children's self-reported physical activity also demonstrated similar associations with submaximal cardiorespiratory fitness as observed in a recent meta-analysis (r = 0.17) (24). It is also possible that some participants changed their physical activity levels or weight status many times during the elapsed period between baseline and follow-up, and that the measures at two time-points do not accurately reflect typical physical activity or weight status over this time. However, other cohorts have found high tracking correlations between physical activity and weight status measured on multiple occasions throughout adolescence, suggesting that physical activity and weight status are reasonably stable during this period (25,26,27,28).

Participants who were relatively more active in childhood than at follow-up as young adults were least likely to be healthy weight maintainers, suggesting that reduced activity resulted in weight gain. However, we cannot exclude the possibility that weight gainers subsequently reduced their physical activity as a result of becoming overweight or obese. Evidence from prospective observational, weight gain prevention and weight loss intervention studies (29,30,31,32,33,34) however supports the inference that reduced physical activity may have preceded weight gain.

Participants in the follow-up study were more likely to be from a higher socioeconomic group at baseline (data not shown) than nonparticipants. A greater proportion of male participants attended independent schools and a greater proportion of females were born in Australia and spoke English at home. Mean BMI at baseline was higher in female nonparticipants (18.4 kg/m2) than female participants (18.1 kg/m2), although this difference was small. The prevalence of smoking at follow-up (20% current daily smokers) was similar to the Australian general population of a similar age (35), as was the prevalence of overweight and obesity (36).

Despite these limitations, this study had a number of key strengths, including its large size, prospective study design, length of follow-up, comprehensive measures of physical activity, measured height and weight, and the opportunity to control for potential confounding factors. In addition, this study quantified the magnitude and the direction of change in physical activity z-scores, which provides an insight into the dose-response relationship. This is one of few large prospective studies internationally with the ability to examine relative change in physical activity from childhood into adulthood and associations with maintaining a healthy weight. The objective measure of physical activity used at follow-up (pedometers) provides advantages over self-report measures, such as better estimating overall physical activity and reducing the likelihood of reporting biases.

In conclusion, compared with those females whose physical activity z-scores displayed large decreases from childhood to adulthood, healthy weight maintenance was more prevalent in females in all other categories of relative change in physical activity. However, the differences were small and no significant associations were observed in males, suggesting that relative changes in physical activity from childhood to adulthood have only a small role to play in the maintenance of a healthy weight. There is still much to learn about the personal, social, and environmental factors that may play a role in healthy weight maintenance from childhood into adulthood.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

