Islam . . . and Epilepsy



Muhammed had epilepsy.

At least according to some medical historians.

This idea that Mohammed experienced epilepsy has its origins, at least in Western allopathic literature, around the late 19th century.   It became a very popular notion that seizures could be related to creativity and genius.  The examples that existed at the time were important historical leaders like Muhammed, and a number of  the most important religious-political leaders in ancient history and the recent post-Renaissance history of the sciences often refer to illness as the reason an individual becomes an exceptional achiever.

Perhaps the most commonly cited illness linked to creativity is Tuberculosis, with the majority of examples of these famous writers producing during the nineteenth century.  In more recent times, we have related depression to creativity and genius, mostly due to the many famous writers who experienced “melancholy” or some sort of multiple state, multiple personality experience like bipolarism or manic-depression.

But the link between genius and creativity, and neurological and/or psychological-psychiatric manifestations, can also have something to do with the “broken mind”.  This occurs because the brain and/or mind is capable of thinking outside the norm, in ways not normally imagined and followed by the normal person.  The new mindset that creative people develop is nurtured by the emotions it is linked to.  Those who experience feel great about it, sense it as being very important and therefore mustn’t be stopped.  Sometimes, the thoughts go “out of control”, they dominate their patient and make him or her a victim, “diseased”, and therefore in need of remedy.  The impacts of this treatment traditionally are to negate the intellectual prowess the afflicted mind experiences. For this reason, patients often refuse their care; they need to experience that “hyperactivity” once again.

In order for biologists and physicians to accept the idea that the mind, or thought process, had some specific reason it could become creative and special, they need to have a theory which would help them argue and understand the natural science reasons for why such behaviors occur in people.  The theory has to be focused upon the physical sciences, a task hard to achieve for knowledge that is theoretical and, as we might term it today, virtual in nature.

The earlier notion that the mind was its own floating entity contained within the body of a person, along with that person’s soul if you go back far enough in the literature, had to be stripped away from the philosophy on who and what people are, why they behave they way they do, what makes them become what they grow up to be.  Psychology had its roots already well established, and those roots, along with the newest technology in medicine (namely the x-ray and the ability to read the electrical behavior of the brain), enabled us to look into the skull of an individual, and assign physical sighting to each of the behaviors we are trying to describe.

By the end of the 19th century, physicians had a more detailed look at the brain and the structure and anatomy of its most delicate materials used to produce a brain.  The striated linear flowing, radiating lines from the center of the brain to the cortex gave them the idea of a specific form of circuitry involved how and why the brain behaves the way it does.  Science was too much dependent upon the physical world to allow a metaphysical theory be used to explain these structures to prevail.  They saw these structures as connections that linked the centermost part where the primitive brain lies, with the outermost section, the cortex, more advanced in its nature, and so large relative to the rest  of the brain.  This new interpretation of the structures they could see was developed due to the ways in which they recorded and interpreted the minds special activities.  They had the EEG to link to their many previous studies and findings about the behavior of the nerve, the muscle, and other electric producing parts of a living being.  Most central to their ideology on how the mind develops, excites and causes the body to behave, is the notion they developed on what relationships exist between the brain and body.

The sympathetic nerve, which traveled along the spine, outside the bony vertebrae protecting the most important spinal cord within, seemed to exist for some major biological reason  They knew this since the 17th century, when the Great Sympathetic Nerve was first talked about by anatomists, scientists, doctors, surgeon and learned religious writers. (I review this in detail at )

This nerve was called the great sympathetic nerve for a reason.  The most important finding about its purposes included a demonstration of its ability to influence the heart in the chest cavity, as well as many other parts of the internal organs live forms possess.  It function in the body as the means to maintain a balance and “sympathy” in the body of one part to another, including the intestines to the heart, the gut to the blood, the liver to the lungs, the lungs to the mind and soul.  Our vital force could be defined using the notion of electricity and the great sympathetic nerve; physicians didn’t need a vital force concept to come up with a viable explanation for consciousness and life. So even during the late 19th century, as the EEG was developed and perfected, and the electric images of what happens in the skull become documented, the physical structure of what these once metaphysical concepts could be connected to, physically made more sense.  The vital force, a philosophy replaced  by galvanics and magnetism ideologies, could now also be related to the electric poles, the volt, the roles of the storage device for these energies, making the brain a battery of sorts, with parts that exist in order to conduct that force of life and thinking along through the rest of the body.

