Research: High Density and Overweight Adolescents

I typically don’t report on scholarly studies in this blog but recently came across a nicely designed study that makes a larger point—that the links between health and place are complex. Fei Xu, JieQuan Li, YaQiong Liang, ZhiYong Wang, Xin Hong, Robert S Ware, Eva Leslie, Takemi Sugiyama, and Neville Owen have produced a report  on the Nanjing High School Students’ Health Survey, published in the Journal of Epidemiology and Community Health in 2010 (64, 1017-1021, http://jech.bmj.com/content/64/11/1017). Titled: “Residential density and adolescent overweight in a rapidly urbanising region of mainland China” the study used data collected from 2,375 adolescents in 2004 to examine the relationship between (large-scale) district level densities and weight.

New high rise behind village
residences in suburban Shanghai, 2010

Densities are reported per square kilometer but in range from 5 persons per hectare to 307.  The study divided the sample into high, medium, low density residents. Youth in the middle and higher density areas (that is over > 35 person per hectare) were roughly  twice as likely to be overweight.

The relationship was reduced a bit but still significant after controlling for TV time, study time, recreation, age, gender, and parents education. Physical activity data came from self reports through a version of the International Physical Activity Questionnaire (IPAQ) that only asked about some kinds  of activities. Given a lack of data on overall physical activity and food intake the authors speculate on a number of reasons for this difference e.g. youth in higher density and potentially higher income areas eating more, higher density areas having less recreational space, or (and here I am paraphrasing quite loosely) that higher density areas may just be too convenient with too much internet access so youth don’t have to expend much energy getting places. The high densities in China are also quite high compared with the US and Australia in particular, where much research has been done. They are also increasing rapidly—the authors report overall densities in Nanjing, presumably including some rural districts, increasing from 14 persons per hectare in 1997 to 23 in 2007.

The study is interesting because in many studies of adults, those in higher densities walk more for transportation (though they may not walk more overall) and some find they are thinner (though not all studies measure this or find it to be true, including my own). In this study of youth, with a large group and fairly good measures, those in higher densities are chubbier. While it is important not to make too much of one study, this is yet another example of the complexity of the relationship between health and place, and the importance of social factors.

Tools: The Pedestrian and Bicycling Survey (PABS)

I’ve recently been involved creating a new survey for measuring transportation behavior, particularly walking and cycling.  The Pedestrian and Bicycling Survey (PABS) is a mail out/mail back survey designed to be an inexpensive means for local governments to learn about nonmotorized transportation use in their communities.

An important component of the design of the PABS was creating a probabilistic sampling approach that would be relatively straightforward to administer and, if desired, could be carried out in house (within municipal agencies). While other sampling approaches—such as snowball sampling across the internet—can achieve a large number of responses, the probability of any person being asked to take the survey is not known making it a challenge to generalize from the sample to the wider population.

Pedestrians in suburban Hong Kong

Documentation includes a report outlining how the survey was developed and the results of reliability (repeatability) sampling. A manual provides step by step guidance about how to use the survey too. The survey is provided at the end of both documents and is available in English and Spanish. Materials are available for download at http://transweb.sjsu.edu/project/ 2907.html (scroll down to find the manual).

Future plans include creating a more modular version so users can mix and match sections to suit the questions they need answered. Updates will be available at http://www.designforhealth.net/health/PABS.html.

Tools: The National Collaborative for Childhood Obesity Research Catalog of Surveillance Systems

The National Collaborative for Childhood Obesity Research is a group spearheaded by the National Institute of Health, Centers for Disease Control, and Robert Wood Johnson Foundation. The Catalog of Surveillance Systems is a new product, also sponsored by the USDA. The catalog provides a searchable database of ongoing monitoring systems related to health. It includes both public systems such as the Census of Agriculture and National Vital Statistics System and private ones such as the Nielsen Homescan and InfoUSA.com. In all, 77 systems are represented.

The database can be filtered by keywords related to the level (person to community to policy), scope (local, state, national), key health variables, age groups, ethnicity, research design, and cost to use.

Each system has an individual entry including information on distinctive characteristics, sampling, key variables, costs, whether information is linked to geographical databases or to other surveys, example publications using the data, and other resources.

This is an extremely useful database and can provide a quick point of entry for those interested in assessing what data are available.