Mapping North America

Ok, now I'm really excited.  One of our central goals here at MPI is to create integrated North American (that is Canada and the US) data-sets. And in our ongoing work, especially in developing the new Canadian edition of Who's Your City? we've been developing maps of North American data on various regional economic and demographic measures. So was I more than delighted to see the map above developed by David Eaves based on an original map I posted from calorielab (via strange maps and Andrew Sullivan). Eaves comments:

If Canadian provinces were ranked along side US States, they would rank 1st (BC), 2nd (QC), 3rd (ON), 4th (AL) and tied for 5th (MB) (YK) as the least obese provinces/states. Colorado would be the first American state placing 7th, with the provinces of NS in 8th and SK in 9th. PEI and NB would appear 15th and 16th and NL would appear 19th. NWT and NU would close out in 30th and 31st position. You can see the original chart at the bottom of this page. Actually even some of the grimmer looking patches of Canada’s map have a silver lining. The Arctic Territories, specifically Nunavut (NU) and the North-West Territories (NWT), appear obese and thus unhealthy. However, Statistics Canada notes that obesity criterion for Inuit populations should be more relaxed since a high BMI does not appear to have the same health risk for Inuit as for non-Inuit.

Comments

[Padraic Ryan]

Wow - my complaint about the last map is it didn't control for wealth (that an obesity per GDP would be more interesting). But given what we know about American vs Canadian GDP per capita, the answer is quite clear...

[Matt]

It's one thing I really noticed when I lived in the States -- statistics end at the 49th parallel and the Rio Grande.

I thought of it again with your recent Allstate accidents post. Sure, Detroit may be the safest place to drive in the US, but it's probably not remarkably safe in any other context. (For example: road mortality in the UK is 5.5 per 100,000; getting down to that level in the US would save 27,400 lives a year. The Allstate report talked about safety factors but didn't even mention one of the biggest: seatbelt use.)

There are plenty of international comparisons in which the US will place well. It's the ones where Americans don't seem to realize they should be aiming higher that scare me.

[Zoe B]

Could you tweak the map to show the stats per mega-region? I'm going to guess that some rural areas will show up towards the red because many of them are full of folks who still eat like old-time farmers, although they no longer need the calorie load to do the job.

[Chris]

It's the border states/provinces that are shocking. You can attribute high obesity in the southern states to regional diets and lifestyles, but there are huge differences between Alaska vs Yukon, Washington vs BC, Manitoba vs North Dakota, and Ontario vs. Michigan. The only two that are close are New Brunswick vs Maine.
Anyone care to guess why Washington State is so much fatter then BC? The only thing I can think of is food portions in Seattle seem to be twice the size of those in Vancouver.

[Whitney Gunderson]

If the United States wasn't so fat, this guy would be out of business....

http://www.nctc.net/counties/harvest/kurtk.jpg

[Michael Wells]

Chris,

I can think of a few possibilities.

1. Universal health care, including preventative.

2. We're not really talking Seattle and Vancouver but rural Washington vs rural BC (or are the proportions of urban/rural different?). However, this doesn't probably hold up for Alaska or North Dakota which have no large cities.

3. As you say, the serving sizes and the mindset that they reflect.

[bh]

It's a great start, but the obvious question is why our neighbors to the south are not included?

[Michael Wells]

I sent the link to a friend who's a nutritionist and here's what she said:

"Obesity directly correlates with socio-economic class--the poorer people are, the heavier. I'm not sure where they got their data, but that would be the general underlying factor.

The southern US has the highest incidence and those states are the poorest.

The way he simplifies "They are the fattest and therefore the least healthy" is a myth. Weight is not an independent risk factor for disease. People who are big but fit have no higher risk of premature death than smaller people who are fit. And there is no way to factor out other effects of poverty--lack of medical care, cigarette smoking, cheap (fast) food, etc. in any data on morbidity or mortality."

[Jeff]

Frankly, I can't tell a difference between Ontarians (in southern Ontario) and US Midwesterners or even up-state New Yorkers. Perhaps Toronto is the land of skinny Cannucks, but from what I've seen in the Niagara Region and Hamilton, there is no visible difference north or south of the border.

My anecdotal observation might reflect the more working-class and working-poor character of this end of the Golden Horseshoe. However, it might also reflect bias in the underlying data. In Canada, regardless of the province, there is some form of universal healthcare. In theory, everyone's weight is recorded in Canada, though in practice I suspect it is is merely a large majority of the population. Still, these data would be biased by excluding those who for whatever reason refuse to enter the medical system. This is probably not the main problem with these data.

Instead, I suspect that the US data are much more biased. In the US, these data are not collected by any centralized authority, no? If these data rely on hospital intake records, then it would include a disproportionate number of people whose health is poor...after all, why else would they visit a hospital? And I suspect a disproportionate share of people who have poor health are also obese. Or are these US data drawn from a representative random sample? I'd trust that more (well, optimally we'd have population data to draw on, but hell will freeze over before my paranoid countrymen would ever accept some centralized system for collecting health information administered by the Federal government).

Can anyone speak to these data sources?

[Matt]

Jeff, you got me curious. And I do agree with one of your theories...

But I don't think it's a methodology difference. First, they appear to both be telephone surveys ( http://www.cdc.gov/brfss/about.htm and http://www.hc-sc.gc.ca/fn-an/surveill/nutrition/commun/index-eng.php ). And at least at the national level, the findings have been confirmed by a survey that used the same methods across both countries ( http://www.statcan.ca/english/freepub/82M0022XIE/82M0022XIE2003001.htm ).

But I do think you're right about regional differences within Ontario. The obesity rate in Hamilton (20% of males, 16% of females) is noticeably higher than in Toronto, where it's 10% for both (Table 105-0409, http://tinyurl.com/6aog2o). The Niagara region does a pretty brisk tourist trade, so it's also possible people you saw were a mix of Americans and Canadians.

[Whitney Gunderson]

bh - Mexico is not included because of a lack of data. If you can help with this, see this blog post....

http://creativeclass.typepad.com/thecreativityexchange/2008/07/fatness-index.html

[Gary Dare]

The big surprises to me are a) WA and OR are 'fatter' (expected pinker) while b) WI, IL and maybe MN are 'thinner' (expected redder). And more light green rather than dark green for Canada.

(Note that I have lived in five Canadian and three American cities, seven of them in the latter half of my life. Right now figuring out return to one of the eight or on to a ninth.)

[Michael Wells]

Gary,

You probably lived in inner city Portland, which is more into exercise and healthy foods. Go out to Gresham (Eastern working class/low income suburb), or even further to the rural East or Southern Oregon and there are lots more heavy folks.

[Gary Dare]

Michael - Yes, I'm downtown by PSU and tend towards the SE (Hawthorne, Belmont) and NW (Pearl, 21st, 23rd, Thurman) socially so that's a younger, fitter and/or upscale crowd but I work down in Wilsonville and go out to Hillsboro and Beaverton on business, and a regular at PDX. Overall, Portland is still a lot fitter than Chicago (and their respective states) so on the map, the overlap should actually indicate Portland towards the left end and Chicago to the right end of that color band. (:

And my Chicago, these days, tends towards the lakeshore from Highland Park down to Hyde Park, with few reasons to go further inland except for business.