Epilepsy, was the manifestation of that energy travelling throughout the corpus (body).  Any half-trained individual could easily make such a conclusion.  If primitive and ancient in nature, that theory states the seizures is a phenomenon that happens when the electricity within us goes along its own decided routes, not a route we send it on through conscious thought and thinking.  In more advanced settings and training, we believe this uncontrollable flow of the force or energy happens because it someone becomes automated within the brain, but once automated and allowed to spread, it follows these predefined routes the body has for it to take, the sympathetic nerve.

By the latter half of the 19th century, the role of the cerebellum was very well understood.  Its purpose in making the body capable of engaging in motor activity, in a controlled, thought out way, was somehow integrated into this sympathetic nerve ideology as well.  The way the sympathetic nerve behaves and the role and activity of the cerebellum are what made the person in a suspended, unconscious state of mind, move about meaningfully when a seizure happened.  The epileptic walked and moved about in sensible ways, without the conscious person we know engaged in those animalistic activities and behaviors.  We could walk, mumble, make simple cerebellar-engaging decisions about moving our muscles and changing our direction for transport, but intellectual ideas were missing.  You couldn’t easily ask that wandering epileptic what day or time it was, and easily get an answer.

This idea on brain function was well stated in an 1861 interpretation of how the “tumour” in the head can cause an epilepsy to erupt.  From it we can see the turn towards philosophy, when anatomy doesn’t seem to make sense for the cause of epilepsy.  Note the distinction made between the “epileptic fit” and “epilepsy” as a condition or diagnosis.

“In such a case no motor nerve fibre going to the various muscles of the head the trunk and the limbs is directly irritated by the tumour. The mode of action of the tumour must therefore be through some peculiar influence exerted over those parts of the base of the brain where exist the motor nerve fibres. It is in fact by an action on a distant part of the nervous centres that the irritation in the cerebellum then acts, just as is the case when epileptic fits are the consequence of an irritation in the bowels. . . The same thing may be said of epilepsy caused by a disease of the cerebral lobes. So that convulsions or convulsive affections symptomatic of disease located in the cerebral lobes or the cerebellum owe their origin to a sympathetic influence of the part altered upon other parts of the brain ”   [C. E. Brown-Sequard.  Lectures on the Diagnosis and treatment of the various forms of paralytic, convulsive, and mental affections . . . . Lecture I. Part IV.  The Lancet, July 27, 1861.  See at ]

Still, the mind had its own roots, larger nerve paths within, that enables thoughts to flow, fairly reasonably.  This enabled the physicians to differentiate seizures into several types.  Some that consisted solely of motor activity like the Jacksonian seizure, and others with the ability of the brain and body to appear to live consciously; others just made you talk or appear to be performing some specific behavior.  Still others make you do unhealthy, uncontrolled activities, the convulsion and ophisthotonic contraction being the prime examples of this.  By the end of the 19th century, the epileptic seizure was defined well enough to allow physicians to begin to link epilepsy to a pattern in Muhammed’s “enlightened” behavior.

Another aspect of the definition of seizures in the late 19th century pertained to the human affect forms of seizures that were being described.  These seizures were based upon the emotional interplay of behavior and seizure related facial expressions produced by people with specific seizure types.  The most common of these were those in which the individual appeared as though he or she were communicating with an alternative being or higher power.  This behavior was documented in the early 1800s and termed divine somnambulism (divine or god-like power generated sleep walking).  By the late 19th century, the different affective states, like fearful, elated, happy, or melancholic, were linked to these seizures.  There was a psychiatric part of the epilepsy that some patients could demonstrate.

Mohammed’s apparent seizures or seizure like activities were not principally or solely motor.  They were of a cognitive nature.  They engaged the emotional centers of the brain.  They made use of the were as much cortical in nature as they were cerebellar and sympathetic in nature. These seizures hadn’t a well defined name for them during his time, except for the ‘epilepsia’ defined by Hippocrates and colleagues, promoted by Galen and perpetuated by Maimonides, who around 1200 claimed epilepsy is “due to pressure on the ventricles of the brain,” possibly a result of the occasional hydrocephaly of the brain  that physicians noted in association with young children with seizures–the association of the seizure with an already documented state of hydrocephaly was very high.  To counter this intracranial pressure problem, Maimonides recommended a healthy lifestyle with proper diet (for more on Maimonides, see Jewish Encylopedia, Vol. 7, p 295 at ).

