Are Londoners Getting Healthier?

Yijing Li

, Sijie Tan, Xiangbo Chang and Xiaohui Chen

CUSP London, Department of Informatics, King’s College London, U.K.

Keywords: Obesity, Mental Health, London, Spatial Pattern.

Abstract: It utilised urban data from multiple sources, to map out the physical health and mental health patterns in

London over space and time. On basis of recognising the spatial patterns trajectory changes, obesity among

adults and children in London areas had been investigated with selected demographic, socio-economic and

environmental factors, to identify the most influential factors in all, and for local community; similarly,

workflow had been again designed to investigate the influential factors for mental health prevalence as well.

Upon comparing the selected models, models considering neighbourhood spillover effect has been found to

be the optimal, to identify significantly influential factors on urban health, such as age group, green space

access, household deprivation, income deprivation and air quality. The findings underscore the necessity for

targeted, location-specific public health interventions to effectively combat obesity; highlight the importance

of spatial heterogeneity, offering detailed insights into regional variations; and suggest tailored strategies for

public health policies. This work fills a critical gap and demonstrates the need for geographically informed

public health strategies.

1 INTRODUCTION

As the capital of UK, health statuses for London

residents have always been the prime topic, and could

be interpreted from mainly two strands, the physical

health and the mental health. The former strand has

been found to be with “overweight and obesity

represent probably the most widespread threat” to

physical health (Department of Health and Social

Care, 2011), requiring for leaders’ implementing

public health measures to reduce obesity, such as the

Mayor’s London Food Strategy to tackle with obesity

and provide a framework for assessing the impact of

these policies (PHE, 2020). The latter strand is a

crucial component of human being’s overall well-

being and is closely linked to physical health (NHS

England, 2020), hence organisations such as PHE and

the GLA regularly publish reports and datasets that for

further investigations (Rosenberg, 2019). It also

inspired empirical studies investigate the significant

socioeconomic disparities among London, suggesting

influential socioeconomic factors may drive the

disparities among London regions (Mayor of London,

2018). However, there is a lack of evidence on status

quo patterns of Londoners’ health: are they becoming

https://orcid.org/0000-0002-9831-0298

healthier over time? Are there any regional variations

for London residents’ health statuses, for either

physical health or mental health? If so, what are the

influential factors driving such differences?

To address above questions, this project aims to

explore the latest obesity and mental health levels

among London residents, on basis of multiple openly

accessed datasets from varied sources, to map out the

health spatial patterns changes in London in recent

decade by age group, to investigate the driving forces

from socio-economic contexts and environmental

measures, so to serve the potential strategic policy

adjustments among relevant stakeholders for

healthcare equity and improvement in the city.

2 BACKGROUND AND

EMPIRICAL STUDIES

2.1 London Demographics

London is formed by City of London and other 32

boroughs, further breakdown into 417 Middle layer

Super Output Areas (MSOAs) or 4835 Lower Super

Output Areas (LSOA) (ONS, 2011), with residents

Li, Y., Tan, S., Chang, X. and Chen, X.

Are Londoners Getting Healthier?.

DOI: 10.5220/0013203300003935

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 11th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2025), pages 15-26

ISBN: 978-989-758-741-2; ISSN: 2184-500X

around 8.9 million (ONS Census 2021, 2022), and is

well-known for its ethnic and cultural diversity in that,

residents in London speak more than 300 languages

and come from a wide range of ethnic backgrounds

(GLA, 2020). High level of population diversity

brings the mayor’s manifesto commitment on a

“fairer, healthier and more equal” London (Mayor’s

Office, 2022) become a prominent but challenging

topic, hereafter the necessity to understand

Londoners’ status quo health levels.

2.2 Obesity as a Measure for Physical

Health

As defined by WHO (2024), obesity is the excessive

accumulation of fat that poses a threat to health,

which commonly measured by body mass index

(BMI), where people with higher BMI may expose

to higher risk of obesity-related health

complications. From WHO’s classification, adults

with obesity are normally categorised into Grade I

obesity (BMI at 30-34.9), Grade II obesity (BMI at

35-39.9), and Grade III obesity (also known as

morbid obesity, BMI at 40 or above); children’s

obesity is defined by age-specific BMI percentiles at

or above the 95

percentile for their age and sex.

Individuals diagnosed with obesity were related

to potential health problems by their age groups,

stages of growth and development. For example,

children are normally thought to experience high

metabolic rate and rapid physical and psychological

changes (Stamatakis et al., 2010), with potentially

long-term health influences into their adulthood. On

the contrary, adults have a relatively lower metabolic

rate and a mature lifestyle and eating habits, hence

obesity developed during adulthood will be

associated with chronic diseases such as heart

disease and diabetes. Their living environments are

taken as influential to the development of obesity

symptoms. For instance, access to recreational

facilities, availability of healthy food, transport

infrastructure, and the built environment are

important shaping residents’ lifestyles, then affect

local obesity rates. The city’s extensive public

transport network, abundant green spaces and

walkability create a unique environment for

assessing the impact of urban design on physical

activity levels and eating habits (ONS, 2019). Recent

research on adult obesity has highlighted the

complex interplay of multiple influencing factors,

including socioeconomic status, dietary habits,

physical activity, and environmental conditions.

Stafford et al. (2010) enriches this discussion by

highlighting that, obesity trends are severe among

population from poorer socioeconomic backgrounds

in developed countries. For example, people with

lower incomes may face food insecurity, leading to

poorer diet quality and higher rates of obesity. These

findings reflect the broader claim that,

socioeconomic disadvantage is a key determinant of

health disparities. Focusing on environmental

factors, Shenassa et al. (2006) highlighted the

significant correlation between sedentary lifestyles

and obesity in Europe. Hobbs (2022) emphasized the

impact of the environment on the level of physical

exercise, which in turn affects the management of

obesity, thus showing the importance of public space

and parks. Furthermore, as highlighted by the 2018’s

Health Survey for England (NHS England, 2019),

the availability of physical activity facilities versus

fast food options illustrates the environmental

opportunities and risks that contribute to obesity.

