Spatial Analysis and Modeling for Health Applications Lars Skog
Doctoral Thesis 1
© Lars Skog 2014 Doctoral Thesis Royal Institute of Technology (KTH) Department of Urban Planning and Environment Division of Geodesy and Geoinformatics SE-100 44 Stockholm Sweden TRITA SoM 2014-03 ISSN 1653-6126 KTH/SoM/14-03/SE ISBN 978-91-7595-040-2 Printed by
2
Abstract Despite the benefits of applying methods of geographic information science (GIScience), the use of such methods in health service planning and provision remains greatly underutilized. Spread of epidemic diseases is a constant threat to mankind and the globalization of the world increases the risk for global attacks from multi-resistant bacteria or deadly virus strains. Therefore, research is needed to better understand how GIScience could be used in epidemiologic analyses and other health applications. This thesis is divided into two parts; one for epidemiologic analyses and one for neighbourhood studies. The overall objective of the epidemiologic part of this research is to understand more about the spatial spread of past pandemics and to find out if there are any common patterns. This overall objective is divided into four specific research objectives; 1) to describe the spatial spread of the Russian Influenza in Sweden, 2) to create models of propagation of the Black Death in Sweden, 3) to establish spatiotemporal characteristics common to past pandemics in Sweden and 4) to visualize the spatiotemporal occurrence of salmonella among animal herds in Sweden. This thesis also discusses some other aspects of health related to place. Are differences in neighbourhood deprivation related to the amount of presence of goods and services? Is the way cities are planned affecting the behaviour within the local population regarding spontaneous walking and physical activity? The specific research objectives for this part are to define how deprivation is related to presence of goods and services in Sweden and to create walkability indices over the city of Stockholm including a quality test of these indices. Case data reported by physicians were used for the epidemiologic studies. The pandemics discussed covered the entire world, but our data is from Sweden only and as regards the Black Death there was no case data at all. The data for the goods and services analyses are from all of Sweden, whereas the walkability indices are based on data from the city of Stockholm. Various methods have been used to clean, structure and geocode the data, including hand written reports on case data, maps of poor geometric quality, information from databases on climate, demography, diseases, goods and services, income data and more, to 3
make this data feasible for spatial analysis, modeling and visualization. Network analysis was used to model food transports in the 14th century as well as walking in the city of Stockholm today. Proximity analysis was used to assess the spatio-temporal spread of the Russian Influenza. The impact of climatological factors on the propagation of the Asian Influenza was analyzed and geographically weighted mean (GWM) calculations were used to discover common characteristics in the spatio-temporal spread of three past pandemics. Among the results generated in the epidemiologic study the following should be noted in particular; the local peaking periods of the Asian Influenza were preceded by falling temperature, the total peaking period for the three pandemics (Russian, Asian and A(H1N1)pdm09) was approximately 10 weeks and their weekly GWM followed a path from southwest to northeast (opposite direction for the A(H1N1)pdm09). From the neighborhood studies one can note that compared to the results measured and reported by tested individuals there is a positive (small but significant) association between neighborhood walkability and physical activity outcomes. The main contribution of this work is that it gives epidemiologists and public health specialists new ideas, not only on how to formulate, model, analyze and visualize different health related research questions but also ideas on how new procedures could be implemented in their daily work. Once the data reporting is organized in a suitable manner there is a multitude of options on how to present important and critical information to officials and policy makers.
Key words: Sweden, spatial analysis, spatial modeling, spatio-temporal spread, epidemiology, pandemic, walkability, health, Russian influenza, Asian influenza, A(H1N1)pdm09, GWM, climate factors
4
Acknowledgements Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning. These words from another old man, about my age at that time, never lose their meaning. To present this doctoral thesis is of course the end of a long process with many challenges to overcome, but it is also hopefully the beginning of a new phase with many more new challenges. In my work with this thesis I have had the opportunity to make friends with numerous interesting and brilliant people, who gladly shared some of their knowledge and wisdom with me. GIScience is still an unknown among most professionals working in the health sector. My hope is that the work conducted will in some way contribute to a broader understanding and use of a technique much needed in our efforts to make this world a little bit better. Much of the work has been done in close cooperation with representatives of the Swedish Institute for Infectious Diseases Control (SMI). Other organizations involved are the Karolinska Institutet in Stockholm, the GeoBiosphere Science Centre at Lund University, the Department of Medical Biosciences at Umeå University, the National Veterinary Institute and the Division of Geoinformatics at the Royal Institute of Technology in Stockholm. There are so many to whom I am extremely grateful. Still I am probably forgetting many of you and for that there is no excuse. First of all I would like to thank my wife Anna-Lena for her continuous support. Without her encouragement and inspiration this work would never have been done. My family means more than anything to me and I am so glad to share this with my children and their families (Emilia, Johan, Matilda, Johanna and Hugo; Karolina, Nicolas, Axel, Eric and Leo; Viktoria, Martin, Astrid, Ingrid and David; Oscar and Maria) and my mother. My father, who was a keen supporter, is of course included too, and so are Tobias, Nina, Elias, Torbjörn, Jenny, Izabelle and Emilie as well as my brother Anders and his family. I wish to thank my supervisor Associate Professor Hans Hauska for being so patient with me and for all his valuable contributions to our 5
discussions. I am also very grateful to Professor Yifang Ban for giving me the chance to undertake this project and for all her warm encouragement and valuable comments. To all my coworkers* in the papers presented in this thesis I would like to express my deep gratitude for taking the time and for their great work. This goes for all of you and Professor Annika Linde in particular for sharing knowledge and inspiration with me. I would also like to thank all my colleagues at Esri Sverige for their willingness to help and discuss whenever I needed support. *Thank you Annika, Hans, Lisa, Anna, Mohammed, Håkan, Kristina, Naomi, Marilyn, Robert, Ulf, Henrik, Daniel, Susanna, Jenny, Helene, Helena, Sharon and Fredrik!!
