Walking and Obesity: The City Life vs the Country Life

May 12 2010 Published by under Health Care/Medicine

Sci rather wishes this study were done in mice, if only so she could write "the city mouse and the country mouse" in her title. But it was done in humans, which was really probably a good thing.
This post has some background. Sci was sitting around with her lab one day, shootin' the breeze like you do when it's Friday and science has you cross-eyed, and we were talking about going to meetings in exotic locales. We were talking about one especially large city, and one person in the group said "you know, what's amazing about cities is how THIN everyone is". And everyone in the group nodded sagely and said it was always shocking to go to large cities and realize how skinny everyone was.
But, being a scientist, Sci wondered...is this really true? Are people thinner in the city and thicker in the country, and if so, why? Is it just our perception, due to some cities featured on TV being full of starving artists and lots of plastic surgery? Is it universal? Is it just among the upper class? What about the urban poor? And is it due to all the walking, or is it just because of increased populations of starving artists?
And that's not all. Sci just hopped up and moved to a new, Very Large City, and for the first week or so, my feet DEFINITELY noticed the walking increase. The blisters on my feet especially. Apparently every other female on the planet knew about those gel insoles to help you wear pretty shoes, but Sci apparently flunked that part of her "how to be a girl" exam.
So relationships between physical activity and obesity? Sci shot off an email to Travis of Obesity Panacea. Travis sent her a paper citation. Let's do this.
ResearchBlogging.org Frank et al. "Obesity relationships with community design, physical activity, and time spent in cars" American Journal of Preventative Medicine, 2004.

(Put on your walking shoes! Sci hates those obligatory pictures of obese people next to pictures of people who are obviously models. This one is better.)

The hypothesis behind this study was simple. Are people more physically active, and are they less obese, when they live in environments which promote physical activity. This isn't just living next to a gym or a park. Rather, it's living in a place that facilitates activity. This doesn't have to be extreme, it's as simple as being able to walk places.
You might say, being able to walk places IS simple! Not always. In the case of many rural and suburban neighborhoods, walking is something you do specifically for leisure or exercise. You don't walk TO places, because you CAN'T. Every place you want to get to is a good 15 minute drive. In contrast, in urban settings it's often easy to walk places, things are nearby, and there are often sidewalks. But not all urban settings are easy to walk in, due to problems like crime and traffic safety.
So for this study, the authors surveyed over TEN THOUSAND people living in and around Atlanta, Georgia. And they looked for some interesting things. While they looked at age, ethnicity, income, and education, they ALSO looked at the number of hours spent in a car, the distance walked, and the LAND USE of the places the people lived. By land use, they checked for what the land was used for (residential vs otherwise), the density of people living there, and the connectivity of the streets. Were there lots of sidewalks? How easy was it to walk around?
And then they put it all together and churned out some correlations.
They found correlations between age and obesity (if you're older you're more likely to be obese), and car time and obesity (the more time you spend in a car, the more likely you are to be obese). They also found negative correlations between education and income and obesity (the more education and money you have, the less likely you are to be obese), and a nice negative correlation between walking distance and obesity (the more you walk, the less likely you are to be obese).
And then they put it all together and mixed it up with some land use data.

What you can see above (if you can get past the cheesy, somewhat difficult background to the figure, yeesh people, that doesn't help!) is the subjects, divided by ethnicity and sex, and curves representing land use. The "mixed use" term refers to when land is devoted to more than one thing (residential and commercial as opposed to purely residential or commercial or industrial). So on the left side of the graph you can see the average mixed land use of 0 (stuff is either residential or commercial, not both, this is relatively average), and as you go rightward on the x axis, you see the land use get more mixed, as in a more urban setting.
The curves are plotted as probability of obesity, and plotted over the land use, and you can see that residents of areas with low mix land usage have a higher probability of obesity than residents of high mix land usage.
Well, now we know land use, but does that necessarily mean that people living in high land mix settings are walking more?