These environmental determinants of obesity

interact intricately with sociodemographic factors.

Children’s development of obesity symptoms

was found to have varied but interrelated factors

from adults, which were mostly the home

environment, school policies, access to recreational

facilities, and nutrition education (Schratz et al.,

2023). The impact of parental obesity and family

eating habits is particularly strong in that, parental

obesity, especially maternal obesity, greatly

increases the risk of childhood obesity due to shared

genetic and environmental factors (Kral and Rauh,

2010). Stafford et al. (2010) found that children from

disadvantaged backgrounds, such as families from

lower incomes or manual labors, are at greater risk

of high obesity rates. Schools also play a vital role in

tackling with childhood obesity in that, schools that

promote physical activity and provide healthy meal

choices can encourage children to make healthier

lifestyle choices, hence greatly reduce obesity rates

(Veitch et al., 2011). Public spaces such as parks

with good accessibility and playgrounds with good

quality of recreational facilities are vital, for

maintaining a healthy weight for children since its

encouragement of children’s proper levels of

physical activity (Van Lippevelde et al., 2012).

Besides, educational programs provided by the city

or local about healthy eating habits, can also

effectively prevent childhood obesity and help

families make informed choices about nutrition and

physical activity, leading to healthier lifestyles

(Story et al., 2009). NHS England (2023) 2022/23

school year report uncovered that, obesity rates were

higher among children from deprived areas, for

example, among Year 6 children 30.2% of children

in the most deprived areas were obese, compared

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

with 13.1% in the least deprived areas. These

insights highlight the importance of national child

and adolescent health programs in guiding public

health policy and ensuring that resources are directed

to the most affected communities. The programme’s

data will be crucial for developing effective health

interventions and promoting healthy lifestyles

among children in England.

2.3 Mental Health

On the other hand, mental health is another strand for

Londoners’ health issue. Chronic physical diseases

increase as mental health levels decline (Yang et al.,

2020). It can directly affect people’s life experiences

and quality of life in that, better mental health allows

individuals to utilize their abilities and align with the

broader values of society (Yang et al., 2006).

However, it was not until the 1990s that people

gradually realised the wider impacts of mental

illness (Zhu et al., 2020). Mental health is defined as

a state of well-being where individuals recognize

their own abilities, can handle daily life stresses,

works productively, and contributes to society

(Daghagh et al., 2019). Mental health consists of the

hedonic dimension, which focuses on well-being,

fulfilment, and avoidance of pain, and the

eudaimonic dimension, which includes personal

growth, personal objectives, and mental functioning

(Henderson and Knight, 2012; Ryan and Deci,

2001). The Public Health Agency of Canada

(PHAC) defines mental health as the capabilities that

every person possesses that enable them to

experience emotions, think critically, and behave in

ways that enhance their life enjoyment and their

capacity to cope with challenges (Heather et al.,

2017), especially in that, positive emotional and

mental health that values the importance of culture,

fairness, social equality, interconnectedness, and

individual dignity (Heather et al., 2017).

To better measure citizens’ mental health, Keyes

(2002) proposed the “Mental Health Continuum”

(MHC) model comprises of emotional well-being,

psychological wellbeing, and social well-being.

Headey et al. (1993) emphasized the dimensions on

such as life satisfaction, positive emotions, anxiety

and depression including factors such as

socioeconomic status, social support, and life events,

while the uneven distribution of these factors in

different geographical areas may lead to significant

differences in mental health status. Dfarhud (2014)

reviewed that there are two dimensions which

influence people’s mental health, which are

endogenic factors and exogenic factors. Endogenic

factors included the biological, cognitive,

personality and ethical, which are challenging to be

quantified, while exogenic factors, such as

greeneries, are convenient factors to evaluate their

influences on mental health. Markevych et al. (2017)

suggested that the greenspace could enhance the

interpersonal communication, relieve the pressure

and protect the environment, all these advantages are

generally related to the mental health. Vires et al.

(2003) runs a test with 10,000 people and found that

greener area benefits human’s mental health,

especially for the elderly and housewives. Gianfredi

(2021) said that the publicly accessible greenspace

in urban area has positive influence on both physical

activity and mental health, and Lachowycz (2013)

suggested that the greenspace area influences people

potentially, even for those people do not walk in the

greenspace they would feel more satisfied.

2.4 Spatial Analytics to Understand

Urban Health

A comprehensive understanding of the various

aspects of physical health and mental health can

support better strategic policies on appropriate

interventions to address areas where poor health is

prevalent among London regions. In assistance of

the development of geographical information system

(GIS) and relevant analytical techniques, geospatial

models and tools had been widely utilised in

identifying the spatial patterns of obesity. For

example, Sun et al. (2020) used spatial regression

models, SAR and ESF, onto childhood obesity

analysis and drew associations between childhood

obesity prevalence and variables such as population

density, race, unemployment rate, and household

income. Cetateanu and Jones (2014) have used

ESDA techniques to identify spatial clusters of

obesity in England, on basis of Global Moran’s I and

Local Indicators of Spatial Association (LISA), they

found important clusters of high rates of obesity in

urban areas and low rates of obesity in rural areas,

further highlighted the need for targeted public

health interventions in specific regions (Anselin et

al., 2006). In London, environmental factors such as

socioeconomic deprivation and access to green

space, has presented similar spatial clustering

patterns (Gaber et al., 2024), which were suggested

to be interlinked with the relationship fast-food

restaurants and supermarkets on obesity rates (ESRI,

2021). Grigoroglou et al. (2019) illustrated that

depression prevalence, to reflect mental health

status, could be identified of their spatial clustering

patterns through Moran’s I measure.

Are Londoners Getting Healthier?