This study was funded by grants from the National Health and Medical Research Council and the National Heart Foundation. The authors gratefully acknowledge the contributions of the Childhood Determinants of Adult Health study's project manager Ms Marita Dalton, all other project staff, and the study participants. Dr Schmidt and Dr Blizzard provided valuable analytical advice.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES
  • 1
    Serdula MK, Ivery D, Coates RJ et al. Do obese children become obese adults? A review of the literature. Prev Med 1993; 22: 167177.
  • 2
    Parsons TJ, Manor O, Power C. Physical activity and change in body mass index from adolescence to mid-adulthood in the 1958 British cohort. Int J Epidemiol 2006; 35: 197204.
  • 3
    Tammelin T, Laitinen J, Nayha S. Change in the level of physical activity from adolescence into adulthood and obesity at the age of 31 years. Int J Obes Relat Metab Disord 2004; 28: 775782.
  • 4
    Yang X, Telama R, Viikari J, Raitakari OT. Risk of obesity in relation to physical activity tracking from youth to adulthood. Med Sci Sports Exerc 2006; 38: 919925.
  • 5
    Hasselstrøm H, Hansen SE, Froberg K, Andersen LB. Physical fitness and physical activity during adolescence as predictors of cardiovascular disease risk in young adulthood. Danish Youth and Sports Study. An eight-year follow-up study. Int J Sports Med 2002; 23 (Suppl 1):S27S31.
  • 6
    Parsons TJ, Power C, Logan S, Summerbell CD. Childhood predictors of adult obesity: a systematic review. Int J Obes Relat Metab Disord 1999; 23 (Suppl 8):S1S107.
  • 7
    Ball K, Brown W, Crawford D. Who does not gain weight? Prevalence and predictors of weight maintenance in young women. Int J Obes Relat Metab Disord 2002; 26: 15701578.
  • 8
    St Jeor ST, Brunner RL, Harrington ME et al. Who are the weight maintainers? Obes Res 1995; 3 (Suppl 2):249s259s.
  • 9
    Davison KK, Birch LL. Lean and weight stable: behavioral predictors and psychological correlates. Obes Res 2004; 12: 10851093.
  • 10
    Ball K, Crawford D. Socio-economic factors in obesity: a case of slim chance in a fat world? Asia Pac J Clin Nutr 2006; 15 (Suppl):1520.
  • 11
    Pyke JE. Australian Health and Fitness Survey 1985. The Australian Council for Health, Physical Education and Recreation: South Australia, 1985.
  • 12
    Dwyer T, Gibbons LE. The Australian Schools Health and Fitness Survey: physical fitness related to blood pressure but not lipoproteins. Circulation 1994; 89: 15391544.
  • 13
    Venn AJ, Thomson RJ, Schmidt MD et al. Overweight and obesity from childhood to adulthood: a follow-up of participants in the 1985 Australian Schools Health and Fitness Survey. Med J Aust 2007; 186: 458460.
  • 14
    Cleland V, Schmidt M, Dwyer T, Venn A. Television viewing and abdominal obesity in young adults: is the association mediated by food and beverage consumption during viewing time or reduced leisure time physical activity? Am J Clin Nutr, in press.
  • 15
    Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 12401243.
  • 16
    Craig CL, Marshall AL, Sjostrom M et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003; 35: 13811395.
  • 17
    Tudor-Locke CE, Myers AM. Methodological considerations for researchers and practitioners using pedometers to measure physical (ambulatory) activity. Res Q Exerc Sport 2001; 72: 112.
  • 18
    Schmidt M, Blizzard CL, Venn A, Cochrane JA, Dwyer T. Practical considerations when using pedometers to assess physical activity in population studies. Res Q Exerc Sport 2007; 78: 162170.
  • 19
    Sallis JF. Age-related decline in physical activity: a synthesis of human and animal studies. Med Sci Sports Exerc 2000; 32: 15981600.
  • 20
    Welk GJ, Corbin CB, Dale D. Measurement issues in the assessment of physical activity in children. Res Q Exerc Sport 2000; 71: S59S73.
  • 21
    Sallis JF, Saelens BE. Assessment of physical activity by self-report: status, limitations, and future directions. Res Q Exerc Sport 2000; 71: S1S14.
  • 22
    Trost SG, Pate RR, Dowda M, Saunders R, Ward DS, Felton G. Gender differences in physical activity and determinants of physical activity in rural fifth grade children. J Sch Health 1996; 66: 145150.
  • 23
    Booth ML, Okely AD, Chey T, Bauman AE, Macaskill P. Epidemiology of physical activity participation among New South Wales school students. Aust NZ J Public Health 2002; 26: 371374.
  • 24
    Trost SG. The association between physical activity and cardiorespiratory fitness in children and adolescents: a meta-analytic review. Report prepared for the University of South Carolina School of Public Health. University of South Carolina: South Carolina, 1998.
  • 25
    Braddon FE, Rodgers B, Wadsworth ME, Davies JM. Onset of obesity in a 36 year birth cohort study. Br Med J (Clin Res Ed) 1986; 293: 299303.
  • 26
    Power C, Lake JK, Cole TJ. Body mass index and height from childhood to adulthood in the 1958 British born cohort. Am J Clin Nutr 1997; 66: 10941101.
  • 27
    Telama R, Leskinen E, Yang X. Stability of habitual physical activity and sport participation: a longitudinal tracking study. Scand J Med Sci Sports 1996; 6: 371378.
  • 28
    Twisk JWR, Kemper HCG, Snel J. Tracking of cardiovascular disease risk factors in relation to lifestyle. In: Kemper HCG (ed). The Amsterdam Growth Study: A Longitudinal Analysis of Health, Fitness and Lifestyle. Human Kinetics: Champaign, IL, 1995.
  • 29
    Erlichman J, Kerbey AL, James WP. Physical activity and its impact on health outcomes. Paper 2: Prevention of unhealthy weight gain and obesity by physical activity: an analysis of the evidence. Obes Rev 2002; 3: 273287.
  • 30
    Saris WHM, Blair A, van Baak MA et al. How much physical activity is enough to prevent unhealthy weight gain? Outcome of the IASO 1st Stock Conference and consensus statement. Obes Res 2003; 4: 101114.
  • 31
    Wareham NJ, van Sluijs EM, Ekelund U. Physical activity and obesity prevention: a review of the current evidence. Proc Nutr Soc 2005; 64: 229247.
  • 32
    Williamson DF, Madans J, Anda RF, Kleinman JC, Kahn HS, Byers T. Recreational physical activity and ten-year weight change in a US national cohort. Int J Obes Relat Metab Disord 1993; 17: 279286.
  • 33
    Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women. JAMA 2003; 289: 17851791.
  • 34
    Koh-Banerjee P, Chu NF, Spiegelman D et al. Prospective study of the association of changes in dietary intake, physical activity, alcohol consumption, and smoking with 9-y gain in waist circumference among 16 587 US men. Am J Clin Nutr 2003; 78: 719727.
  • 35
    White V, Hill D, Siahpush M, Bobevski I. How has the prevalence of cigarette smoking changed among Australian adults? Trends in smoking prevalence between 1980 and 2001. Tob Control 2003; 12 (Suppl 2):ii67ii74.
  • 36
    Cameron AJ, Welborn TA, Zimmet PZ et al. Overweight and obesity in Australia: the 1999–2000 Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Med J Aust 2003; 178: 427432.