(Click the following link)


(address: https://www.surveymonkey.com/r/JW88X3F)

With Managed Care receiving limited support from the masses this presidential election year, one has to wonder if the current PPACA (“Obamacare”) program will survive another year.

When the presidential race is over this coming fall, if either of the two most supported candidates is elected, there will more than likely be changes brewing in how to resolve the issues that hold our healthcare system ransom in the United States.


Same image, different rotations

For example, my personal speculation is that if Hillary Clinton get elected, that the attempts might be to continue riding upon the system already in place, under the claim that some things need improvement, and the financing needs for effective management and control, but in the long run, we will simply take what it now in place and change it to make it more effective, more consumer friendly.

If Donald Trump is elected, the resolutions formed in my mind are less convincing or indicative of attempts to maintain some status quo and stability.  In the worst scenario, healthcare will be placed more into the hands of private businesses again, and those programs devoted to the underpriviledge lower class fortified as they have always beenm being available to those most needy, but considerably more thought our in terms of determining those who are eligible.


This means that in-migrating people will not easily acquire such coverage, those moving here for work related reasons, be it permanent or temporarily, will also have more “hoops” to jump through to earn their coverage.  That will service the current system greatly, in terms of finances, but re-open old wounds as the previously uncovered masses, who briefly entered some sort of healthcare status under Obamacare, will once again be left to fend for themselves.  Some of this impact can be lessened, with certain eligible members maintaining their eligibility due to such things as young mother and child status, or having to deal with some life threatening chronic diseases state.  But the least qualified people who were fortunate enough t obtain coverage in recent years, may again be out of luck by the time a new program is put in place, should the next President-Elect be Donald Trump.


Fortunately, it will likely take one and a half to two years for any new health insurance managed care program to be implemented.  That will lessen to brunt of change this nation might endure, whichever way we go.

Some evidence suggests that Hillary Clinton might opt to produce some formidable changes in the Obamacare plan, were she elected.


This begs the question:

What might the best changes be in such a system, were it to be established as an extension in the current program that exists, yet meet the financial and service industry needs in ho the newer way to manage care gets developed?

For one thing, eliminating “Obamacare” is easy.  Change the program, get rid of that title that now has a tremendous amount of hatred and regret attached to it.


But we might only need to eliminate a very small of the PPACA plan in order to satisfy the public masses.

The most important elements to PPACA are the transfer, change and implementation of healthcare into a high technology supported industry, complete with information technology never perfected before by any single healthcare plan (and yes, none have perfected it, no matter how grandiose their thinking and self-judgment appears to be to colleagues).

The information technology systems in healthcare today, known specifically as health information technology or HIT (like this label avoids to scary innuendo of the title IT), are still struggling now more than 15 years into managed care, and nearly twenty years into their complete development and implementation.

Generation after generation of IT fears and barriers have surfaced and resurfaced, demonstrating that the healthcare insurance companies and masters of the health economics  industry aren’t as wise or able as they want us to think.  A change in technology occurs quite readily and in months to a year longer, through simple acceptance of the technology and willingness to test its application.  But one could argue that insurance companies and businesses fear IT advancement due its financial impact upon their individual financial securities, and morality targeting impacts upon the leaders, people and companies that have to use this new technology.


Insurance companies lose considerably once they see solid evidence for overcharging (we can leave the fraudulent billers and overbillers out of this argument even).  Insurance companies will lose greatly once their additional charges and unnecessary charges are uncovered.  The waivering tendency for insurance to assign value and price to specific procedures will raise some red flags, more than they have been doing since this issue was described to the public several years ago–for example, why is the PET costing $175 for medicaid, but $675 for privately insured individuals?


With presidential-elect changes inevitable in the next year or two, how will IT itself — that skill set which insurance companies have failed to support or engage in —   determine the lifespan of the long-lasting stubborn companies unwilling to engage in the EMR, EHR, IT, HIT, HITECH Act requirements used to define PPACA?

The truth is, these IT developments are here to stay, whether you like them or not.  They have already demonstrated their value, and provided evidence to necessitate their further development and use in the healthcare billing and insurance industries.  Even if meaningful measures lessen to some extent, leaders and common sense workers will not engage in lessening or reducing their production of proofs that changes are happening.

With IT/HIT development, we will learn more and more about effective and ineffective certain companies, programs, facilities, neighborhoods, insurance plans, mail pharma companies, allied health programs (bogus or not), and counseling and family support service agencies will be when included as part of the broader, long term minded, preventive-palliative healthcare programs many places are trying to develop.

So IT/HIT must stay for these companies to document, and brag about their successes.


Adding GIS to this only strengthens whatever arguments are going to be made, and making it easier to convince you audiences, in whatever direction you want.

What prevents spatial health work from becoming a part of the managed care  system is not just lack of intellect about what it is and what it can do (everyone knows a map can portray, or lie).  What prevents spatial health from becoming a part of the healthcare system is the lack of knowledge, desire, experience, and skillset on behalf of leaders, to put forth the option of developing such a program in order to save their own business significant amounts of money.

There are a few companies out there that are experimenting with spatial analysis.