Underweight men were found to have higher

rates of anxiety disorders compared to normal-

weight men and women (McLaren et al., 2008), and

obese women have higher rates of mood disorders

compared to normal-weight women. Daghagh Yazd

et al. (2019) also agreed that female farmers suffer

greater mental distress than male counterparts, and

marital status is associated with poorer mental health

(Daghagh et al., 2019). Houlden et al. (2019)

examined the linear relationship between the amount

of green space and mental health by developing a

univariate ordinary least squares (OLS) regression

model exploring relevant factors of mental health

indices. The GWR method computes a local

regression where the coefficients can vary spatially

(Brunsdon et al., 1996; 1998), hence could provide a

more nuanced understanding of how different factors

influence mental health (SMI) in local areas, such as

the work of Cruz et al. (2022) who constructed a

Bayesian spatial regression model incorporating

random effects applied to the log-transformed mean

SMI prevalence, to offer a precise comprehending of

the spatial pattern and determinants of mental health

problems, in recognising that high risk of SMI was

associated with unemployment, low income, low

education level, and living in a high-crime area.

Besides of the multifaceted nature of mental health

linking to socioeconomic status, environmental

factors such as greenery, air pollution, and climate

also have been examined (Mueller et al., 2023;

Houlden et al., 2019; Bakolis et al., 2021). The

findings highlight the huge impact that climate and

air quality have on the planet. In conclusion, these

studies reveal the complex impact of environmental

factors on obesity and mental health, highlighting the

importance of further research and consideration of

multiple variables.

3 DATA AND METHODOLOGY

3.1 Data

The datasets used in this study are derived from

multiple open sources and integrated into the

geographical boundaries for UK Census at Middle

Layer Super Output Area (MSOA) and Lower Layer

Super Output Area (LSOA) levels respectively. Data

fusions are realised through joining into the

boundaries’ shapefiles by the shared MSOA or LSOA

codes for following 4 datasets:

Firstly, measures for obesity and relevant socio-

economic factors mainly come from MedSat data,

under CC BY-SA 4.0 license and could be found at

TUMMedia (data sharing service from Technical

University Munich) (Šćepanović et al., 2023),

includes a wide array of variables (Figure 1) such as

population density, age, gender, ethnicity, religion,

marital status, employment, commute to work,

residence, self-reported health, language, and the

index of multiple deprivation (IMD), air quality,

greenery, climate, and land cover, as well as medical

prescriptions data encompasses conditions such as

diabetes, hypertension, asthma, depression, anxiety,

opioid prescriptions, and overall total prescriptions.

Figure 1: Structure of Medsat dataset (Access the dataset at:

https://github.com/sanja7s/MedSat).

The second dataset, Access to Healthy Assets and

Hazards (AHAH), is collected from the Consumer

Data Research Centre (CDRC), includes a variety of

variables measuring access to both healthy assets and

environmental hazards, such as the distance to the

nearest gambling outlet, fast food outlet, GP practice,

hospital, dentist, pharmacy, and leisure center, as well

as measures of air quality (nitrogen dioxide,

particulate matter) and green/blue space (NDVI value

for passive green space), which were collocated from

various national organizations such as ONS, NHS

England, NHS Scotland, LDC, DEFRA, and

OpenStreetMap (Figure 2).

The third dataset on adults and children obesity is

sourced from the NHS Digital's Quality and

Outcomes Framework (QOF), aggregated at the

MSOA level (GLA, 2023). This dataset combines

GP-level data with the 2019 population estimates

from the Office for National Statistics to estimate the

percentage and number of obese adults in each

MSOA. Similarly, the childhood obesity dataset

identifies MSOAs with the highest levels of

overweight children in reception and Year 6, which

are viewable as below interactively.

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

Figure 2: Detailed variables of CDRC dataset (Access the

dataset at: https://data.cdrc.ac.uk/dataset/access-healthy-

assets-hazards-ahah)

The data for mental health modelling is Small

Area Mental Health Index (SAMHI), which is a

composite annual indicator of population mental

health at LSOA level from 2011 to 2019 (Daras and

Barr, 2020). SAMHI combined multiple sources of

mental health data, such as NHS-Mental health

related hospital attendances, GP Patient Survey, and

Prescribing Data, where a positive SAMHI indicates

relatively severer mental health issues in the area,

whilst a negative SAMHI indicates for fewer

symptoms of residents’ mental health issues. On top

of it, 11 independent variables had been selected for

modelling, include household income, employment,

physical health, education, environment, air

pollution, green area, the percentage of the population

over 65 years old, the percentage of men, population

density, and the percentage of the population aged 16-

29, derived from the English indices of deprivation

(IMD) in 2015 and 2019 (ONS, 2024), and the total

population, population density at LSOA scale,

greenspace coverage, air pollution concentration data

from London Data Store.

3.2 Research Design and Methods

The research workflow design diagram is identical for

obesity and mental health, hence they are running in

parallel following the same workflow shown in

Figure 3, taking obesity as example for illustration.

Geographic boundary files and datasets upon merging

adult and childhood obesity rates were used to map

obesity patterns in London, followed by appropriate

exploratory spatial data analysis, spatial patterns

recognitions (i.e., Global Moran’s I and Local

Moran’s I clustering), and varied influential factors

investigation on obesity among adults and children,

with step-by-step methods adoptions included in the

lower half of Figure 3.3. The project employs spatial

regression models, including SAR and SEM models,

to identify the influential factors for whole city, but

uses localised model Geographically Weight

Regression (GWR) to find their varied influences

among London areas.

Anselin et al. (2006)’s Global Moran's I and Local

Moran’s I indices were firstly calculated to examine

the existence of spatial autocorrelations for health

data, as well as the potential spatial clusters and

outliners (Equation 1), so that to decide which

regression models will be the optimal for this study.

𝑦𝛽



+𝛽



𝑥



+𝛽



𝑥



+⋯+𝛽



𝑥



+𝑢 (1)

Traditional regression model Ordinary Least

Square (OLS) is going to be used for preliminary

regression, as one of the most classical types of

regression analysis (Yang et al., 2016), which

predicts the output variable as a linear combination of

the input variables; but with more recognition of the

existence of spatial autocorrelation, spatial regression

models are to be tested as well. For example, if taking

the mental health measure SAMHI as dependent

variable, then the model will be expressed as in

Equation 2.