For other interpretations (mostly the same) of this aspect of Muhammed’s health see: 

Why Muhammad thought he was a “prophet” (Temporal Lobe Epilepsy)

Was Muhammed an Epileptic?

Did Prophet Muhammed (pbuh) suffer from Epilepsy? 


The notion that Muhammed has epilepsy, in the beginning, is not culturally denied.  The seizure or epileptic fit was different from Hippocrates’ Epilepsy diagnosis.  The ‘epileptic fit’, as perceived by an Islamic scholar at the time, may have been likened to the fits that other priests and prophets had according to the Old testament, in which, as traditional  stories state, “God” had conversed with them.

For example, in the July 16, 1910 The Reform Advocate, a paper on religion, during a discussion on how the Koran authored by Muhammed came to be, it was mentioned that Muhammed used the epileptic fits to gain the “enlightenment” needed for parts of his writing.


Page 1131, The Reform Advocate, July 16, 1910, at

In the Karl Baedeker series on Travels, his Egypt. Handbook for Travellers, this interpretation of Muhammed’s life experience was discussed extensively in its forward.  A footnote to this forward includes a note penned by “Prof. Socin”, in Section III, on the “Doctrines of El-Islam”.  The footnote to the first page is as follows:


To make clear the main point that he spent the latter part of his life as an epileptic:

[W]hen Mohammed was about forty years of age, he too was struck with the vanity of idolatry. He suffered from epilepsy and during his attacks imagined he received revelations from heaven. He can scarcely, therefore, be called an impostor in the ordinary sense. A dream which he had on Mt Hira near Mecca gave him the first impulse and he soon began with ardent enthusiasm to promulgate monotheism, and to warn his hearers against incurring the pains of hell. It is uncertain whether Mohammed himself could read and write. His new doctrine was called Islam or subjection to God.  [Source:  Alternate source: ]. 


Muhammed’s seizures could have been separated from those considered to have full-fledged or “true epilepsy”.  They, therefore, could just labelled epileptic, meaning they resemble epilepsy, but lack the repetitiveness and debilitating aspects that were so clearly pronounced in people whose seizures resulted in convulsions and the great sickness or “grand mal” version of seizures.  Such a clarification might minimize the cultural judgments made of the individual, setting the stage for how society would react to this person’s natural history and “purpose” in society.


Muslim Prayer ritual, from Baedeker.

In a 1907 discussion of the same topic concerning Muhammed and epilepsy by W. S. Monroe, in his book on Turkey and The Turks, and Account of the lands, the peoples, and the institutions of the Ottoman Empire (source ), several explanations are given for Muhammed’s state:

“Mental pathologists on the other hand have explained the contradictions in the life of the Prophet by theories of epilepsy and hallucination. Dr. William W. Ireland, an eminent Scotch alienist, believes that Mohammed was subject to some nervous disease accompanied by hallucinations. Theophanes Honoras and contemporary Greeks assert that he was afflicted with epilepsy. After a careful examination of all the evidence touching the mental health of the Prophet, Doctor Ireland concludes that it seems likely that Mohammed at the commencement of his mission was subject to hallucinations of hearing and sight, which, taking the tone of his deeply religious feelings, and his dislike to the idolatry and polytheism of the people of Mecca, were interpreted by him as messages from God. In this belief, he was prompted by his wife Kadija and some of his relations and was thus induced to commence his remarkable course of apostleship. How far these hallucinations accompanied the remaining twenty one years of his life it would be difficult to say.”

Thus, was Matthew Wood’s story about three epileptics, each of whom had a profound effect upon the world meaningful and correct.  In Wood’s In Spite of Epilepsy, being a review of the lives of three Great Epileptics,–Julius Caeser, Mohammed, Lord Byron,–the Founders Respectively of an empire,  religion, and a school of poetry (New York: Cosmopolitan Press, 1913 – – accessible at ), he becomes another example of the many authors trying to understand the inner workings of the brain, . . . the content, ingredients,  and substance or makings of the mind, . . . the power of human feeling, and emotions, . . . and the attachments of each of these to life and the potential of some high power or ‘the universe’.

