December 2013

See on Scoop.itEpisurveillance

Faculty Project:  2013-2014

PI: Andrew Curtis
Collaborators: Wei-An Andy Lee, Kreck Medical Center of the University of Southern California

Brian Altonen‘s insight:


This project is designed to understand the diabetes landscape of vulnerable populations in the Boyle Heights area of Los Angeles and to help facilitate intervention and management strategies for a participating diabetes clinic. The objective of this work is to develop a geospatial tool and/or approach that is easy to implement and is transferable amongst other resource challenged clinics serving vulnerable populations. To achieve this, the GIS Health & Hazards Lab is currently (1) Providing a comprehensive summary review of health, social science and planning research previously conducted in Los Angeles, (2) Evaluating the feasibility of using spatial analysis of health and census data for the targeted neighborhood of Boyle Heights, and (3) Piloting the use of mobile mapping to capture the fine scale built environment of patients in Boyle Heights.

See on

See on Scoop.itGlobal Health Care

Faculty Projects:  

2013-2014PI: Andrew Curtis
Collaborators: Jason K Blakcburn, Spatial Epiemiology & Ecology Research Laboratory (SEER), Department of Geography, University of Florida; Jocelyn M Widmer, Urban Affairs and Planning, School of Public and International Affairs, Virginia Tech University; J Glenn Morris Jr., Emerging Pathogens Institute, University of Florida

Brian Altonen‘s insight:


Fine-scale and longitudinal geospatial analysis of health risks in challenging urban areas is often limited by the lack of other spatial layers even if case data are available. Underlying population counts, residential context, and associated causative factors such as standing water or trash locations are often missing unless collected through logistically difficult, and often expensive, surveys. The lack of spatial context also hinders the interpretation of results and designing intervention strategies structured around analytical insights. This project offers a ubiquitous spatial data collection approach using a spatial video that can be used to improve analysis and involve participatory collaborations.

See on

See on Scoop.itMedical GIS Guide

We love Open Data. CartoDB is a geospatial database on the cloud to allow development of location…

Brian Altonen‘s insight:

Examples of how to apply Theissen’s mid-19th century polygon mapping routine  to your work, and more.

See on

See on Scoop.itNational Population Health Grid

Some weeks ago I visualised the Swiss cantons (states) and their population numbers using what information visualization scientists call a linked view. You can click through to the actual, interact…

Brian Altonen‘s insight:

One web site that effectively demonstrates the value of hexagonal grid mapping GIS for medical surveillance is Spatialists by Ralph Straumann.  On this page he displays a useful technique for displaying public health using hexagonal grids.  This method produces more accurate results, enabling more correct interpretations of health and space to ensue, which in turn can can be used to produce intervention programs.  


Small area spatial analysis using hexagonal grid techniques allows us to focus more on the details of public health, to define where and upon whom our new interventions must be developed and targeted.  

Most current applications of medical GIS focus on retrospective studies, a review of the past in order to determine which activities we wish to engage in to make changes in the future, afterwhich, we don’t always monitor our performance using the same GIS approach.    

See on

See on Scoop.itMedical GIS Guide

A number of ways to map cholera were use during the mid-19th century epidemic.  Most of these methods were completely correct.  Yet only one was accepted as the most important at the time–that of London epidemiologist John Snow.  Of the remaining four disease mappers, one other was equally accurate and another from the United States very valuable due to the association its author made between alkalinity and the disease.  

Brian Altonen‘s insight:

What do the accomplishments of past disease cartographers and medical geographers teach us?


If we spend the time reviewing their methods, logic, and in some cases, important findings about the spatial distribution of people, communities, organisms and disease, we are provided with important insights into the more modern uses for medical GIS.


Recently, a considerable amount of celebration in England focused the 160th anniversary of John Snow and his famous map of cholera. Snow used this map to prove and prevent the transmission of this disease throughout a part of London.  He also used it to demonstrate the value of hygienic living practices and the importance of a clean water supply.  


Today, we are still very much concerned about the cleanliness of our drinking water, and our risk of exposure to toxic chemicals, bacteria and other environmental health hazards.  

For this reason, another map produced before Snow’s 1854 map deserves recognition.  In 1850, John Lea of Cincinnati, Ohio published a map and pamphlet on the local cholera epidemic, claiming the cases he observed were due to the use of well water contaminated by non-alkaline (non-calcareous) soils.  He recommended rainwater be used instead.  Much of the public agreed, and a number of businesses were established to produce rainwater barrels.  

John Lea’s map is more detailed than Snow’s and yet was completely forgotten.  Lea correctly associated the relationship between cholera and alkaline wells.  John Snow never supported his theory, and claimed that Lea’s observations were correct but his reasoning was somehow wrong (see my International Journal of Epidemiology article on this–link below).