SAMHI =𝛽

+𝛽

income score +𝛽

employment score +

𝛽

education score + 𝛽

health score +𝛽

environment score

+ 𝛽

concentration+ 𝛽

canopy tree + 𝛽

65aged + 𝛽

male

ratio + 𝛽

density + 𝛽

16 29ratio + u

(2)

Figure 3: Research Design flowchart for Obesity (or Mental

Health) Patterns in London

Spatial Lag Model (SLM) is deployed to account

for spatial dependencies by incorporating a spatial lag

of the dependent variable into the regression model,

embracing the influence of neighbouring areas on the

Are Londoners Getting Healthier?

outcome of interest, and be expressed in Equation (3):

𝑌= 𝜌𝑊𝑌+𝑋𝛽+𝜖 (3)

where Y is the dependent variable, ρ is the spatial

autoregressive parameter, W is the spatial weights

matrix, X represents the independent variables, β

denotes the coefficients, and ϵ is the error term. By

including 𝜌𝑊𝑌.

Spatial Error Model (SEM) addresses spatial

autocorrelation in the error terms rather than in the

dependent variable. This model is suitable when the

spatial dependence arises from unobserved factors

that affect the error term, and can be formulated as in

Equation (4):

𝑌 = 𝑋𝛽 + 𝜖，𝜖 = 𝜆𝑊𝜖 + 𝜂 (4)

where η is a normally distributed error term with

mean zero and 𝑊𝜖 represents the spatially lagged

error term. Here, λ is the parameter that measures the

degree of spatial autocorrelation in the errors. The

SEM helps to correct for biases that might be

introduced by spatial correlation in the error terms,

leading to more accurate estimates of the

relationships between the dependent variable and the

independent variables.

Geographically Weighted Regression (GWR) as

the last model, to capture local variations in the

relationships between variables, offering a more

detailed understanding of spatial dynamics. Unlike

traditional regression models, which assume a

uniform relationship across all locations, GWR

calculates different regression coefficients for each

area, addressing spatial differences in the data.

The formula of the GWR model is displayed in

Equation 5:

𝑦

𝑖

= 𝛽

(𝑢

𝑖

,𝑣

𝑖

) + 𝛽

(𝑢

𝑖

,𝑣

𝑖

) 𝑥

1𝑖

+ 𝛽

(𝑢

𝑖

,𝑣

𝑖

) 𝑥

2𝑖

+ + ···

𝛽

𝑘

(𝑢

𝑖

,𝑣

𝑖

) 𝑥

𝑘𝑖

+ 𝜖

𝑖

(5)

It detects spatial changes in relationships in the

model and can reveal local features and trends in

spatial data and handle spatial autocorrelation in data.

Drill into the research design and methods for

influential factor investigations, hierarchy of the

selected factors are reflected in Figure 4, and similar

for mental health measure (to replace the dependent

variable from measure for obesity to measure for

mental health), informed by empirical studies in

Section 2.

Figure 4: Variables for Modelling Obesity Patterns in

London (similarly for Mental Health).

4 RESULTS

4.1 Spatial Patterns of Londoners

Health

4.1.1 Physical Health – Obesity

Londoners’ obesity patterns vary by age group in that,

as reflected in Figure 5 on obesity rates among

London regions, the redder in the map the higher

values and vice versa. From left to right, maps are

obesity patters for adults, children in Reception (ages

4-5), and children in Year 6 (ages 10-11) respectively.

Figure 5: Londoners’ Obesity Patterns by Group (Adult,

Reception and Year 6).

It is apparent that, areas in the outer East (coloured

in dark red) have higher obesity rates such as

boroughs Barking and Dagenham and Havering.

Besides, there were more adults with obesity in the

western borough Hillingdon as well; whilst obesity

rates among younger children in reception (aged 4-5

years old) were more dispersedly distributed, with

areas in central London, such as Newham and Tower

Hamlets, and outer areas such as Brent in the North;

obesity among bigger children in Year 6 (ages 10-11)

were also high in areas such as Barking and

Dagenham, Tower Hamlets, and Hackney, but with

wider distribution than the younger counterpart.

To identify the potential spatial autocorrelation,

Global Moran’s I had been calculated for each of the

three age groups’ obesity rates, in return with

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

significant values as adult obesity at 0.6112, younger

children at 0.2093 and older children at 0.2689. It is

obvious that adults with obesity in London tend to

cluster more in certain outskirt areas, whilst children

with relatively much less pronounced clustering

spatial pattern, but mostly in Northeast London. It

implies that obesity issues among London adults may

be more influenced by the neighbouring areas and

local environment, hence leading to the localised

spatial analysis visualised by LISA maps (Figure 6).

In LISA maps, areas in red are HH clusters with high

obesity rates and surrounded by similar neighbouring

areas and predominantly located in the northeast parts

of London regardless of age groups with consistently

high obesity rates. On the contrary, areas in blue

indicates LL clusters areas with low obesity rates

surrounded by low obesity neighbours, indicating

central and southwestern regions tend to be healthier

physically. Meanwhile, areas with high obesity rates

but neighboured by low obesity areas are colored in

orange, whilst light blue areas are the opposite.

Adults and bigger children also experience higher

obesity in some western boroughs, whilst younger

children have higher obesity in central-west London

for limited area.

Figure 6: LISA clusters among adult, reception and year 6

(left to right).

4.1.2 Mental Health

Londoners’ mental health indices distribution had

been presented in Figure 7 for 2011 to 2019

respectively for comparison, but received similar

spatial patterns. Redder areas with lower index values

indicated better performance regarding residents’

mental health, comparatively the bluer areas with

positive mental health index values are those with

worst performance of residents’ mental health. From

2011 to 2019, residents’ mental health has improved

significantly with expanded redder areas.

Figure 7 Mental Health Indices Patterns (left: 2011; right:

2019).

Upon calculating the global Moran’s I measure, it

was found that mental health index has significant

spatial clustering patterns with a Moran’s I value at

0.4875 in 2011 and 0.5341 in 2019. Localised LISA

cluster mappings (Figure 8) further highlighted the

clustering patterns with high-high (HH) areas

indicating for those with worse mental health

performance, and low-low (LL) areas with better

mental health performance.

