Tuesday, January 06, 2009

Submission to the Preventative Health Taskforce's Discussion Paper ''Australia: the healthiest country by 2020''

EXECUTIVE SUMMARY

In summary the National Preventative Health Taskforce's Discussion Paper:

  1. downplays the positive role individual choices can play in the health sphere,
  2. pays little attention to the rights of individuals to consume legal products of their choosing, and for commercial vendors to provide consumers with those legal products,
  3. fails to interrogate the extent to which the management of individual risk should be appropriated by the state,
  4. neglects to properly assess the evidence base of its policy prescriptions, and
  5. presents policies that fail to live up to the framework of evidence-based public policy.


INTRODUCTION

This submission is divided into two sections.  The first section tackles some of the specific issues raised by the Preventative Health Taskforce's Discussion Paper, and outlines some of the ways the Discussion Paper's approach to policy making has been deficient.  The second section discusses the Discussion Paper's conceptual framework, dealing specifically with two issues — the use of the term and concept of "public health", and the issue of individual risk management, and the role that risk management should be subject to government intervention.

In an effort to address as many of the issues raised by the Discussion Paper, attached to this submission are two articles which cover some of the topics at hand.  While these do not directly deal with the Preventative Health Taskforce's Discussion Paper or technical documents, they raise points which cut across many of the proposals within.



POLICY-MAKING IN PREVENTATIVE HEALTH

The need for evidence-based policy

The Discussion Paper's recommendations are not sufficiently grounded in the basic principles of evidence-based policy making.  While it suggests that building the evidence base of preventative health measures is an important priority, it neglects to apply an evidence-based framework to its own policy prescriptions.

"Evidence-based policy" sounds uncontroversial — what policy could possibly be instituted without evidence to support it? — but it is in fact an increasingly important (and contentious) approach to policy design.  Evidence-based policy is policy based on evidence of that policy's efficacy at achieving its goals.  "Evidence" is taken to mean not merely research-based, but rather indicates the adoption of a scientific approach to public policy analysis — the purpose of evidence-based policy making is to discover in the most rigorous way the costs and consequences of policy design and implementation.  To do so, evidence-based policy makes use of combinations of systematic or randomised trials, benchmarking, cohort studies and the results from these trials and studies are ranked in a systematic hierarchy to determine their reliability. (1)

The Discussion Paper's recommendations are starkly deficient when seen through such a framework.  The Discussion Paper's policy proposals are justified by reference to a wildly varying standard of evidence.  And the policy goals are a complex patchwork hierarchy of "targets", "priorities", "imperatives", "actions", and "capacities to do".  In the tobacco technical paper alone, there are 42 of these categories, with additional references to measures and performance indicators in the document.  Overall, there are 79 "points of consideration" that the Discussion Paper urges the government to consider.  The technical paper on obesity paper targets 21 "initiatives", with dozens of "actions" nested within them.

By contrast, rarely throughout the Discussion Paper or in the technical papers are policies considered in their entirety.  The goals of individual policies are rarely clearly articulated.  Alternative proposals are not canvassed.  Policies are not presented in a cost-benefit framework, let alone coupled with the necessary analysis.  The criteria for policy success is not defined beyond the vaguest of motherhood statements, and metrics by which that success could be measured are not offered.  Unfortunately, the Discussion Paper's evidentiary emphasis has been on describing the extent of health problems, or, for example, secondary "problems" such as the increasing hours Australian children spend watching television or the frequency of advertisements for "non-core" foods, rather than looking at the evidence that lies behind their specific policy proposals.  (The specific issue of advertising and food is examined below.)

Furthermore, a fundamental lack of seriousness of the Discussion Paper's policy recommendations is demonstrated by the "future outlook" sections within the Discussion Paper, which set a combination of near impossible goals and vague assertions.  For example, the Discussion Paper asserts that "Australia has the capacity to:  ensure no tobacco products are sold to children". (2)  Tobacco sales to children are already illegal.  But the implementation and enforcement challenges that have meant that this goal has not already been achieved are never going to go away, despite the well-intentioned reforms proposed in the Taskforce's papers.  The sales of cigarettes to children will always be above zero because enforcement can never fully reach 100 per cent.  Similarly improbable or vague goals are given for alcohol, such as the claim that "If we act now, by 2020 ... there will be less drunken behaviour on the street" and "commercial activity in city centres, particularly at night, will become more diverse and prosperous." (3)  These statements undermine the Taskforce's serious purpose and unfortunately reveal a distinct lack of policy realism.

In many cases, to achieve these vague goals, the Discussion Paper recommends little more concrete than the institution of new regulatory agencies, a National Prevention Agency, education campaigns, health interventions and worker training, research and development, town planning schemes, tax measures and other regulations.  The Discussion Paper suggests vague work programs, but spends little time dealing with the administrative costs of such programs.

Lacking a strict evidence-based approach, the Discussion Paper risks recommending regulatory and spending measures which may not achieve their goals at all, are open-ended, and are riddled with consequences unintended by their designers.


Specific policy proposals

With the Discussion Paper's enormous array of policy options, it is not possible in this space to fully tackle each recommendation listed in the Discussion Paper or the technical documents systematically.  I have a long track record of dealing with individual policy proposals as they enter the public debate.  My Nanny State research has illustrated some of the flaws with proposals such as alcohol tax increases, advertising bans, obesity-related public policy, and broader issues to do with the impact of regulation on social capital and the philosophy and governance of the risk society.