But I have yet to see one inkling of evidence that any industry out there is able to analyze its entire popular for the 15,ooo ICDs and ICD groups, the 4 fold as many visits as patients, the 10 to 20 fold number of procedures these people engage in, for 10 to 20 years of long term EMR/EHR history held secure as part of the older medical records systems.

A few companies have 20+ years of EMR on file (I work for one).  GIS demonstrates outcomes that these data produce, that no other MC can produce through much of the country.  Likewise, insurance companies and health businesses that manage the insurance aspects as well, such as KP, have developed partially functional spatial analytic systems that may have the ability to perform full reporting for a full healthcare program.

The question is: what will be the first company to emerge that is able to create a full-fledged GIS/Spatial analysis  based version of the QI improvement report, detailing thousands, not hundreds of the most important population health features?


These features must include the following:

  1. Human genomic data and disease history and spatial analysis
  2. Emergent/Urgent care data, past and present, in reference to age, gender, etc.
  3. Religious group data in relation to specific ICDs and medical events, compliance and non-compliance, and social and moral public health related disease or injury related risks
  4. Race-Ethnicity derived reviews of its regions and communities in relation to population health, place, economics, and cultural traditions.
  5. Socioeconomic status and predicted employment pattern related health care coverage, disease and success in relation to place, time and money
  6. Chronic Disease (CD) management databases, devoted to hundreds of CDs, engaging in predictive modeling of these costly populations
  7. Standard endemic disease pattern public health surveillance
  8. Atypical epidemic and foreign born disease patterns and travel in relation to local regional health matters (local Legionnaires, Ebola, Chikungunya, etc. types of studies and prediction models)
  9. The differentiation of standard linear versus hierarchical diffusion modeling behaviors and patterns pertaining to locally-specific disease patterns
  10. A remodeling of epidemiological transition theory to best meet the needs of the local populations being served, with specific attention paid to economics, business and health, using the sequent occupancy model proposed on other pages posted at this site.

This survey is developed to determine how ready the healthcare profession is for the spatial analysis of health and health-related costs by a managed care system.




(address: https://www.surveymonkey.com/r/JW88X3F)




A depiction of how diseases were theorized around 1812, wtih expected locations of diseases over the Lake Michigan shoreline, Great Lakes area.

At times I get a little self-assured with the way my predictions have panned out over the past five years.

Back in 2013 I stated that measles was due to re-emerge, after discovering the dense cluster of non-immunized families and children along the west coast, centered in the Pacific Northwest.


Immunization Refusal peaks in the Pacific Northwest, 2000-2010.  Individuals in the national dataset evaluated for this work, who had a Measles history.

Like I said in the many posting on that topic, before, during and after the measles outbreak in the Palisades area of New York, followed by the much broader, nationally spreading disease that erupted out of California.  To date, most of the blame for that outbreak has been directed both at people from other countries coming into the U.S. for a short while, unimmunized and carrying the disease or bringing a child who has the disease, or we have blamed the numerous families into alternative medical philosophies and traditions in California, who felt the world was safe, and whp had that malingering phobia about their child getting autistic due to vaccinations–just MMRs mind you–“but hey, what the heck, we might as well skip them all.” I hear them say.

CDC and other major epidemiology, surveillance groups are focused on the California nidus based cause for the measles outbreak.  They are wrong mind you, since the higher likelihood remains untouched due to this mis-directed focus down south.

And that is often what epidemiologists do–jump to conclusions, determine what’s likely to happen, and take action based upon that guess.


For decades, if not centuries (depending upon how far you read into the past), this basic layout of the world was used to define epidemic patterns, especially for fevers.  Yellow fever was natural to the Torrid Zone, but during warm spells migrated into Temperate Zones.  If in the upper north (for the southern hemisphere, lower south), the colder half of the temperate zone kept the fever present either by warmer weather, or by dense populations of people who could become infected.  This important piece of American history is reviewed in more detail at https://brianaltonenmph.com/gis/historical-disease-maps/yellow-fever/a-new-war-yellow-fever/

So, the new Yellow Fever concern is bringing us another of these returning diseases that we are going to perhaps have to deal with.

See recent news article: Lydia Ramsey’s “Another deadly virus is poised to go global“. 

If yellow fever hit the U.S., in epidemic fashion, I can predict that several big mistakes are going to happen and be published.

A section of William Aitken’s 1872 map of the U.S., depicting the Yellow Fever regions of North America   [Link to Aitken’s map]

First, some of the past “yellow fever” outbreaks that weren’t yellow fever at all may be brought up.  Yellow fever is very much a seasonal epidemic that erupts in the northern latitudes during the late summer, early fall–or as one Quaker physician Shadrach Ricketson stated around 1803 or 1804, it can become more likely when summer returns in the months of October and early November– that period of time we call “Indian Summer.”

