Category Archives: Inequalities

Neighbourhood problems lead to depression, but effects vary across countries

Neighbourhood characteristics are linked to mental health in older age, but the magnitude of effects might differ across countries. Using data from 16 different countries, our paper in the American Journal of Epidemiology indicated that country-level social, political, economic and environmental characteristics may explain differences in how neighbourhood affects mental health.

Depression, local area and macro-level context

Depression is a common mental disorder with substantial disability and economic burden worldwide. More than 10% of adults aged 50 years and older present depressive symptoms with an even greater proportion of the population suffering in older age groups. Older adults spend more time in their neighbourhood, so that social and physical features of the residential area become increasingly important for them.

Signs of neighbourhood disorder, such as crime, vandalism and exposure to rubbish, are thought to indicate problems in the residential environment, by elevating the level of stress and fear among residents. On the other hand, positive aspects of neighbourhoods such as social cohesion – which is often characterised as a sense of community, help and support among neighbours – have the potential to buffer the effects of stress and contribute to healthy ageing. Both neighbourhood disorder and social cohesion has been linked to mental health problems.

Neighbourhoods are situated within macro-level environments capturing larger geographic areas such as local authorities, counties or even countries (Figure 1). Policies, as well as social, economic and environmental characteristics of macro environments can affect citizens’ life and health, and also shape the physical and social characteristics of local communities, where people live and age. It is important, as both local and macro-level environments are modifiable, presenting opportunities for improving population mental health and contributing to healthy ageing.

Figure 1: Local and macro-level determinants of depression

Neighbourhoods influence mental health among older adults

We investigated the associations between neighbourhood disorder, lack of social cohesion and depression among adults aged 50 and over. Longitudinal information on perceived neighbourhood characteristics and depressive symptoms across 16 high-income countries were utilised, including 32000 older adults. Results showed that living in an area with significant neighbourhood disorder increased the chance of developing depression (Figure 2), and so did lack of social cohesion (Figure 3). If people were already in retirement, effects became stronger.

Figure 2: Neighbourhood disorder impacting depression across 16 high-income countries. Odds Ratios <1.0 indicate decreased odds of depression by exposure to neighbourhood disorder; Odds Ratios >1.0 express increased odds. Small black diamonds are country-specific estimates, larger empty diamonds are pooled effects across group of countries. Abbreviations: ELSA, English Longitudinal Study of Ageing; HRS, Health and Retirement Study; SHARE, Survey of Health, Ageing and Retirement in Europe.

Neighbourhood effects differ across countries

As the magnitude of the associations varied across the 16 included countries, we further explored country-level differences.

Figure 3: Lack of social cohesion impacting depression across 16 high-income countries. Odds Ratios <1.0 indicate decreased odds of depression by lack of social cohesion; Odds Ratios >1.0 express increased odds. Small black diamonds are country-specific estimates, larger empty diamonds are pooled effects across group of countries. Abbreviations: ELSA, English Longitudinal Study of Ageing; HRS, Health and Retirement Study; SHARE, Survey of Health, Ageing and Retirement in Europe.

Amongst other, we found that in countries with higher population density, lack of social cohesion was more detrimental for mental health. Particularly in the oldest age groups, where limited mobility is more likely present, neighbours can be an important source of social and emotional support. In countries where people live closer to each other, not having proper social ties to neighbours can lead to social isolation and higher risk of developing mental health problems.

Also, in countries with higher pension spending, the adverse effect of neighbourhood disorder on depression was buffered for individuals already in retirement. It is plausible that by providing material resources, more generous welfare states equip older people to deal with stressors arising from less safe and deteriorated residential neighbourhoods.

Macro-level context and policy recommendations

Understanding how larger context can influence mental health inequalities across neighbourhoods has the potential to inform policy, and provide more tailored recommendations. For example, tackling crime and vandalism in countries with lower pension spending would be particularly beneficial for mental health, as older people has less material resources to protect themselves from the negative effects of neighbourhood disorder. Supporting social ties and improving social capital in densely populated areas may stronger contribute to healthy ageing and lead to better mental health among older adult.