One consistent theme that my research has found is that regulatory policies from all areas of government can have impacts on seemingly unrelated issues.  This is particularly made clear in my article "Your child is a wuss" (4) (Appendix A) that details the ways that overly risk-averse school administrators are unintentionally undermining exercise and physical activity in playgrounds.  But the obesity technical paper, when it discusses the school setting, focuses on diet and school layout, rather than on deliberate and already-existent public policies which are acting to limit playground exercise.

Indeed, it is a condemnation of the Taskforce's approach to the issue of obesity that in the 64 page technical paper, the word "exercise" appears only five times, and is even downplayed in those contexts.  To virtually ignore one major contributory factor towards obesity prevalence only illustrates the Taskforce's predisposition to attribute health problems to collective and social causes, rather than the choices made by individuals.  Physical activity, in the Taskforce's approach, is less a deliberate choice by individuals — "exercise" — and more a consequence of political decisions, hence the concentration of the Discussion Paper on environmental factors and urban planning initiatives.  My article indicates a positive step the National Preventative Health Strategy could pursue to encourage children to exercise.

Policy settings can also act as unintended consequences in other ways.  The tobacco technical paper raises the issue of smokeless tobacco, but, despite acknowledging the wealth of new evidence that suggests that smokeless tobacco is substantially less harmful than cigarettes, does not recommend any significant investigation or change of approach to these products.  The restriction on smokeless tobacco is a clear example in Australia of a policy, which, designed to protect individuals from harmful products, actually limits the ability of Australians to act in ways that are less harmful.  As the objectives of the National Tobacco Strategy include encouraging and assisting as many smokers to quit as soon as possible and to eliminate harmful exposure of tobacco smoke among non-smokers, the Taskforce has a responsibility to allow tobacco users the full array of harm reduction products provided in an open marketplace.


Obesity and advertising

Given the prominence of advertising restrictions and bans in the public debate over alcohol, tobacco and obesity policy, it is worth looking at in some detail.  Unfortunately, the Taskforce's approach to this issue is critically flawed.

The Discussion Paper reverses the conclusions of the federal government's most detailed study into the impact of food advertisement restrictions and body weight, the Australia Communications and Media Authority's 2008 review of Children Television Standards. (5)

That 2008 review was informed by the ACMA's earlier study "Television Advertising to Children" which concurs with a British Ofcom report that it was "difficult to determine the relative contribution of television advertising compared with all the other factors that influence children's food choice and health". (6)  This conclusion was reiterated in the ACMA's Children's Television Standards review.  The ACMA cited an analysis reviewed by Ofcom which concluded that "advertising/television exposure accounts for some two per cent of the variation in "food choice/obesity".  Given the vagaries of such a claim — "advertising" and "television exposure" are not the same thing — Ofcom was careful to point out that estimates vary.

But the Discussion Paper has instead discarded the ACMA/Ofcom conclusions by referring only to an appendix of the ACMA's 2008 review that analysed only the economic impact of an advertising ban.  The Taskforce's Discussion Paper pointed out that the cost/benefit analysis used in that economic study has been since superseded by subsequent data.  The Discussion Paper does not detail how that subsequent data may alter the analysis.  Nevertheless, even if it significantly altered the analysis of the economic impact of advertising bans, the new data would do nothing to challenge the original ACMA/Ofcom conclusion that it is nearly impossible to parse out the relationship between advertising and obesity.  The Discussion Paper has inappropriately dismissed the ACMA/Ofcom conclusion, and conducted an only rudimentary literature review that does not have the same depth and breadth as the more comprehensive ACMA study it discarded.

The early 20th century retailer John Wanaker summed up the conceptual and analytic problems in this area when he exclaimed:  "I know half my advertising is wasted, I just don't know which half."  This famous quip has formed the basis of a century's worth of investigation into the impact of advertising on consumer demand and brand awareness.

Contemporary economic analysis of advertising emphasises the practice's informative rather than manipulative qualities.  Advertising is a device used by producers to inform consumers of the availability of certain products.  This argument emphasises that consumer demand is less forged than discovered.  Consumers may not know that they want certain products until they are informed about them, but that does not mean that the act of informing was manipulating their demands — rather, that there was an undiscovered demand waiting to be fulfilled.  This is a much richer and nuanced view than the simplistic belief that advertising somehow "creates" demand, robbing individuals of both their free will and their heterogeneous range of preferences. (7)  This model also explains how advertising for existent products tends to focus on brands rather than the product themselves;  rather than being able to increase the pool of total demand, advertisers seek to shift brand preferences.

This is a critical point which is unrecognised by the Discussion Paper.  Children who identify advertised brands when they "pester" their parents for junk food may have already — independently of the brand's advertising — wanted some sort of junk food.  For example, just because a child pesters their parents for Cadbury chocolate, it does not necessarily follow that the child did not want chocolate before the advertisement was aired.  It defies credibility that children only want sweet food because they have been manipulated by advertising.