This figure certainly shows that walking, across all demographics, reduces the probability of obesity. Sci would LOVE to see this laid over some land use data so you could see whether the people in the high mix land use areas were walking more and were also less likely to be obese. There was a significant effect of this, but they didn't graph it.
And finally, the rough one:

This figure shoes the relationship between probability of obesity and the number of minutes spent in a car per day. First of all, some of these people were spending 8 hours a day in a car (I really hope they were taxi or bus drivers, but that doesn't make it much better). But regardless, there was an increase in the probability of obesity associated with increased time spent in the car.
So, to recap:
1) The more the land use is mixed where you are, the less of a probability you have of being obese. This is presumably related to walking more, but the correlation was only effective for African-American females.
2) The more you walk, the less probability you have of being obese.
3) The more time you spend in a car, the MORE probability you have of being obese.
Sounds pretty simple, don't it? But this isn't the easiest thing. Many people HAVE to drive to work, and often do not have enough leisure time outside of it to make up the car time with other physical activity. In addition, many people will walk more when they have somewhere to go, and suburban residential neighborhoods don't really go in for that kind of thing. But it DOES provide some interesting data for people looking to plan new residential communities. If you make things more walkable (especially work and necessities), maybe people will walk more, and maybe that will translate to smaller probabilities of obesity and improvements in health. Maybe those people planning those overly picturesque walkable communities are on to something.

(Sci wonders if, when all communities are walkable, everything will magically appear in soft watercolor. We have sepia-toned memories, behold the watercolor future!).
FRANK, L. (2004). Obesity relationships with community design, physical activity, and time spent in cars American Journal of Preventive Medicine, 27 (2), 87-96 DOI: 10.1016/j.amepre.2004.04.011

30 responses so far

  • razib says:

    the disjunction between new york city and say the south is really evident in this. new yorkers walk around a fair amount between public transport nodes, and they walk fast, and if you have an obese friend arrive from a part of the country where this isn't typical...it can cause awkwardness (e.g., had a friend whose high school friend visited, she was obese, and kept paying for cabs or car services to get to transit points!). but, i think there's low hanging fruit even in cities. as you imply, walkability means that driving is facultative, not obligate. but a lot of people still make that choice, and they're usually the obese ones. but as i allude to above, there's the issue of being "trapped" in a particular equilibrium state. an obese person in new york in the summer may simply find it uncomfortable to keep up in the heat with locals, and so make recourse to the cab more often (that's basically what happened in the example i gave above).

  • Anonymous says:

    "Every place you want to get to is a good 15 minute drive."
    That's a ridiculous rationalization. It means that it's probably a 40 minute walk or a 20 minute bike ride. In my mind, whenever you get in the car, it's a choice.

  • simba says:

    Footpaths make a huge difference as well- I walk about 2 hours a day for fun while in the city, whereas in the country (where I live) it's harder to walk simply because you feel like you're going to get knocked down all the time. The boy racers for example tend to pull in and speed up when they see you, just to give you a fright and make the dogs jump into the ditch.

  • ASWatch says:

    @2, I don't know what the average speed limits are in rural US, or what the average walking speed is there, but here in Finland the most common speed limits in rural towns and nearby areas are 50 or 60 kph (roughly 30-40 mph, usually bumping up to 50 mph when you get outside the official town limit, and 70 mph (in the summer) on the main roads), so a 15 minute drive will get you 10-25 km (6-15 miles) away (presuming street lights and heavy traffic are uncommon in rural areas). I'm no slouch on foot but I'm certainly not fast enough to cover that in 40 minutes.

  • IasasaI says:

    "This figure shoes the relationship between probability of obesity and the number of minutes spent in a car per day."
    Typo of the day!

  • Scicurious says:

    razib: I also think it's not just that walking is easier in cities, in many major cities, driving is discouraged. It's really expensive to have a car, it's impossible to park it close to your destination, with the time and money you spent driving and paying for parking, you might as well have walked.
    simba: yes. Anonymous doesn't realize that many people living in suburbs would be walking along on the side of the road if they tried to walk anywhere. There are no sidewalks and people are often driving fast down country roads. In addition, I'm not lying about the spread out bit. Sci's middle school (for example) was a good 15 minute drive across town from where she lived. WAY more than 40 minutes on foot.