By Gergő Baranyi, School of Geosciences, University of Edinburgh

New ESRC funded study on Tobacco and Alcohol

We are delighted to announce that we have been awarded funding from the ESRC for a project exploring tobacco and alcohol environments in Scotland. The project ‘Change in alcohol and tobacco availability, population health and the lived experience’ will be funded for 3 years, beginning December 2019 for a total of £761, 470. The project will be led by Professor Niamh Shortt with co-investigators from the University of Edinburgh (Professor Jamie Pearceand Dr Tom Clemens), Glasgow Caledonian University (Professor Carol Emslie) and the MRC/CSO Social and Public Health Sciences Unit, University of Glasgow (Professor Richard Mitchell).

This research will measure change in the availability of alcohol and tobacco in Scottish neighbourhoods over time and explore how this change relates to health outcomes and how residents experience the availability of alcohol and tobacco in their neighbourhoods. The findings will be important because smoking and alcohol consumption are leading causes of illness and death. In Scotland smoking causes one in every 5 deaths and one in 20 deaths is related to alcohol. This harm is not equally shared; those on the lowest incomes suffer the greatest harm. These illnesses and deaths are preventable. The World Health Organisation recommends that nations prioritise interventions that reduce the supply of alcohol and tobacco.

Why might neighbourhood supply of alcohol and tobacco matter? Research suggests that when there are a lot of outlets in a neighbourhood this impacts upon consumption in three ways. 1. The outlets may be more competitive and drive prices down to attract customers. 2. Oversupply may normalise the products when they are sold alongside everyday commodities, such as bread and milk. 3. Tobacco and alcohol may simply be easier to buy in areas where there are more outlets.

In order to explore the relationship between supply, behaviour and harm we need data on the location of every outlet in Scotland selling tobacco and/or alcohol. Ideally, to be able to say something about whether the relationship may be causal, we need this data over time. We have already collected data on the specific location of every outlet selling tobacco and licensed to sell alcohol over multiple time periods (2012 and 2016 (nationwide alcohol and tobacco – see paper here) and 2008 (alcohol in four cities – see paper here)). As part of this project we will collect updated data for 2019/20. This will allow us to measure this change over time. Using an approach called trajectory modelling we will group neighbourhoods that have had a similar degree of change; some neighbourhoods may have lost, or gained, local shops or pubs, whereas some may not have changed at all. We will then identify features of these neighbourhoods that may be driving this change, for example the age profile of the population or poverty levels. This will help policy makers understand the drivers of change in our neighbourhoods.

To measure the relationship between changing supply and harm we will link these trajectories, and our measures of availability at each time point, to alcohol and tobacco health outcomes (behaviour, illness and death). We will use statistical models to see whether areas experiencing an increase or decrease in outlets have seen a corresponding increase or decrease in these outcomes. This will allow us to get a better understanding of whether an over supply of alcohol and tobacco is related to smoking and alcohol consumption and harm. These findings will provide important evidence related to the provision of such commodities in our neighbourhoods.

Although these statistics are important to report we also need to understand why an oversupply of alcohol and tobacco may influence behaviour and harm. Whilst the literature suggests the pathways listed above, we know little about the experiences of individuals living in neighbourhoods with contrasting availability. We don’t understand the individual experience of any of these pathways. Professor Carol Emslie will lead a qualitative work package and researchrs will meet with groups of individuals, in neighbourhoods of contrasting trajectories, to talk to them about the supply of alcohol and tobacco. We will explore their experiences of neighbourhood and assess how their perceived notions of their neighbourhood availability contrast with our statistical measures. Finally, we will meet with residents, retailers and policy stake-holders to explore potential interventions related to supply. Policies at this level require public, retailer and political support. We will discuss the priorities held by various groups, present our quantitative results and gauge attitudes towards potential interventions.

Throughout the project will be committed to knowledge exchange, public events and speaking with non academic partners. If you wish to know more about this research, or would like one of the researchers to come to your organisation to provide you initial findings (once the project is under way) then contact the Principal Investigator here: niamh.shortt@ed.ac.uk

Children in deprived areas encounter shops selling tobacco six times more frequently than those in well-off areas

By Dr Fiona Caryl.

