Impact of neighborhood resources on cardiovascular disease: a nationwide six-year follow-up
Calling et al. BMC Public Health
Impact of neighborhood resources on cardiovascular disease: a nationwide six-year follow-up
Susanna Calling 0 3
Xinjun Li 0 3
Naomi Kawakami 2
Tsuyoshi Hamano 1
Kristina Sundquist 0 3 4
0 Department of Clinical Sciences, Center for Primary Health Care Research, Skåne University Hospital, Lund University, Clinical Research Centre (CRC) , Building 28, floor 11, Jan Waldenströms gata 35, 205 02, Malmö , Sweden
1 Center for Community-based Health Research and Education (COHRE), Organization for the Promotion of Project Research, Shimane University , Izumo , Japan
2 Waseda Institute of Sport Sciences, Waseda University , Tokorozawa Saitama , Japan
3 Department of Clinical Sciences, Center for Primary Health Care Research, Skåne University Hospital, Lund University, Clinical Research Centre (CRC) , Building 28, floor 11, Jan Waldenströms gata 35, 205 02, Malmö , Sweden
4 Stanford Prevention Research Center, Stanford University , Stanford, CA , USA
Background: Living in a socially deprived neighborhood is associated with lifestyle risk factors, e.g., smoking, physical inactivity and unhealthy diet, as well as an increased risk of cardiovascular disease, i.e., coronary heart disease and stroke. The aim was to study whether the odds of cardiovascular disease vary with the neighbourhood availability of potentially health-damaging and health-promoting resources. Methods: A nationwide sample of 2 040 826 men and 2 153 426 women aged 35-80 years were followed for six years for first hospitalization of coronary heart disease or stroke. Neighborhood availability of health-damaging resources (i.e., fast-food restaurants and bars/pubs) and health-promoting resources (i.e., health care facilities and physical activity facilities) were determined by use of geographic information systems (GIS). Results: We found small or modestly increased odds ratios (ORs) for both coronary heart disease and stroke, related to the availability of both health-damaging and health-promoting resources. For example, in women, the unadjusted OR (95 % confidence interval) for stroke in relation to availability of fast-food restaurants was 1.18 (1.15-1.21). Similar patterns were observed in men, with an OR = 1.08 (1.05-1.10). However, the associations became weaker or disappeared after adjustment for neighborhood-level deprivation and individual-level age and income. Conclusions: This six year follow-up study shows that neighborhood availability of potentially health-damaging as well as health-promoting resources may make a small contribution to the risk of coronary heart disease and stroke. However, most of these associations were attenuated or disappeared after adjustment for neighborhood-level deprivation and individual-level age and income. Future studies are needed to further examine factors in the causal pathway between neighborhood deprivation and cardiovascular disease.
Cardiovascular disease; Follow-up study; Multilevel analysis; Neighborhood
Cardiovascular disease (CVD) is still the leading cause of
death in Sweden and many other Western countries,
even if there has been a marked decrease in the incidence
rates during the last decades [
]. The reasons behind CVD
are multifactorial, including metabolic, sociodemographic
and lifestyle factors, e.g., smoking, unhealthy diet and
physical inactivity [
]. The prevalence of obesity has
markedly increased during the last decades . Efforts
have been made to counteract the unfavorable trend by
interventions to increase physical activity and reduce
intake of unhealthy food in the population [
studies have shown associations between modifiable risk
factors, e.g., physical inactivity and obesity, and living in a
socially deprived neighborhood [
Furthermore, neighborhood-level socioeconomic
deprivation is associated with CVD, and the reasons behind
this association are not fully explored [
risk factors could lie in the pathway, but the
underlying mechanisms are poorly understood. Studies linking
neighborhood environment and CVD risk factors have
yielded mixed results and have most often been limited by
cross-sectional designs and same-source bias [
6, 7, 13, 14
It has been debated whether deprived neighborhoods
have higher density of health-damaging resources, such
as fast food restaurants, and lower access to
healthpromoting resources, such as health care services and
physical activity facilities . Previous studies have
found that the density of fast-food restaurants is
associated with neighborhood-level deprivation [
However, no clear associations have been found between lack
of health-promoting resources and neighborhood-level
]. Moreover, several cross-sectional
studies have found that individuals living in areas with
high density of fast-food restaurants were more likely
to be obese, but some studies could not confirm this
]. Contradictory results have been
reported for a potential association between density of
physical activity facilities and level of physical activity
Studies trying to disentangle a possible association
between neighborhood resources and CVD as endpoint
have found no clear associations [
most studies did not have a longitudinal multilevel
approach. In 2011, Kawakami et al. performed a
multilevel analysis of a large Swedish study sample during
two years of follow-up, which showed weak associations
between coronary heart disease (CHD) and neighborhood
availability of potentially health-damaging and
healthpromoting goods, services and resources, but the
associations disappeared after controlling for neighborhood-level
deprivation and individual-level age and income .