  • yud says:

    Where I live, the two nearest (tiny) strip malls are 6 and 8 miles away, along narrow curving 2-lane roads with no shoulders where cars drive 40-50 mph, and along a divided highway for part of the way. There are no sidewalks anywhere, and once you get to the commercial areas, you'll have a nice busy 4+ lane road with no crosswalks to get across.

  • Shoghi says:

    Throughout your post (and presumably the article as well), there's the implicit assumption that the exercise of walking causes a reduction in obesity, which is not necessarily the case. It's definitely a possibility, and usually accepted as naturally intuitive and common sense, but it could also be that people who are obese simply lack the energy, or choose not to walk for other reasons such as joint pain. Just because neighbourhoods are more walkable doesn't mean people are going to walk them. The limitations of correlation strike again.

  • Eric Lund says:

    @2,4: In a typical US suburban area you will encounter traffic lights and moderate traffic, so a realistic average speed would be around 20 mi/hr. Fifteen minutes of that gets you five miles away. A typical pedestrian walks between 2.5 and 3 mi/hr, so even if you had useable sidewalks (often they don't exist, and even when they do they may not be maintained in winter) you would be looking at 90-120 minutes each way. On a bicycle averaging around 10 mi/hr, you are looking at 30 minutes each way, and that assumes that (a) a safe bicycle route exists (often you will find yourself on the same roads as cars that are going much faster and may be unwilling or even unable to give you an adequate buffer zone) and (b) you have a safe and secure place to park your bike at your destination (also not a given).
    As for rural or semirural areas: I live in one in the US. The towns neighboring mine are about 4 miles south (10 minutes), 6 miles north (15 minutes), and 12 miles east (20 minutes via roads with higher speed limits). So the estimates given in #4 above for Finland are also roughly applicable to my area.

  • Moopheus says:

    People in New York walk fast? I lived in Brooklyn for 10 years, and that wasn't the impression I had. Most places I went, the average sidewalk speed seemed pretty slow to me. Sure, neighborhoods were walkable, and you could get to most of the things you needed in a short walk--but it was just that, a short walk. People walk more in cities, but they don't necessarily go very far. There's still way more dependence on cars than there needs to be.

  • Dr Becca says:

    People in Manhattan walk fast. I walk fast in Brooklyn, too, because I am usually late.
    It's funny, when I first moved to NYC I thought that I might shed a pound or two because of all the walking (and stair climbing out of the subway!). And I do walk a ton, especially if I'm out and about on the weekends. But one thing people in cities (especially NY) also do a lot is eat out, which usually means you're going to consume more calories than if you'd made yourself a meal at home. So while the walkability of city life may have certain cardio benefits, whether that translates into waistline differences may be another story.

  • *Which* city probably also makes a difference. Some cities I've lived in are really pedestrian-friendly with broad sidewalks (to accommodate all the walkers!) and frequent crossing lights. As razib alluded to above though, in some cities there really isn't a cultural expectation that people get out of their cars. My personal experience suggests that very hot cities as you might find in the south, and very car-obsessed cities such as LA are both prone to this sort of cultural anti-walking mentality. So although things are technically within walking distance, the absence of sidewalks and crossing lights (and prevalence of asshole drivers) along one's route to the grocery store necessitate bodily peril should one choose to walk or pedal rather than drive. (Being a bit of an adrenaline junkie, I do take my life into my own hands and walk or cycle to work. Makes me feel alive every time I get run off the road.)