Our new study  looking at exposure of children to tobacco retailing, recently published in the journal Tobacco Control, shows that an average 10-to-11-year-old child in Scotland comes within 10m of a shop selling tobacco 43 times a week. This rises to 149 times a week for children living in the poorest areas—six times more than the 23 encounters a week experienced by children living in affluent areas. This demonstrates an unexpectedly large inequality in the amount of times children are exposed to tobacco sales. Unexpected because in the same study we showed that tobacco outlets are 2.6 times more common around the homes of children living in the most deprived areas than the least. Yet we found a six-fold difference in exposure because we used GPS trackers (fully consented and ethics-approved, of course) to follow exactly where children moved through their environments. We found that most exposure came from convenience stores (41%) and newsagents (15%) on school days, with peaks before and after school hours. At weekends, we found most exposure came from supermarkets (14%), with a peak around midday.

Why does this matter? This may not sound like a lot of exposure, and we might ask if a child is really exposed to tobacco just by being in or near a shop selling tobacco, especially after the ban on point-of-sale (POS) tobacco displays. But then we don’t actually know how many micro-exposures it takes to make a child think that smoking is a normal, acceptable and widespread behaviour rather than a major cause of premature death. Research into advertising suggests that the mere-exposure to indirect and incidental stimuli can influence attitudesnon-consciously when they’re repeatedly presented. The ban on POS tobacco displays has reduced children’s susceptibility to smoking, but children still notice tobacco on sale. In fact, recent research shows that the conspicuousness and prominence of tobacco in shops varies considerably between areas of high and low deprivation. And the difference in prominence has been increasing since the POS ban.

The difference in the number of times children in poor areas are in or near to places selling tobacco is most concerning when you consider the pathways leading people to start smoking. Most adult smokers start when they are teenagers, and the availability of tobacco products is a key factor in in why people start to smoke and why they find it hard to give-up. Our findings raise important questions about when and where tobacco products are sold and the messaging this is sending to children.

Mental health problems are common among prisoners in low-income and middle-income countries

In a major international review, we found very high rates of psychiatric and substance use disorders among prisoners in low-income and middle-income countries. The results from the systematic review and meta-analysis published in Lancet Global Health showed that the prevalence of psychosis, depression, and drug and alcohol use disorders is much higher in this marginalised population in comparison to the community, pointing to unmet needs and calling for action in research and policy.

Mental health and substance use problems are common among individuals involved in the criminal justice system. Incarcerated men and women often come from disadvantaged socioeconomic and family background, and frequently have a life history of victimisation and substance use, making them more vulnerable to mental health problems. While in prison, they often remain undiagnosed and untreated. Prisoners with unmet mental health needs have higher mortality, especially by suicide, and greater risk of recidivism and reoffending after release in the community, leading to multiple imprisonments.

Although 70% of the worldwide prison population are residing in low-income and middle-income countries (LMIC), almost all scientific evidence is coming from studies conducted in high-income countries. It is an important limitation as recommendations from a mostly Western context might not be applicable or generalizable to poorly resourced settings. Prison conditions in LMICs are usually very harsh, characterised by overcrowding, poor nutrition, and sanitation, and limited or complete lack of access to basic health care. To provide evidence for future research and policy making, we conducted a systematic review and meta-analysis on prison prevalence studies in LMICs.

After screening 6000 titles and abstracts from 17 electronic global databases, we identified 23 relevant publications based on 14,527 prisoners from 13 different LMICs. Considering a one year time interval, approximately 6.2% of the prisoners had psychosis, 16.0% major depression, 3.8% alcohol use disorders, and 5.1% drug use disorders. To illustrate the immense burden of mental health problems, we quantified the difference between the prevalence among prisoners in each sample and in the sex-matched general populations of the respective countries. Prevalence rates among prisoners were 16 times higher for psychosis, 6 times higher for major depression and illicit drug use disorder prevalence, and twice as high for alcohol use disorders, indicating a significant public health concern and large unmet health care needs in this marginalised group. Moreover, rates in prison populations of LMICs might be even higher than in high-income countries.

Based on our findings, we presented several implications and recommendations for research and policy.

  1. While a review from 2012 on prison mental health in high-income countries identified over 100 samples, we were able to find only 23 studies from a much larger and diverse group of countries. There is a need further evidence from LMICs to adequately plan interventions for prisoners with mental disorders, especially from regions underrepresented in research such as Central and East Asia, and Central America.
  2. Because correctional facilities in LMICs often lack basic health care, the implementation of cost-effective interventions and scalable treatments for individuals with mental health problems is crucial.
  3. Imprisonment could present an opportunity to treat people with mental health and substance use problems who otherwise would be difficult to reach for health services. National governments in LMICs should move the responsibility for prison health care from prison administrations to the national health services.
  4. Since human right violations, and physical and psychological abuse are more common in resource-poor correctional settings, increasing mental health literacy among staff and protecting the rights and health of people with mental illnesses should be a priority for penal justice policies.