Over the next several years, John Lea fought many of the arguments against his alkaline theory, and never won the support of the profession.  (A lot of his other lines of reasoning were wrong.)  

As a result of this forgotten discovery, a Japanese epidemiologist received recognition for the same decades later, demonstrating that pH and alkalinity determined whether or not the bacteria that caused cholera could survive.


Other maps depicting the mid-19th century cholera epidemic in the illustration include one published by Hector Gavin of London (ca. 1848), which made use of miasma theory to explain how the undernourished poor could become so sick.


In 1850, the American Medical Association (AMA) published a map depicting the cases near Philadelphia, but unlike Lea’s map was very difficult to interpret.


Also important to note is the cholera map published the same year as Snow’s famous map by British epidemiologist Henry W. Acland.  He produced a map of equal caliber to Snow’s, one that was more colorful.  


For more on this see 

my article published in International Journal of Epidemiology . . . 


Brian Altonen.  Commentary: John Lea’s Cholera with Reference to Geological Theory, April 1850 .  International Journal of Epidemiology 2013 42: 58-61.   Link:  JohnLea-Cholera_IJE-Article


or . . .


and William Farr’s elevation and cholera paper at

See on

See on Scoop.itGlobal Health Care

Using GIS to Improve Health Resource Allocation in Yemen

Brian Altonen‘s insight:

A very basic Step-by-Step example of how to strategize healthcare . . . 


Research question for this study:

What are the gaps in availability and accessibility of immunization services in Amran Governorate, Yemen?


This is an example of a health GIS sample analysis was used to identify gaps in immunization services in Amran Governorate.


The value of the map in strategizing healthcare is in enables you to more precisely target manpower, intervention activities and expenses.    

The maps for this project were generated using ESRI’s ArcGIS, but many of these same tasks can be done using other open access GIS tools out there.  Buffer and flow or travel analysis is possible in some of the freeware or shareware packages.  The hardest step to locate a tool for is the unique Theissen Polygon process (show in step 2, highlighted in step 4).


These planners utilized a World Health Organization packet made available known as Yemen Health Analyzer ArcView GIS Tool, or Health Analyzer, which works as an extension to ArcGIS.  It is available for download at



Descriptions accompanying the above set of maps:


Step 1: Plot population areas by size (PURPLE points) and identify health facilities with immunization services and vaccine refrigerator using the Facility Survey Analyzer (YELLOW points).


Step 2: Analyze the immunization services network by dividing area according to closest proximity to service provider (LT GREEN area) using Service Network Provider tool.


Step 3: Analyze 60-min walking accessibility/coverage area (BLUE) using the Facility Accessibility Mapper.


Step 4: Analyze population-weighted gaps in immunization coverage and target prioritized areas or health facilities for intervention (RED vs. YELLOW or GREEN areas) using the Healthcare Gap Analyzer.


Step 5: Evaluate immunization gaps compared to an analysis of 60-min accessibility area for immunization services added to user-selected existing facilities (WHITE area) and/or initiated at potential new facilities (GRAY area).


For more on the Yemen Health Analyzer ArcView GIS tool, you can email —


This information is from:


See on

See on Scoop.itMedical GIS Guide

Ecological fallacy is when we produce a map that suggests one thing, but in reality, due to spatial errors generated by the mapping technique, is guilty of a Type 1 or Type 2 error.”

Brian Altonen‘s insight:

Dasymetrics mapping enables us to take into account other features related to disease and health care that normally we ignore when relying on just census  data and the mapping of patient distribution.


Census data produces statistics that are defined based on number of people within predefined census zones block groups or blocks  We are typical left with an underestimate of spatial distribution for census areas where people reside in a very small section of the census area.


Dasymetrics is a valuable way to improve the spatial accuracy of your results when grid methods are not desired.  Dasymetrics allows users to dissect large areas like census zones and blocks down to much smaller areas that depict more accurately the region our data represents.  

At the site "Free and Open Source GIS Ramblings" by Anita Graser (aka Underdark), the application of this technique using Open Source mapping is provided–

Intersection and buffering routies are two very easy ways to make spatial epidemiology results more accurate.  

More on applying this skillset to managed care and meeting your Meaningful Use goals can be found at: [Poster Session] [another example of application] [wiki site]



See on

See on Scoop.itMedical GIS Guide

Brian Altonen‘s insight:

One of the hardest parts of studying historical medical geography is the time it takes to find your documentation.  Maps are figures, usually included as addenda of figures to published writings.  This means they are poorly indexed.