  • Anni says:

    I definitely notice this. We live in Chicago, but were raised in the suburbs where our parents still reside. Whenever we have friends or family come to visit they say "it's HOW far of a walk?" and then pay for a cab.
    We actually have a car here in the city, but it's not for trips within the city. Unless we're really running late and are going somewhere we know there's free parking (which is rare) or need to fill up our 5 gallon water bottles. We use it more to leave the city for special trips. Even with that, walking/public transportation is just easier here. If you want to go to the Loop from our neighborhood, the time on the train may be longer than the time in a car, but walking off the train versus finding parking? No comparison.
    On other thing that may be worth looking into - the general health habits of people who live in the city. Here we have just as many Whole Foods as other grocery stores. There are vegan and vegetarian options everywhere. For every fast food staple, there is a different, fresher option. We don't go through drive throughs because, well, we don't have them. Our recreational activities tend to be whatever free ones are available - so while our suburban counterparts may go sit in a movie theatre, we run along the lake, walk through Lincoln Park Zoo, or go to the dog park.

  • red rabbit says:

    @ anon #2: So, you live urban or mixed residential commercial suburban. I live in Farmville.
    My nearest supermarket is 18km away, but the issue is: it's 80km/h highways and a 350m climb to come home. Not to mention that 14 foot snowbanks are the norm in winter.
    My workplace is 37 km away, and the average elevation difference is the same again.
    Yeah, getting into the car is not always a choice, and physical activity really is something you have to work into your life if you live rural. Counter-intuitive, but true.
    When I lived in the city (Montreal), it was public transit and bicycling. It was a lot easier to keep the weight down.

  • Kevin says:

    @ anon #2.
    I agree with red rabbit.
    I used to commute into NYC and got a LOT of unintentional exercise, just walking to and from the train station, to and from lunch places, shopping, etc. etc. etc.
    When I moved to a suburban community, ALL of that went away. And I gained 25 pounds...even though I moved away from the city specifically to get MORE intentional exercise (hiking, etc).

  • Ford says:

    It's remarkable that living in a mixed-use area is so much more beneficial than walking 10 km per day, especially since it's presumably easier to eat out in mixed-use areas. Could it be that self-reported walking distances aren't that accurate, while people are quite accurate about where they live?

  • Alteredstory says:

    I've lived in cities where walking or biking were not only possible, they were the easiest ways to get around, both in terms of work and shopping - a 30 minute commute by bike does wonders for one's health.
    I grew up where everything's half an hour drive away, and unlike Anonymous #2, I don't walk at 15-20mph, so it's not a choice between driving half an hour one way and walking an hour one way, it's a choice between driving or two to three hours walking, depending on the number of mountains in the area. Sure, it's a choice, but if you spend four to six hours of your day just on the trip to and from the grocery store, you will be in very good shape and have no time for anything but work and shopping (assuming you're that close to your job).

  • Aquaria says:

    I've lived in areas where you were anywhere from 7 to 20 miles from the nearest town with a grocery store. To walk there, would take either half or all the day, using the shoulders of roads with 60-70 mph speed limits for cars.
    Not no but hell no would I have walked far on them. When I was a kid I'd walk the quarter mile to my great-aunt's house. But that's because I knew that rather than walking on the shoulder of the state highway I had to take, I could cut across my grandpa's cousin's front yard, up the hill along my schoolteacher's side yard, the parking area of the church at the top of the hill, cross the highway to the lot of the weird building that seemed to change owners every six months, cut through the other church's lot and then down the blacktop road where my aunt lived. Heaven help me if the drunk across the road from her was coming home from a bender, which was entirely possible at 11 a.m.
    If I was really adventurous, I'd make lavender sachets for my other great aunt across the street, bum some lemonade and snickerdoodles from her, hike through her back property, sneak through the drunk's back yard (hoping he wouldn't fire at me), and then I'd be at my other great aunt's house. It was faster, but there were those lavender sachets and the drunk's shotgun to contend with.