In a similar Swedish study in 2013, Hamano et al.
showed weak associations between stroke and
neighborhood resources, which partly remained after
adjustments for individual-level sociodemographic factors
and neighborhood-level deprivation [
]. In these two
studies, it was concluded that a longer follow-up was
needed to further disentangle the possible association
between neighborhood resources and CVD. The aim of
the present study was to clarify whether neighborhood
resources make any contribution to cardiovascular risk
during a six-year follow-up and it represents the largest
and longest follow up so far using the “hard” outcomes
coronary heart disease and stroke.
The novelty of the present study is the availability of a
nationwide sample of more than 4 million individuals
with an extended follow-up period of six years, and a
longitudinal multilevel approach. The first aim was to
examine, in men and women separately, whether the
risk of CHD and stroke was associated with
neighborhood availability of potentially health-damaging resources,
such as fast-food restaurants and bars/pubs, as well as
health-promoting resources, such as physical activity
facilities and health care services. The second aim was
to study whether controlling for neighborhood-level
deprivation and individual-level age and income affected
The study population included a nationwide sample of
2 040 826 men and 2 153 426 women aged 35–80 years.
The same study population was used in our previous
] but the present study was based on an
extended follow-up of six years. Data were retrieved from
national Swedish registers, containing information on
nationwide individual-level medical diagnoses from the
Swedish Hospital Discharge Register (obtained from the
National Board of Health and Welfare) and the Cause
of Death Register. These data were linked to registers
obtained from Statistics Sweden, the Swedish
Governmentowned statistics bureau, which provided individual-level
data on socioeconomic factors, such as age and income.
All individuals were followed between 1 January 2005
and 31 December 2010 for their first hospitalization of
CHD or stroke as well as mortality for CHD or stroke.
Men and women with either pre-existing CHD or stroke,
defined as hospitalization ≤ 5 years before the start of
the study, were excluded. In an additional analysis, we
excluded those individuals with pre-existing CHD or
stroke ten years prior start of the study. The disease codes
were based on the 10th version of the International
Classification of Diseases (ICD-10) and included the following
I20, angina pectoris; I21, acute myocardial infarction;
I22, subsequent myocardial infarction; I23,
complications owing to acute cardiac infarction; I24,
other acute forms of CHD; and I25, chronic CHD.
I60, subarachnoidal haemorrhage; I61-62, hemorrhagic
stroke; I63 cerebral infarction; I64, acute cerebrovascular
disease not specified as hemorrhagic or infarction;
I65-I65, occlusion and stenosis of precerebral and cerebral
arteries; I67-I68, other cerebrovascular diseases; and I69,
late effects of cerebrovascular disease.
Four categories of neighborhood goods, services and
resources were selected to represent potentially
healthdamaging and health-promoting resources [
1. fast-food restaurants (e.g., pizzerias and hamburger
3. health care facilities (e.g., health care centers, public
hospitals, dentists, and pharmacies)
4. physical activity facilities (e.g., swimming pools,
gyms, and ski facilities).
The ready-to-use nationwide GIS (Geographic
Information Systems) dataset of business contacts (i.e., goods,
services and resources) was provided to us by the Swedish
company Teleadress, which is a leading provider of
Swedish contact information [
]. Teleadress provides
information on practically all businesses and services in Sweden
with a registered telephone number and/or businesses that
have provided information about their existence to the
company. Both government and private entities are
included. The database is updated continuously and has a
high level of completeness of the data [
]. Data was
drawn from 2005.