Chances are, for every 100 or more articles I have read, reviewed and/or hunted down in the 19th century medical journals, only one or two percent will have decent geographical illustrations about disease.  The majority of books and articles of such value in Google Books lack the map due to the nature of its inclusion, normally as a pocketed insert or a fold out.   So it is usually important that either a decent copy be found in hardcopy form, or that interlibrary loan services be used to obtain a photo of the map and/or the book/article itself. generally fairs better with the illustrations, spending the time needed to include these in the variety of electronic forms of the documents that they provide.  


The above maps from the 1890 Census are detailed fairly extensively at my page


The point however being made here is that by reviewing historical maps we develop insights into how to better interpret the maps of disease we produce today.  


The above maps are from my use of the Census maps to analyze what was becoming one of the poorest parts of the country–Appalachia.  Even with large area choropleth mapping, we can still see where latitude, aspect (east vs. west face of the mountains), large scale area-specific landuse patterns (Coastal, vs. Great Lakes, vs. South, vs. Great Plains) and even population density (the early NY-to-Ohio, north to Boston, south to Washington DC megalopolis then developing).


For example East versus West had little to do with impacting how croup could be spread, whereas north vs. south and population density appear to play a major role in Scarlet Fever activities.  Stillborn deliveries occur no matter where you reside.  Enteric fever, from contaminated dairy products mostly, is more a southern phenomenon.   Heart Disease and Old Age have higher rates on the east side of Appalachia (the megalopolis, with developed towns and cities).


This same method of looking at major geographic features i nrespect to disease and disease maps can provide us with important insights into the value of spatially presenting you public health and epidemiological findings.  Programs that lack these valuable spatial techniques and insights into their findings are not going to provide us with as much value as a GIS study of the same important topics.


For more examples of how the maps of the past can help current spatial health researchers, epidemiologists and medical geographers, go to my blogsite on this topic at or visit my Pinterest page, where some of the more interesting examples of my findings are posted.


The five boards in Pinterest which review this work are as follows:

See on Scoop.itEpisurveillance

July 1973. From page 72 of “The Future Society: Aspects of America in the years 2000″ American Academy of Political and Social Science annual meeting. “Health Challenges of the Future” lecture by George E. Ehrlich.

Brian Altonen‘s insight:

A little more than 40 years ago, George E. Ehrlich gave a lecture at Temple University on July  of 1973 entitled “Health Challenges of the Future”.   This lecture was part of the annual meeting of the American Academy of Political and Social Science devoted to “The Future Society: Aspects of America in the years 2000.”

Then Professor of Medicine at the Temple University School of Medicine, Ehrlich predicted the depersonalization of medicine which the computer might result in.


However, we are falling short of one of his visions about the direction in which the field of medicine was heading due to the invention of the computer.


Ehrlich thought that by 2000 we would be fully engaged in making the best use of the computer and the storage of patient records, thereby create tremendous improvements in people and population health.  He speculated that with the computer, diagnoses could be made more rapidly, lab orders and clinical testing could be automated, with the results generated and then posted in a timely manner, and that we could therefore understand the best options for care we had available to us, all in a very short time.


Ehrlich’s major concern with these technological advancements was the further reduction of the human contribution that could ensue–a reduction of interactions that normally occurred between patients and care givers.


Unfortunately, many of today’s practitioners, allied healthcare givers, and patients agree with Ehrlich’s last statement.


Even more unfortunate however, the failure of the system to more quickly and more effectively make the best use of its technology to provide patients with more health care value for their money.


This latter failure has nothing to do with the technology itself, only with those responsible for the best use of that technology–those responsible for employing it within the health care system with the best long term interests in mind.


George Ehrlich could not foresee the increasing split that has occurred between the rich and poor since the 1970s.   But he would probably agree and be incredibly surprised to see how that, in spite of technological achievements and advancements, the human side of providing care and making care accessible has not changed in more than forty years.


The recent resistance to change and improvements in healthcare, are a repeat of these same events unforeseen by Ehrlich.  The ongoing resistance to change due to financial managers and CFOs of these systems offers little explanation for the tremendous acceptance these companies have for their lack of progress during the past 40 years.


The failure of insurance companies to implement EFFECTIVE, cost savings population health analytics programs into their systems is an example of what Ehrlich refers to with his criticisms.


Conformity is not always to our benefit when it comes to  healthcare.  The attached quality of life and financial benefits of receiving more effective care are opportunities missed due to poor management and the corporations’ resitance to change.


Ref:  George E. Ehrlich, (Publ. in The Annals of the American Academy of Political and Social Sciences, Vol. 408, July 1973, pp. 70-82.)

See on