Figure 8 Mental Health Indices LISA maps (left: 2011;

right: 2019).

Boroughs such as Islington, Camden and Hackney kept

poorer mental health over time, while boroughs such

as Enfield, Sutton and Lewisham improved in 2019

and no longer being hot spot clusters. On the other

hand, boroughs such as Brent, Hounslow, Redbridge,

Newham, Harrow and Wandsworth were better

performed over time, but Havering became getting

worse in 2019 from its good performance in 2011.

4.2 Influential Factors from Regression

Modelling

4.2.1 Influential Factors for Physical Health

To better investigate the contributing factors for

London residents’ obesity, OLS, SLM and SEM

regressions had been conducted with results

summarised in Table 1.

Table 1: Regression Results for Obesity Modelling.

Variables OLS SLM SEM

Adult

Bigger

Children

Adult

Bigger

Children

Adult

Bigger

Children

Intercept 14.7687 33.5218 2.2114 958.0314 ----- 1817.1489

Percent of

Age 19～29

-0.0005 -0.0002 -0.0002

Percent of

Age 40～49

--- ---- -----

Percent of

Age 50～64

0.001 0.0007 0.0006

Percent of

Age 85+

-0.0058 -0.0043 -0.0038

Distance to

nearest Fast

food outle

---- -47.1416 ----- 38.4173 -0.0742 ----

Distance to nearest

Green Space

---- -1316.4992 0.8584 -656.2771 ---- -879.9875

Net Annual

Income

-0.0002 -0.0237 -0.0001 -0.0171 -0.0001 -0.0159

Percentage of

unemploymen

-3.3465 ---- ----- ----- ----- 953.2364

Percentage of

household

deprivatio

------ 47910 41.7626 40950.0536 40.5875 43658.535

Neighbourhood

spillove

0.9857

R-square 0.3346 0.3645 0.7662 0.5038 0.2356 0.3546

Are Londoners Getting Healthier?

Comparing the 3 selected models, it is apparent

that SLM is outperformed with R-square at 0.7662 for

adults and 0.5038 for bigger children, indicating for

the inclusion of neighbouring areas’ influences

(0.9857), especially for adults’ obesity pattern.

Income level exerted significant mediating effects

onto obesity issue, regardless of age groups, but the

influences were stronger to children than to adults.

Percentage of residents at 40-49 years old is not

related to regional adults’ obesity level, but other age

groups will be significantly related, especially the age

group 50-59 years old, will drive mental health index

higher. For children, community and family

environment, such as the distance to the nearest fast-

food outlet, the distance to the nearest green space,

and the percentage of household deprivation had

substantial effects on obesity rates, indicating that

more convenient fast-food accessibility and more

deprived families will drive children’s obesity rate

higher, while greenspace accessibility can mediate

such obesity through easier exercise and outdoor

activities.

4.2.2 Influential Factors for Mental Health

Similarly, London residents’ mental health index and

selected independent variables are modelled by OLS,

SLM and SEM regressions for data in 2011 and in

2019 comparatively, with results summarised in

Table 2.

Table 2: Regression Results for Mental Health Modelling.

OLS SLM SEM

2011 2019 2011 2019 2011 2019

R Square

0.5983 0.5499 0.7238 0.7443 0.5390 0.5045

Income Score -0.5020 -0.4295 -0.3528 -0.2147 -0.2098 -0.0605

Employment

Score

0.8584 0.8837 0.6874 0.6272 0.5968 0.5301

Education Score -0.1516 -0.1191 -0.1407 -0.1246 -0.0539 -0.0510

Health Score 0.5239 0.4034 0.3919 0.2683 0.5391 0.3746

Environment

Score

0.0989 ----- 0.0501 -0.0306 0.0478 0.0115

Air Pollution

Score

0.1139 ----- 0.0675 0.0396 ------ ------

Canopry Tree 0.0407 0.0302 ----- ------ ------ -------

65age

0.2513 0.2931 0.2124 0.2356 0.1467 0.2326

Male Ratio -0.1106 -0.1173 -0.0527 -0.0358 ----- --------

Densit

----- ---- -0.0330 ----- ----- -0.0333

16-29ratio -0.0607 0.0114 -0.0546 ----- -0.050 -0.0400

Spatial Weight ----- ---- 0.5020 0.5924 0.7978 0.8145

It is obvious that SLM model still have the highest

R squared value for both 2011 and 2019 data, hence

should be the optimal model to be selected.

Residents’ mental health was mostly driven by their

deprivations of employment and health, but can be

mediated if be with a lower deprivation of income.

Similarly, deprivation on education and the

percentage of male residents could also help to

alleviate local mental health issues. However, the

neighbouring areas’ mental health level is influential

more and more greatly over time, as well as the

percentage of aged populations (over 65). It also

called our attention that at global scale, greenery

didn’t play significant influence onto Londoners’

mental health, but air pollution concentrations are.

4.2.3 Localised Influences - GWR Model

To better understand the varied influences and model

fitting among London boroughs, Geographically

Weighted Regression (GWR) model had been applied

to map out the factors’ explanation ability for obesity

and mental health measures. GWR allows for the

examination of local variations in predictors, in

Figure 9 it provides a deeper understanding of factors

influencing obesity among London regions, by

visualizing how well the model explains obesity rates

across the study area.

Figure 9: Local R-Squared map of GWR model among

adults and children.

The local R-squared values for adults (left in

Figure 9) are higher in East London especially the

southeast such as Bromley, suggesting that the model

explains obesity rates better in these regions. In

contrast, the western areas show lower R-squared

values, indicating that the model’s explanatory power

is weaker. For children (right in Figure 4.5), the local

R-squared values are higher in the South and some

northern areas, implying that the model performs

better in explaining childhood obesity rates in these

regions, but is weaker in explaining young people’s

obesity in central, and central east London areas.

It can also exhibit the varied influences onto

dependant variable among London regions for each

identified factor. For example, in Figure 10 presented

the deprivation of income in London exerted varied

influences onto mental health levels in 2011 and

2019, where outskirt areas were more driven by such

economic inequality; on the other hand, it tends to

cluster towards certain areas such as in Bromley over

time.