The invited comment on our paper gives a valuable and very practical recommendation on how to improve mental health services in correctional facilities. Training prison health workers by mental health professionals using the WHO’s Intervention Guide for mental disorders (mhGAP-IG), could be a cost-effective and valuable programme for addressing the treatment gap among prisoners in LMICs and coming closer to the Sustainable Development Goals declared by the United Nations.

By Gergo Baranyi, PhD Student in Human Geography and Marie Sklodowska-Curie Early Stage Researcher, The University of Edinburgh

What is ‘on the ground’ in a city linked to levels of inequality in life satisfaction

In a European-wide study of 63,554 people from 66 cities in 28 countries, we found links between urban design and levels of inequality in life satisfaction. This is the first study to theorise and examine how the entire urban landscape may affect levels of and inequalities in wellbeing in a large international sample.

Cities with an even distribution of facilities, housing and green space were linked with lower levels of inequality between residents’ life satisfaction levels, suggesting that more equal access to a range of facilities and types of land may help reduce the gap in life satisfaction between the most economically-deprived and most affluent residents of a city.

There was a strong link between higher life satisfaction and living in cities with homes surrounded by natural, green space. However, lower life satisfaction was linked to living in cities that had more wasteland, more space dedicated to housing, and more space in which all the land is concrete or tarmacked.

Implications for policy and planning.

The findings of our study suggest that urban planning has a role to play in addressing inequalities in cities. Our finding that more equal distribution of land cover/use is associated with lower levels of socio-economic inequality in life satisfaction supports the idea that city environments could be equigenic – that is, could create equality.

What people want or need from their city varies moment to moment, day to day, life stage to life stage. If a city is varied enough to offer people what they need, when they need it, it is likely to support a higher quality of life.

Three Generation Out Walking

Why understanding city design is important for population health and well-being?

With the United Nations reporting that more than half the world’s population residing in urban areas and this proportion rising, it is important to understand how well-planned urban environment might improve, and reduce inequalities in, quality of life. Therefore, understanding the influence urban environments can have on all aspects of health and wellbeing is increasingly important. Cities are continuously evolving and there is ample opportunity to ensure these are healthier and happier place to live.

What we did:

We applied theory and methods from landscape ecology to explore associations between cities’ land cover/use, residents’ reported life satisfaction and within-city socio-economic inequalities in life satisfaction. We joined individual-level responses to the European Urban Audit (EUA) Perception Surveys (2012 and 2015) with city-level data from the European Urban Atlas classifying land cover/use into 26 different classes. Our sample included 63,554 people from 66 cities in 28 countries.

The study, ‘Are urban landscapes associated with reported life satisfaction and inequalities in life satisfaction at the city level? A cross-sectional study of 66 European Cities’ is published in Social Science & Medicine (Open Access). The work was funded by The Medical Research Council (MRC) and Scottish Government Chief Scientist Office.

By Rich Mitchell, Natalie Nicholls & Jon Olsen , Neighbourhoods and Communities programme, MRC/CSO Social and Public Health Sciences Unit, University of Glasgow.

New evidence linking availability of tobacco & smoking

The connection between the local availability of tobacco products and smoking behaviour has been underlined in new research from the CRESH team this week. Published in the journal Tobacco Control, we show how moving into an area of Scotland where tobacco products are more readily available can significantly increase the risk of smoking while pregnant. We estimate that a pregnant woman living in an area with the highest tobacco availability is 70% more likely to smoke than when she was living in an area with the lowest availability of tobacco products.

Why is this important? Firstly, smoking during pregnancy is a vital Public Health issue and is recognised as a key priority area for UK health policy. It is harmful for both the mother and the developing fetus and the effects for social and health outcomes can persist into childhood and adulthood. Since smoking is so strongly associated with poverty and deprivation, it also has an important role to play in the persistence of health inequalities across generations.