Relating this to contemporary health care, Islamics have three causes for epilepsy noted in their combined medical-religious philosophy. Muhammed could experience this because he is young and “gifted”.  It could be a test of his purpose in life. Or it could be a condemnation of who he is or is soon to become, but more importantly, this same message can be interpreted by his parents as some form of punishment to them, for reasons never made clear.

In such a case, if your child is young and you suspect “epilepsy” due to a seizure, you go to the physician immediately with hopes of learning if this is due to Allah’s discontent with you or not. If it is Allah’s test of the child, and not just you, you will more than likely find out later, but still need to know now.  And equally important, especially in the modern medical world, you wish to know if the seizure is due to those scientific ideas circulating about for its cause–is this simply a manifestation of the physical world influencing the child’s body?

The answer to this question about the three options for why seizures and/or epilepsy are manifesting in someone (like Muhammed) are lacking for the moment.  They may even be forgotten . . . briefly.

The notion that Muhammed himself had epilepsy is essentially absent from their contemporary writings, but may be present in scholarly Muslim works on him.  This philosophy was essentially promoted, if not generated and promoted largely by Western European scholars, an irresistible reason for Imahs to ignore its publication by other parts (the bulk) of the world (see Criticism of Muhammed, on wikipedia at ).

Muhammed developed his epilepsy (if that assumed diagnosis is right) in his thirties or late thirties.  He was not born with it.  He did not obtain it as a hydrocephaly or any other function or physical birth related defect.  Thus, it was not the Epilepsy of Hippocrates, and it was not produced deliberately by Allah.  The production of a child with epilepsy is believed to be one generated of Allah’s intent; that which will result in a child born with the epilepsy or who develops signs of being born with it early on in life.  So, the first theory of why Allah allows disease to happen–Allah allows the person to be born into a life with it, for whatever reason–does not pertain to Muhammed’s history.

There is also the theory that epilepsy develops in someone because Allah allows it, and thus it becomes a test of the individuals strength, wisdom, and other talents needed to overcome this malady.  Children who develop seizures and adults would certainly relate to this theory, which is the theory most closely related by Muslims to Muhammed’s reason for possessing it (if they believe).  Thus, epilepsy is like other cultures state–a test of one’s strengths in life, the route to developing a much needed wisdom.

There is also the third cause for epilepsy which focuses on the physical world theories, and approximates the western medical paradigm for why seizures and epilepsy happen.  Although such a theory of Muslim writing and belief may not focus on the same paradigms taught by western scientists about the intricacies of the brain, there is enough of this physical and anatomical, functional knowledge in Muslim doctors writers to allow them to produce their own “take” or “spin” on theory for why epilepsy happens.  Since it is of Muslim origin, and because it is complex to many, like the western paradigms that barely a non-medically trained professional could understand, this Islamic theory about the cause and reason for seizures or epilepsy become irrefutable.

Now, relating this to contemporary medicine . . . a recent review of a significant number of Islamic patients demonstrated that Islamic children today are being brought to the family physician at the age of 1, 2 and 3.  This is occurring more for Islamic children than for any other race or religiously related raised child.

This expenditure of time and money for medical services in a young child, due to concerns for epilepsy diagnosis, stem from the Islamic belief that this seizure is a test of who the child is, who the parents are or might be, and whether or not Allah is pleased with these people or not.  Their visit of a doctor for a child so young is needed, in order to know your fate, your child’s fate, as soon as possible in life.

In my next review, I will go over the contemporary literature on this, and the Islamic medical writings I have reviewed over the past 25 years on this subject.



FGM: is it already here? Part 1: A history


The numbers of Articles (and a few books and book reviews) on Female Genital Mutilation published over the years, as noted in a) Google Scholar based upon queries for separate years, 1600-1899, and b) David M. Westley. (1999).  Female Circumcision and Infibulation in Africa.  EJAB: Electronic Journal of Africana Bibliography, v. 4. Search carried out on 6-4-16.