Neighborhood availability of the four categories was
measured in 2005 as counts per predefined area called
SAMS (Small Area Market Statistics, provided by
Statistics Sweden, see below) by the use of GIS. Availability
was defined as the presence within the SAMS unit of at
least one feature for the category in question. Presence
yes/no was chosen to define neighborhood availability
rather than linear density because the large majority had
no access to the studied resources [
]. The correlation
between the presence of the different types of resources
was low and varied between 0.2 and 0.4.
Neighborhood of residence was provided to us by
Statistics Sweden via data from the National Land
Survey. The home addresses of all individuals had been
previously geocoded, allowing us to identify the SAMS
units in which the participants lived. However, the
researchers in the present study had no access to the
home addresses of the individuals, in order to ensure
all individuals’ integrity. SAMS units were used as proxies
for neighborhoods and has been used previously [
Each unit contains an average of around 1000 residents in
all Sweden and around 2000 in Stockholm. This study
examined only those SAMS units that overlap with
‘localities’ or urban areas. In Sweden, ‘localities’ (which are
defined by Statistics Sweden every fifth year) represent
any village, town or city with a minimum of 200 residents
and adjacent areas where the houses are no more than
200 m apart . We chose to include only SAMS
units overlapping with localities because more rural
SAMS units have very few goods, services and
resources. In 2005, 1940 Swedish localities were identified
by Statistics Sweden. GIS were used to overlay the
SAMS boundaries with the locality boundaries. Of a
total of 9 617 SAMS units in Sweden, 7 945 overlapped
with localities and were therefore selected. SAMS units
with fewer than 50 people were excluded on the basis that
they might yield unreliable statistical estimates in the
calculation of the neighborhood deprivation index. A final total
of 7 033 SAMS units was included in the present study.
In order to investigate the individuals’ immediate
neighborhood, we also used buffer zones. Statistics Sweden
provided information about the individuals’ land coordinates
in squares of 100 × 100 m, ensuring the integrity of the
individuals. For each individual, a buffer zone with a radius
of 1000 m was applied, i.e., a distance that most people
are willing to walk. Most previous studies use ½ mile
(~800 m) and/or 1 mile (~1600 m), which corresponds
relatively well to the present study [
]. The number of
goods, services and resources within the buffer zones was
calculated using GIS. Availability was defined as the
presence within the buffer zone of at least one feature for
the category in question.
All covariates were measured at baseline in 2005.
Individual-level age was categorized as 35–44, 45–54,
55–64, 65–74 and 75–80 years.
Individual-level family income was calculated from the
annual family income divided by the number of family
members and took into account the ages of the people
in the family. The variable was categorized as empirical
quartiles of the distribution.
Neighborhood-level deprivation was included in the
analysis as a covariate, as previous research has shown that
it is associated with an increased risk of CHD [
This is in line with previous studies [
]. A summary
measure was used to characterize neighborhood-level
deprivation. We identified deprivation indicators used by
past studies to characterize neighborhood environments
and then used a principal components analysis to select
deprivation indicators in the national registers. The
following four variables were selected for those aged
25–64: low educational status (<10 years of formal
education); low income (income from all sources,
including that from interest and dividends, defined as
less than 50 % of individual median income);
unemployment (not employed, excluding full-time students, those
completing compulsory military service, and early
retirees); and social welfare recipient. Each of the four
variables loaded on the first principal component with
similar loadings (+.47 to + .53) and explained 52 % of the
variation between these variables. A z score was calculated
for each SAMS unit, and these scores were summed to
create the index. The z scores, weighted by the coefficients for
the eigenvectors, were then summed to create the index.