  • Tim Ellis says:

    "On other thing that may be worth looking into - the general health habits of people who live in the city. Here we have just as many Whole Foods as other grocery stores. There are vegan and vegetarian options everywhere. For every fast food staple, there is a different, fresher option."
    I think this is a frequently overlooked point that bears a lot of thinking about. When I lived in Awfulland, my walkable food choices were pizza and wings, or Dairy Queen. There was a convenience store where I could get microwavable food. And there was a diner that was open for breakfast only. And I lived "downtown" - the rest of our little village was just acres and acres of farms.
    Contrast that with the city I live in now, with vegetarian, vegan, and generally healthy options on most every block, and a bustling farmers market downtown (how ironic that I now find it so much easier to get fresh farm produce than I did when I lived next to 30 farms).
    Nutrition is a big chunk of fitness, and we exercise the options that we have.

  • Hinemoana says:

    A small amount of unintentional excercise can have huge benefits. For two years I rode my bike to and from university. My city is rather hilly and there are virtually no road concessions for cyclists (we technically share bus lanes, but some bus drivers play a game of trying to pass us as closely as possible). But the traffic in my area meant that riding too about the same time as bussing, was cheaper, and counted as excercise (I could not and still connot afford the gym or time to exclusively exercise). I lost five kilos when I started riding. Now I work too far away to ride, and I have gained back that 5 kilos... and another ten thanks to having a more sedentary job too.

  • Silver says:

    I'm with Red Rabbit @#14 and the other rural/agrarian residents. I think there are varying definitions of rural, and I don't always see them broken out in these studies. My definition involves "hauling water in a truck to livestock in the winter" and "buying hay by the 10-ton load a few times per year and stacking it." I am more physically active overall out here than I was in town, but that is because I am feeding and shoveling the products of feeding all the time. My next door neighbor, however, has a total of two Yorkies and drives his 13 year old 3/4 mile to the end of the road to wait for the bus (!). Very different lifestyles.
    By the way, at what point did parents start doing this? I see lots of parents waiting with their kids at the end of the road every day. I understand it when it's -20F, although I remember that I was told to bundle up and not forget my gloves (as if) and take a flashlight... Am I old and cranky?
    When I lived in the city, which has an extensive trail network, I could go 6-10 miles on a different route 3 days per week, and I cross-country skied all the time in the winter. Supporting trails is not a priority for our local government, apparently on the theory that we have quaint backroads. I have about a 2 mile loop on my side roads, and for anything longer, I have to go about 1.5 miles on a 55mph road to connect.
    I would like to be a person who rides my bike uphill 14 miles to the grocery store, but I'm not that person. And, as it turns out, the store is directly on my route home from work. I only bicycle with the dog out here in my 'hood, I would not DARE ride on the actual roads.
    You know, though, my patients in the actual city who do live in some of the few mixed-use areas (newer, western town) and are reliant on (bad) public transportation... are not less likely to be obese (or more likely to be fit, the more useful point for me) than their relatives on the other side of town, which is not mixed-use. At least not in the kinship groups I'm thinking of right now. I'll have to go pull charts. We have patients wearing pedometers and tracking them in diabetes group classes (prizes are involved for everyone who completes this cycle; mind you, the pedos may just be on top of the washer in spin mode for all I know), so I will look at that.

  • Christina says:

    Not only are distances closer in cities, but walking is more tedious in rural or suburban environments. I can - and sometimes do - walk a mile and a half to the grocery store, but it's almost entirely residential neighborhoods. For almost the entire route, nothing but boringly monotonous front yards and probably 95% of the time not meeting one single person on the way. So, I pretty much only do it when I feel the need for a walk to clear my head. In a city, there's a varied environment, lots of different stores and so forth, and lots of people. It feels more comfortable walking with other people doing the same than being the ONLY person on foot.
    Personally, I can't wait to get out of the suburbs and into a city.