This intuitive view is backed up by the literature, and accounts for much of the ambiguity in empirical analysis of the relationship between food advertising and obesity.  Nevertheless, as the lead editorial of a 2004 edition of the Journal of the Royal Society of Medicine argued, "there is no good evidence that advertising has a substantial influence on children's food consumption and, consequently, no reason to believe that a complete ban on advertising would have any useful impact on childhood obesity rates."  It went on to argue that "the claim that food advertising is a major contributor to children's food choices and the rising tide of childhood obesity has obvious appeal, but as an argument it does not stand up to scrutiny." (8)

Unfortunately, the sections of the technical papers that deal with this issue appear to be intending to squeeze the evidence into their deliberate policy preference.  The critique of the claim that advertising bans in Sweden and Quebec have not positively impacted the obesity rates in those jurisdictions is dismissed by the Discussion Paper with a weak response that the efficacy of the Swedish and Quebecan policies has been hampered by external factors, such as television broadcasts originating outside the jurisdictions and other legal technicalities.  But such issues with the policies are, at best, hypothetical.  Instead of urging caution about policy implementation or the need for further study, the Discussion Paper claims that the negative consequences of the advertising bans have been comprehensively "refuted" by these unempirical and theoretical objections.  The Swedish and Quebecan cases are the only real-world examples of the Discussion Paper's proposed policies in action.  Given that they have failed — whatever the cause of that failure might be — the Discussion Paper could afford to be little less certain about its recommendations.  This sort of argument is an unfortunate consequence of the Discussion Paper's failure to adopt a rigorous evidence-based policy framework.

Similar critiques can be made of the approach to alcohol and tobacco advertising which form key parts of each respective technical paper.  The proposals concerning alcohol promotion and marketing, and the proposed regulatory body to oversee tobacco marketing and labelling do not adequately explore the relationship between marketing and choice.  Establishing the extent and limits of that relationship has to be a key task for the government before it acts on any of the Discussion Paper's recommendations.



CONCEPTUAL PROBLEMS WITH THE DISCUSSION PAPER'S APPROACH

What constitutes "Public health"?

It is worth briefly addressing the use of the phrase "public health" in discussions about preventative health measures.  According to the Discussion Paper, "one of the greatest public health challenges confronting Australia and many other industrialised countries is the obesity epidemic", (emphasis added) and the Taskforce uses the phrase to similarly describe alcohol and tobacco control measures.

George Rosen's definitive 1958 book The History of Public Health (9) described the historical role that government action has taken in managing health.  Some health problems are categorised as "public" health problems because they are most usefully tackled as collective problems, primarily because they involve the use of collective goods.  Public health problems are problems such as contaminated water, sanitation, epidemics — in other words, those health areas covered by the title of the Hippocratic book On Airs, Waters and Places.  Where problems are collective, the government has historically forged a role — for instance, in the draining of swamps to prevent spread of malaria through mosquito breeding, or in the construction of city-wide sewerage systems.  (It should be noted that there are many examples where collective problems can be efficiently and effectively tackled by non-state agents such as the private sector.  But for our purposes here, government can have a legitimate role dealing with collective problems.)

By contrast, most of the problems which the Preventative Health Taskforce seeks to tackle are not "public health" problems in any sense of the phrase.  The individual choices which contribute towards health problems from alcohol, tobacco and obesity do not, for the most part, have any collective attributes at all.  An individual's consumption of fatty food does not cause the group to consume fatty food.  While it is possible to project externalities from the consumption of such food, or, of course, the "second-hand" impact of tobacco, those externalities that do exist are in no way indicative of a "community" problem.  Just because a health problem occurs widely in a given population does not mean that it is a "public health" problem — to describe it as such is to mischaracterise it.

This point may appear semantic, but it is an important one to recognise while trying to conceptualise the problems caused by tobacco, alcohol and obesity.  For example, the Discussion Paper states that "Australia needs to ... ensure that policy directions to tackle overweight and obesity as a major public health issue have a population-wide focus".  Apart from the tautology — by definition, public health problems are population-wide problems — this argument illustrates a key flaw in the intellectual framework by which the Taskforce is addressing these health issues.  To classify something as a public health problem is to immediately imply government responsibility for that problem.  The Taskforce has tangled up the notions of "public" and "population-wide" in order to justify government action.

Such confusion is important because the Taskforce has to develop a public policy justification for government intervention in these areas.  Most of the Discussion Paper's policy proposals do not meet any of the economic criteria under which governments have intervened in individual choices.

There is no market failure in the areas of obesity, alcohol and tobacco.  And there is no obvious public good that is being underprovided, unlike, as we have seen, in the cases of sanitation and epidemic management.


Health spending and public health

One potential cause of this confusion is the relationship between public health spending and private health decisions.  Many advocates of preventative health regulation argue that private health decisions become public problems because the costs of those private decisions are, through the health system, borne by the taxpayer.  The Discussion Paper appears to partly endorse this view, citing the "massive economic burden on the community and the health system" of chronic diseases caused by obesity, alcohol and tobacco.  But there are a number of problems with this argument.

The Discussion Paper uses assessments of the costs of alcohol, obesity and tobacco that conflate costs privately incurred, costs incurred by the health system, and costs incurred by "society".  For example, on top of the direct costs to the health care system, the Discussion Paper and technical papers often include assumed aggregate productivity losses.  In the case of alcohol, the data used by the Discussion Paper goes so far as to include the cost of policing, property damage, insurance administration, lost productivity of those incarcerated due to offences caused after the consumption of alcohol, and the cost of lost household labour.

But every human action incurs some costs — the act of choice is, in fact, the act of deciding which sort of costs one would like to incur, if only costs from lost opportunities.  It is easy to add up all imaginable costs and compare it to an ideal alternative, but the results are not particularly insightful.  Individuals will never act in the lowest cost, most socially optimal, ideal.