The index was categorized into three groups: below one
standard deviation (SD) from the mean (low deprivation),
above one SD from the mean (high deprivation), and
within one SD of the mean (moderate deprivation). Higher
scores reflect more deprived neighborhoods. The rationale
for this categorisation was that it is easier to interpret
results in categories and accordingly identify populations for
intervention. The neighborhood deprivation index was
based on those individuals living in the neighborhood aged
25–64 years as individuals in those ages are supposed to be
more socioeconomically active and thus have a stronger
impact on the neighborhood than, for example, young
persons and retirees. The variables used in the index are most
relevant among those in working ages, which is around
25–64 years of age in Sweden. All individuals in the
study population (ages 35–80 years) were assigned a
Age-standardized incidence proportions (proportions of
subjects who became cases among those who entered
the study time interval) were calculated separately for
men and women by direct age standardization using
10-year age groups, with the entire Swedish population
of men or women aged 35–80 as the standard population.
Multilevel (hierarchical) logistic regression models
were created with incidence proportions as the outcome
variables. The analyses were performed using MLwiN.
Multilevel logistic regression models were used in the
computing process to help our multilevel models to
converge, as the large dataset did not converge when we
used multilevel Cox proportional hazards models. These
models are a good approximation of multilevel Cox
proportional hazards models under certain conditions,
such as ours, i.e., a large sample size, low incidence
rates, risk ratios of moderate size and relatively short
]. Previous studies have used a similar
]. For comparison, we made a sensitivity
analysis by creating ordinary Cox proportional hazards
models (Additional file 1: Tables S1 and S2). These
models gave similar results. First, we created models
that only included neighborhood availability of each of
the four categories of neighborhood goods, services and
resources in order to determine the crude odds ratios (OR)
for CHD and stroke with 95 % confidence intervals (CI).
Next, we created a model that included
neighborhoodlevel deprivation. The third and final model included
neighborhood availability, neighborhood-level deprivation
and individual-level age and income.
Random intercept multilevel logistic regression
models were used to allow for the clustering of individuals
within neighborhoods and to estimate the variance in
CVD risk that is attributable to neighborhood
characteristics. This approach was used to estimate the intraclass
correlation coefficient (ICC), the proportion of variance in
the outcome attributable to differences between individuals
in different neighborhoods in contrast to differences
between individuals within the same neighborhood
]. The ICC was estimated by applying the latent
variable method as exemplified by:
ICC ¼ V n þ π2=3
where Vn is the variance between neighborhoods and
π2/3 is the variance between individuals.
The proportion of the second level variance explained
by the different variables was calculated as:
V EXPLAINED ¼
V 0−V 1
where Vo is the age adjusted variance in the initial
model and V1 is the second level variance in the different
Cross-level interaction tests were performed. The
interaction tests were performed to examine potential
interactions between neighborhood deprivation and
availability of the four categories of resources. The results
showed no meaningful interactions (data not shown).
Table 1 shows the distribution of the study population
and the number of CHD and stroke events in relation to
neighborhood deprivation and availability of the four
types of resources, in men and women separately. In
men, 12.7 % had access to bars/pubs in their
neighborhood, 45.4 % to fast food restaurants, 38.5 % to
healthcare services and 41.0 % to physical activity facilities. In
women, similar proportions were found. Incidence of
CVD increased with higher level of neighborhood
deprivation in all subgroups; however, there were no
large differences in relation to availability of the four
resources. In men, the OR (95 % CI) in relation to high
neighborhood deprivation compared to low deprivation
was 1.39 (1.35–1.42) for CHD, and 1.28 (1.24–1.32) for
stroke, adjusted for individual-level age and income. The
corresponding ORs in women were 1.60 (1.55–1.66) for
CHD and 1.36 (1.31–1.41) for stroke (data not shown
Table 2 shows the associations between incidence of
CHD/stroke and neighborhood availability of the four
resources in the SAMS units. The ORs with 95 % CIs
are presented for CHD and stroke, in relation to
neighborhood availability. In the unadjusted models
(Model 1), availability of all resources, both
healthpromoting and health-damaging, showed statistically
significant increased ORs for both CHD and stroke.
The associations were stronger in women than in men.