Female Genital Mutilation: an ethnomedical history


If we consider genital “mutilation” a specific form of genital change or “modification”, the process of changing the form or appearance of this part of the human body has an ancient history.  Along with changing appearance, these changes were also meant to control or modify the sexual activities that are engaged in.  The reasons these modifications might be wished for are numerous.  But they typically focus  on just one behavioral change –to reduce the desire to engage in sexual activity, experienced by a man, or a woman, or both.

Genital mutilation is just of many activities mankind engages in to control and modify the sex drive and activities.  For each culture, there are numerous expectations for when, how, why and what to perform as a part of this behavior that generally requires two or more people.  For much of history, these interpersonal behaviors have engaged a man and a woman, with each playing an assortment of different roles in the rituals expected by a culture.  These ritualistic activities may involve one sex or the other as an aggressor, and so to modify these behaviors when they are not wanted for some reason, the instigator of this change has to decide whether to prevent the aggressor from starting such activities, the respondent from accepting, further supporting and then engaging in these activities, or if it more appropriate, to simply make such activities harder to engage in harder to pursue, harder to complete.


The genital modification process is one method used to control how a person will beginning engaging in, following through, or finishing up with the sexual activities expected by each of the parties involved.   The purpose of genital manipulation or mutilation with regard to this specific interpersonal process is, more often than not, to be sure the possible participants cannot participate in their behavior of choice, for culturally defined reasons related to the initiator and/or instigator or recipient of such events.  Genital changes are made an accepted by a culture and society in order to better ensure the survival of their own type, their own social group.  There are certain expectations that a culture has to help define what these genitalia related events might be.

Culture defines what it is people can do when it comes to sex.  Genital manipulation, change, modification and mutilation are culturally approved procedures that are used to maintain control of this otherwise completely personal, biologically driven phenomenon between two people.  Since the healthcare process is often provided by cultural, societal leaders, with much of its activity and expectations defined by the culture and society’s belief system, be they documented on paper or not, the process of engaging in genitalia changes as a part of the healthcare “practice” is intended to make for a “healthier” more emotionally sound and appropriately active community.  When this process fails to meet cultural requirements and demands, the consequences of such actions can be perceived as more than just a failure by the society involved, but sometimes as a failure of the cultural system itself to be able to survive.  For example . . .


“Monstrosities” 16th C. to 19th C. – Babies born due to poor temperament, heredity, inappropriate sexual encounters, and a host of culturally defined moral causes.

World History of Genital Modification Processes

The global history of GM or genital modification (later genital mutilation) processes did not begin as lopsided or unidirectional in how it targets people genderwise, as it does today.

In some of the oldest tales about this practice by the Ancient writer Seneca, it is said that the purpose of this method of controlling people was targeted most directly at soldiers, so they would retain their male aggression and the strength and agility needed should another war become part of their history.  Seneca’s tales refer to the use of infibulation practices, involving the cutting of flesh from the male genitalia, for the purpose of reducing their desires for engaging in others forms of aggressive and active energy expenditure.

The culturally defined reasons for these changes and their related ethnic reasoning, good or bad, typically bear just a few common elements per generation or few generations period.  A number of related procedures were developed as well to ensure their purpose and activities were justly directed towards the needs of whom they served.  The castration of men was very common, perhaps more than female infibulation practices in some cultural settings.   The development of eunuchs ensured those who were hired met the needs of their employer, instead of the needs of competitors, spies, or enemies.


This eroticus-generated means for protection of the forces was different for the women’s side of politics and the military.  The non-treated males of a given society were like any others in any other societies–meaning the women intended for leaders and royalty to become involved or engaged with were not necessarily “protected” from whatever the other options were in their life experience.  The means for controlling these processes, from the women’s perspective, involved equally aggressive processes focused on manipulating their body form, functionality and desirability, as well as finding the means to still physically control such activities even when the opportunities surfaced for  these behaviors to ensue.

The latter issue was an easier issue to contend with, and involved the well known option of producing a chastity belt in order to deter the potential male (or sometimes female) partner(s).

By the dawn of the eighteenth century, with the scientific revolution well underway, the knowledge of anatomy and physiology helped scholars, and sometimes physicians, develop useful theories for understanding the interplay of human consciousness, thought processes, emotions, and the desire for sex.  Astrologers had a long history of trying to define this unique interplay of supposed natural “forces” between two different beings or objects.   During the 17th century, astrologers’ ideas were still circulating about the medical and early human psychology fields, but were often intertwined with the similar ideologies posed to scholars by alternative cultures and their own unique philosophies.