All public policy should be subject to economic assessment.  But the use of costs to the health care system as a justification for curbing individual behaviour is a slippery slope.  In the United Kingdom, the spiralling costs of the National Health Service have led to discussions about excluding individuals who have engaged in unhealthy behaviour or made lifestyle choices deemed poor from being able to use the health system.


Risk assessment by individuals and governments

The Taskforce's approach to alcohol, tobacco and obesity is one in which it is the government's responsibility to alter individual preferences — even, in the case of alcohol, to alter an entire culture — to meet its goals.  In other words, the Taskforce seeks to shift the management of long-term or low risks from individuals and groups to the state.

Whether consciously or sub-consciously, individuals make thousands of risk assessments every day.  Indeed, every individual action involves risk — from minor risks such as those risks involved in eating prepared foods, to major risks such as those involving global catastrophe.  Individuals who choose to use a mobile telephones have implicitly made an assessment of the risks doing so — are there known or unknown dangers involving electromagnetic radiation?

Driving involves thousands of risk assessments;  not just choosing to drive to a destination, but more mundane and arguably more risky choices such as the timing of lane changes or how loud a car radio can be before it impairs the driver's attention.  When individuals choose to go faster or slower, they are weighing the risks of faster driving — that is, the potential to get in an accident or be ticketed for speeding — with the potential benefits of arriving at their destination sooner.

Similarly, the consumption of food can be weighed against the long term risks involved for an individual's health and lifespan, as well as more mundane risks about taste and enjoyment.  An individual purchasing any good takes the risk that their purchase might not be subjectively worth the price they paid.

Academic definitions of risk usually take into account decision-making within a framework of probability, and this level of risk is presented in terms of its statistical occurrence — for example, the risk that action A would lead to result B is 50 per cent.

Nevertheless, popular usage of the concept of risk tends to see risk levels as binary — an activity is either risky, or it is not risky.  Eating at a restaurant is not risky;  base-jumping is risky.  Driving at half the speed limit is not risky;  driving at twice the speed limit is risky.  But this dichotomy is value-laden.  Classing an activity as "risky" is itself a characterisation of an unacceptable level of risk.  Similarly, an individual participating in an activity that they would describe as "not risky" still involves a certain level of (apparently acceptable) risk.

As this indicates, assessing whether an activity is "risky" involves highly subjective individual judgment.  Furthermore an activity that is risky may still be pursued — if the individual is willing to wear that risk for a potential benefit.  It is easy to see how some individuals may view base-jumping as an unacceptably risky activity, and yet others — even if they acknowledge a high level of risk — perceive the benefits to be high enough to accept that level.  Assessing the level of risk is only the first part of the decision to pursue a course of action — that assessment has to be processed by an individual's subjective views of the balance between risks and benefits.

Crucial to risk assessment is information.  Risks can only dissuade individuals from action if they perceive those risks.  This observation is particularly important for risk politics.  That an activity carries with it a level of risk has no political consequence;  risks can only become politicised if individuals perceive those risks.  Knowledge about risk — whether accurate or erroneous — is the dominant currency in what the sociologist Ulrich Beck has influentially termed the "risk society". (10)

A large part of the challenge of living in a risk society is reconciling different claims about relative risks.  Alcohol is both unhealthy and healthy.  Regular alcohol intake cuts down the incidence of heart attacks in men, (11) but it also raises blood pressure. (12)  As Radley Balko points out, alcohol consumption has been variously shown to also reduce the incidence of Type 2 diabetes, the stiffening of the arteries, prostate cancer, skin cancer and the common cold. (13)  But to oppose that list, the Cancer Institute of New South Wales argues that alcohol consumption has been positively linked with breast cancer, liver cancer, upper aero-digestive tract cancer, colorectal cancer and stomach cancer, as well as heart disease. (14)  These medical claims are further confused by claims about the social consequences of risky behaviour, which are commonly reported as simple cause-and-effect stories that classify certain activities as the exclusive cause of unwanted results.  Recent studies have claimed that binge drinking in women leads to unsafe sexual practices, (15) or that there is a positive correlation between neighbourhoods with high numbers of alcohol vendors and high numbers of injuries sustained by local children. (16)

For the non-scientifically trained public, assessing the validity and applicability of these studies is further complicated by the prisms through which the media reports on them, and the way that politicians, lobbyists and advocates harness the evidence to defend their arguments.

But for those "experts" (Beck's term) who professionally hold the knowledge that allow them to interpret relative risk levels, the uneven awareness in the public and the media is a never-ending source of frustration:

The "irrationality" of "deviating" public risk "perception" lies in the fact that, in the eyes of the technological elite, the majority of the public still behaves like engineering students in their first semester.  They are ignorant, or course, but well-intentioned;  hard-working, but without a clue.  In this view, the population is composed of nothing but would-be engineers, who do not possess sufficient knowledge.  They only need to be stuffed full of technical details, and then they will share the expert's viewpoint ... Protests, fears, criticism, or resistance in the public sphere are a pure problem of information.  If the public only knew what the technical people know, they would be put at ease — otherwise they are just hopefully irrational. (17)

But, while this belief is widespread, risk assessment on the level of the individual involves mgathering technical data about health and safety costs.  Risks are only perceptible within an individual's value system;  and the effect of this value system on risk assessment is not something which can be deduced by expert studies.  Values determine risk.  As Beck writes:

Even in their highly mathematical or technical garb, statements on risks contain statements of the type that is how we want to live — statements, that is, to which the natural and engineering sciences alone can provide answers only by overstepping the bounds of their disciplines.  But then the tables are turned.  The non-acceptance of the scientific definition of risks is not something to be reproached as "irrationality" in the population, but quite to the contrary, it indicates that the culture premises of acceptability containin scientific and technical statements on risk are wrong.  The technical risk experts are mistaken in the empirical accuracy of their implicit value premises, specifically in their assumptions of what appears acceptable risk in the population.  The talk of a "false, irrational" perception of risk in the population, however, crowns this mistake;  the scientists withdraw their borrowed notions of cultural acceptance from empirical criticism, elevate their views of other people's notions to a dogma and mount this shaky throne to serve as judges of the irrationality of the population, whose ideas they ought to ascertain and make the foundation of their work. (18)

In other words, the third party risk assessments provided by experts cannot possibly substitute the risk assessments performed by individuals, as highly subjective values play such a key part.  Risks can only be determined on the level of the individual, not on the level of groups or societies.

The Preventative Health Taskforce's Discussion Paper downplays the role individuals play in assessing their own risks, instead focusing on the risks perceived by the health advocates, who — by virtue of their choice of profession — have different concepts of what constitutes "acceptable" risk.  The Taskforce needs to recognise that these questions are subjective, rather than objective.

While individuals may choose to pursue activities that the Taskforce considers excessively risky or detrimental, the Taskforce needs to also recognise that it does not necessarily share the same value system as all of those individuals they seek to regulate.  Many Australians may be happy to forgo some health benefits in order to use alcohol, tobacco, or consume fast food.  Considering the widespread knowledge of the health impact of these unhealthy practices, it would be reasonable to assume that this sort of conscious decision is one which is widespread in Australia.



CONCLUSION

The issues raised in this submission are by no means the only problematic issues contained within the Discussion Paper and its associated technical papers.  Preventative health, as the Taskforce has conceptualised it, can impact on the autonomy individuals have over their own decisions, the limits or otherwise of individual responsibility, and, as elaborated above, the concept of individual risk management and risk assessment.

Individuals need to retain the autonomy to consume products of their own choosing, and the autonomy to accept responsibility for those choices.

Unfortunately, the Discussion Paper does not recognise any limits of government interference in individual choices.  This has led to some poorly made policy recommendations.  But perhaps more importantly, it has led to an assumption that the most effective way to tackle health issues is through government intervention — that regulation, taxation and government bodies should be allowed to overrule individual choice and responsibility.



APPENDIX A

Your child is a wuss

When Belgian Gardens, a Townsville state primary school, banned cartwheels and handstands in August, it ignited a media frenzy.  But as bizarre as it is, the handstand ban is only one incidence of a widespread trend affecting all Australian schools.

Carlton Gardens Primary School has removed its monkey bars, St Michael's Primary School has banned football and soccer during recess, St Peter Chanel Primary School now allows students to play football and soccer only if there is no tackling, and Ascot Vale West Primary has banned all games that are deemed "too rough".

Many schools have even instituted birthday cake bans to prevent children bringing cakes to school to share with their classmates;  cakes are too unhealthy and raise food safety concerns.

The Belgian Gardens cartwheel ban drew attention from both the Queensland and federal governments.  Federal Education Minister Julia Gillard promised to look into the ban, while Queensland Premier Anna Bligh and Education Minister Rod Welford made it clear responsibility for the decision rested on the principal's shoulders.

The media never picked up on, and the politicians never admitted, that the Principal, Glen Dickson, was following Queensland Activity Risk Management Guidelines when instituting the ban.  He correctly classified cartwheels as a level 2 (medium risk) activity requiring adult supervision.  He followed the rules and in return has been given the cold shoulder by politicians who should ultimately be responsible for these ridiculous state department guidelines.

In addition to the general risk management guidelines, there are 133 curriculum activity risk management modules that Queensland teachers must follow.  The modules cover nearly every activity a child could undertake at school, and some they likely will not.  There are even modules for candle making, tractor driving, shooting and bait gathering.

The NSW Department of Education and Training's Guidelines for Safe Conduct of Sport and Physical Activity in Schools is 284 pages long, with detailed requirements for 61 different sports from abseiling to water polo.

The document is full of useful advice including diagrams to locate the seat, handlebars and pedals when conducting a bicycle safety check.  Prohibited sports include bungee jumping, break dancing, tobogganing, rodeo, hang-gliding, multi-pitch rock-climbing, boxing, and rock-fishing.

NSW students should remain in the shade if possible between 11am and 3pm daylight savings time and padding must be 2m high on goal posts.  Students aren't allowed to throw curve balls at baseball until they are in Year 9.  Finishing tape is banned from certain running events for fear of choking runners.  Teachers are recommended to confirm that students can swim before letting them onto the high diving platform.

The Victorian Government's Schools Reference Guide fairs little better, it has 66 pages dedicated to student safety and risk management.  Paragraph 4.4.2.3.1 recommends young children should wear identity tags if taken outside school premises.

In case this wasn't enough, the federal government released its own guidelines for Children's Safety in Sport and Recreation.  The Rudd Government contributed over $300,000 to Sport Medicine Australia to develop the guidelines and subsequently plans to distribute over a million brochures and 40,000 copies of the guidelines across Australia.  Alarmingly, it advocates pre-participation screening of students before physical activity using a questionnaire to ascertain the medical history of all participants as well as family, school, other sporting and social commitments.