Periods in Epidemiological Transition, according the the Sequent Occupancy method of interpretation.  For basic background, see  https://brianaltonenmph.com/gis/applying-new-methods-with-gis/sequent-occupancy/    In 1786, the famous patriot Benjamin Rush wrote about the ways in which society evolved in regions, producing a Human Species-based Evolution Theory that preceded Sequent Occupancy.  (for which see  https://brianaltonenmph.com/gis/historical-medical-geography/1786-benjamin-rush-an-early-rendering-of-the-sequent-occupancy-philosophy/ )

A very early possible outbreak, isolated on an Island south of Rhode Island is discussed in an article on an “Extraordinary Disease”.  The geographical theoreies as to how an epidemic can strike an area, based on its levels of development, is also presented on this page that is referenced.  This economic geography interpretation of disease explains the epidemiological transition patterns and behaviors we often see, in more detail and in a much more concise way that the traditional epidemiological transition theory.  This philosophy that fits the epidemic changes better than epidemiological transition is called Sequent Occupancy Theory.  (For “Extraordinary Disease” see https://brianaltonenmph.com/gis/historical-disease-maps/yellow-fever/1763-the-extraordinary-disease-of-marthas-vineyard-and-nantucket/ )

German medical cartographer Friedrich Berghaus produced this 1848 map, the second global disease map to be published.  The yellow sections (barely visible) are the yellow fever areas of North America.

But the biggest mistakes US epidemiologists could very well make is to predict the entry of this disease by traditional mid-southern and southeastern routes into the United States.  Some will believe it can return in a hierarchical way, via the largest, most heavily traveled cities.

Congressman-to-be Samuel Anderson theorized the possibility that yellow fever was being carried by ships.  On one of the ships he managed as a ship surgeon, he documented the conditions that existed and he suspected may have been related to an epidemic brought up from Curacao to Delaware.  This is from my page on this topic https://brianaltonenmph.com/gis/historical-disease-maps/yellow-fever/samuel-anderson-and-the-mystery-of-yellow-fever-on-board/

But my review of yellow fever 6 years ago demonstrated more cases striking the northern states than the southern ones.  That can happen in one of several ways.

From one of my studies of Seaman’s maps of the 1790s-e1800 Yellow Fever in New York City (source: https://brianaltonenmph.com/gis/historical-disease-maps/valentine-seaman-1804-the-black-plague-or-yellow-fever-in-new-york-city/ )

First, the disease managed to come into the U.S. in spite of epidemiological surveillance in its natural endemic settings to the south.  In other words, people bypassing the public health security part of the southern border of the U.S. will never be caught, and therefore make their way, as quickly as possible, to the northernmost parts of the US, via the Mississippi River and Great Plains travel routes.

Chiclero’s Ulcer ( see  http://youtu.be/5a08oS23f4Q    )

Of all things to relate to yellow fever, Chiclero’s Ulcer or Ear, an infection of the external ear indiced by the heavy humidity of Yucatan, also makes its way into this country via this route.  The same migration route into the United States is followed by another Mexican-originated ecological disease–Pinta (pinto), a relative of STDs Falcoparum.


♥ (link)  <

So, Yellow Fever could make its way into the U.S by a classical in-migration route, that one would think the U.S. government can secure and watch over.

Another route that Yellow Fever could take in is the route in first took as it made its way into this country–by traditional shipping routes.  Though this may not be so likely today as it was between 1740 and 1815, when shipping was a primary way to get around.  Shipping routes enabled the disease to follow the inland waterways as well, meaning that in order to infect parts of inner Canada the way it did, yellow fever had to enter by way of the St. Lawrence Seaway, head across the Great Lakes, and then head in every direction along major tributaries heading west and north into Canada, and south and west into the United States.

The first disease with numerous topographic and geographic associations was the yellow fever, finally put onto a map by New York’s Valentine Seaman, as part of his investigation of the New York City outbreaks.

National Yellow Fever cases in the 2000s (personal research)

Numerous more maps of yellow fever behaviors have been published.  Unlike Malaria or Dengue Fever, two other mosquito related foreign diseases to  hit the United States, yellow fever has more of an ecological niche on this continent, and of the three, it is therefore more likely than them to return, and acclimate to the warming climate and the warmer southern shores of the United States, now managing to become a part of the ecosystems here, unlike back in the late 18th and early 19th century.

An Early Pinta 3D rotating image video I produced using my sql and sas formulas.

The third possible route for yellow fever into this country that is most unusual is akin to the last example just mentioned with the St. Lawrence Seaway.  Today’s peaks of yellow fever seen between 2000 and 2010 possible relate to lesser security to the north, due to lesser expectations for yellow fever to enter the U.S. by way of a northern route.

And so, that’s the hypothesis I am testing here, by watching how the yellow fever behaves if it reaches this country.  If there are weaker security measures for this disease at our north border, accompanied by lesser expectation by Canadian Public Health officials for a tropic disease to enter the heart of their country and establish a seasonal-only nidus there.


For more examples of Ecological Disease Cluster research using GIS, see my NPHG page: https://nationalpopulationhealthgrid.com/applications/2-ecological-disease-clusters/





The figures on this part are from my 2000 to 2006 studies of the Ecology of West Nile Fever.  ESRI GIS, Idrisi and several raster-grid and imagery tools were employed over the years for this work.   The research methods presented on this page and the site I provide links for, pertain mostly in New York, and to a lesser extent in Colorado.  To date, the full data set I developed for west nile ecology work has only been partially reviewed.