The highest ORs were found in women for fast-food
restaurants (OR for CHD = 1.17 [95 % CI: 1.13–1.20], ICC
0.043) and for health care facilities (OR for CHD = 1.22
[95 % CI: 1.19–1.26], ICC 0.042). After adjustments for
neighborhood-level deprivation (Model 2) and
individuallevel age and income (Model 3), the associations became
weaker or disappeared. The associations for stroke were
generally somewhat stronger than for CHD. For CHD, the
increased ORs remained only for neighborhood
availability of fast-food restaurants and health care facilities
in women, after full adjustments.
ICC was generally low and varied between 0.004 and
0.010 in the fully adjusted models, indicating that the
majority of variance in CVD risk was attributed to
within-neighborhood rather than between-neighborhood
Table 3 shows the associations between incidence of
CHD/stroke and the “immediate” neighborhood
availability of the four resources in the buffer zones (radius
1000 m). The associations were similar to the results in
the SAMS units, but the ORs in the buffer zones were
somewhat stronger. For stroke, there were after full
adjustments (Model 3) statistically significant increased
ORs in both men and women, (e.g., OR in men = 1.06
[95 % CI: 1.04–1.08]) for fast-food restaurants and OR =
1.07 [95 % CI: 1.05–1.09] for health care facilities), with the
exception of OR for stroke in relation to physical activity
resources in women, where no association was found.
We also made an additional analysis with only mortality
from CHD and stroke as an outcome. The results were
overall similar to the main results; however, no
associations were found for stroke mortality in women.
In a separate analysis, we excluded individuals with
hospitalization of CVD ten years prior to the start of the
follow-up; however, the results from this sensitivity analysis
were very similar to the main results in the present study.
The main finding of the present study with six years’
follow-up is that neighborhood availability of potentially
health-damaging as well as health-promoting resources
contributes to small but increased odds of CHD and
stroke. However, the major part of the increased odds was
related to neighborhood-level deprivation and
individuallevel age and income, which is in line with the results of
our previous studies with shorter follow-up [
results indicate that vicinity of health-damaging resources,
i.e., fast food and bars/pubs, may make a small
contribution to an increased cardiovascular risk, probably through
an unhealthier lifestyle. However, the effect seems to be
small and this research area is not fully understood. Earlier
studies have shown that the built environment can affect
obesity and physical activity, but the relationship is
complex and operates through many mediating variables, such
as socio-economic characteristics, personal and cultural
variables, and safety in the built environment [
Neighborhood availability of fast food restaurants has been shown
to be related to increased fast food consumption [
density of fast-food restaurants is associated with CVD
25, 27, 28
]. However, we do not know whether the
studied individuals choose resources that are not in their
immediate neighborhood. As the studied health-damaging
resources were only weak predictors of CVD, other
neighborhood mechanisms may more strongly influence the
increased risk of CVD in socioeconomically deprived
neighborhoods, e.g., social disintegration (criminality and
], low social capital [
] and stress [
Concerning potentially health-promoting resources,
the present study showed that high neighborhood
availability, especially of health care facilities, was also
associated with increased risk of CVD. This supports the
conclusion that the increased cardiovascular risk in
deprived neighborhoods is mainly associated with individual
socioeconomic characteristics as well as the
neighborhoodlevel deprivation itself rather than the availability of
neighborhood resources, and that it is unlikely that
neighborhood resources have a large influence on
cardiovascular risk. Generally, the associations between neighborhood
availability of the studied resources and CVD were
somewhat stronger for stroke than for CHD. Previous
research indicates that genetic influence on stroke risk is
weaker than for other cardiovascular manifestations
], suggesting that lifestyle and other modifiable
risk factors may have stronger influence on the disease
development. However, when studying stroke mortality
separately we found no associations in women. As
mortality is an outcome that comes later than lifestyle
factors and cardiovascular disease, it is also more difficult
to interpret. For example, for mortality, genetic factors
may play an important role.
Strengths and limitations
Our study had several strengths. We were able to follow
a large study sample of the entire Swedish population
for CVD, which so far only has been possible in our
nationwide study population. In the present study, we were
able to extend the follow-up period to six years to better
follow the cardiovascular risk than in our previous studies
]. Six years of follow-up is still short enough to
ensure that most people stayed in their neighborhood, so
that any potential changes in availability of resources were
minor. Preliminary unpublished data from Sweden shows
that only 1 % of middle-aged people change neighborhood
socioeconomic status during a 10-year period.