  • Aquaria says:

    Nothing can ruin a walking regimen like moving to San Antonio.
    It's tough to get motivated to walk outside when the temp is 100+ at 7 p.m. Well, that's the weather, that can't be helped.
    My current neighborhood has an astounding number of walkers, but if we leave our immediate neighborhood, there are sidewalks on only one side of an extremely busy main street. Worse still, you have to cross the street to have the sidewalk for a while, then cross back to have sidewalks again. Even when you have the s/w, at some points, there's a sidewalk, but no curb, meaning that drivers can drift over to you with nothing to slow them down or jar them at all. Or they can make that right turn very hard, and nearly run over you while you're waiting for them to pass. I have to run backwards not to get hit by a right-turning vehicle at least once a week.
    My other favorite: The crosswalk for one intersection isn't on the side of the street with the nice curb. Nope, it's on the other side of the street, where you have no curb, and must stand in a drainage ditch next to a guard rail. Yeah--that's really safe.
    Who the f plans these things????

  • caia says:

    Question about recap point 1: Maybe I'm reading the first graph wrong, but it looks like mixed land use correlates with decreased likelihood of obesity in black males about as much as black females, with a slightly shallower drop for white males and a shallower drop yet for white females. Was the correlation only statistically significant for black females?

  • Oikoman says:

    I'd be curious to see what the results of this survey would be if applied in Europe or the UK. I've found that walking or biking in rural Holland to be ridiculously easy, thanks to their extensive network of walking and cycling paths. Rural england isn't quite as amenable, but has a network of footpaths and many roads have sidewalks, even in the middle of nowhere. In this situation I suspect what would be seen is a cultural divide on whether or not to walk (or bike) given similar opportunities across different european countries.

  • simba says:

    Oikoman- I think it depends on the country. I'm in Ireland and sidewalks can be hard to find in most rural areas . Towns generally have them, even small ones, but the sidewalk of a small town will only make up a tiny fraction of your journey.
    Good point, Christina.

  • Urban Nostalgic says:

    A second factor that I think matters a lot is stairs. The victorians I grew up in had the bathroom on the second floor (the plumbing stack fed the kitchen on the first), so I had a trip up and down the stairs every time I needed to pee.
    In the spread-out suburbs, everything is one story. I really miss stairs. This is correlated with density, so it's a possible confounding factor.
    I know it's good for wheelchair accessibility, but for those of us whose legs work, using them is the best way to exercise.

  • Greg Laden says:

    Very nice writeup of an interesting study. I started out making a comment, and it got longer, and I looked stuff up, and my comment grew a bibliography, so I made it into a post: PING

  • Jim B says:

    Does anybody else find the curves the graphs a little too clean? Sure, there are 10,000 people in the survey so you get good data averaging. Nevertheless, it seems too pat.

  • Erasmussimo says:

    The data presented in this study is WAY too neat to be believable. Even with 10,000 respondents, I'd expect something other than mathematically perfect curves. These are the best-fit curves, I'm sure, and they probably are close to the data, but I'd still prefer to see the data -- with error bars -- and not just the author's interpretation of the data.
    Also, I think you are conflating country with suburbia. Yes, walking is discouraged in suburbia, but I do an enormous amount of walking around my land. It's 120 feet to the workshop, 150 feet to the duck pen, 100 feet to the garage, 300 feet to the barn, and so on. Just feeding the animals requires walking about a thousand feet, and when I'm working outside I can walk several miles just getting tools, refueling the chainsaw, and so on. Rural is not the same as suburban!
    Also, there are a variety of additional factors to consider. For example, cities attract younger people, while older people tend to live in more rural environments. Since older people tend to be more obese than younger people, the residence patterns based on age affect the obesity patterns.
    Here's another factor to consider: urban residents tend to eat out more often than rural residents (it's much easier for the urban resident to eat out, and the number of restaurant seats per capita seems much higher to me than in rural environments). My impression is that most restaurant meals are more fattening than most home-cooked meals -- although the two bell curves do overlap considerably. In my own case, I have a rule that, if I eat out, I eat only half the meal, taking the rest home, and I have to reduce my caloric intake at home afterwards to counterbalance the caloric load of the restaurant meal.
    I do not believe that this study is sufficient to suggest that it's healthier to live in cities than in rural environments.