Distance walked to rail stations on an average weekday for all trip purposes, Sydney GCCSA (source data: NSW BTS)

Here’s a finding that goes against the common wisdom that public transport users are healthier because they walk more. It’s an extract from a new working paper, Transit makes you short, by University of Minnesota researchers, Alireza Ermagun and David Levinson:

(The) effect of transit mode share and accessibility by transit on general health, body mass index, and height are investigated, while controlling for socioeconomic, demographic, and physical activity factors… (We) found that the transit mode share and accessibility by transit are not practically significant, and the power of large-sample misrepresents the effect of transit on public health.

Note this finding doesn’t mean you and I won’t get more exercise from walking to the bus stop or the rail station rather than (say) driving. What it throws doubt on is the claim that the walking inherent in public transport delivers important health benefits across the whole community and, accordingly, that policy on public transport should take those benefits into account.

The link between the physical environment and health outcomes like obesity and diabetes at the social level is a contested issue; I’ve discussed it a number of times before (e.g. see herehere and here) .

Unfortunately the paper doesn’t provide enough information to assess if the methodology is reasonable; but it prompts an intriguing question. Assuming for arguments sake that the authors are right, how could it be that greater public transport use doesn’t translate into better community health outcomes? (1)

One reason might be that public transport users are skewed toward the young, who tend to be healthy anyway in spite of their sometimes careless habits. A few hundred or even a few thousand metres walking per day mightn’t make a measurable difference to their health status.

Another might be that for many users the amount of walking involved in getting to their nearest stop isn’t that much. Dedicated public transport users tend to select housing and jobs near stops; for example, 64% of train passengers in Sydney who walk to the station cover less than 500 metres one-way and 79% less than 1,000 metres.

Even a 500 metre walk in each direction only involves around 1,250 steps a day relative to the recommended daily minimum of 10,000. Bus and tram stops have even smaller catchments (buses carry more people than trains in Sydney). Some travellers “power walk” their way to the station/stop; others adopt a more relaxed gait.

Or it might be that the extra time required to travel by public transport gives users less time for for active recreation (e.g. playing with the kids, gardening) or for intensive exercise (e.g. gym, sport). The average train trip is considerably longer than the average car trip for all purposes e.g. see Is driving quicker than taking the train?

Perhaps another explanation is travellers with poor eating habits are over-represented among public transport users. For example, students tend to eat fast food because they’re often out and about and have limited cash. In the US, this tendency would be reinforced by the high use of transit by very low income groups.

The authors note that good access to public transport might lead to some short trips that would otherwise be taken on foot being replaced by train, bus or tram i.e. some travellers walk less in total when there’s a faster option available, like the frequent free trams in Melbourne’s CBD.

For example, see this comment on an unrelated topic by reader Waffler.

I have a yearly Myki and find it is the same with PT. I don’t care how much I use it and hop on the tram or bus to travel only a stop or two if it is handy (yeah I know I am lazy!) because it doesn’t cost me any extra.

The authors also suggest the exercise benefits of walking to the station might be offset for some people by the mental stresses – e.g. greater anxiety – of travel by public transport. This might also be linked to poor eating habits.

Finally, it might be that the sorts of destinations that driving makes accessible provide on average a degree of walking and exercise that offsets some or all of the foregone walk to the local station e.g. regional shopping malls, regional parks, beach. Not all drivers have basement car parks at work either.

These are only conjectures and some seem more plausible than others. As noted, I can’t say if the paper supports its findings satisfactorily or not. This is only one study so it wouldn’t be prudent to put too much weight on it.

My view – which I’ve set out a number of times before – is that far too much weight is given to the influence that the physical environment has on health outcomes.

As I noted here, I think Henry OvermanDiego Puga and Mathew Turner were right when they pointed out in their paper, Fat city: questioning the relationship between urban sprawl and obesity, that sorting rather than causation is the primary mechanism that drives observed differences within cities on many socioeconomic variables.

It follows immediately from our results that recent calls to redesign cities in order to combat the rise in obesity are misguided. Our results do not provide a basis for thinking that such redesigns will have the desired effect, and therefore suggest that resources devoted to this cause will be wasted. The public health battle against obesity is better fought on other fronts.

Update 26 August 4:30 pm. This is how silly it gets; just published today: Miracle tram to cure obesity.


  1. The limited information on methodology is explained by the subtitle: On health impact assessment of transportation and the built environment. The paper’s key purpose is a (unflattering) critique of the methodological practices of many researchers in this field. If you know what p-hacking is you might want to read it.