Another factor, deprivation on employment has

been investigated as well (Figure 11) in finding its

expanded influences in the South and West London

from 2011 to 2019, where central London and South

London such as Croydon were always the most-hit

areas for employment inequality driven mental health

issues.

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

Figure 10: Local Influences on Mental Health from

Deprivation of Income (left: 2011; right: 2019).

Figure 11: Local Influences on Mental Health from

Deprivation of Employment (left: 2011; right: 2019).

In all, each factor can be mapped across London

areas on their varied influences onto local mental

health levels, either be driving forces or mediating

effects. From a global perspective, greenery is not

significantly influencing mental health, but it became

influential locally in certain boroughs such as

Hounslow. Income, education, employment,

physical health, environment, greening, air pollution

concentration, population density, and the percentage

of young and elderly populations are important

factors affecting mental health. These findings are

conducive to designing localised intervention

measures according to the featured circumstances of

varied boroughs.

5 CONCLUSIONS

Obesity is a significant public physical health issue,

leading to health risks, including developing chronic

conditions such as heart disease, type 2 diabetes, and

certain types of cancer. To understand and mitigate its

potential impacts on Londoners’ health could be

meaningful to the society’s development. The

findings highlighted significant factors influencing

obesity, differing between adults and children,

underscoring the need for targeted, location-specific

public health interventions. Specifically, for adults,

age group compositions, green space access,

household deprivation, and income were significant;

while for children, access to fast food outlets, green

space, household deprivation, and income levels were

critical. Recommendations included promoting

weight management programs at local community,

improving green space access, and addressing socio-

economic disparities. The GWR model highlights the

significance of contextual and geographical factors in

obesity. By focusing on local variations, more

effective policies can be developed. For instance, in

eastern boroughs like Newham and Tower Hamlets,

strategies could include enhancing green spaces and

providing financial support to low-income

households. In central areas such as Islington and

Camden, where childhood obesity is a major concern,

interventions might include reducing fast food outlets

near schools, expanding recreational facilities, and

promoting nutrition education. Tailoring these

strategies to regional needs can significantly improve

public health outcomes and address obesity more

effectively.

Mental health issue is not a sole challenge in

London, instead serious mental illness (SMI) affected

millions of people all over the world (James et al.,

2018). In England, nearly three million adults were

listed on local GP registers with depression in 2013-

2014, and around half a million were diagnosed with

a serious mental illness (SMI), which refers to people

with schizophrenia, bipolar disorder or other

psychotic illnesses (Grigoroglou et al., 2020). In

London it is especially crucial to understand the

mental health status over areas and time. The spatial

patterns of London’s mental health indices from 2011

to 2019 were largely consistent but varied at finer

granularity for local distribution. Comparatively, the

overall mental health indices in 2019 have improved

significantly. Some boroughs, such as Islington,

Camden and Hackey, kept poorer mental health

performance from 2011 to 2019, whilst some other

borough (Enfield, Sutton and Lewisham) exhibited

great improvement over the time. On the contrary,

outskirt boroughs such as Brent, Hounslow,

Redbridge, Newham, Harrow and Wandsworth

maintained good mental health performance over this

period, but with exceptional deterioration in Havering

from 2011 to 2019.

The spatial lag models (SLM) were found to be

optimal in capturing maximum information hence

explaining the dependant variables. The regional

deprivations on income, employment, education,

physical health, environment, air pollution, the

proportion of the young and elderly population,

population density, and the proportion of males

significantly impacted mental health regardless of the

year. Geographically, Hounslow, Redbridge and

Harrow are boroughs consistently with good mental

health, where Hounslow benefits from good

employment equality, Redbridge and Harrow both

benefit from better environmental equity, as well as

better education in the latter. However, Havering

Deprivation on Income (2011)

Deprivation on Income (2019)

Deprivation on Employment (2011)

Deprivation on Employment (2019)

Are Londoners Getting Healthier?

stood out as an example for mental health

deterioration due to severe income deprivation and air

pollution pressures. Environmental factor greenery

didn’t exert a significant impact on London’s mental

health, but significantly affected mental health locally

in specific boroughs. For example, Havering’s mental

health deteriorated significantly from 2011 to 2019

when driven by widened income gaps, high level of

air pollution and less access to greeneries.

Mental health is affected by a mixture of multiple

socioeconomic and environmental factors, future

trajectory studies could be suggested to investigate

the long-term impacts. Qualitative research can also

be recommended to complement the findings

interpretation from residents’ point of view by

interviews. The impact of each variable on mental

health varies significantly across regions and years,

reflecting the spatial heterogeneity of different

regions. For example, Greening can mediate the

mental health in Hounslow, but deprivation of

employment could drive mental health level up. On

the other hand, some factors only were influential in

the area for certain time, such as deprivation on

income and education were only negatively related

with mental health in 2011, but no longer being

significantly influential in 2019. Such variations

might be the results from local policy interventions

hence being worthwhile to get evaluated.

The project successfully addresses several

limitations from empirical studies in that, it fills the

gaps by simultaneously analyzing both adults and

childhood obesity, offering a more comprehensive

understanding of obesity across different age groups;

compared the trajectory changes of mental health

spatial patterns from 2011 to 2019 over both space

and time, and highlights the necessity of region-

specific policies through various spatial regression

methods, demonstrating the importance of

considering local context in public health

interventions for both physical and mental health.

Future research could benefit from integrating socio

economic and environmental factors influencing

obesity and mental health, such as air quality, dietary

habits, and physical activity levels. Expanding the

study to include longitudinal data or qualitative

research data could provide insights into causal

relationships. Incorporating a spatial-temporal

analysis would allow for examining how childhood

obesity transitions into adult obesity, mental health

had been influenced by neighbouring areas, and

revealing long-term trends and patterns. This

approach would involve tracking cohorts of

individuals over time to observe how health related

challenges develop. By linking multi-sourced data,

researchers can identify critical periods and factors

that influence the health trajectories, providing a

deeper understanding of how society conditions

affect long-term health outcomes for citizens.