The multi-level approach and adjustments for
neighborhood-level deprivation and individual-level
covariates measured at baseline are also strengths of the
present study. Moreover, we had access to valid
geographic information about neighborhood availability of
several different resources. The data on neighborhood
resources are very complete, as Teleadress provides
information on practically all businesses in Sweden [
Finally, our use of small neighborhood units (SAMS and
individuals’ buffer zones) increases the probability that the
residents were actually exposed to the studied resources.
However, our study also had certain limitations. We
do not know whether the residents actually utilized the
studied resources in their neighborhoods, or whether the
individuals changed neighborhoods during the follow-up
period. Another limitation was that the neighborhoods
were based on geographic areas and may therefore not
have corresponded perfectly with the individuals’ social
environment. Moreover, it is likely that residual confounding
exists. We had no data on lifestyle factors, e.g., unhealthy
diet, smoking and physical inactivity, which are important
and modifiable cardiovascular risk factors.
In this large-scale study with six years follow up, we
found increased but small odds of CHD and stroke
associated with neighborhood availability of potentially health
damaging fast-food restaurants and bars/pubs, as well as
availability of potentially health-promoting healthcare
services and physical activity facilities. Although planning
of residential areas should aim to promote neighborhood
resources that encourage a healthy lifestyle, the present
results suggest that neighborhood resources are unlikely to
make an important contribution to variations in
cardiovascular disease by neighborhood deprivation. More studies
in this area are needed to further examine potential causal
factors in the pathway between neighborhood deprivation
Additional file 1: Table S1. Models showing associations between
coronary heart disease and stroke and neighborhood availability of potentially
health-damaging and health-promoting goods, services and resources in the
“small area market statistics”; Cox regression. Table S2. Models showing
associations between coronary heart disease and stroke and neighborhood
availability of potentially health-damaging and health-promoting goods,
services and resources in the “buffer zones”; Cox regression. (DOCX 22 kb)
CHD, coronary heart disease; CI, confidence interval; CVD, cardiovascular disease;
GIS, geographic information systems; ICC, intraclass correlation coefficient;
OR, odds ratio; SAMS, Small Area Market Statistics; SD, standard deviation
This work was supported by grants to Kristina Sundquist from The Swedish
Research Council and ALF funding from Region Skåne. Research reported in
this publication was also supported by the National Heart, Lung, And Blood
Institute of the National Institutes of Health under Award Number R01HL116381
to Kristina Sundquist. The content is solely the responsibility of the authors
and does not necessarily represent the official views of the National
Institutes of Health.
Availability of data and materials
Data will not be publically shared, as these are owned by Statistics Sweden
and the National Board of Health and Welfare.
SC, XL, NK, TH, KS worked on conception of the study; XL wrote the initial
statistical analysis plan; SC, NK, TH and KS contributed to the statistical
analysis plan; XL analysed the data; SC, NK, TH, KS contributed to the analysis
and interpretation of the data; SC drafted the paper; XL, NK, TH, KS worked
on further drafting and revising the paper critically. All authors read and
approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
The study was approved by the Ethics Committee of Lund University.
Consent to participate was not applicable as the study was based on
anonymous database material.
1. National Board of Health and Welfare . Causes of Death 2011 [cited 2016 ]. http://www.socialstyrelsen.se/statistics/statisticaldatabase/causeofdeath. Accessed 20 Jul 2016 .
2. Libby P , Ridker PM , Hansson GK . Progress and challenges in translating the biology of atherosclerosis . Nature . 2011 ; 473 : 317 - 25 .
3. World Health Organization. Global Atlas on cardiovascular disease prevention and control . Publication date 2011 . http://www.who.int/ cardiovascular_diseases/publications/atlas_cvd/enAccess date: July 20 , 2016 . Accessed 20 Jul 2016 .
4. Finucane MM , Stevens GA , Cowan MJ , Danaei G , Lin JK , Paciorek CJ , et al. National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants . Lancet . 2011 ; 377 : 557 - 67 .