Swings, see-saws, flying foxes and roundabouts are also now banned from NSW and Victorian public schools.  Playground equipment must meet strict building standards which regulate falling heights, impact absorbing surfaces and construction materials.

Litigious parents are taking the blame for these bans after Education Queensland released figures showing that 93 compensation claims were brought against the Queensland Government last year over school yard injuries.  No consideration has been given to the fact that these guidelines actually invite litigation;  setting unrealistic and complex safety requirements is throwing ammunition to litigious parents and encouraging stupid bans from fearful principals.  What principal will want to admit at court that they chose not to follow safety guidelines?

The real losers are the children.  Not only are they being harassed by the fun police, but experts are saying these bans are having long term consequences on children's development.  The child and adolescent psychologist Dr Michael Carr-Gregg believes that sanitising children's playing environments will create a generation of wimps who lack decision making skills and resilience to life's setbacks:  "it's all part of this 'wussification' syndrome that we're seeing in contemporary Australia where schools have been forced to bow to the great God of occupational health and safety."

Long and complex education department guidelines are encouraging a culture of risk adversity in schools.  If governments wish to continue denying liability for these maternalistic bans, they must remove these absurd guidelines and restore true autonomy to teaching staff.



APPENDIX B

Big fat beat up

Arnold Schwarzenegger is obese.  He's been obese his entire adult life, including the seven times he won the Mr Olympia bodybuilding title.

Despite obesity not impacting his own, clearly excellent, health, Governor Schwarzenegger has introduced many anti-obesity programs in an attempt to combat what he regards as California's obesity epidemic.  In doing so, Governor Schwarzenegger becomes just another in a long line of policy makers attacking the wrong problem with the wrong solutions.

Before any public policy responses to obesity can even be considered, a single, surprisingly controversial, question has to be answered:  when does being fat become a health problem?  Only then can the second, also controversial, question be asked:  what, if anything, can policy makers do about obesity?


Carry that weight

According to the World Health Organisation anybody with a Body Mass Index above 30 is considered obese, and anything over a BMI of 25 is classified as overweight.  At his bodybuilding peak Arnie was 1.88m (6'2") and weighed 107kg, leaving him with a BMI of 30.2.

So where did the magic number of a BMI of 25 come from?  What makes a woman of average height (164cm/5'4½") fat at 67.5kg but normal at 67kg?  Has she actually increased her risk of diabetes, cancer, or heart disease by gaining half a kilo?  The use of BMI, and the classifications, come from life insurance tables of the 1940s.  But the adoption of 25 as the magic number is credited to the International Obesity Taskforce (IOTF) whose members are also prominent on the US National Institute of Health and World Health Organisation obesity panels.  It was the adoption of 25 as the cut-off by the WHO that led to that number becoming the international standard of fatness.  Eric Oliver in his book Fat Politics demonstrates that the IOTF is funded by diet drug companies such as Hoffman-La Roche.

The dirty secret of the BMI is that health experts know the BMI is a deeply flawed way to measure whether someone is too fat.

If the classification for overweight was moved to a BMI of 30 then 52 per cent of Australians — some 10 million people — would immediately be classified as having a "normal" weight.  And there is considerable evidence that the greatest public policy intervention legislators could make in this area is to do exactly that.  When the US moved the obesity goal posts the other way in 1998 and overnight reclassified 37 million previously normal weight Americans as overweight, the immediate impact was to stigmatise many newly overweight women as too fat — for no measured public health benefit.

But the BMI has one big advantage over accurate methods to measure body fat — it's cheap and easy.

Obviously, diet pill companies have a big incentive to make more people think they are fat — it's good for business — but the drug companies are by no means alone in creating the "obesity epidemic".  Obesity researchers in Australia have proved highly media savvy in getting out the message that obesity is killing us.  In the crowded bazaar of medical research, where it's hard to get your message heard above the clamour of competing illnesses and causes, it helps to proclaim, as Access Economics did earlier this year, that your "disease" affects seven in 10 adult Australians and costs the country $58.2 billion a year.


So what's wrong with being a bit tubby?

It is therefore surprising to learn the evidence does not support a BMI of 25 as an important marker of health outcomes.  In their 2006 book Diet Nation, Patrick Basham, Gio Gori, and John Luik demonstrate that the evidence shows higher mortality rates don't become prominent until beyond a BMI of 35 (about 118kg for a 183cm/6' man).

It is clear that the extremely fat (those who have a BMI over 40) face a range of health risks and problems and there is some evidence that the relatively fat, (BMI 35-40) face some additional health risks.  The very fat (the morbidly obese) and the very thin (the anorexics) have appreciably higher mortality rates than the rest of the population but, according to Diet Nation, what is not clear is at what point fatness causes disease rather than any other factor such as smoking, age, metabolic disorder or family history.

Nevertheless, over and over researchers trumpet links between disease and obesity.  Being even slightly overweight is meant to cause diabetes, some cancers, heart disease and strokes.  How can it be that the evidence doesn't appear to confirm the health risks except for the very fat?  Researchers are beginning to re-examine the evidence on weight and health, and coming to some dramatically different conclusions than those which obesity lobbyists rely upon.

Critically, health researchers rarely measure health problems "caused" by fatness along the full range of body weights.  This approach fails to distinguish whether a BMI of 22, 32 or 42 is the tipping point for whatever disease excess fat is meant to cause.  For this reason, the Australian Institute of Health and Welfare, in its report Burden of Disease and Injury in Australia 2003 notes that its methods inflate the risk attributed to high body weight.  Yet the authors continue to assert being even slightly overweight is a cause of a range of medical conditions.  This research flaw is found widely in the obesity literature.