6
Table of Contents 1
2
Introduction ................................................................................ 11 1.1
Background .......................................................................... 11
1.2
Research Objectives............................................................. 14
1.3
Thesis organization .............................................................. 17
Literature review ........................................................................ 22 2.1
Spatial analysis and modeling of epidemic diseases ........... 22
2.2
Neighborhood accessibility, walkability and health ............ 34
3
Study areas and data description ................................................ 39
4
Methodology .............................................................................. 44
5
6
7
4.1
Data preparation................................................................... 45
4.2
Spatial Analysis ................................................................... 47
4.2.1
Spatial analysis for epidemiology ................................ 48
4.2.2
Spatial analysis for neighborhood studies .................... 52
4.3
Spatial modeling .................................................................. 57
4.4
Visualization ........................................................................ 61
Results & Discussion.................................................................. 66 5.1
Results and discussion of epidemiologic study ................... 66
5.2
Results and discussion of neighborhood study .................... 72
Conclusions and future research................................................. 75 6.1
Conclusions.......................................................................... 75
6.2
Future research..................................................................... 78
References .................................................................................. 80
7
List of Figures Figure 1. Organization of the selected papers in the thesis ................ 19 Figure 2. Mechanism of ripple-spreading (after Liao et al. 2013) ..... 25 Figure 3. The epidemic states and how they relate (after Liao et al. 2013) .................................................................................. 26 Figure 4. Linroth’s case data and railroad map from 1889 ................ 40 Figure 5. Medieval road network ....................................................... 40 Figure 6. Case data and climate data .................................................. 41 Figure 7. Methodology flow chart ...................................................... 44 Figure 8. Thiessen area selection. Thiessen polygons surrounding each of the locations where cases had been reported by local physicians. The small polygons represent parishes. .......... 48 Figure 9. Interpolated temperature for week 44 in 1957 .................... 50 Figure 10. Interpolated precipitation for week 44 in 1957 ................. 51 Figure 11. Steet crossings before buffering........................................ 54 Figure 12. Street crossings after buffering ......................................... 54 Figure 13. Categories in land use mix ................................................ 55 Figure 14. Walkability classified in deciles ....................................... 56 Figure 15. Test areas for selection of individuals .............................. 57 Figure 16. Parishes, centroids, churches and water bodies ................ 59 Figure 17. Estimated numbers of cases per Thiessen area ................. 61 Figure 18. Map with bar charts representing the spread of the Russian Influenza ............................................................................ 62 Figure 19. Combined model ............................................................... 63 Figure 20. Kernel Density presentation of Russian Influenza status . 64 Figure 21. Salmonella infected pig herds in Sweden 1993-2010 ....... 65 Figure 22. Three models for the spread of the Black Death ............... 66 Figure 23. Weekly number of cases centered on the peaks ................ 67 Figure 24. GWM of incidence per municipality for three pandemics 68 Figure 25. Z-scores for temperature and number of cases ................. 69 Figure 26. Operations Dashboard for the Asian Influenza ................. 77
8
List of Tables Table 1. Known influenza pandemics (after Wikipedia 2013a) ......... 12 Table 2. States used in epidemic model ............................................. 26 Table 3. Walkability index by Frank et al. 2006 ................................ 36 Table 4. Number of pandemic influenza cases ................................... 46 Table 5. Pattern for local development of the Black Death ............... 58 Table 6. Salmonella serotypes ............................................................ 71 Table 7. Descriptive statistics on individuals included in the study .. 73
9
List of Papers This thesis is based on the following papers: I.
Skog L., Hauska H., Linde A., 2008. The Russian Influenza in Sweden in 1889-90: An Example of Geographic Information System Analysis, Eurosurveillance
II.
Skog, L. and Hauska, H., 2013. Spatial Modeling of the Black Death in Sweden. Transactions in GIS, 17: 589–611. doi: 10.1111/j.1467-9671.2012.01369.x
III.
Skog L., Linde A., Palmgren H., Hauska H., Elgh F., 2013. Spatiotemporal Characteristics of Pandemic Influenza, Submitted to BMC Infectious Diseases
IV.
Lewerin S., Skog, L., Frössling J., Wahlström H.. 2011. Geographical distribution of salmonella infected pig, cattle and sheep herds in Sweden 1993-2010, Acta Veterinaria Scandinavica 2011 53:51
V.
Kawakami N., Winkleby M., Skog L., Szulkin R., Sundquist K., 2011. Differences in neighborhood accessibility to healthrelated resources: A nationwide comparison between deprived and affluent neighborhoods in Sweden, Health & Place.
VI.
Sundquist K., Eriksson U., Kawakami N., Skog L., Ohlsson H., Arvidsson D., 2011. Neighborhood walkability, physical activity, and walking behavior: The Swedish Neighborhood and Physical Activity (SNAP) study, Social Science & Medicine.
The papers are referred to in the text by their respective roman numerals.
10
1 Introduction 1.1 Background “On 7 September 2004, a mother arrived at her 11-year-old daughter’s bedside in a provincial hospital in Thailand. She sat down, hugged and kissed her child, and wiped secretions from her mouth. Little did she know that her daughter’s breathing difficulties were caused by the deadly H5N1 avian influenza virus. The next day, the girl was dead and, nearly two weeks later, so too was the mother, making a strong case for the virus passing from one human to another” (Ungchusak et al. 2005). The influenza virus, however, has to find a more efficient way to reach humans to cause a pandemic. It has leapt from birds to humans, and it has killed about 50% of those who were infected. But we need not be really worried until the virus finds a genetic combination that allows it to jump easily from human to human. Only then will we face the pandemic that the world's media would have us believe is imminent (Nichols 2006). In 2003 the highly pathogenic strain (H5N1) started to spread throughout Asia and the avian influenza reached Europe in 2005. The Middle East and Africa were reached the following year (Monke 2006). In 2006 domestic as well as migratory birds infected with the deadly virus of type H5N1 were found at several locations in Sweden. Table 1 (Wikipedia 2013a is the only source found here) shows a list of the known viral pandemics. The Spanish flu is worst with more than 20 million deaths, but still not as severe as the Black Death, the plague, caused by the bacterium Yersina pestis, in the 14th century. Killing more than 30% of the population wherever it reached the Black Death is the worst catastrophe in mankind.
11
Table 1. Known influenza pandemics (after Wikipedia 2013a) Name of pandemic Russian Influenza
Date
Deaths
1889– 1890
1 million
Spanish Flu
1918– 1920 1957– 1958 1968– 1969 1977– 1978 2009– 2010
20 to 100 million 1 to 1.5 million 0.75 to 1 million no accurate count 18,000
Asian Influenza Hong Kong Flu Russian flu 2009 flu pandemic
Case Subtype fatality rate involved 0.15% possibly H3N8 or H2N2 2% H1N1
Pandemic Severity Index NA
0.13%
H2N2
2
0 for this particular day
60
2. Export the selection layer to a shape file and add it to the map document 3. Create a dissolved buffers layer for all selected radial distances 4. Intersect the buffers with a hydrographical map of Sweden The buffers created in this fashion depict the spread day by day and have been displayed in many different ways. The mortality parish by parish was modeled as a function of latitude and population density.
4.4 Visualization During the past 5 years there has been a significant development in techniques for communication of spatiotemporal data. Paper I was published in a journal primarily read by epidemiologists and even though animations (now at http://larsskog.wordpress.com/) were created, the readers of the journal had to rely on a figure with a map series (Figure 17)
Figure 17. Estimated numbers of cases per Thiessen area and a map with bar charts (Figure 18) to experience the spatiotemporal progress of the outbreak of the Russian Influenza in Sweden. The map shows the intensity of the Russian Influenza, week by week. One colour represents one week. Green hues are for the first weeks and red hues for
61
the last weeks. The largest segments represent the local peaks in intensity.
Russian Influenza 1889-1890
Figure 18. Map with bar charts representing the spread of the Russian Influenza 62
The results of the calculations in Paper II could be presented in many different ways. In the supplementing information there are animations to show the spatiotemporal spread of the Black Death in Sweden according to the three models developed. Figure 19 shows the results in a single map of the propagation over the whole period for the combined model.
Figure 19. Combined model 63
Maps and animations (available as supplementary information to Paper III) were created displaying the geographical spread of the three pandemics, week by week in total numbers as well as incidence per municipality (cumulative number of cases per week divided by number of inhabitants). Figure 20 demonstrates a Kernel Density estimation (from known observations with 30 kms search radius) of what the spread of the Russian Influenza in Sweden looked like the week before Christmas in 1889.