With Zika now present on the United States continent’s mainland, numerous questions must be considered for the events that may soon follow.

Some mosquito-born diseases like Malaria fail to reach a state where the pathogen, vector, and hosts other than people enable it to become a part of the local ecology.  Just what are the features that prevent such an event from happening?  What parts of the potential ecosystem in which the vectors could survive is incapable of providing the requirements for its pathogens–plasmodia?

With yellow fever, there was limited possibility for the pathogen-vector-host/carrier relationship to exist as part of the natural setting.  The limiting factor much of the time was weather, in particular seasonal climate changes that made it difficult if not impossible for pathogens to remain alive in much the U.S. over the winter.  And although such is not the case in the deep south of the U.S., yellow fever never became the pathogen that it still is within its native tropical settings.  So there’s more at play with this form of mosquito-vectored disease, than just the pathogen-vector-host relationship.  Something in the ecosystem extinguishes the yellow fever virus, as well as limits its principally Aedes and Haemagogus vectors.


If we ignore for a moment the historically important Dengue Fever linked to a similar vector dependency, the next most important mosquito vectored pathogen to strike the U.S. is the West Nile Fever related virus.  Although it also arose from a very different climate and topographic setting, like yellow fever and malaria, West Nile was able to establish itself as part of the natural ecology.  Why did it accomplish this?


The traditional reasoning, from traditional epidemiologists, tends to focus on the human-host-vector-pathogen relationship, paying limited attention to the macrocosm of the pathogen and peoples’ world–the surrounding ecosystem.  West Nile had a number of very distinct ecological behaviors that to date, with the exception of some of the work I published more than ten years ago, have not been developed into a potential method for evaluating and spatially analyzing mosquito-vectored imported disease patterns.


The following are the most important spatial methods for evaluating disease ecology pertaining to mosquito-borne diseases:

  • Plant Ecological research provides insights into the ecosystems that certain species of mosquitoes favor, and/or stay away from in terms of finding the niches for potentially infected mosquito populations.
  • Understanding the relationship that exists between landuse form, plant type, vegetation region or biome type, and the effects of these upon the microclimate of vector-breeding areas, provides a unique set of insights into how to evaluate the ecology of the vector-host relationship, and provides us with a way to initially define, more accurately, high risk regions.
  • As demonstrated by a field study I performed around 2002, using a light sensor, subcanopy light influences the behaviors of vectors and can be used to define the most susceptible natural settings for a vector to engage in specific feeding, mating, breeding and migration activities.
  • The application of normalized difference vegetation index (NDVI) imagery to studying species patterns and the possible locations of swarms or dense clusters of populations provides additional insights into how a region at the small area level becomes disease prone or not. [See https://brianaltonenmph.com/west-nile/6-remote-sensing/ ]
  • Satellite imagery may be used to further this method of research on the influences of canopy upon host-vector-pathogen relationships. [ https://brianaltonenmph.com/west-nile/6-remote-sensing/ ]
  • Competition between different species in a shared territory influences how they position themselves within the ecosystem, at the elevation and canopy-related level, and in terms of the distance from the water body that they may reproduce and aggregate the most.


My most successful series of studies demosntrated the ecological links between certain species and population density per species, the potential impacts of species competition on the exposure of people to disease carriers, and the important roles of understanding landform relationships with the behaviors of potential mosquito-carriers, a feature that is very microecological nature, and can vary from one moderately sized region to the next.


Most important to this study was the value it demonstrated for the use of NLC imagery and the value of applying other free  (then, not necessarily now) spatial imagery data to this research process.

My most understandable and therefore successful outcome was of course related to the aerial photo imagery project, the third image above, for describing and demonstrating the spatial relationship between a cluster of positive testing animal hosts, and the probably location of a positive testing pool of vectors.  These vectors overwintered the following seasons, and re-emerged in April of the following year, demonstrating the ability of vectors to retain an active pathogen or virus into the following years.

Important to note here – – for Zika to become ecologically stable within certain areas of the United States, and to become a part of the local ecology, this behavior has to be duplicated.    


With IDRISI (and its subsequent products, IDRISI 32 onward), one is able to evaluate important topographic, elevation-related, and surface-reflection/solar pattern linked features to the behaviors of a disease.  To identify the most susceptible, effective surveillance sites, the SLARs and related elevation datasets available to remote sensing staff can be used to define the more successful topographic regions for trapping possibly infected vectors.  This worked in the field and for lab results (producing positive testing hosts and vectors) in several ways.   The most important was it correlated back to the elevation-distance from water edge study I performed, by setting mosquito traps along a transect of the creek’s flood plain.  It also explained why one creek was not very productive with positive testing species, versus the other.  The two creeks appear the same, but topographically and ecologically, according to elevation and topographic images, their forms influenced canopy and solar energy aspect features, making one creek very stable ecologically and not a carrier due to high vector species diversity, whereas the other was less species rich and therefore more likely to harbor infectable vectors species, predominantly in regions where the disease would be spread by human ecological means.