The mid 17th century was when science had to explain the role of the nerves aligning the spine, which they defined as the source for sympathetic activities in the body that take place to modify and controls its reactions to everything internal and external to the corpus in general–they named it the sympathetic nerve.  Over the next few decades, the sympathetic nerve was related to the energy of heat, the element of fire in the body, and these could be likened to the oriental philosophy of chi (a term they did not use in their writings), which in turn many felt reconfirmed much older hippocratic teachings.


Daniel Turner, donated a set of his books to the Yale Medical Library when it opened.  Author of books on the mother and her influences on the health and formation fo the child in utero, such as The Force of the Mother’s Imagination upon her Foetus in Utero, Still Farther Considered (1730)

From these insights came other theories about the body’s energy, its animated powers, its emotions and metaphysical responses.  By 1700, the notion of a life energy in the human egg and sperm was known, and the notion of seminal power in the production of life a popular metaphysical construct used by physicians to explain the inexplicable nature of humans to engage in certain behaviors as if they were out of their conscious control.  The “sperm” of the egg, its energy, could be born by either man or women, but the nature of science often led its followers to believe that the “sperma” produced by man gave the ova of women the life force it needed. Thus, when it came sexual activity, performance, and physical events produced by way of these activities, the male was the producer of life, but woman the vessel.  This further detailed and separated the roles that each gender had in sex and reproduction, the production of the life force, by merging the metaphysics of one with that of the other, only in a different form.

Between 1720 and 1750, this dominated some of the mental concepts continuously being shared between physicians and philosophers.  During the late 1600s, the remaining philosophies of alchemists were transposed into more related religious and philosophical concepts, by chemists Robert Boyle and Johannes von Helmont, and chemist-physician Herman Boerhaave, personified in their metaphysical value by Paracelsus and Jakob Boehme (entia), and transformed into a very much God-given Christian philosophy by the Harvard natural philosopher (to us a scientific “medium”) Charles Starkey (ens veneris).  This made mankind seem uncontrollable due to his/her “passions”, and the nature by which their energies wished to flow freely, without physical restraints.


When we look at the popular writings for this time of a “scientific”, scholarly, and philosophical nature, we find their mention of sexual desire and activity to be much along these same philosophies of the then past and present.  But when it came to controlling these behaviors, the physical procedures needed to change or modify the human behavior were important.  These dictionaries commonly explained to their readers the purpose of the castration, clitorectomy and infibulation. Treating this as a fair non-preferential human sexual topic, men’s need for these processes were explained almost as fairly as the women’s need for their equivalents.

The more historical nature of the need to control male in Western societies led this particular philosophy along a different route than the same for a woman’s “lack of control.”   The religious influences now prevail, and the notion of prostitution and infidelity become important concepts by the 1770s, as the philosophy of controlling sexual vigor finally reached at state that was desperately in need of change.

In the late 1760s and 1770s, two common themes prevailed for writings pertaining to women–circumcision and infibulation.  For men, these two terms could also relate, but the main means for controlling masculinity was castration, a practice by then atypical of western societies, except when the dire need for it exists, or when it becomes the result of an “accident”.  By this time as well, the published notes about the practice of the same purportedly by Native Americans were in the public press.  These behaviors reminded some of the nearly identical medical practices engaged in by African and some “Oriental” (including Middle Eastern or Muslim) communities.

During the 1780s and 1790s, this transition of infibulation into a mostly female related process continued to develop.  Reminiscences of the ancient Greek and Roman practices of male infibulation are noted in the historical writings here and there, but as for the contemporary practice of this process, it was slowly becoming intended mostly if not only for women.

The male circumcision process now had its culturally defined places for being practiced, as well as infibulation, castration, “generation” or “degeneration”, and  “skoptziism”.  Their causes were different for the infibulation and clitorectomy procedures performed on women.



The Skoptzy are an unmentioned culture that thrived in certain parts of Eastern Europe.  Yet they are the equivalent to the women of African engaged in the ritual process of clitorectomy and infibulation.  They voluntarily undergo a ritual castration of external sexual organs, thus removing their ‘semenic’ power and reducing the “pleasure” sensing parts of their body, which otherwise might distract them from their cause and purpose.