REFERENCES

Anselin, L., Syabri, I. and Kho, Y. 2006. GeoDa: An

introduction to spatial data analysis. Geographical

Analysis 38(1): 5-22. doi: https://doi.org/10.1111/j.00

16-7363.2005.00671.x.

Bakolis, I., Hammoud, R., Stewart, R., Beevers, S., Dajnak,

D., MacCrimmon, S., Broadbent, M., Pritchard, M.,

Shiode, N. and D. Fecht. 2021. Mental health

consequences of urban air pollution: prospective

population-based longitudinal survey. Social psychiatry

and psychiatric epidemiology 56: 1587–1599.

Brunsdon, C., Fotheringham, S. and Charlton M. E. 1996.

Geographically weighted regression: a method for

exploring spatial nonstationarity. Geographical

analysis 28(4): 281–298.

Brunsdon, C., Fotheringham, S. and Charlton, M. 1998.

Geographically weighted regression. Journal of the

Royal Statistical Society: Series D (The Statistician)

47(3): 431–443.

Bultena, G., Lasley, P. and Geller, J. 1986. The farm crisis:

Patterns and impacts of financial distress among iowa

farm families. Rural Sociology 51(4): 436.

Cetateanu, A. and Jones, A. 2014. Understanding the

relationship between food environments, deprivation

and childhood overweight and obesity: Evidence from

a cross-sectional England-wide study. Health & Place

27: 68-76. https://doi.org/10.1016/j.healthplace.2014.0

1.007.

Cruz, J., Li, G., Aragon, M. J., Coventry, P. A., Jacobs, R.,

Prady, S. L. and White, P. C. 2022. Association of

environmental and socioeconomic indicators with

serious mental illness diagnoses identified from general

practitioner practice data in England: A spatial

Bayesian modelling study. PLoS medicine 19(6):

e1004043.

Daghagh Yazd, S., Wheeler, S. A. and Zuo, A. 2019. Key

risk factors affecting farmers’ mental health: A

systematic review. International journal of

environmental research and public health 16(23): 4849.

Daras, K. and Barr, B. 2020. Small Area Mental Health

Index (SAMHI). Data Catalogue DOI: 10.17638/data

cat.liverpool.ac.uk/1188.

Department of Health and Social Care. 2011. Healthy

Lives, Healthy People: A call to action on obesity in

England. https://www.gov.uk/government/publicati

ons/healthy-lives-healthy-people-a-call-to-action-on-

obesity-in-england.

ESRI. 2021. UK COVID-19 Cases and Mortality Rates.

BMJ Nutrition, Prevention & Health.

https://nutrition.bmj.com/.https://www.arcgis.com/ho

me/item.html?id=206b01a5a13541ef9fd067cfcaf3cdb

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

7&view=list&sortOrder=desc&sortField=defaultFSOr

der#overview. [Accessed: Jun. 3, 2024].

Gaber, M., Benlamkaddem, M., Jayyousi, R., Kurdi, L. and

Yassin, S. M. 2024. Effects of the COVID-19

Pandemic on Physical Activity among the Population

of Qatar. International Journal of Environmental

Research and Public Health 19(19): 12572. Available:

https://www.mdpi.com/1660-4601/19/19/12572.

[Accessed: Jun. 3, 2024].

Gianfredi, V. et al. 2021. Association between Urban

Greenspace and Health: A Systematic Review of

Literature. International Journal of Environmental

Research and Public Health 18(10). doi:

10.3390/ijerph18105137.

Greater London Authority (GLA). 2020. Diversity in

London. Available: https://www.london.gov.uk.

Greater London Authority (GLA). 2023. Super Output Area

Population (LSOA, MSOA) London. Available:

https://data.london.gov.uk/dataset/super-output-area-

populationlsoa-msoa-london.

Headey, B., Kelley, J. and Wearing, A. 1993. Dimensions

of mental health: Life satisfaction, positive affect,

anxiety and depression. Social indicators research 29:

63–82.

Grigoroglou, C., Munford, L., Webb, R. T., Kapur, N.,

Ashcroft, D. M. and Kontopantelis, E. 2020. Prevalence

of mental illness in primary care and its association with

deprivation and social fragmentation at the small-area

level in England. Psychological Medicine 50(2): 293–

302.

Heather, O., Julie, V., Jennifer, D. and Gayatri, J. 2017.

Measuring positive mental health in Canada: construct

validation of the mental health continuum—short form.

Health promotion and chronic disease prevention in

Canada: research, policy and practice 37(4): 123.

Henderson L. and Knight, T. 2012. Integrating the hedonic

and eudaimonic perspectives to more comprehensively

understand wellbeing and pathways to wellbeing.

Higgins, V. 2022. Ethnic density and area-deprivation

effects on central obesity among ethnic minority

people in England—A cross-sectional, multi-level

analysis. Frontiers in Epidemiology [Online].

https://www.frontiersin.org/articles/10.3389/fepid.202

2.1000155 [Accessed: May. 22, 2024].

Houlden, V., de Albuquerque, J. P., Weich, S. and Jarvis,

S. 2019. A spatial analysis of proximate greenspace and

mental wellbeing in London. Applied Geography 109:

102036.

James, S. L., Abate, D., Abate, K. H., Abay, S. M.,

Abbafati, C., Abbasi, N., Abbastabar, H., Abd-Allah,

F., Abdela, J. and Abdelalim, A. 2018. Global, regional,

and national incidence, prevalence, and years lived with

disability for 354 diseases and injuries for 195 countries

and territories, 1990–2017: a systematic analysis for the

global burden of disease study 2017. The Lancet

392(10159):1789 - 1858.

Keyes, C. L. 2002. The mental health continuum: From

languishing to flourishing in life. Journal of health and

social behavior: 207–222.

Kral, T.V.E. and Rauh, E.M. 2010. Eating behaviors of

children in the context of their family environment.