But there are other reasons why the research is important. Much of what we know, including previous research from CRESH, is based on information from a single point in time. While these studies are crucial in establishing the strength of associations, they are less useful for determining mechanisms. A key question that remains is whether high availability is the cause of smoking behaviour or whether retailers preferentially locate in areas of high demand. Both pathways are plausible but both carry very different conclusions and policy recommendations. Our latest research is able to address this question using information on smoking during pregnancy which is collected routinely as part of Scotland’s hospital maternity records. By looking at multiple pregnancies to the same individual, we were able to relate changes in smoking behaviour between pregnancies to changes in exposure to tobacco retailers from residential moves. This approach provides strong evidence that availability is causally linked to behaviour.

The policy implications are clear. As more and more countries move towards a “Tobacco Endgame” policy this, and other research, highlights how a focus on tackling the local availability of tobacco products will be crucial. In a week where the UK government has suggested that preventing poor health lies with “people choosing to look after themselves better, staying active and stopping smoking” our findings are a timely reminder of the importance of considering the wider set of structural factors that shape our health of which our residential environment is one important component.

Do people actually use the facilities in their home neighbourhood?

This blog explores a key question in neighbourhood and health research: if there is a facility or amenity close to someone’s home, is it OK to assume they use it? Surprisingly, this assumption is at the heart of a lot of health and environment research.

We often have data which tells us where facilities and amenities are, and we tend to make the assumption that proximity means use. So, for example, if we see that some neighbourhoods have more parks or more leisure facilities, we expect the people who live in that neighbourhood use them more. Understanding local amenity and facility use is important because we want to know whether / how these things affect health.

With technological advances in recent years, studies have started to collect precise data which tell us exactly where people go using global position system (GPS) devices. We no longer have to assume, for example, that if there’s a park close to a child’s home, they will visit it. The GPS tracks we collect will tell us if they did or not. That presents an opportunity to test our assumptions.

Do children use facilities they have access to in their home neighbourhood?

Our team is interested in children’s use of facilities in and around their homes and to test whether we need GPS to research this we conducted an analysis of facility availability and facility use for 30 10-year-old children living in Glasgow. We used data from GPS devices worn by the children for eight days. These children were part of our ‘Studying Physical Activity in Children’s Environments across Scotland’ Study (SPACES).

The diagram below shows what we did. Our key finding was that facility availability in the home neighbourhood is not a good indicator of facility use; the children used facilities from across a much wider area in the city, even if they had a facility close to their home.  For example, 18 of the 30 children (60%) had a leisure centre within their ‘neighbourhood’ (which we defined as 800m around their home). Only 3 of the 18 actually visited that facility (as identified by their GPS tracks). Of those 18 children, 8 actually visited a leisure centre outside of their ‘neighbourhood’. We saw the same kind of pattern when exploring availability and visits to playing fields, public parks and libraries

Blog graphic

Are our results similar to other research?

Yes, other studies that used GPS devices have found that children do spend time outside of their immediate home area for specific purposes. For example, a 2017 study by Chambers and colleagues in Wellington, New Zealand analysed leisure time GPS data (before and after school) in 114 children aged 11 to 13 years from 16 schools, and found that 38% of their leisure time was spent outside of the home neighbourhood (using a 750m buffer around the home). Time outside of the home neighbourhood was mostly spent visiting their school, other residential locations, and fast food outlets.

These results, and those from similar studies, show that it is important not to treat what’s in someone’s immediate home neighbourhood as a good measure of what they do, or in epidemiological language ‘what they are exposed to’. We must challenge the idea that residential neighbourhood is an adequate way to capture the socio-environmental factors which contribute to health. Many people, including children, can and do access environments well beyond their immediate home neighbourhood. We think that a much wider geographic area should be considered when we’re asking questions about how environment affects health and we call this the city-wide landscape.

What does this mean for future research?

It’s clear that the ‘traditional’ approach which uses someone’s neighbourhood (often defined by a distance around their home, or an administrative area in which their home sits) to assess their access to facilities or exposure to environments is seriously flawed.

  • Other methodological approaches are required to measure ‘exposure’ to environment;
  • We must move beyond traditional fixed neighbourhood-health relationships (although we can’t ignore them);
  • We should embrace and integrate innovative technology to explore mobility (e.g. GPS and accelerometer).

Of course, even when we’re able to see exactly where people go and what they do, we still need to understand the decisions people make about whether or not to visit or spend time at different places.

By Jon Olsen, Research Associate with the Neighbourhoods and Communities programme, MRC/CSO Social and Public Health Sciences Unit, University of Glasgow.