5. Lau DC , Douketis JD , Morrison KM , Hramiak IM , Sharma AM , Ur E , et al. 2006 Canadian clinical practice guidelines on the management and prevention of obesity in adults and children [summary] . CMAJ . 2007 ; 176 : S1 - 13 .
6. Brennan LK , Brownson RC , Orleans CT . Childhood obesity policy research and practice: evidence for policy and environmental strategies . Am J Prev Med . 2014 ; 46 : e1 - 16 .
7. Hanson S. Active living research in light of the TRB/IOM report . J Phys Act Health . 2006 ; 3 Suppl 1 : S258 - 66 .
8. Cubbin C , Sundquist K , Ahlen H , Johansson SE , Winkleby MA , Sundquist J . Neighborhood deprivation and cardiovascular disease risk factors: protective and harmful effects . Scand J Public Health . 2006 ; 34 : 228 - 37 .
9. Sundquist J , Malmstrom M , Johansson SE . Cardiovascular risk factors and the neighbourhood environment: a multilevel analysis . Int J Epidemiol . 1999 ; 28 : 841 - 5 .
10. Diez Roux AV , Merkin SS , Arnett D , Chambless L , Massing M , Nieto FJ , et al. Neighborhood of residence and incidence of coronary heart disease . N Engl J Med . 2001 ; 345 : 99 - 106 .
11. Sundquist K , Malmstrom M , Johansson SE . Neighborhood deprivation and incidence of coronary heart disease: a multilevel study of 2.6 million women and men in Sweden . J Epidemiol Community Health . 2004 ; 58 : 71 - 7 .
12. Winkleby M , Sundquist K , Cubbin C. Inequities in CHD incidence and case fatality by neighborhood deprivation . Am J Prev Med . 2007 ; 32 : 97 - 106 .
13. Boehmer TK , Hoehner CM , Deshpande AD , Brennan Ramirez LK , Brownson RC . Perceived and observed neighborhood indicators of obesity among urban adults . Int J Obes . 2007 ; 31 : 968 - 77 .
14. Fisher KJ , Li F , Michael Y , Cleveland M . Neighborhood-level influences on physical activity among older adults: a multilevel analysis . J Aging Phys Act . 2004 ; 12 : 45 - 63 .
15. Macintyre S. Deprivation amplification revisited; or, is it always true that poorer places have poorer access to resources for healthy diets and physical activity? Int J Behav Nutr Phys Act . 2007 ; 4 : 32 .
16. Macdonald L , Cummins S , Macintyre S. Neighbourhood fast food environment and area deprivation-substitution or concentration? Appetite . 2007 ; 49 : 251 - 4 .
17. Pearce J , Blakely T , Witten K , Bartie P. Neighborhood deprivation and access to fast-food retailing: a national study . Am J Prev Med . 2007 ; 32 : 375 - 82 .
18. Macintyre S , Macdonald L , Ellaway A . Do poorer people have poorer access to local resources and facilities? The distribution of local resources by area deprivation in Glasgow, Scotland . Soc Sci Med . 2008 ; 67 : 900 - 14 .
19. Pearce J , Witten K , Hiscock R , Blakely T. Are socially disadvantaged neighbourhoods deprived of health-related community resources? Int J Epidemiol . 2007 ; 36 : 348 - 55 .
20. Holsten JE . Obesity and the community food environment: a systematic review . Public Health Nutr . 2009 ; 12 : 397 - 405 .
21. Mehta NK , Chang VW . Weight status and restaurant availability a multilevel analysis . Am J Prev Med . 2008 ; 34 : 127 - 33 .
22. Pearce J , Hiscock R , Blakely T , Witten K. A national study of the association between neighbourhood access to fast-food outlets and the diet and weight of local residents . Health Place . 2009 ; 15 : 193 - 7 .
23. Diez Roux AV , Evenson KR , McGinn AP , Brown DG , Moore L , Brines S , et al. Availability of recreational resources and physical activity in adults . Am J Public Health . 2007 ; 97 : 493 - 9 .
24. Lee RE , Cubbin C , Winkleby M. Contribution of neighborhood socioeconomic status and physical activity resources to physical activity among women . J Epidemiol Community Health . 2007 ; 61 : 882 - 90 .