Different species of vectors have different elevation related behaviors.  Idrisi was also used to produce DEM derived images with overlays, to demonstrate how elevation above water surface does influence the species captured.  The most revealing images and findings about vector species ecology came from my various studies of plant species, canopy type and light penetration, in relation to small water body locations, potential and infected raptor host species that had recently deceased, and the use of algorithms developed for testing the relationships between plants and small areas.


This first study is of a small shady yeard setting, with a nearby pond, a positive testing host, several related dead hosts, and a unique multispecies canopy setting.  The NDVI was used to demonstrate the results of a supervised classification process used to reprocess the original image of this setting.


The trap sites on this property were added as point, converted to grid data, and then overlain of the reclassified NDVI.  Buffers around the trap sites were identified and that data pulled from all subsequent images analyzed.   The results of this analysis was then related back to the entire image to produce an entire map depicting the risk of each grid cell area.  A Theissen Polygon analysis was performed of these trap areas, for use if defining the content of each trap site in terms of landuse, NDVI,  canopy and vegetation related grid cell findings.


Several ecological tools already present in Idrisi were also applied, focusing on biodiversity and species richness, and tested as possible indicators of potential west nile-bearing vector-host settings.


Four types of regions were defined by Idrisi, each with different types of ecological relationships between the plants and other materials or belongings found in those particular regions.  The hot spot was overlain on these four classes of disease inducing locations, and the point related best with Classes 5 and 6. (Above).  This implied that Deciduous trees played an important role in producing the ecological setting that the vector species most desired, followed by a low lying moist vegetative region adjacent to it.  It also demonstrated that the heavy canopy settings had minimal influence on vector swarm patterns.  The heavy canopies at this site were planted evergreens, mainly 30-40 foot tall Fir and Pine trees.


Applying this back to the entire image, a cross-sectional risk area pattern could be produced, using several formulas or dimensions by which risk can be assessed.   Linear, quadratic and cuboid formulas were applied. The cuboid model demonstrated the best fit, especially for Class 5 area, again, reconfirming the point-area derived summarization deduced based simply upon visual inspection of Class 5 of the classification images.


For more on the method for mapping mosquito cector ecology, see my pages at the following links:

West Nile Surveillance



An Automated 3D Mapping process (courtesy NPHG technology).  

Recent results of a new technology for spatial medical cartography that I developed, requiring minimal engagement at the IT developer’s end, including an elimination of time requirements related to the development of videos


We are in the initial stages of implementing GIS for managed care based population health monitoring.  There are a number of software programs out there than enable you to include maps in the displays or presentations you develop as part of the monitoring process of your workstation.  As leaders in the field, the production of “dashboards”, “scorecards” and “reports” have enabled us to increase our understanding of the population we serve and develop the knowledgebase needed to better manage the various types of healthcare programs we provide.

Surveillance is a major reason to establish as GIS workstation that has specific public health and population health and safety related metrics defined.  Most of the uses of GIS to date have been for research purposes, with a few projects actually evolving into standard intervention, health safety, health security and even cost effectiveness reviews of our programs.

There are literally thousands of conditions, events, diagnoses that fit the description of being important to national health security concerns.  Aside from the hundreds or more zoonotic and rare infectious diseases,  there are culturally linked and bound health conditions for which the knowledge of their notations in the EHR or EMR should lead to the raising of a “red flag.”

Included in these issues are those related to domestic violence, spouse or child related mistreatment, malnutrition, unsanitary living conditions, criminal activities, illegal or immoral seclusion to an inside home-setting, etc. etc.    This particular event is one of the most controversial to appear in the US EMR/EHR data.  Its controversy in part related to its reasons for practice, as well as its reasons for continuous practice in spite of international laws passed prohibiting it from being performed.

This new technology enables spatial analysts to define the sociological or sociocultural “hot spots” in health related issues.  The recent re-eruption of measles, the in-migration of mosquito-born diseases like zika, the possible planting of new forms of food-spread antibiotic resistant bacteria in certain parts of the U.S., the increasing density of certain culturally-linked genetic disease traits due to cultural growth, combined with the ongoing forming of you families with shared genetic traits, set the stage for the development of a medical GIS by all managed care institutions.  When such an HIT-GIS station is developed in association with the local public health program activities utilizing spatial health analytics tools, we develop a better understanding of our local population health, and can more quickly use this knowledge to monitor, survey and even predict the health changes expected for a region.

Genetic Disease Carriers, for a very common malady within a heavily populated urban setting.  Note: this was not performed using a GIS, but a simple set of SQL-SAS algorithms

Fortunately, technology does catch up with some of the programs I write.  One of those programs in the combined SQL and SAS needed to automate reporting for managed care companies.

The limiter to producing an automated ‘MAPP’ Program, as I call it (Mine-Analyze-Produce-Present) is the amount of work you put your system through.

My first attempt to go through this process entailed a production of about 3500 popualtion pyramids for most of the major ICD9 disease classes, many broken into specific age groups, in order to determine the highest risk gender-age groups in one year increments for very special social disease patterns, such as anorexia, wife beating, pyromancy, infibulation, and the varous forms of child abuse.  This evaluation took me nearly a half-year to complete, b ut inspired me to automate my processes some more for producing more effective products, more quickly.

