Peter Gallagher

It is sur­pris­ingly dif­fi­cult to find an answer to these ques­tions in the ABS data. The Bureau pub­lishes only age-stratified preva­lence data from its sam­ples; not basic sta­tis­tics such as the mean and vari­ance of the adult pop­u­la­tion BMI. I can’t tell from their pub­li­ca­tion how the BMI isdis­trib­uted across the whole pop­u­la­tion or how the ‘cen­tral ten­dency’ of that dis­tri­b­u­tion changed between the mea­sured sam­pling in 1995 and 2008. This is cru­cial infor­ma­tion, how­ever, because it is likely to show what is actu­ally going on in the trend and whether the appar­ent increases in the pop­u­la­tion indices are due mainly (as I sus­pect) to the greater vari­ance in the sam­ple or due to an alarm­ing ‘grossing-up’ of the aver­age Aus­tralian (as some inter­ested groups want us to believe).

Peo­ple are liv­ing longer and BMI tends to increase with age, as the ABS preva­lence data shows. So it’s quite pos­si­ble that a larger pro­por­tion of the total pop­u­la­tion has a higher BMI just because there is a big­ger ‘spread’ (vari­ance) in the fre­quency dis­tri­b­u­tion curve. If that were so, the ‘mean’ (cen­tral ten­dency) of the dis­tri­b­u­tion may not have changed to the extent that the increase in the pro­por­tion of ‘over­weight and obese’ in the sam­ple seems to indi­cate. In other words, not so much ‘fat­ter on aver­age’ as ‘fat­ter at the extremes’.

So let’s accept, for the moment, the pre­miss that there has been an explo­sion of fat. What can we do to check whether we should be alarmed? The obvi­ous thing is to check whether the ‘risks’ that are said to be increas­ing are vis­i­ble in the chang­ing inci­dence of mor­tal­ity. That is, has there been an alarm­ing increase in actual deaths from the dis­eases that over­weight and obe­sity are said to cause?

It turns out that there’s no appar­ent cor­re­la­tion between the claimed trend in the obe­sity risk-factor and actual deaths from obesity-associated dis­eases. Here’s the rel­e­vant infor­ma­tion on mor­tal­ity trends from the Aus­tralian Insti­tute of Health and Welfare’s Australia’s Health 2008 publication.

“Australia’s mor­tal­ity from cere­brovas­cu­lar dis­ease has been declin­ing in recent decades… since the mid-1970s, con­sis­tent declines have been noted for both males and females. Age-standardised death rates for cere­brovas­cu­lar dis­ease fell by 37% (males) and 35% (females) over the period 1996–2005.”

“There have not been major changes in the death rate from dia­betes (as an under­ly­ing cause) over the last 25 years… For males, the death rate rose by an aver­age of 0.7% per year. In con­trast, the rate for females fell by an aver­age of 0.5% per year.”

“From 1997 to 2005, the mor­tal­ity rate for CKD [chronic kid­ney dis­ease] as the under­ly­ing cause of death fell from 13 to 11 deaths per 100,000 population.”

This absence of cor­re­la­tion between the obesity-trend and the mortality-trend in Aus­tralian now looks less mys­te­ri­ous, thanks to a unique and detailed United States study pub­lished in Feb­ru­ary by the senior sta­tis­ti­cians of the United States Cen­ters for Dis­ease Con­trol and Pre­ven­tion. Using a very large sam­ple of the U.S. pop­u­la­tion, and tak­ing BMI and death data since 1988 into account, Fle­gal and Graubard found BMI is not sig­nif­i­cantly cor­re­lated with excess deaths from all-causes, from ‘obesity-related’ causes of death or from other causes. Looks like the same might be true in Australia.

But what about the dis­ease bur­den? Excess mor­tal­ity is the only final arbiter of the degree of risk, but the bur­den of dis­ease on indi­vid­u­als and the com­mu­nity (indi­vid­u­als, tax­pay­ers, car­ers) is also impor­tant. What’s going on there?

As far as I can tell from the semi-obscure way that ABS presents the data, the story for the dis­ease bur­den is con­sis­tent with the much-less-than-alarming mor­tal­ity data. I went back to the ear­li­est report of the National Health Sur­vey data avail­able on the ABS site: as it hap­pens, the 1995 sum­mary report that con­tained the last mea­sured BMI data. That report also con­tains some ‘com­par­i­son’ data on dis­ease preva­lence from 1989–90.

Here’s a sum­mary of what I found:

The dia­betes inci­dence seems to have risen sharply—possibly because of more active diag­no­sis and because the deter­mi­na­tive blood-sugar thresh­olds have been low­ered in that time—but the inci­dence of cir­cu­la­tory dis­ease has appar­ently fallen (while every­one was get­ting fat).