Figure 20. Kernel Density presentation of Russian Influenza status. Higher incidence is represented by darker colours. 64
In Paper IV maps like the one in figure 21 were produced of infected herds, stratified on animal species as well as salmonella serotype. Based on ocular inspection of all maps, some were collapsed and some used separately. Data were also examined for temporal trends. Animations are available at www.larsskog.wordpress.com
Figure 21. Salmonella infected pig herds in Sweden 1993-2010
In Paper V and VI GIScience methods are mainly used as tools to test certain hypotheses. The results are presented in tables rather than in maps.
65
5 Results & Discussion In this chapter the main results of papers I-VI are summarized and the contributions of each study are discussed.
5.1 Results and discussion of epidemiologic study The main result of Paper I is the calculated estimation and visualization of how the Russian Influenza propagated in Sweden in time and space. The methods described and the basic data from 1890 collected by Dr. Linroth and his colleagues (Linroth 1890) has been made available to epidemiologists all over the world for further interpretation and analysis, one example of which is the article on possible epidemic sources by Buscema et al. (2013). The results of Paper II are the models described in chapter 3.3 and the visualizations described in chapter 3.4. The spatial modeling enabled us to describe and visualize epidemiologic processes in space and time. Which one of the models on the Black Death propagation in Sweden that gives the best description of the process is hard to tell as they are all based on assumptions and a limited number of facts. A comparison for one day of the three models is made in figure 22. The radial model shows the modeled result for August 22, 1350, whereas the cost and the combined model refers to August 25, 1350
Figure 22. Three models for the spread of the Black Death
66
There is no detailed evidence saying exactly when in 1350 the Black Death reached any particular place in Sweden. However we have found some statements describing how the Black Death affected different locations. These statements and their locations can be found in a web map service available in the supplementing information to Paper II. In Paper III we looked for common spatiotemporal characteristics of pandemic influenza based on case data from three past pandemics that affected Sweden. Much has changed in the society over the 120 years from the time of the Russian Influenza (1889-1890) to the recent A(H1N1)pdm09. Even though the infrastructure and the social contact patterns are totally different the characteristics in the spread of pandemic influenza seem to vary less. Common for all three pandemics is a peaking period of approximately 10 weeks (Figure 23) and that the GWM (geographically weighted mean) follows a path from southeast to northwest, even though the A(H1N1)pdm09 follows the same path in the opposite direction as shown in figure 24.
Figure 23. Weekly number of cases centered on the peaks
67
Figure 26. GWM of incidence per municipality for three pandemics
Figure 24. GWM of incidence per municipality for three pandemics
Figure 25. GWM of incidence per municipality for three pandemics
68
The meteorological data we were able to collect for the period 19571958 facilitated an analysis of the impact of climate factors on the propagation of the Asian Influenza in Sweden. Comparing the zscores for temperature with the z-scores for the number of influenza cases in 1957 it was obvious (figure 25) that peaking periods were preceded by falling temperature.
Figure 27. Z-scores for temperature and number of cases The findings in Paper III on temperature fall preceding local outbreaks are very much in line with recent findings by Atchison et al. (2009) and Tamerius et al. (2013) who studied global variations in climate and 69
onset of influenza outbreaks. Their analysis stresses the covariance between specific humidity (the ratio of the mass of water vapor in air to the total mass of the mixture of air and water vapor) and temperature at high latitudes. Temperature or specific humidity drops below their respective thresholds result in an increased influenza activity. Previous studies have indicated that specific humidity may be a stronger predictor of virus survival and transmission than temperature (Shaman et al. 2010). Chunara et al. (2012) discuss how new technologies for reporting realtime emergent infections have led to a new paradigm in disease surveillance. Advances in computation and communications can provide exact positioning and increase speed and automation of collection, to make the data more accessible to healthcare professionals as well as to the public. Many internet-based reporting tools allow information to flow directly between parties, thereby bypassing the traditional hierarchical structure of public health systems. The time for reporting and control measures can consequently be reduced. HealthMap (Freifeld et al. 2008) is such a system, bringing together different data sources. The system includes online news feeds, eyewitness reports, blogs and validated official reports to achieve a common view of the current global state of infectious diseases. Using mobile phones disease information can be reported and used for intervention in real time. One example is the trial in Botswana where over 1000 malaria notifications were reported with GPS coordinates. The data transmission time reduced from 3–4 weeks to just minutes. When plotted on a map, the observations suggested over 80 potential outbreak sites (Integrated Regional Information Networks, 2011). Smart portable biological sensors of e.g. oral fluids can also be used for onsite health tests. According to Gartner (2013) the convergence of four powerful forces: social, mobile, cloud and information continues to drive change in the coming years and top strategic technology trends for 2014 include mobile device diversity, the Internet of Things and mobile apps. The main result of Paper IV is the visualizations of the total numbers of infected herds identified during 1993-2010 in different animal species. No geographical clustering was observed for ovine or porcine cases. Cattle herds infected with Salmonella Dublin were mainly located in the southeast region and cattle herds infected with Salmonella 70
Typhimurium in the most southern part of the country. Analyses of data on salmonella infected herds revealed some spatial and temporal patterns for salmonella in cattle, but available data was not sufficient for further analyses. The most common serotypes identified in different animal species are shown in table 6. The salmonella cases in pigs seemed to be geographically associated with the density of pigs. Table 6. Salmonella serotypes Animal Most common 2nd most species serotype common serotype
Other serotypes isolated
Cattle
S Dublin (124) S Typhimurium Agona, Diarizonae, Dusseldorf, (45) Enteritidis, Jangvani, Livingstone, Montevideo, Mbandaka, Oritamerin, Reading, Sandiego, Stanleyville, Tennessee, Teshie
Pigs
S Typhimurium (31)
S Infantis (6) S Yoruba, Putten, Newport, Mbandaka, Derby (5) Muenster, Java, Cubana*; Arizonae, Anatum, Agona
Sheep
S Diarizonae (10)
S Typhimurium Reading (2)
The most common salmonella serotypes isolated from different animals species in Sweden 1993-2010. *Except for the large outbreak in 2003, S Cubana was only isolated in one pig herd in 1997
71
5.2 Results and discussion of neighborhood study The main findings from the work in Paper V were that high- and moderate-deprivation neighborhoods had a significantly higher prevalence of all types of goods, services, and resources than lowdeprivation neighborhoods. These results do not support previous research that hypothesizes that poorer health among people in deprived neighborhoods is explained by a lack of health-promoting resources, although a higher presence of health-damaging resources may play a role. The potential significance of our study is high because few studies have been conducted in a European context or been able to include a broad range of goods, services, and resources from an entire country. Our findings do not support the hypothesis of “deprivation amplification”, suggesting that the consistent differences in health between individuals in deprived neighborhoods and those in affluent neighborhoods are not explained by poorer accessibility to healthpromoting goods, services, and resources in deprived neighborhoods. The results in Paper VI for individuals living in highly walkable neighborhoods, as compared to those living in less walkable neighborhoods, were:
77% higher odds for walking for active transportation and 28% higher for walking for leisure
50 min more walking for active transportation/week
3.1 min more MVPA (Moderate to Vigorous Physical Activity)/day
Descriptive statistics for all 2269 individuals participating in the survey are summarized in table 7. The findings in Paper VI stress that future policies concerning the built environment must be based on contextspecific evidence, particularly in the light of the fact that neighborhood redevelopments are time-consuming and expensive.