Physiology & Behavior 100 (5): 567-573.

https://doi.org/10.1016/j.physbeh.2010.04.031

Lachowycz, K. and Jones, A. P. 2013. Towards a better

understanding of the relationship between greenspace

and health: Development of a theoretical framework.

Landscape and Urban Planning 118: 62–69. doi:

10.1016/j.landurbplan.2012.10.012.

Mayor of London. 2018. The London Food Strategy.

Available: https://www.london.gov.uk [Accessed:

Apr. 15, 2024].

McLaren, L., Beck, C. A., Patten, S. B., Fick, G. H. and

Adair, C. E. 2008. The relationship between body mass

index and mental health: A population-based study of

the effects of the definition of mental health. Social

psychiatry and psychiatric epidemiology 43: 63–71.

Mueller, M. A. E., Midouhas, E. and Flouri, E. 2023. Types

of greenspace and adolescent mental health and well-

being in metropolitan London. Cities & Health, 7(3),

378–397. https://doi.org/10.1080/23748834.2023.217

5410.

NHS England. 2019. Health Survey for England 2018

[Online]. Available: https://digital.nhs.uk/data-and-

information/publications/statistical/health-survey-

forengland/2018 [Accessed: May. 22, 2024].

NHS England. 2020. Type 1 and type 2 diabetes and

COVID-19 related mortality in England. Available:

https://www.england.nhs.uk/publication/type-1-

andtype-2-diabetes-and-covid-19-related-mortality-in-

england/ [Accessed: Mar. 3,, 2024].

NHS England. 2023. National Child Measurement

Programme, England, 2022/23 School Year.

[Online].Available: https://digital.nhs.uk/data-and-

information/publications/statistical/national-child-

measurement-programme/2022-23-schoolyear.

NHS England. 2024. National Child Measurement

Programme." https://digital.nhs.uk/data-and-informati

on/publications/statistical/national-childmeasurement-

programme.

Office for National Statistics (ONS). 2019. Regional gross

disposable household income, UK: 1997 to 2018.

[Online]. Available: https://www.ons.gov.uk

[Accessed: Aug. 4, 2024].

ONS. 2024. Lower super output area population density.

https://www.ons.gov.uk/populationandmigration/popul

ationestimates/datasets/lowersuperoutputareapopulatio

ndensity.

Public Health England (PHE). 2020. Health Matters:

Obesity and the food environment. Available:

https://www.gov.uk/government/publications/health-

matters-obesity-andthe-food-environment

Rosenberg, M. 2019. Population Density Information and

Statistics. Available: https://www.thoughtco.com/popu

lation-density-overview-1435467 [Accessed: May. 13,

2024].

Ryan, R. M. and Deci, E. L. 2001. On happiness and human

potentials: A review of research on hedonic and

eudaimonic well-being. Annual review of psychology

52(1): 141–166.

Are Londoners Getting Healthier?

Šćepanović, S., Obadic, I., Joglekar, S., Giustarini, L.,

Nattero, C., Quercia, D., Zhu, X. 2023. MedSat: A

Public Health Dataset for England Featuring Medical

Prescriptions and Satellite Imagery. TUMMedia (data

sharing service from Technical University Munich),

under CC BY-SA 4.0 license. https://doi.org/10.14

459/2023mp1714817.

Schratz, L.M., Larkin, O., Dos Santos N. 2023. Caregiver

Influences on Eating Behaviors in Children: An

Opportunity for Preventing Obesity. Curr Atheroscler

Rep 25: 1035-1045. https://doi.org/10.1007/s11883-

023-01171-6.

Shenassa, E. D., Liebhaber, A. and Ezeamama, A. 2006.

Perceived Safety of Area of Residence and Exercise: A

Pan-European Study. American Journal of

Epidemiology 163 (11): 1012-1017 https://doi.org/

10.1093/aje/kwj142.

Stafford, M., Brunner, E. J., Head, J. and Ross, N. A. 2010.

Deprivation and the Development of Obesity: A

Multilevel, Longitudinal Study in England. American

Journal of Preventive Medicine 39 (2):130-139.

https://doi.org/10.1016/j.amepre.2010.03.021.

Stamatakis, E., Wardle, J. and Cole, T. 2010. Childhood

obesity and overweight prevalence trends in England:

Evidence for growing socioeconomic disparities.

International Journal of Obesity 34(1): 41-47.

https://doi.org/10.1038/ijo.2009.217.

Story, M., Nanney, M. S. and Schwartz, M. B. 2009.

Schools and Obesity Prevention: Creating School

Environments and Policies to Promote Healthy Eating

and Physical Activity. The Milbank Quarterly 87: 71-

100. https://doi.org/10.1111/j.14680009.2009.00548.x.

Sun, X., Hu, Y., Huang and T. O. Chan. 2020. Spatial

Patterns of Childhood Obesity Prevalence in Relation

to Socioeconomic Factors across England. ISPRS

International Journal of Geo-Information 9 (10): 599.

doi: https://doi.org/10.3390/ijgi9100599.

Van Lippevelde W. et al., 2012. Associations between

family-related factors, breakfast consumption and BMI

among 10- to 12-year-old European children: the cross-

sectional ENERGY-study. PLoS One 7(11): e49534.

Veitch J. et al., 2011. Children's perceptions of the use of

public open spaces for active free-play. Children's

Geographies 9(4): 421-435.

World Health Organization (WHO). 2024. Obesity.

https://www.who.int/healthtopics/obesity#tab=tab_1

Yang, J., Zheng, Y., Gou, X. et al. 2020. Prevalence of

comorbidities and its effects in patients infected with

SARS-CoV-2: A systematic review and meta-analysis.

International Journal of Infectious Diseases 94: 91-95.

Yang, J. K., Feng, Y., Yuan, M.Y. et al. 2006. Plasma

glucose levels and diabetes are independent predictors

for mortality and morbidity in patients with SARS.

Diabetic Medicine 23(6): 623-628.

Zhu, L., She, Z. G., Cheng, X. et al. 2020. Association of

blood glucose control and outcomes in patients with

COVID-19 and pre-existing type 2 diabetes. Cell

Metabolism 3(6):1068-1077.

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management