For example, recent research casts significant doubt on the supposed link between obesity and heart disease risks.  Research published this year in The Archives of Internal Medicine found over half of those classified as overweight have normal blood pressure and cholesterol levels.  More importantly though, is that study found weight was not the major risk factor.  In all weight groups, inactivity, smoking and age were linked with higher risk factors for heart disease than waist circumference.  People with normal weight but larger waist circumference had a greater risk than fatter people with smaller waists.

Another paper in the same journal found no difference in insulin sensitivity between a normal weight group and an obese group leading them to conclude that not all obese people face higher risks for developing diabetes.  Similar studies exist for all other supposed diseases of fatness.  Increasingly researchers are being faced with evidence that being merely overweight does not cause the raft of diseases attributed to it.  A key part of that evidence is from Katherine Flegal and colleagues in a paper published in April 2005 in the Journal of the American Medical Association which noted the continued increase in life expectancy and the continuation of the fall in deaths from heart disease and stroke despite more than 25 years of increased weight in the US population.

For sure, there is substantial evidence that being very heavy — those with morbid obesity — is correlated with a range of health problems and risks.  Hefting around additional weight puts stress on the joints causing increased demand for hip and knee replacements as well as osteoarthritis.  The very fat are more likely to suffer from diabetes, heart disease, some cancers and strokes.

Continuing research does point to one worrying trend, the super fat — the morbidly obese and even fatter — are both getting fatter and increasing in number.  However, even the country with the most massive people, the US, only has 4.8 per cent of its population in the morbidly obese or worse category.  Yet the very fat are unlikely to be properly targeted for assistance, not least of which because if the money made available to "fight the obesity epidemic" is spread across 70 per cent of the population, it is unlikely the substantial sums required to treat the multiple health and behavioural problems experienced by the extremely obese will be made available.

As Oliver writes in Fat Politics, the reason why a majority "are overweight is because a nineteenth-century astronomer, a twentieth-century insurance actuary, and a handful of contemporary scientists concocted some ideas about what a normal weight should be".  The original definition of a normal weight came from a Belgian astronomer who measured the most common weight for their height of Scottish and French army conscripts — and he did so in the 1830s, a time when life expectancy was about 40 as a result of chronic malnutrition and disease.  No wonder their average weight was so low.  The other major source of data for what is "normal" came from measuring men to develop life insurance tables in America in the 1940's.  Because of their different physiology, it is likely women gain additional protective benefits from carrying more weight than men yet the "ideal" weight is deemed to be the same for a man and woman of equal height.

Given the increasing evidence that the categories of overweight and mildly obese have fewer elevated health risks, and definitely lower health risks than the underweight, is it not possible that the healthy ranges are set too low, particularly for women?  It is hard to see any benefit in public health terms or for the individuals concerned by stigmatising large swathes of the population as overweight or mildly obese if the huge effort of them losing weight will not appreciably improve their health or life expectancy.


Neither cause nor cure are correct

The way to reduce fatness is to eat fewer calories and exercise more.  This approach has the great advantage of matching common sense.  We see the contestants on Australia's Biggest Loser go on a diet and do huge amounts of physical activity and — Hey Presto! — they all lose weight.  It is therefore very surprising to learn that the extensive research literature on exercise and dieting shows virtually nobody loses weight through low calorie dieting, that the overweight do not eat more than the lean and exercise is not correlated with weight loss.

Although all experts agree imbalance between energy in and energy out causes weight gain, it seems the actual amount of food needed varies so dramatically between people that no general diet recommendation works across the population.  Short of locking all the overweight up with an individual dietician and trainer, the obesity industry has no workable way to help most people get to the weight the experts deem ideal.

The obesity industry — that is, those researchers and drug companies that rely on the notion of an "obesity epidemic" — therefore face an increasing amount of evidence suggesting that the links between fatness and various diseases are weaker than believed, and that the industry's proposed solution to obesity — a low calorie/high carbohydrate diet with exercise — does not work.  Moreover, if current definitions of overweight are set too low — so that attaining a "normal" weight requires constant dieting — then this in itself may be leading many people to feel discouraged when the ideal is unattainable.

Survey after survey tells us that most people — including the fat themselves — blame fat people for their size.  If only all the fatties could get a grip on themselves, they wouldn't be fat.  Being over-weight or obese is typically seen to be a result of lack of control.

This is also the message pushed by the weight loss companies and gyms as well as the writers of diet books and sellers of weight loss additives.  Even though weight loss, particularly to the weight range deemed ideal, is recognised as very difficult to achieve, losing weight is still seen by most as a private decision.

Increasingly policy makers and activists are out of step with this response.  Increasingly the Nanny State intrudes into obesity policy.  Instead of obesity simply being the result of an energy imbalance caused by an individual's eating patterns it is some unquantified combination of genes, metabolism disorders and an "obesogenic" environment.

Despite the failure of research to identify either that moderate fatness causes other health problems or, more importantly, that there are cures for this supposed health crisis, increasingly a disparate number of researchers and activists think they have the public policy answers to the "epidemic".  Unlike the public's preference for individual self-control, the obesity activists favour extensive government intervention and a severe restriction of the civil liberties of the entire population, regardless of whether they are too fat or not.