72
Table 7. Descriptive statistics on individuals included in the study ALL
TYPE OF NEIGHBORHOOD HW
HW
LW
LW
HI
LI
HI
LI
Median Std Median Std 41 23 47 23
Median Std 39 25
Median Std Median Std 39 23 40 23
Time in 10-minute bouts of moderate-tovigorous physical activity (min/day)
14
18
18
12
18
14
18
Walking for active transportation (min/week)
125
275 180
287
150
300
100
254 100
267
Walking for leisure (min/week)
60
222 90
225
68
248
60
216 60
208
Gender
n
%
n
%
n
%
n
%
n
%
• Male
1014
45
201
42
162
39
378
48
273
47
• Female
1255
55
278
58
252
61
411
52
314
53
• 20–30
251
11
82
17
68
16
42
5
59
10
• 31–40
461
20
115
24
88
21
139
18
119
20
• 41–50
645
28
104
22
111
27
242
31
188
32
• 51–66
912
40
178
37
147
36
366
46
221
38
• Low
766
34
205
43
220
53
152
19
189
32
• Middle
959
42
179
37
173
42
325
41
282
48
• High
544
24
95
20
21
5
312
40
116
20
• Single
590
26
186
39
154
37
106
13
144
25
• Married/Cohabiting
1679
74
293
61
260
63
683
87
443
75
Moderate-to-vigorous physical activity (min/day)
17
13
19
Age (years)
Family income
Marital status
HW = High Walkability LW = Low Walkability HI = High Income LI = Low Income
It must of course not be forgotten that the results are also depending on the way the walkability indices are defined. Frank et al. (2006) defined their index as 73
WI = Zresidential density + 2 x Zstreet connectivity + Zland use mix + ZFAR whereas the walkability index in Paper VI is defined as WI = Zresidential density + 1,5x Zstreet connectivity + Zland use mix The definition of residential density and street connectivity in Paper VI adheres to the definitions of Frank et al. (2009). To make the walkability indices comparable, in the absence of floor area figures, the multiplying factor was reduced from 2 to 1.5 in Paper VI. The reason is twofold; 1) In Paper VI the path and bicycle network is added to the street network 2) The relative importance of street connectivity in the first definition of WI is 40% and, with the reduction, 43% in the second.
74
6 Conclusions and future research 6.1 Conclusions The research has demonstrated how spatial analysis and modeling can be used in epidemiology and for other health related applications where the geographic position is of importance. The contributions to address the scientific questions stated in section 1.2 are listed below. 1) Could the spatiotemporal spread of the Russian Influenza in Sweden 1889-1890 be visualized using GIScience? Yes, the spatio-temporal spread of the Russian Influenza in Sweden was depicted in tables, in map series, in animated video sequences and in a single map document. Using the map as an interface to the epidemiological data it is shown how everyone involved in disease prevention and crisis management may have easy access to vital information, presented in an intuitive way. 2) Could the spatiotemporal spread of the Black Death in Sweden 1349-1350 be modelled despite total lack of case data? Yes, although there was no case data available, the spread of the Black Death in Sweden could be spatially analysed. Three different models were created, based on epidemiological findings, historical reviews, estimated population data and other basic geographic data. The models were presented as dynamic maps. Other pandemics could very well be modeled and visualized with the methods described. 3) Are there any common patterns in the spread of pandemics in Sweden? Looking at the results of the studied pandemics one can conclude that Each pandemic had a peak lasting for approximately 10 weeks The weekly geographically weighted mean (GWM) of the incidence of the Russian and the Asian Influenza pandemics was mainly moving along a line from 75
southwest to northeast. The weekly GWM for the A(H1N1)pdm09 influenza pandemic moved in the opposite direction, but along the same line The peak of the Asian Influenza was preceded by falling temperatures Epidemiologists have been given new information to be used in the development of new models for preparedness, prediction and analysis 4) Are there any common patterns in the spread of salmonella among herds in Sweden? Analyses of data on salmonella infected herds revealed some spatial and temporal patterns for salmonella in cattle. However, despite using 18 years' of data, the number of infected herds was too small for any useful statistical analyses and no patterns could be observed in pigs or sheep. 5) Does accessibility to different goods and services differ by level of neighborhood deprivation? No, it doesn’t as the hypothesis of “deprivation amplification” was not supported in the present study. All types of goods, services, and resources were more prevalent in moderate and high-deprivation neighborhoods than in low-deprivation ones. 6) How can local walkability indices be defined for the city of Stockholm? Using different spatial analysis methods walkability indices over the city of Stockholm were created. The findings show a positive (small but significant) association between neighborhood walkability and physical activity outcomes. The rapid development in IT in general and GIS in particular now give epidemiologists and public health officials excellent opportunities to use the methods demonstrated in their daily work. Development of new methodology still requires skills in GIScience, but the deployment of
76
new methods can be arranged in such a way that non-experts can perform complicated analyses without any prior knowledge of GIS. The main contribution of this work is that it may give epidemiologists and public health specialists new ideas, not only on how to formulate, model, analyze and visualize different health related research questions but also ideas on how new procedures could be implemented in their daily work. Once the data reporting is organized in a suitable manner there is a multitude of options on how to present important and critical information to officials and policy makers. The epidemiological dashboard shown in figure 28 using data from the study in Paper III could e.g. easily be implemented and made accessible to anyone or to a restricted group, should that be desired. A similar approach is already implemented at the Centres for Disease Control and Prevention, CDC (2013) in the US with additional data representation and navigation between different outputs.
Figure 28. Operations Dashboard for the Asian Influenza Communication of health status has always been highly controversial. On one hand, the more detailed information we have the better 77
diagnosis, treatment and follow up. On the other hand, the more detailed information we have, the larger the risk that we disclose personal information, potentially very harmful to the individual. The legislation in most countries so far is very much in favor of personal integrity. One consequence of that is that many research projects using spatial information have to be based on aggregated data. The methods used in the research projects in this work have all been executed using a desktop GIS, which is an excellent environment for evaluation and design of different processes. It is however not a platform that can be widely used by non-experts. Luckily the rapid development of cloud computing, internet services, smart phones and tablets have made it possible to easily convert processes developed on the desktop to web services to be reached by any device at any time. Methods and research regarding spatial modeling will benefit from development along many of these trends and it is to be expected that self-reporting will increase dramatically over the years to come. The big challenge will be to evaluate and control the quality of all the data to be collected. We will never know exactly how a new pandemic is likely to propagate but using the historical examples we are much better prepared to test and practice for different scenarios. Then we will create knowledge on how to prepare for the next pandemic. It is only a matter of time…
6.2 Future research Despite the considerable potential benefits of GIScience, its use in health service planning and provision remains greatly underutilized (Shaw 2012). This work is yet another proof of the potential for spatial analysis and spatial modeling for health applications and there is of course a multitude of possible directions for future research. One of the most obvious ones is a future study of the impact of local climate variations on the spread and outbreak of influenza pandemics or epidemics. Better positioned case data and more climatological parameters are now available for analysis. The main task for future research on GIScience for health applications must however be to develop, implement and educate simple self-explanatory methods for everyday use at health institutions in our society. The technical development discussed in the previous chapter will assist us in 78
developing the methods for everyday use, but we have to be much more observant on the real needs out in the field to make it happen. Another study should be established to determine whether climatological factors in any way influenced the spread of the A(H1N1)pdm09 in Sweden. As can be understood from the discussion, the results of the walkability calculations are very much depending on the choice and weighting of the parameters in the walkability indices. This is a field that should be addressed in a future study.