My second attempt entailed evaluating regional disease patterns.  To accomplish this, I broken the US down ultimately into about 25 regions.  It seemed the NCQA, NIH, US Census and USGS ways of breaking the US population and states down into regions, based on income and/or expeditures and insurance related patterns, wasn’t exact enough to demonstrate the varying family sizes I noted for unique areas, in more unique parts of this country.  The Midwest and Great Lakes areas for example, I divided into north-south and east-west and combined N-S-E-W quadrant patterns to determine where the most statistically significant differences existed across state line/regional or subregional borders.  That project fortunately took only a week or two to perfect, and led me to develop the grid mapping of the entire United States, in detail.


Over the past five years, the status of Medical GIS as practiced within the Managed Care profession has remained at a 5.50-5.75 level.  We don’t use GIS to improve our HEDIS results, nor event to routinely monitor the HEDIS requirements, or event the Obamacare “Meaningful Use” requirements.  There are numerous programs underway to try to convince managers to implement some form of spatial monitoring process.  But due to disinterest, and/or lack of knowledge for this form of research and exploration, this more productive form of Health Information Technology (HIT) management has not become a standard part of any managed care system.  It remains, in what I like to call, an “Experimental Use” category for HIT.

My grid mapping algorithms begin with the zip code mapping style, in which two kinds of maps are developed.  The first is a raw data related zip code choroplethic or 3D columnar like demonstration of where the distributions exist for each particular metric.  These metrics are, like before, of diseases and disease groups based upon ICD9, but also included evaluations of human behavioral patterns like late refills on necessary chronic disease prescriptions, or incidence/prevalance rates for some of the more classical population based disease distributions.

My second way of 3D mapping uses the grid modeling, of one or more resolutions.  I tends towards the “best resolution” models determined using standard comparison analyses of the ability of different grid cell sizes to cluster series of adjacent cases to each other, producing the best fit half-bell curve of smoothly changing frequency distribution half-bell graphs.    Each have their values when used to produce 3D mapping of large regional health care statistics.

My recent years have been spent applying these processes to local disease mapping.  In particular, larger urban areas were needed to test the models that were produced.  This allows the cartographer to test outcomes, and then to quantify the utility of each process relative to the spatial-temporal distribution patterns of diseases or measured events.  Some processes work incredibly rare on rare disease events.  Some appear to be under representative of high density patterns, like diabetes, obesity, hypertension, atrial fibrillation.

The use of prevalence incidence metrics for the spatial grid modeling technique is still very important to work with.  But in too many cases (about one third of the time), this traditional use of “rates” as determined by demographers, epidemiologists, and public health specialists to be the chosen method, are unrealistic and useless at the business level.  It would be a tremendous waste of company money to direct your spending to the higher prevalence/incidence places, which some of these values relate to low population counts, not any true need to healthcare facility or products and care management development.


There are several essential ways to develop your EMR data for automating the managed care population health reporting process.  With the proper use of macros, you can develop the programming needed to make these calculation processes happen.  Most importantly, this methodology doesn’t require the introduction of new software and work station settings to the system.

Granted, these standards to evaluating populations will be around for a while, and unfortunately, they distract from the directions in which more effective evaluations need to be made.  The standard Cognos, Sharp and other systems out there are productive, but too slow.  These processes evolved from the processes I developed are a great deal more productive, and produce lengthy detailed reports, on a daily basis if you wanted.

In the past year, the following algorithms were developed, tested, utilized and produced into report-producing programming end products:

  1. Population pyramid and standard barchart graphical depictions, for detailing standard metrics (demographies, diseases, costs, savings per member, cost, predicted costs, etc.)
  2. Age-Gender Population Pyramid produced statistical comparisons of matched or comparable regions, for any health-related metric
  3. Race-Ethnicity defined grouped data and regrouped data (binomial grouped) differential analytics, per metric
  4. Religion and Health regrouping algorithm used to compared traditional and various non-traditional religion and religion-health profiling methods for evaluating areas, regions, neighborhoods, in relation to treatment patterns, refusal for care, disease outbreak patterns, and performance of preventive care programs.
  5. A method for defining the least healthy patients in any program or population
  6. A method for defining the most important (highest numbers, incidence, costs)  ICD groups for a given population, based upon race, ethnicity, etc.
  7. A method for applying and then combining or merging the three major risk scoring formulas into a single population-area health analysis methodology, specific down to the 1-year age increments for assigning a risk score, for each patient, and then reporting the summary of these data at the age-gender depicting race, ethnicity, religion, regional, program-defined, facility-defined levels [8 x 3 x 10 x n1 regions x n2 insuranceprograms or insured groups (MCD, CHP, MCR, COM, etc.)].
  8. A method for mapping these data (numerically, or fraction/ratio related):
    1. institutionally
    2. facility or office/provider based
    3. by network types (MCD, MCR, CHP, COM, BCBS, Obamacare)
    4. by industry distributions (urban hierarchical modeling processes)
    5. by socioeconomics data, race, ethnicity, and religion
    6. by specific disease classes (ICD, V and E codes)
    7. by specific human behavior patterns (based upon criminal or consumer data)
    8. by potential investor types


