POVERTY:

Two different neighborhoods within the same city exhibit strikingly different levels of public smoking, drinking, and physical activity, much more than that predicted by survey and census data, with possible implications in perpetuating health disparities along socioeconomic lines.

In so-called developed nations, lower socioeconomic status (a polite term for poverty) is correlated with increased smoking and drinking, as well as decreased physical activity. The likelihood of successfully adopting healthy behaviors is reduced in this group of people, as is that of successfully rejecting unhealthy behaviors.

Fully elaborating the reasons for such socioeconomic discrepancies may help reduce these differences. The research of Daniel Nettle (Newcastle University, United Kingdom) brings us closer to this goal.

Previous research suggests that the actions of those around a person strongly influence this person's behavior (e.g. the 2008 confirmation of the broken window theory). For example, seeing others littering increases the likelyhood of further littering.

This may extend to public health behaviors, but research on this question tends to focus on surveys rather than direct observation. Consequently, much pertinent information, such as indoor vs outdoor behavior, and behavior in private relative to that with company, is either not recorded or not fully reliable.

Furthermore, conflicting variables (e.g. differences in neighborhood geography) add uncertainty to the analysis. These limitations must be addressed in order to fully understand (and remedy) socioeconomic health disparities.

The experiments.

Dr. Nettle's study sites were in Newcastle upon Tyne (United Kingdom). The two sites were of similar ethnic composition, centrality, layout, and size, with a shopping street backed up by residences (a full elaboration and quantitation of similarities as well as differences is given in Table 1 of the original manuscript, cited below, open access).

The big difference between the two neighborhoods, less than 3.5 miles apart, is socioeconomic status. One neighborhood had 74% of residents in the top socioeconomic class (defined by census), and in the other, 16% of residents.

Using a digital voice recorder, Dr. Nettle recorded information on everyone close enough to identify (not within buildings). The information of interest was gender, age (adult, child, baby), and whether they were smoking, drinking, or running (the same people were not recorded more than once).

Dr. Nettle recorded his data over the course of mid-April to early July of 2010, obtaining data for every minute covering the 9AM to 9PM timespan. Different 30 minute time periods were recorded each day (arriving at a "composite day").

However, the same time frame observed in one neighborhood was observed in the other neighborhood at most four days apart. In summary, this procedure averaged observations over a wide range of days without biasing the results (further details are provided in the original manuscript, cited below, open access).

Dr. Nettle calculated expected health behavior disparities based on a national survey conducted in 2008 and a census conducted in 2001. As previously noted, surveys may be unreliable, and it was of interest to compare them to direct observations.

Observed health disparities.

The major results were that 3.4 times more adults were observed smoking, and 4.3 times fewer adults were running, in the less privileged relative to the more privileged neighborhood. Furthermore, a child in the social group had little correlation with smoking frequency in the former, while it was correlated with over 10 times less smoking in the latter.

No adults were drinking in the more privileged neighborhood, so percentage differences were not calculable. Notably, such unhealthy public health behaviors in the less privileged neighborhood were always at least twice that predicted based on survey and census data.

The socioeconomic discrepancies are not explainable by differences in the number of people observed. While it's not surprising that the discrepancies exist, they are much more than what non-observational data suggests.

Implications.

Dr. Nettle estimates that a typical person in the less privileged neighborhood may observe thousands more people smoking and drinking, and a small fraction of the people running, over the course of a year relative to a person in the more privileged neighborhood. It's reasonable to expect that this greatly influences lifestyle choices, and may help to entrench socioeconomic health disparities.

In my opinion, the best way to get around this challenge is to redistribute wealth more evenly. Unfortunately, this is unlikely to happen in the near-term, suggesting that public health officials must somehow counter the destructive feedback contributing to reduced health among people who just happen to make less money than others.

NOTE: Dr. Nettle received no funding for his research.

Daniel Nettle (2011). Large Differences in Publicly Visible Health Behaviours across Two Neighbourhoods of the Same City PLoS ONE, 6 (6) DOI: 10.1371/journal.pone.0021051