79
7 References Ahmed S.S.U., Ersbøll A.K., Biswas P.K., Christensen J.P., Toft N., 2011. Spatio-Temporal Magnitude and Direction of Highly Pathogenic Avian Influenza (H5N1) Outbreaks in Bangladesh. PLoS ONE 6(9): e24324. doi:10.1371/journal.pone.0024324 Alegana V.A., Atkinson P.M., Wright J.A., Kamwi R., Uusiku P., Katokele S., Snow R.W., Noor A.M., 2013. Estimation of malaria incidence in northern Namibia in 2009 using Bayesian conditional-autoregressive spatial-temporal models. Spat Spatiotemporal Epidemiol. 2013 Dec;7:25-36. doi: 10.1016/j.sste.2013.09.001. Epub 2013 Sep 17. Anderson, R.T., Sorlie, P., Backlund, E., Johnson, N., Kaplan, G.A., 1997. Mortality effects of community socioeconomic status. Epidemiology 8,42–47. Andersson-Palm L., 2001. Livet, kärleken och döden: Fyra uppsatser om svensk befolkningsutveckling 1300–1850. Solna, Sweden, Nykopia Global Print Anyamba A., Chretien J., Small J., Tucker C., Formenty P., Richardson J., Britch S., Schnabel D., Erickson R., Linthicum K., 2009. Prediction of a Rift Valley fever outbreak. Proc Natl Acad Sci USA. 2009 Jan 20;106(3):955-9. Epub 2009 Jan 14. Arcaya M., Brewster M., Zigler C., Subramanian S., 2012. Area variations in health: A spatial multilevel modeling approach, Health & Place 18 (4) (Juli): 824–831. doi:10.1016/j.healthplace.2012.03.010. Atchison C., Tam C., Hajat S., van Pelt W., Cowden J., Lopman B., 2009. Temperature-dependent transmission of rotavirus in Great Britain and The Netherlands, doi: 10.1098/rspb.2009.1755 Proc. R. Soc. B 2010 277, 933-942 ArcGIS, 2012. http://www.esri.com/software/arcgis/index.html (Accessed 12 Oct 2012)
80
ArcGIS Resource Center 2013, http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#/H ow_Kernel_Density_works/009z00000011000000/ (Accessed 11 Sep 2013) Beale L., Abellan J., Hodgson S., Jarup L., 2008. Methodologic Issues and Approaches to Spatial Epidemiology, Environmental Health Perspectives, 2008, Vol.116 (8), p.1105-1110 Bengans historiesidor, 2011. http://wadbring.com/historia/undersidor/vagar.htm (Accessed 5 Jan 2011) Berrang-Ford L., Odiit M., Maiso F., Waltner-Toews D., McDermott J., 2006. Sleeping sickness in Uganda: revisiting current and historical distributions, African Health Sciences, v6(4), dec 2006, PMC 1832067 Bian, L., 2012. Spatial Approaches to Modeling Dispersion of Communicable Diseases – A Review. Transactions in GIS. doi: 10.1111/j.1467-9671.2012.01329.x Bing Maps, 2012. http://www.bing.com/maps/ (Accessed 12 Oct 2012) Boulos M., Resch B., Crowley D., Breslin J., Sohn G., Burtner R., Pike W., Jezierski E., Chuang K., 2011. Crowdsourcing, citizen sensing and sensor web technologies for public and environmental health surveillance and crisis management: trends, OGCstandards and application examples http://www.ijhealthgeographics.com/content/pdf/1476-072X-10-67.pdf (Accessed 5 Oct 2013) Boumédiène F., Druet-Cabanac M., Marin B., Preux P., Allée P., Couratier P., 2011. Contribution of geolocalisation to neuroepidemiological studies: Incidence of ALS and environmental factors in Limousin, France, Journal of the Neurological Sciences 309 (1–2) (Oktober 15): 115–122. doi:10.1016/j.jns.2011.07.002. 81
Brouwers L., Boman M., Camitz M., Mäkilä K., Tegnell A., 2010. Micro-simulation of a smallpox outbreak using official register data . Euro Surveill. 2010;15(35):pii=19651. Available online: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19 651 (Accessed 5 Feb 2013) Brouwers L., Cakici B., Camitz M., Tegnell A., Boman, M. 2009. Economic consequences to society of pandemic H1N1 influenza 2009 – preliminary results for Sweden. Euro Surveill. 2009;14(37):pii=19333. Available online: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19 333 (Accessed 5 April 2013) Brownstein J., Holford T., Fish D., 2005. Effect of Climate Change on Lyme Disease Risk in North America, Ecohealth. 2005 Mar;2(1):38-46. Burchfiel, C. M., Sharp, D. S., Curb, J. D., Rodriguez, B. L., Hwang, L. J., Marcus, E. B., et al. (1995). Physical activity and incidence of diabetes: the Honolulu Heart Program. The American Journal of Epidemiology, 141(4), 360e368. Buscema M., Grossi E., Bronstein A., Lodwick W., Asadi-Zeydabadi M., Benzi R., Newman F., 2013. A New Algorithm for Identifying Possible Epidemic Sources with Application to the German Escherichia coli Outbreak, International journal of Geo-information 03/2013; ISPRS Int. J. Geo-Inf. 2013(2):155200. DOI:10.3390/ijgi2010155 Carpenter T., 2011. The spatial epidemiologic (r)evolution: A look back in time and forward to the future, Special Issue: GEOVET 2010 2 (3) (September): 119–124. doi:10.1016/j.sste.2011.07.002. Casalegno J., Ottmann M., Bouscambert Duchamp M., Escuret V., Billaud G., Frobert E., Morfin F., Lina B., 2010. Rhinoviruses delayed the circulation of the pandemic influenza A, European Society of Clinical Microbiology and Infectious Diseases, CMI, 16, 326–329 82
CDC, 2013. http://gis.cdc.gov/GRASP/NCHHSTPAtlas/main.html?s_cid=nc hhstp-atlas-006 (Accessed 4 Nov 2013) Charlton M., Fotheringham S., Brunsdon C., 2006. Geographically Weighted Regression, ESRC National Centre for Research Methods, NCRM Methods Review Papers, NCRM/006 http://www.ncrm.ac.uk/research/outputs/publications/methodsre view/MethodsReviewPaperNCRM-006.pdf Chowell G., Munayco C., Escalante A., McKenzie F., 2009. The spatial and temporal patterns of falciparum and vivax malaria in Peru: 1994-2006. The Malaria Journal 8:142 (2009) Chowell G., Viboud C., Munayco C.V., Gómez J., Simonsen L., et al., 2011. Spatial and Temporal Characteristics of the 2009 A/H1N1 Influenza Pandemic in Peru. PLoS ONE 6(6): e21287. doi:10.1371/journal.pone.0021287 Chowell G., Towers S., Viboud C., Fuentes R., Sotomayor V., Simonsen L., Miller M., Lima M., Villaroel C., Chiu M., Villaroel J., Olea A., 2012. The influence of climatic conditions on the transmission dynamics of the 2009 A/H1N1 influenza pandemic in Chile. BMC Infectious Diseases 12:298. Christakos G., Olea R., Serre M., Yu H., and Wang L., 2005. Interdisciplinary Public Health Reasoning and Epidemic Modelling: The Case of Black Death. Berlin, Springer-Verlag Chunara R., Freifeld C., Brownstein J., 2012. New technologies for reporting real-time emergent infections, Parasitology. 2012 December; 139(14): 1843–1851. Published online 2012 August 16. doi: 10.1017/S0031182012000923 Clarke, G., Eyre, H., Guy, C., 2002. Deriving indicators of access to food retail provision in British cities: studies of Cardiff, Leeds and Bradford. Urban Studies 39, 2041–2060.