We expect some of what we see on the above diagram.  As people get older, they potentially get sicker and require more care.  Those who are the sickest experience increases in care related needs, relative to their number of chronic disease patterns.  Visits include all visits to doctors, hospitals, labs, referrals, ERs, counselors, anytime you walk into a place and are billed for that visit. Procedures are actions taken by whom you isit, like your lab tests, your annual x-ray, PET, MRI or mammography, your routine drug levels screening for seizure control, your monthly drug urinalysis.  VPR is Visits to Patients (per patient) ratio, per 1 year increment of age. ProcVR is the procedures to Visits ration; expected to increase as you get older and more procedures need to be peformed for more reason.  ProcPtR is Procedure Patient Ratio, which is a product of ProcVR and Visits to Patient ration of average visits per patient, by age increments (1 year).  The left column is for all standard office visit procedures; the middle column for Emergency Visits only; the third is of a population with a specific medical history using the ER for a specific reason.  What is most important to note  is the flate VPR that exists throughout the ER visits, relative to age.  Younger people come in for different reasons, but that seems to balance out with the newer reasons that lead older people to come into ER for in order to receive care.  The small spike at 0-2 years of age is for post-delivery problems, that often cause deaths in some children.



A review of this website performance over the past 11 months suggests, as expected, the academic year is an indicator of visitors.  Both winter and summer breaks define the lulls for its use.

As of July 4th, more than 360,000 visits were documented at this site since 2009.  It is currently receiving about 90,000 visits per year, representing about 60,000 visitors. The most visited site is page one.


A review of the monthly reporting for pages visited demonstrated there are repeats for certain pages as being the most visited for the month.  Tabulating this data produced a table of 24 most visited sites.  The two that stood out the most during the past year are my detailed review of the first US Census maps of disease and the historical medical geography article on an outbreak at Martha’s Vineyard in 1763.


If the ratio of visitors and views relative to the views of these Top 24 sites are calculated, the following numbers are generated based on these three monthly sums.


There are approximately 925 more pages that can be viewed at this site.  Therefore, a review of these percentages was made relative to total visitors for all pages.

This produced the following percentages of visits that each of the 24 sites represents, on a monthly basis.


A closer view of the lower values is provided:



From these results we can determine the most popular topics in historical medical geography and historical medicine.


My site is saturated with coverage of the history of Colonial medicine, centering on the New York area, and so may of these pages pertain to that subject

  • Small Pox and the Cree
  • Throat Distemper in Kingston, 1735 (early Diphtheria?)
  • The Diagnosis of James River Ringworm (Slavery and disease, Thomas Jefferson’s estate; a disease never before evaluated in the historical literature)
  • 1763 – the Extraordinary Disease
  • Revol War Doctor,
  • Jane’s [Colden] Plants . . . ,
  • Valentine Seaman,
  • the Fowler [initiators of Phrenology] Estate,
  • Divine Psychiatric Truth [religion, philosophy and psychology]).

Midwestern and Far West exploration and development are also reviewed extensively, those pages  being:

  • Cholera on the Oregon Trail (my 2000 MS thesis)
  • Plants along the Trail (Oregon Trail medicine)
  • 1808 May’s Lick Kentucky (an early geographic interpretation of the Midwest)
  • 1851-1917 Cattle Drives and Texas Fever

Historical Mapping

  • Valentine Seaman (the first ever disease maps, of NYC yellow fever)
  • The diagnosis of James River Ringworm
  • Yellow Fever revisited . . . again
  • 1851-1917: Cattle Drives and Texas Fever
  • 1890 – The Census disease maps
  • Four Prussian Disease maps (1890s map)

Several cultural topics received a lot of attention, especially:

  • Small Pox and the Cree
  • Medicine (Prayer) Stick (Indian medical philosophy)
  • Chicle: The History of Chewing Gum (an important Hispanic Heritage study I did from 1988 to 1992)

A stand alone piece I produced was on the history of the late 1890s-1930s ammunitions seller, Francis Bannerman and Bannerman’s Island, which I have family photos and relics from, an important piece of Hudson Valley history

Another important standalone topic is my work on combining qualitative and quantitative research practices for more thorough cultural reviews of community and population health (my current occupation).

This leaves the mention of hexagonal grid modeling, a technique I developed in GIS and applied in winter 2003/4 (unemployed), for mapping environmental chemical exposures in the state of Oregon.  I developed the math and theory behind this after beginning work for Medicaid/Medicare in Denver Colorado in 2004.  I posted it on this site in late 2009 or early 2010.  The popularity of this page immediately jumped to the top of my lists of pages visited.


Due to the increase popularity of this topic, I produced the DOWNLOAD page for students to access the excel I used in 2004 to produce my hexagonal grids. (newer versions of this are perhaps due for release; I have an SQL/SAS related version as well developed.)  This method of modeling continues to rise in popularity.