25. Alter DA , Eny K. The relationship between the supply of fast-food chains and cardiovascular outcomes . Can J Public Health . 2005 ; 96 : 173 - 7 .
26. Chaix B . Geographic life environments and coronary heart disease: a literature review, theoretical contributions, methodological updates, and a research agenda . Annu Rev Public Health . 2009 ; 30 : 81 - 105 .
27. Daniel M , Paquet C , Auger N , Zang G , Kestens Y. Association of fast-food restaurant and fruit and vegetable store densities with cardiovascular mortality in a metropolitan population . Eur J Epidemiol . 2010 ; 25 : 711 - 9 .
28. Morgenstern LB , Escobar JD , Sanchez BN , Hughes R , Zuniga BG , Garcia N , et al. Fast food and neighborhood stroke risk . Ann Neurol . 2009 ; 66 : 165 - 70 .
29. Kawakami N , Li X , Sundquist K . Health-promoting and health-damaging neighbourhood resources and coronary heart disease: a follow-up study of 2 165 000 people . J Epidemiol Community Health. 2011 ; 65 : 866 - 72 .
30. Hamano T , Kawakami N , Li X , Sundquist K. Neighbourhood environment and stroke: a follow-up study in Sweden . PLoS One . 2013 ; 8 , e56680 .
31. Kawakami N , Winkleby M , Skog L , Szulkin R , Sundquist K. Differences in neighborhood accessibility to health-related resources: a nationwide comparison between deprived and affluent neighborhoods in Sweden . Health Place . 2011 ; 17 : 132 - 9 .
32. Teleadress . https://www.bisnode.com/sverige/tjanster/teleadress/. Accessed 20 Jul 2016 .
33. Sundquist K , Winkleby M , Ahlen H , Johansson SE . Neighborhood socioeconomic environment and incidence of coronary heart disease: a follow-up study of 25,319 women and men in Sweden . Am J Epidemiol . 2004 ; 159 : 655 - 62 .
34. Statistics Sweden. Localities 2010 : Buildings and Urban Structures . 2012 . ISSN 1654-3823.
35. James P , Berrigan D , Hart JE , Hipp JA , Hoehner CM , Kerr J , et al. Effects of buffer size and shape on associations between the built environment and energy balance . Health Place . 2014 ; 27 : 162 - 70 .
36. Callas PW , Pastides H , Hosmer DW . Empirical comparisons of proportional hazards, poisson, and logistic regression modeling of occupational cohort data . Am J Ind Med . 1998 ; 33 : 33 - 47 .
37. Snijders T , Bosker R. Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling . Thousand Oaks: Sage Publications; 1999 .
38. Boone-Heinonen J , Gordon-Larsen P , Kiefe CI , Shikany JM , Lewis CE , Popkin BM . Fast food restaurants and food stores: longitudinal associations with diet in young to middle-aged adults: the CARDIA study . Arch Intern Med . 2011 ; 171 : 1162 - 70 .
39. Sundquist K , Theobald H , Yang M , Li X , Johansson SE , Sundquist J . Neighborhood violent crime and unemployment increase the risk of coronary heart disease: a multilevel study in an urban setting . Soc Sci Med . 2006 ; 62 : 2061 - 71 .
40. Sundquist J , Johansson SE , Yang M , Sundquist K. Low linking social capital as a predictor of coronary heart disease in Sweden: a cohort study of 2.8 million people . Soc Sci Med . 2006 ; 62 : 954 - 63 .
41. Daniel M , Moore S , Kestens Y. Framing the biosocial pathways underlying associations between place and cardiometabolic disease . Health Place . 2008 ; 14 : 117 - 32 .
42. Calling S , Ji J , Sundquist J , Sundquist K , Zoller B . Shared and non-shared familial susceptibility of coronary heart disease, ischemic stroke, peripheral artery disease and aortic disease . Int J Cardiol . 2013 ; 168 : 2844 - 50 .
43. Meschia JF , Tournier-Lasserve E . Advances in stroke: genetics 2012 . Stroke. 2013 ; 44 : 309 - 10 .