To demonstrate their faith and their willing to adhere to its teaching about fidelity only to one, they undergo a ritualistic physical castration of both phallus, most of the scrotum, and its testicular contents.  In modern societies, such as practice is rare, if it indeed still exists.  Many modern “castrations” that are performed are engaged in medically, and for medical reason, using unusual chemical and pharmaceutical castration processes.



From a missionary pamphlet, for education of travelling missions students. (The small two-pointed spot is the belly button, beneath which is the belly and then pubic area (hair possibly included); the left and right lower parts are right and left thigh, respectively, with the skin creases and joints between these parts.)

In most of western European post-renaissance history, nearly every generation had one or a few reasons for considering a faith-proving intervention.  The first and by far the broadest reason shared by cultures stems from the adolescent desire for promiscuity and sex, involving both genders.  The second reason for genital modification, in terms of cultural beliefs, is the focus on the desire to control the mind and emotions, most often privately expressed in the form of “onanism”, as they termed it during the 19th century, or masturbation, as it is referred to today.  Like the concerns for uncontrollable desire and promiscuity, the fear of “onanism” also pertained to both genders.

Whereas the above two sexual desires pertain to both genders, the third reason often given focuses on just the nature of the female body.  It is considered the receptacle and instigator of the opposite sex.  The desire to control it was important to the minds of community leaders.

To control this problem, in the physical sense, two different preventive measures could be taken.  The first was to prevent the sex acts from occurring, by closing up the orifice.  The second involved producing a physical defense against potential perpetrators.

Such logic of course pertains mostly to the traditional western philosophy pertaining to chaste and virginity.  Whereas a simple physical device served this purpose in many cultures, the means for protecting the body from unwanted advances without such a device required extensive modification of the body/reservoir of such unwanted people.  Due to cultural beliefs, this practice makes the infibulation and circumcision process unique in that is is mostly a practice argued for a defined by cultural beliefs, as held by the family, parents and/or individual willing to engage in this activity.

This need to control the attractive feature, if performed on an adult, may have the mental health required for it to be allowed by its recipient.  But in most traditional settings where this practice is engaged in, the lack of sufficient reasoning for the need of infibulation and/or clitorectomy practices makes this unwarranted by most outside cultures.  The lack of understanding that children have about this procedure, and the level of maturity that they are at,  turn them into passive, forced recipients of this medical procedure.

This cultural split in the ideology of this procedures has been evident since the late 18th and early 19th century.  However, during those decades, we see several transitions in Western philosophy take place that temporarily support one of more reasons why these practices might be allowed.


In Western Europe, much of the Christian interpretation of onanism was in full control of these activities; with few exceptions, we find this reasoning prevail over others in the writings published during this time.  But around the end of the 18th century, Malthusianism was having an influence upon much of what society was thinking about population growth and crowding.  This led a number of leaders and governing heads to consider the role of infibulation and even castration or the promotion of “eunuchism” as methods for prevent the Malthusian predictions of the future.

And not that surprisingly, this belief system did perpetuate for several decades, almost to the mid-19th century, but it is infrequently mentioned.  When it is argued, the emphases are placed once again on maintaining family integrity, the control of human emotions control, preventing onanist behaviors from fostering sick and unhealthy children.  By this the mid to late 1800s, we find the practice of infibulation has become a mostly female-targeted procedure, and a major moral issue for western, non-Islamic societies, around the world.

The development of this practice in Africa and parts of the Middle East since the early to mid-1800s helped to stabilize its cultural meaning, resulting even in its ability to exist because it is a culturally specific behavioral practice with belief systems different from western traditions, that enabled these practices to continue.  For political and financial reasons, it enabled the woman to become more powerful through attractivity, but less powerful in terms of position within society.  The idea that prostitution and lust were to be prevented gave medical professionals even more reasons to promote the female genital modification procedures like clitorectomy and infibulation.  It was also during this time that the definitions of three types of procedures became well known and published, and their inventors or instigators well defined and targeted by the anti-gender discrimination groups developing in the western world.