Based on number/percent of visitors to it for downloading the excel file, it continues to increase, with primary interest perhaps being expressed by students enrolled in GIS and/or urban planning and development programs.  Based on the feedback I get, most of the support for hex grids remains mostly a practice of western European spatial analysts; for the Americas, its frequent visitors and users appear to be Canadian.

(Time for US urban planners and spatial analysts to catch up!)



Over the past few weeks I have been ranting about the fact that the current managed care system has advanced little over the past 10 to 15 years.  In fact, the first articles on the barriers to developing an electronic medical records system so essential to the managed care environment were published fifteen years ago.  Today, many of these problems remain.  The only thing that has changed is that information technology software programs and packages have improved, the amount of data that a data warehouse can store has increased significantly, and the speed at which ‘Big Data’ analyses can be carried out has improved substantially.

When you turn to the LinkedIn posting on IT, you’re left with the impression that this is a rapidly advancing field, with the ability to bring US health care programs to the next generation of due to progress.

Well, if you take a look at the accomplishments of most programs, you are more than likely going to see managed care programs still struggling to demonstrate success with their programs.  Should this success be in the form of highly successful meaningful use measures, or important changes in population health features, there had better be a few of these “accomplishments”.  After all, a typical program is evaluating between 50 and 150 metrics to document its accomplishments.  Only some of these are reported.  None of these programs (to my knowledge over the past 4 years) produces a thorough ‘Quality of Care’ or ‘Quality Improvement’ analytic program, measuring all or many of the accomplishments of the past years worth of healthcare efforts, and then maps the most significant of these findings.  (Such is a reasonable goal, and product, for a HIT-GIS program.)

Knowing how to program your system to make such measurements is 90% of the work; its focus should be on the quality of care/quality of service (QOC/QOS) process for a majority of disease patterns in the region.   The reasons such steps are not taken relate mostly to poor planning and administration.  These processes could have been decided upon, planned, implemented, expanded and made more efficient in 2005, the year that the content of a valuable meaningful use program could be defined, a time when such a program referred to as a quality improvement [QI] program became a requirement.

There are several reasons managed care programs have not implemented a data warehouse setting in such a way that a spatial analysis technique can be applied, or added to an institution’s QI program with full scale implementation of a GIS.

First, there is no official policy or recommendation that was ever put into place for the implementation of spatial analysis techniques as part of a new meaningful use, quality improvement program.

Second, the skillset for implementing such a program is lacking at the leadership and administrative level.  This lack is because directors, managers and administrators failed to hire the type of human resource needed to implement such an innovative program.

Third, even if individual exist in a healthcare system whom are capable of producing highly useful monitoring and intervention maps, the know-how for implementing a new program based upon these findings is also lacking, a blame we may once again lay on the administrators.

So meanwhile, as the Health Information Technology [HIT] departments at most institutions struggle to make old and new outcomes match, and develop a database that is not only consistent but highly worthwhile for monitoring managed care population health, it is best for those already working in their system to take on the next most important research question in managed care.

Can religion be used to evaluate population health, and if so, how do we make use of this unique form of datum?

Religion provides us with equally valuable if not more insight when compared with ethnicity and race data.  Ethnicity can be used to search for social inequality and language related barriers involving the most predominant patient ethnic group in the United States.  Race provides us with insights into how socioecononic status/poverty, and race and race-linked genetics can impact a populations health features.

What religion tells us can serve as a further clarification of race-related findings for illness and disease.  Religion also adds to the insights we receive from the ethnicity  work engaged in.

I have identified 9 physical science, 7 social science and 5 behavioral science and/or mental health topics that are closely linked to diseases and the medical disorders people are diagnosed with, in such a way that if these lists were related to the therapeutic processes engaged in for any ICD-defined medical/health state, the results of this interpretation of each ICD should enable us to define those which are most linked to a particular religious culture.

Elsewhere on the web, I have identified the various types of cultural disease patterns that exist, based on the current ICD systems.  There are well defined culturally bound diseases, rarely mentioned culturally-linked disease patterns, frequently mentioned culturally-related health conditions and disease patterns involving the physical body.

Specific religious groups infrequently overlap with their disease patterns in United States cultural settings.  There is the majority of diagnoses that are ubiquitous to population health, which individuals of all race, culture and religion experience.  But there are also specific diagnoses and health problems that occur due to culturally-related human behavior habits.  These are the topic of this review.

Using the standard research theory models, like Health Belief Model, to understand how people react to a medical condition, the above figure can be related to that model and the primary dimensions, subclasses and behaviors linked to a disease can be identified. Then a more effective and thorough intervention program may be developed and specifically targeted.

A “meaningful” managed care system assesses the entire population for all of its healthcare needs.  This model enables planners to develop a balanced program devoted to physical, behavioral, mental and social care processes.   It can be used to define when and where focus groups are needed, what populations to target with a survey, where to look for unexplored ICD related topics for you particular population, and how to improve the intervention planning process for conditions that aren’t effectively treated at the moment.

Consider each of these elements a question that has to be explored in order to fully understand the causes for certain non-physical, behavioral, social and cultural behavioral disorders and/or disease patterns.


Get every new post delivered to your Inbox.

Join 4,152 other followers