83
Cromley, E. and McLafferty S., 2002. GIS and Public Health, New York: Guilford Press RA566.C764x 2002 Crowl A., Crist T., Parmenter R., Belovsky G., Lugo A., 2008. The spread of invasive species and infectious disease as drivers of ecosystem change, Front Ecol Environ 2008; 6(5): 238–246, doi:10.1890/070151 Cubbin, C., Sundquist, K., Ahlen, H., Johansson, S.E., Winkleby, M.A., Sundquist, J., 2006. Neighborhood deprivation and cardiovascular disease risk factors: protective and harmful effects. Scand. J. Public Health 34, 228–237. Cummins, S.,Macintyre, S.,2002.Fooddeserts—evidence and assumption in health policy making.Br.Med.J.325,436–438. Demografiska databasen, 2013. www.ddb.umu.se (Accessed 3 Feb 2013) Diez Roux, A.V., Merkin, S.S., Arnett, D., Chambless, L., Massing, M., Nieto, F.J., Sorlie, P., Szklo, M., Tyroler, H.A., Watson, R.L., 2001. Neighborhood of residence and incidence of coronary heart disease. N. Engl. J. Med. 345, 99–106. Duncan D., Aldstadt J., Whalen J., Melly S., Gortmaker S., 2011. Validation of Walk Score® for Estimating Neighborhood Walkability: An Analysis of Four US Metropolitan Areas, Int J Environ Res Public Health. November; 8(11): 4160–4179. Published online 2011 November 4. doi: 10.3390/ijerph8114160 ECDC, 2013. http://ecdc.europa.eu (Accessed 4 April, 2013) Eisen R., Glass G., Eisen L., Cheek J., Enscore R., Ettestad P., Gage K., 2007. A spatial model of shared risk for plague and hantavirus pulmonary syndrome in the southwestern United States. Am J Trop Med Hyg. 2007 Dec;77(6):999-1004. Eisen R.., Mead P., Meyer A., Pfaff L., Bradley K., Eisen L., 2008. Ecoepidemiology of Tularemia in the Southcentral United States, Am J Trop Med Hyg 2008 78:586-594 84
Fenner F., Henderson D., Arita I., Ježek Z., Ladnyi I., 1988. Smallpox and its Eradication, History of International Public Health, No6, World Health Organisation Forsyth A., D’Sousa E., Koepp J., Oakes J., Zimmermann J., 2007. Twin Cities Walking Study, Metropolitan Design Center 1 Rapson Hall, 89 Church Street SE, Minneapolis, MN 554550109USA http://www.designcenter.umn.edu
[email protected] (Accessed 6 November 2013) Fotheringham A., Bransdon C., Charlton M., 2002. Geographically Weighted Regression: The Analysis of Spatially Varying Relationships, Wiley, Hoboken, NJ, 2002 Frank, L. D., Andresen, M., & Schmid, T., 2004. Obesity relationships with community design, physical activity, and time spent in cars. American Journal of Preventive Medicine, 27 (2), 87–96. Frank L., Sallis J., Conway T., Chapman J., Saelens B., Bachman W., 2006. Many Pathways from Land Use to Health: Associations between Neighborhood Walkability and Active Transportation, Body Mass Index, and Air Quality, Journal of the American Planning Association, 72:1, 75-87, DOI:10.1080/01944360608976725 Frank L., Sallis J., Saelens B., Leary L., Cain K., Conway T., Hess P., 2009. The development of a walkability index: application to the Neighborhood Quality of Life Study, Br J Sports Med 2010;44:924-933 doi:10.1136/bjsm.2009.058701 Freifeld C., Mandl K., Reis B., Brownstein J., 2008. HealthMap: global infectious disease monitoring through automated classification and visualization of Internet media reports. Journal of American Medical Informatics Association. 2008;15:150–157. Epub 2007 Dec 20. [PMC free article] [PubMed]
85
Gagliardi H., Furlan L., Gagliardi H., da Silva F., Pisa I., Alves D., 2006. Automata Network Simulator Applied to the Epidemiology of Urban Dengue Fever Lecture Notes in Computer Science, 2006, Volume 3993/2006, 297-304, DOI: 10.1007/11758532_41 Gartner, 2013. www.gartner.com/symposium/us (Accessed 12 Dec 2013) Gatrell A., Bailey C., Diggle J., Rowlingson S., 1996. Spatial point pattern analysis and its application in geographical epidemiology Transactions of the Institute of British Geographers, Volume 21, Number 1, March 1996 , pp. 256-274(19) Google Dengue Trends, 2012. http://www.google.org/denguetrends/ (Accessed 12 Dec 2012) Google Flue Trends 2012, http://www.google.org/flutrends/ (Accessed 12 Dec 2012) Graham A., Atkinson P., Danson F., 2004. Spatial analysis for epidemiology. Acta Tropica, 91, (3), 219-225. (doi:10.1016/j.actatropica.2004.05.001). Harrison D., 2000. Stora döden: Den värsta katastrof som drabbat Europa. Stockholm, Sweden, Ordfront Hinman S.E., Blackburn J.K., Curtis A., 2006. Spatial and temporal structure of typhoid outbreaks in Washington D.C., 1906-1909: evaluating local clustering with the G i* statistic International Journal of Health Geographics 2006 Mar 27;5:13 hitta.se, 2009. www.hitta.se (Accessed 10 August 2013) Holmes E., 1997. Basic epidemiological concepts in a spatial context. In Tilman D and Kareiva P (eds) Spatial Ecology. Princeton, NJ, Princeton University Press: 111–36
86
Hu X., Di Paolo E., Barnett L. 2008, Ripple-Spreading Model and Genetic Algorithm for RandomComplex Networks: Preliminary Study, https://www.google.se/search?q=RippleSpreading+Model+and+Genetic+Algorithm+for+Random+Com plex+Networks:+Preliminary+Study&ie=utf-8&oe=utf8&rls=org.mozilla:sv-SE:official&client=firefoxa&gws_rd=cr&ei=QAR8UoyiHOqc4wTWvYGQCg, (Accessed 7 Nov 2013) Huang I., Li W., Sui J., Marasco W., Choe H., Farzan M., 2008. Influenza A Virus Neuraminidase Limits Viral Superinfection, J. Virol. 2008, 82(10):4834. DOI: 10.1128/JVI.00079-08 Published Ahead of Print 5 March 2008 Integrated Regional Information Networks, 2011. Cambodia: Using text messaging as weapon in malaria war. IRIN Humanitarian News and Analysis 2011 Jacquez G., 2000. Spatial analysis in epidemiology: Nascent science or a failure of GIS? Journal of Geographical Systems Volume 2, Number 1, 91-97, DOI: 10.1007/s101090050035 Krist F. and Brown D., 1994. GIS Modelling of Paleo-Indian Period Caribou Migrations and Viewsheds in Northeastern Lower Michigan Photogrammetric Engineering & Remote Sensing, Vol. 60, No. 9, 1994, pp1129-1137 Lafferty K., 2009. The ecology of climate change and infectious diseases, Ecology, 90(4), 2009, pp. 888–900, Ecological Society of America Landers J., 199.2 Historical epidemiology and the structural analysis of mortality Health Transition Review National Centre for Epidemiology and Population Health, The Australian National University Liao J., Hu X., Wang M., Leeson M., 2013. Epidemic Modelling by Ripple-Spreading Network and Genetic Algorithm, Mathematical Problems in Engineering, Hindawi Publishing Corporation, http://www.hindawi.com/journals/mpe/aip/506240/
87
Linde A., Rotzén-Ostlund M., Zweygberg-Wirgart B., Rubinova S., Brytting M., 2009. Does viral interference affect spread of influenza?, Euro Surveill. Oct 8;14(40). pii: 19354. Lindgren E., Tälleklint L., Polfeldt T., 2008. Impact of climatic change on the northern latitude limit and population density of the disease-transmitting European tick Ixodes ricinus. Environ Health Perspect. 2000 Feb;108(2):119-23. Linroth K., 1890. Influensan i Sverige 1889-1890 enligt iakttagelser af landets läkare, på Svenska läkaresällskapets uppdrag skildrad af Klas Linroth, Curt Wallis och F. W. Warfvinge. Del I: Influensan i epidemiologiskt hänseende. [Influenza in Sweden 1889-1890 according to observations of the country's doctors, description on request the Swedish medical society. Part 1: Influenza in epidemiological terms] [In Swedish]. Svenska Läkaresällskapets Nya Handlingar. Serie III. 1890:1-92. Lowen, A. C., Mubareka, S., Steel, J. & Palese, P. 2007. Influenza virus transmission is dependent on relative humidity and temperature. PLoS Pathog. 3, 1470–1476. (doi:10.1371/journal.ppat.0030151) Macdonald, L., Ellaway, A., Macintyre, S., 2009. The food retail environment and area deprivation in Glasgow City, UK. Int. J. Behav. Nutr. Phys. Act. 6, 52. Macintyre, S., 2007. 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. 4, 32. Manson, J. E., Nathan, D. M., Krolewski, A. S., Stampfer, M. J., Willett, W. C., & Hennekens, C. H. (1992). A prospective study of exercise and incidence of diabetes among US male physicians. Journal of the American Medical Association, 268(1), 63e67.
88
Margonari C., Freitas C.R., Ribeiro R.C., Moura A.C., Timbó M., Gripp A.H., Pessanha J.E., Dias E.S., 2006. Epidemiology of visceral leishmaniasis through spatial analysis, in Belo Horizonte municipality, state of Minas Gerais, Brazil. Mem Inst Oswaldo Cruz. 2006 Feb;101(1):31-8. Epub 2006 May 12. May, R.. and M. Nowak., 2001.Virus Dynamics: Mathematical Principles of Immunology and Virology. Oxford: Oxford University Press. 256 pp.; 48 line illus. Mayne D., Morgan G., Willmore A., Rose N., Jalaludin B., Bambrick H. Bauman A., 2013, An objective index of walkability for research and planning in the Sydney Metropolitan Region of New South Wales, Australia: an ecological study, International Journal of Health Geographics, 12:61, doi:10.1186/1476-072X12-61 McLafferty S., Freeman V., Barrett R., Luo L., Shockley A., 2012. Spatial error in geocoding physician location data from the AMA Physician Masterfile: Implications for spatial accessibility analysis, Spatial and Spatio-temporal Epidemiology 3 (1) (April): 31–38. doi:10.1016/j.sste.2012.02.004. Monke J., 2006. Avian Influenza: Agricultural Issues, CRS Report for Congress, Order Code RS21747 Myrdal J., 2003. Digerdöden, pestvågor och ödeläggelse. Stockholm, Sweden, Sällskapet Runica et Mediævalia Nakazawa Y., Williams R., Peterson A., Mead P., Staples E., Gage K., 2007. Climate change effects on plague and tularemia in the United States. Vector Borne Zoonotic Dis. 2007 Winter;7(4):529-40. National Veterinary Institute, 2013. http://sva.se/en/ (Accessed 10 Jan 2013) Nicholls H., 2006. Pandemic Influenza: The Inside Story. PLoS Biol 4(2): e50. doi:10.1371/journal.pbio.0040050
89
Nordström M., 2001. Geographical Information System (GIS) as a Tool in Surveillance and Monitoring of Animal Diseases, Acta Veterinaria Scandinavica 2001, 42(Suppl 1):S79-S85 Nuckols J., Ward M., Jarup L., 2004. Using Geographic Information Systems for Exposure Assessment in Environmental Epidemiology Studies, Environmental Health Perspectives 112 (9) (Juni 1): 1007–1015. doi:10.2307/3838102. Ohlander, E., Vikstrom, M., Lindstrom, M., Sundquist, K., 2006. Neighbourhood non-employment and daily smoking: a population-based study of women and men in Sweden. Eur. J. Public Health 16, 78–84. Ordnance Survey, 2009. Network Analysis, http://www.ordnancesurvey.co.uk/oswebsite/gisfiles/section4/pa ge4.html (Accessed 10 March 2012) Ostfeld R., Glass G., Keesing F., 2005. Spatial epidemiology: an emerging (or re-emerging) discipline, TRENDS in Ecology and Evolution Vol.20 No.6 June 2005 Owen N., Cerin E., Leslie E., duToit L., Coffee N., Frank L., Bauman A., Hugo G., Saelens B., Sallis J., 2007. Neighborhood Walkability and the Walking Behavior of Australian Adults, American Journal of Preventive Medicine, Volume 33, Issue 5, November 2007, Pages 387-395, ISSN 0749-3797, PAR, 2013. http://www.par.se (Accessed 15 May 2013) Patz J., Githeko A., McCarty J., Hussein S., Confalonieri U., de Wet N., 2012. Climate change and infectious diseases http://www.who.int/globalchange/publications/climatechangech ap6.pdf (Accessed 22 April 2013) Patz J. and Norris D., 2004. Land Use Change and Human Health, Ecosystems and Land Use Change, Geophysical Monograph Series 153
90
Peterson A., Papes M., Kluza D., 2003. Predicting the Potential Invasive Distributions of Four Alien Plant Species in North America, Weed Science Society of America, Vol 51, No 6, NovDec, 2, http://www.jstor.org/stable/4046739 Pickle L., Carr D., 2010. Visualizing health data with micromaps, GEOMED Conference 1 (2–3) (Juli): 143–150. doi:10.1016/j.sste.2010.03.007. Puget Sound Regional Council, 1999. Puget Sound household travel survey. Austin, TX: NuStats Research and Consulting. Ratnasinghe, L. D., Modali, R. V., Seddon, M. B., & Lehman, T. A. (2010). Physical activity and reduced breast cancer risk: a multinational study. Nutrition and Cancer, 62(4), 425e435. Rattan A., Campese A., Eden C., 2012. Modeling Walkability, ArcUser, http://www.esri.com/news/arcuser/0112/modelingwalkability.html (Accessed 27 May 2013) Riksarkivet, 2012. Nationell Arkivdatabas. WWW document, http://www.riksarkivet.se/default.aspx?id=26257&ptid=0 (Accessed 12 Jan 2013) Rinaldi L., Musella V., Biggeri A., Cringoli G., 2006. New insights into the application of geographical information systems and remote sensing in veterinary parasitology, Geospat Health, 1 (2006), pp. 33–47 Rytkönen M., 2004. Not all maps are equal: GIS and spatial analysis in epidemiology Int J Circumpolar Health 2004 Mar;63(1):9-24. Sallis, J. F., Saelens, B. E., Frank, L. D., Conway, T. L., Slymen, D. J., Cain, K. L., et al., 2009. Neighborhood built environment and income: examining multiple health outcomes. Social Science & Medicine, 68(7), 1285e1293. SAMS, 2013. http://www.scb.se/Pages/Standard.aspx?id=333368 (Accessed 7 Jun 2013)