These activists blame processed food companies, modern agricultural methods and fast food companies for obesity.  Television watching (especially presumably plasma TV watching), car driving and urban sprawl are also highly cited.  Activists argue that modern agriculture and processing has made food very cheap, and that this cheap food is then shovelled out in huge portion sizes at the supermarket and by fast food companies to people who live in suburban fringe estates with no footpaths;  who drive their cars through drive-throughs so they can return to eat in front of the TV;  where they watch increasing amounts of advertisements that trick them into repeating the process the next day.

Over all this they have little or no control.


Forgive your upsize

For some obesity activists the public-spirited response to this is to ban junk food advertising, ban suburban fringe development, build new public transport services, increase labelling requirements, ban super-sized servings, tax junk food, and increase the number of bariatric surgeries done on Medicare.

And these are responses to reducing the number of adults who are obese and overweight.  There are far more draconian options proposed for childhood obesity.

One commonly seen proposal is the restriction of fast food outlets in low income areas.  Supporters of this heavy-handed restriction on civil liberties imply that poor people are incapable of making good food choices, and so the only option is to restrict their choices to good food.  Yet recent research shows that while there is a greater preponderance of fast food outlets in low income areas, there are also more supermarkets.  The choice is already there.

Another popular proposal is a fat tax — a tax on high calorie dense foods such as soft drinks, confectionary and much fast food.  The idea is if these types of foods are taxed to the point that reduces consumption and the proceeds are used to subsidise "too expensive" healthy foods, then the poor will be able to eat better.

But beyond the practical problems with this idea — do we really want to make food more expensive for poor people?  And do we really want to set individual tax rates for every single food type?  There is something deeply repugnant about expecting people to vote, to raise their children, to hold a job or to serve on a jury yet at the same time use policy settings to forcibly change what people eat.

Moreover, it is not clear at all that fast food is the culprit of the obesity "crisis".  A number of research studies have failed to find a link between fast food consumption and obesity.  Those dreadful news clips of fat people chowing-down on chips, fried chicken and burgers while slurping huge buckets of coca-cola, do not, obviously, tell the whole story.

The studies show people compensate over the day so that overall very few additional calories are consumed compared to people not eating fast food.  Another claim made by supporters of a fat tax is that supposedly the poor live on soft drink and chips because they are much cheaper than a healthy alternative.

But it does not stand up to scrutiny.  Consider this hypothetical — a one litre bottle of coke from the supermarket costs about $2.10, a 200g packet of chips $3.80 or a large fries from McDonalds $2.75.  Total meal cost around $5 or $10 for a couple.  For much less money, that couple could have purchased ingredients to cook pork chops and vegetables, pasta with tuna, chicken breast and vegetables or risotto with chorizo and peas, or any number of other meals.  The argument that the poor — perhaps with the exception of those who live in the most remote locations — cannot afford unprocessed fresh food is simply not true.

Policy makers of all political persuasions like to be seen to be doing something, especially when there's an apparent crisis, the "obesity epidemic".  Some are open in claiming it is the right and duty of the state to dictate what a good life should look like.  Others are more circumspect, hiding behind disputed science and drug company funded reports, to justify a supposed cost-benefit of limiting people's basic life choices.

In either case, Nanny State proposals to solve a crisis that may not exist, with proposals that are known not to work, embodies the worst excesses of the government interventionism in a democratic society.



ENDNOTES

1. For more information on the principles and challenges posed by evidence-based policy, see Richard J. Wood, "Evidence-based Policy and Public Sector Innovation", Occasional Paper, November 2008.

2. Australia: the healthiest country by 2020, p22.

3. as above, p33.

4. Richard J. Wood, "Your child is a wuss", November 2008. Available at Appendix A.

5. Australian Communications and Media Authority, "Children's Television Standards — Report for the Review", 2008.

6. Australian Communications and Media Authority, "Television Advertising to Children", June 2007

7. For a further overview of the economic theory of advertising, see Alberto Mingardi, "Advertising", in John Meadowcroft (ed.) Prohibitions, Institute of Economic Affairs, London, 2008

8. "Food advertising and childhood obesity", Journal of the Royal Society of Medicine, Feb 2004, v97n2

9. George Rosen, A History of Public Health, MD Publications, New York, 1958

10. Ulrich Beck, Risk Society: Towards a New Modernity, 1992, SAGE, London

11. K. Mukamal, K. Conigrave, M. Mittleman, et al. "Roles of drinking pattern and type of alcohol consumed in coronary heart disease in men", New England Journal of Medicine, 2003, vol. 348

12. R Zilkens, V Burke, J Hodgeson, et al. "Red Wine and Beer Elevate Blood Pressure in Normotensive Men" Hypertension, 2005, vol. 45

13. Radley Balko, "Back Door to Prohibition: The New War on Social Drinking", Cato Policy Analysis, No. 501, December 5, 2003.

14. "Alcohol Consumption and Cancer Risk", Cancer Institute of New Wales Position Statement, May 2008.

15. "The Relationship between Recent Alcohol Use and Sexual Behaviors: Gender Differences among STD Clinic Patients". Alcoholism: Clinical & Experimental Research, November, 2008

16. "Neighborhood Alcohol Outlets Tied to Kids' Injury Risk", HealthDay News, September 4, 2008.

17. Ulrich Beck, Risk Society: Towards a New Modernity, 1992, SAGE, London, p58

18. as above, p58

No comments: