The Scottish Government

« Previous | Contents | Next »

Listen

CHAPTER FOUR: ARE ENVIRONMENTAL INCIVILITIES IMPORTANT FOR HEALTH?

4.1 In the previous chapter, we showed considerable differences between those living in different types of area in the perceived incidence of environmental incivilities. However, the importance of this apparent inequity depends, at least in part, on the consequences, if any, of exposure to these incivilities. If, for example, such exposure has an adverse effect on people's health then this would provide one good reason why the reduction of environmental incivilities should be an objective of public policy. This is the question we address in this chapter. In the subsequent chapter we look at the relationship between environmental incivilities and people's feelings about the community in which they live.

4.2 As indicated in Chapter One, there is more than one way in which environmental nuisance might affect health. One possibility is simply the physical consequence of exposure to toxic or infectious substances. However, the particular focus of this study is on the extent to which exposure has a psychosocial effect on health by, for example, influencing levels of stress or anxiety, or by affecting health behaviours. It is on this possibility that this chapter concentrates.

Psychosocial health and the local environment

4.3 A number of previous studies have shown that individual health is associated with the extent to which residents perceive their residential environment to be unpleasant or threatening (Sooman and Macintyre, 1995; Collins et al, 1998; Molinari, 1998; Ellaway and Macintyre, 2001; Ellaway et al, 2001; Cohen et al, 2003). In this study, we aimed to capture feelings of anxiety, depression, and self rated general health, using questions that have been asked by previous surveys. First of all, we asked respondents, 'During the past twelve months, how often if at all have you been bothered by feelings of anxiety?' and, 'During the past twelve months, how often if at all have you been bothered by feeling sad or depressed?' In both cases respondents were asked to choose one of five possible options: 'very often', 'fairly often', 'sometimes', almost never', and 'never' with an additional option of 'can't choose'. In addition respondents were asked to rate their own health compared to someone of their age and were asked to choose one of the following options: 'very good', 'good', 'fair', 'bad' or 'very bad'.

4.4 Furthermore, as well as looking at people's feelings about their health, we also asked respondents two questions about their health behaviour. The first was whether they currently smoke or not, so that we could establish whether people who are more stressed by their environment are more likely to smoke (Ellaway et al, 2000). The second was an indicator of physical activity; respondents were asked whether they had taken a walk of around a mile or more during the last year, including walking for pleasure as well as to and from work or the shops. Our hypothesis here was that those who do not find their local environment congenial are less likely to want to walk in it (Ross, 2000; Ross and Mirowsky, 2001).

4.5 In order to examine the relationship between these indicators of health and perceptions of the local environment, we again combined our more commonplace incivilities into three groups - street level, infrastructural and absence of goods - in accordance with the factor analysis described in chapter two (see Table 2.2). For each of these three groups of incivilities we calculated an additive index score using exactly the same procedure outlined previously in chapter three; thus in each case a low score indicates a more positive assessment of the local environment. For ease of presentation, we have then collapsed respondents' scores on these three indices into quartiles. Those with the lowest 25% of scores comprise the first quartile, and are the group with the most positive view of their local areas on that set of incivilities. Those with the next 25% of scores comprise the second quartile, etc.

4.6 Table 4.1 shows the resulting analysis of the relationship between our indicators of health and the street level incivilities scale. Thus, for example, in the first column we can see that amongst those with the lowest 25% of scores on the scale (that is, those least likely to regard street level incivilities as a problem), 11% said that they feel anxious very or fairly often. In contrast this figure rises to 20% amongst those with the top 25% of scores (that is, those most likely to consider incivilities a problem). This pattern is typical. Levels of depression are highest amongst those reporting most problems with street level incivilities, while this group is also a little more likely to say that they smoke and that they are not in good health. They are not, however, less likely to report having taken a walk of a mile or more over the last year; indeed if anything the opposite is true.

Table 4.1 The distribution of perceived 'street level incivilities' and health

Notes to table:

Percentages do not necessarily add up to 100 due to rounding.

4.7 A similar analysis of the relationship between infrastructural incivilities and our health indicators is shown in Table 4.2. While those with the highest levels of incivilities are the most likely to report anxiety, depression, poor health or smoking, in each case the difference across the four categories of our table is small. For example, those with a high score on our scale are just four points more likely than those with a low score to say they are anxious very or fairly often, less than half the equivalent difference of nine points in Table 4.1. Meanwhile once again we discover that those who report the highest levels of incivility comprise the group most likely to say they have taken a walk of a mile or more.

4.8 There are, however, rather bigger differences when we undertake the equivalent analysis of our absence of goods scale (see Table 4.3). Indeed, they are larger differences than we saw in respect of street level incivilities. So there is, for example, no less than a 14 point difference between those with a high score and those with a low score in the frequency with which they report being anxious very or fairly often. Moreover, in contrast to our two previous analyses, in this case those who report the highest level of incivility are slightly less likely to say they have taken a walk of a mile or more. Given that one of the questions in this scale asks about the availability of somewhere pleasant to walk, it would be indeed surprising if this were not the case.

Table 4.2 The distribution of perceived 'infrastructural incivilities' and health

Notes to table:

Percentages do not all add up to 100 due to rounding.

Table 4.3 The distribution of perceived 'absence of goods' with health

Notes to table

Percentages do not all add up to 100 due to rounding.

Multivariate analysis

4.9 It appears then that there is some relationship between our health indicators and the reported incidence of incivilities, at least so far as street level incivilities and absence of goods are concerned, that is the two sets of incivilities which, in Chapter Three, showed the most marked inequities in their distribution. However, this analysis has not taken into account any of the other possible influences on health such as gender, age or socio-economic position. We have already seen in the previous chapter that age and socio-economic position are correlated with the degree to which many incivilities are reported as problems. Perhaps once we have taken these into account we will find that most of the differences in Tables 4.1 to 4.3 are not significant.

4.10 To account for this possibility we ran a series of logistic regression models in which gender, age and social class have been included as well as each of our three incivilities scales, again divided into quartiles. For this purpose we have divided our three five-point measures of anxiety, depression and general health into binary variables. Our models thus estimate the odds of feeling anxious or depressed at least sometimes rather than never or almost never, while in the case of general health they show the odds of someone saying their health is no better than fair rather than very good or fairly good. The results of the models are reported in the form of odds ratios, which show the impact of being in the category in question as compared with those in the base category. An odds ratio of two, for example, means that people in that category ( e.g. a high incivilities score) are twice as likely than are those in the base category to experience whatever is being measured as the dependent variable ( e.g. feeling anxious 'very/fairly/sometimes').

4.11 This analysis confirms that those with the highest level of street level incivilities do suffer from poorer health than do those with the lowest level. As Table 4.4 shows, they are twice as likely to report anxiety, 1.8 times more likely to say they get depressed and 1.7 times more likely to say they are not in good health. All of these differences are statistically significant. Those with the highest level of street level incivilities are also 40% more likely to say that they smoke. However, as we might have anticipated there is no significant relationship between having taken a walk and our street level incivilities. We should note too that even in the case of our other four measures, those in the third, 'higher medium', quartile are not significantly different from those with the lowest score while those in the second 'lower medium', quartile sometimes are. The relationship is thus not a straightforward linear one.

Table 4.4 Health and street level incivilities after controlling for sex, age, and social class

Note to table

ns: not significant at the 5% (0.05) level. Odds ratios show the impact of being in that category as opposed to the reference category (shown in brackets)

4.12 Equally, our multivariate analysis confirms our earlier impression that there is little relationship between infrastructural incivilities and health (see Table 4.5). Only depression is significantly more commonly reported by those with a high score on this scale as opposed to those with a low score. Meanwhile we find that those with a high score are actually more likely to have taken a walk. It may well be of course that people become more aware of the problems of their local environment as a result of walking through it.

Table 4.5 Health and infrastructural incivilities after controlling for sex, age, and social class

Note to table

ns: not significant at the 5% (0.05) level. Odds ratios show the impact of being in that category as opposed to the reference category (shown in brackets).

4.13 Finally, our multivariate analysis confirms our earlier conclusion that the relationship between the health indicators and absence of goods is particularly strong (see Table 4.6). Those who report the highest levels of such incivilities also report significantly higher levels of anxiety, depression and poor general health, while they are also significantly more likely both to smoke and not to undertake a walk of a mile or more. In the case of anxiety, for example, those with the highest scores are two and half times more likely to say they get anxious, even higher than the equivalent odds ratio of 2.0 in respect of street level incivilities. Indeed all of the odd ratios for those with high scores are higher in Table 4.6 than are the corresponding figures for street level incivilities in Table 4.4.

Table 4.6 Health and absence of goods after controlling for sex, age, and social class

Note to table

ns: not significant at the 5% (0.05) level. Odds ratios show the impact of being in that category as opposed to the reference category (shown in brackets).

Conclusions

4.14 It appears that 'street-level incivilities' and perceived absence of 'environmental goods' are related to health and health behaviours. Those who experience such incivilities are more likely to suffer depression, anxiety and poor health, thereby appearing to confirm the psychosocial impact of environmental incivilities on health. They are also more likely to smoke. This pattern is, however, largely absent so far as 'infrastructural problems' are concerned. In line with earlier chapters, our results underline the relative importance of street level incivilities and environmental goods to people's wellbeing. These are of course the very incivilities to which some sections of the population are far more likely to be exposed than others.

4.15 However, there is a need for some caution in interpreting our results. What we have shown is the existence of cross sectional associations between health and perceptions of the local environment. This does not necessarily prove that the latter causes the former. In particular, it is possible that those with low self reported health and who feel their environment is poor may simply be reflecting a general underlying pessimism i.e. they feel negative about all aspects of their lives (Stansfeld et al, 1993). However, this potential bias is less likely to account for the relationship we have uncovered between experience of incivilities and smoking behaviour, as smoking behaviour is a reported action rather than a subjective assessment of well-being. Moreover if this potential bias were simply to account for our results then we should not find the association between our indicators of health and infrastructural incivilities is weaker than the association between our indicators and both street level incivilities and absence of goods. It thus seems wisest to conclude that high exposure to incivilities does indeed have some adverse impact on health.

« Previous | Contents | Next »