91
SCB, 2013. http://www.scb.se/default____2154.aspx (Accessed 15 Jul 2013) Sen K. and Bonita R., 2000. Global health status: two steps forward, one step back Lancet 2000 Aug 12;356(9229):577-82 Sesso, H. D., Paffenbarger, R. S., Jr., & Lee, I. M. (2000). Physical activity and coronary heart disease in men: the Harvard alumni health study. Circulation, 102(9), 975e980. Shaman J., Pitzer V., Viboud C., Grenfell B., Lipsitch M., 2010. Absolute Humidity and the Seasonal Onset of Influenza in the Continental United States. PLoS Biol 8(2): e1000316. doi:10.1371/journal.pbio.1000316 Shaw N., 2012. Geographical Information Systems and Health: Current State and Future Directions, Healthc Inform Res. 2012 June;18(2):88-96. http://dx.doi.org/10.4258/hir.2012.18.2.88pISSN 2093-3681 • eISSN 2093-369X Sickweather, 2012. http://www.sickweather.com/ (Accessed 11April 2013) Silverman B., 1986. Density Estimation for Statistics and Data Analysis, 1986, Monographs on Statistics and Applied Probability, London: Chapman and Hall Smallman-Raynor M. and Cliff A., 2001. Epidemic Diffusion Processes in a System of U.S. Military Camps: Transfer Diffusion and the Spread of Typhoid Fever in the SpanishAmerican War, 1898, Annals of the Association of American Geographers, Online ISSN: 1467-8306 Smallman-Raynor M. and Cliff A., 2013. The geographical spread of the 1947 poliomyelitis epidemic in England and Wales: spatial wave propagation of an enigmatic epidemiological event, Journal of Historical Geography, Volume 40, April 2013, Pages 36–5 http://dx.doi.org.focus.lib.kth.se/10.1016/j.jhg.2012.12.003 92
SMHI, 2012. Sjöar, vattendrag och avrinningsområden. WWW document, http://www.smhi.se/klimatdata/hydrologi/sjoar-ochvattendrag (Accessed 5 March 2013) SmiNet, 2013. http://www.SmiNet.se/ (Accessed 12 Aug 2013) Song Y. and Rodríguez D., 2005. Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and Accessibility (Gravity with Competition) Indices. SOU 2007:60, 2007. Sverige inför klimatförändringarna - hot och möjligheter, http://www.regeringen.se/sb/d/8704/a/89334 (Accessed 5 Oct 2011) Spatial Join (Analysis), 2009. ArcGIS Desktop 9.3 Help, http://webhelp.esri.com/arcgisdesktop/9.3/index.cfm?id=1339&pid=1 335&topicname=Spatial_Join_(Analysis) (Accessed 9 Apr 2009) Sundquist, J., Malmstrom, M., Johansson, S.E., 1999. Cardiovascular risk factors and the neighbourhood environment: a multilevel analysis. Int. J. Epidemiol. 28, 841–845. Sundquist, K., Qvist, J., Sundquist, J., & Johansson, S. E. (2004). Frequent and occasional physical activity in the elderly: a 12year follow-up study of mortality. American Journal of Preventive Medicine, 27(1), 22e27. Sundquist, K., Winkleby, M., Ahlen, H., Johansson, S.E., 2004. Neighborhood socioeconomic environment and incidence of coronary heart disease: a follow- up study of 25,319 women and men in Sweden. Am. J. Epidemiol. 159, 655–662. Sundquist, K., Johansson, S. E., Qvist, J., & Sundquist, J. (2005). Does occupational social class predict coronary heart disease after retirement? A 12-year followup study in Sweden. Scandinavian Journal of Public Health, 33(6), 447e454.
93
Sundquist, K., Qvist, J., Johansson, S. E., & Sundquist, J. (2005). The long-term effect of physical activity on incidence of coronary heart disease: a 12-year follow-up study. Preventive Medicine, 41(1), 219e225. SVA, 2013. http://sva.se/en/ (Accessed 10 Aug 2013) Tamerius J., Shaman J., Alonso W., Bloom-Feshbach K., Uejio C., Comnie A., Viboud C., 2013. Environmental Predictors of Seasonal Influenza Epidemics across Temperate and Tropical Climates. PLoS Pathog 9(3): e1003194. doi:10.1371/journal.ppat.1003194 Teleadress, 2009. http://www.teleadress.se/htdocs/omoss/english.htm (Accessed 8 Aug 2009) Tellier R, 2009, Aerosol transmission of influenza A virus: a review of new studies, J. R. Soc. Interface 2009 6, doi:10.1098/rsif.2009.0302.focus first published online 22 September 2009 Thrusfield M., 2007. Surveillance, Veterinary epidemiology, Blackwell, Oxford/Ames (2007) pp. 168–187 Tobler W., 1970. A Computer Movie Simulating Urban Growth in the Detroit Region, Economic Geography, 46, 2 (1970), pp. 234240. Tomalty, R., Haider, M., Smart Growth BC, 2009. BC sprawl report: walkability and health, Smart Growth BC, Vancouver, B.C. Ungchusak K., Auewarakul P., Dowell S., Kitphati R., Auwanit W., nPuthavathana P., Uiprasertkul M., Boonnak K., Pittayawonganon C., Cox N., Zaki S., Thawatsupha P., Chittaganpitch M., Khontong R., Simmerman J., Chunsutthiwat S., 2005. Probable person-to-person transmission of avian influenza (H5N1). N Engl J Med 352: 333–340.
94
WalkScore, 2011. Find your neighborhood - Redlands. http://www.walkscore.com/score/100-new-york-st-redlands-ca92373 (Accessed 08.03.2014) WHO, http://www.who.int/topics/physical_activity/en/ , (Accessed 08.05.2010) Wikipedia, 2013a. http://en.wikipedia.org/wiki/Influenza_pandemic#Asian_Flu_.2 81957.E2.80.931958.29 (Accessed 12 Nov 2013) Wikipedia, 2013b. http://en.wikipedia.org/wiki/Voronoi_diagram (Accessed 3 May 2013) Winkleby, M., Cubbin, C., 2003. Influence of individual and neighbourhood socioeconomic status on mortality among black, Mexican–American, and white women and men in the United States. J. Epidemiol. Community Health 57, 444–452. Winkler, E., Turrell, G., Patterson, C., 2006. Does living in a disadvantaged area mean fewer opportunities to purchase fresh fruit and vegetables in the area? Findings from the Brisbane food study. Health Place 12, 306–319. Zhou X., Yang G., Yang K., Wang X., Hong Q., Sun .L, Malone J., Kristensen T., Bergquist N., Utzinger J., 2008. Potential impact of climate change on schistosomiasis transmission in China. Am J Trop Med Hyg. 2008 Feb;78(2):188-94.
95