From about 1850 to 1875, the first versions of the modern interpretation of genital “modification” or “mutilation” were published, referred to by their previous names for the time, without the cultural inferences linked to a term like “mutilation”.


The first US Medical Journal article on infibulation.  This complete article may be reviewed at

United States History

The first detailed description in United States medical literature of what we today refer to as female genital mutilation (FGM) is found in the Medical Repository, one of two of the earliest medical journals published in the newly formed United States during the late 18th century.  This article was penned by a physician, in the form of a retrospective case study drawn up from memory by a physician who was hired by a plantation owner in the mid-Atlantic region to manage the health of local slaves (see my full page on this, with the article, at ).


Title, figure 1, “Infibulation” and dedication page, from the 60 page tome entitled (as per rough translation) “The Hidden Truths of Ethnography: on the circumcision of women, virginity, infibulation, aging, eunichs, skopticism, padlocks and belts.”   In the figure, a mature lady is presented, possibly with a “hottentot apron”, but also alluding to preparation for infibulation.

A number of accounts of this practice were noted in the journals kept by travelers of Africa during the late 18th and 19th century.  One of the first fairly explicit ethnomedical essays on this procedure and several related cultural medical practices is French Navy Commandant Émile Duhousset’s  Les huis-clos de l’ethnographie : de la circoncision des filles, Virginite, Infibulation, Generation, Eunuches, Skopizis, Cadenas, Ceintures. (Londres, 1878.  Accessed at ), published under the penname “E. Ilex” (see also Google version at ).

Prior to 1960 (1900-1960, perhaps as early as 1865), many of the items published on infibulation were primarily anthropological in nature and provided the medical anthropologist’s perspective.  Medical cases came to be reported by the 1950s, resulting in a shift in philosophy about moral and ethical issues regarding this practice.  This cultural perspective however, being western based, had little if any influence on the actual practice of these rituals.  Until the 1960s, very little was known within the medical profession about this process, and with few if any patients presenting with these conditions, the understanding of this condition and its health consequences remained unlearned material for nearly all physicians, except foreign trained attending physicians.

As an example of this dilemma in health care, in a fairly early publication on this process in the British Medical Journal dated 1964, a brief description of the procedure was given along with the following two black and white images of the affected parts.  Its purpose was to detail what to nearly all physicians in the United States was a new and unessential therapeutic process capable of causing numerous health risks and repercussions.  Over the next few years, this now primarily female targeted diagnosis or state became of global interest.  (Dewhurst, C. J., & Michelson, A. (1964).



In 1962, 1966, and 1967,  pivotal articles for the African medical ethnomedicine field were published on this topic, the most influential British Journal of Obstetrics and Gynaecology (Mustafa, 1966), and by the Sudan Medical Journal (Shandall, 1967).  The first was a general overview of this practice, the second a more thorough clinical review in which three consequences of this process were mentioned: urinary retention, hemorrhaging and the post-procedural/post-surgical shock induced by pain and infection-related physical, emotional and/or mental consequences (Shandall, 1967).  About the same time of this review, a presentation was made on the history of these procedures on local sociocultural and economic stability and their influence upon future international relations (Modawi, 1967).


Dewhurst, C. J., & Michelson, A. (1964). Infibulation complicating pregnancy. British Medical Journal, 2(5422), 1442. )

Modawi, S. (1967). The impact of social and economic changes in female circumcision. [Unpublished Presentation].

Mustafa, A. Z. (1966), FEMALE CIRCUMCISION AND INFIBULATION IN THE SUDAN. BJOG: An International Journal of Obstetrics & Gynaecology, 73: 302–306. doi:10.1111/j.1471-0528.1966.tb05163.x

Shandall, A. A. (1967).  Circumcision and infibulation of females: a general consideration of the problem and a clinical study of the complications in Sudanese women. Sudan Medical Journal, 5(4), 178-212. Accessed at ).

Published Bibliography

Westley, David M.  (1999).  Female Circumcision and Infibulation in Africa.  Accessed at EJAB: Electronic Journal of African Bibliography.  Vol. 4. Accessed at




Why Map Religion?


Religions in the United States.  Source:


Currently, managed care programs are struggling trying to make the best of of the available electronic medical records data.  One of the most helpful series of insights into population health comes from a study of the religious background of your patients.  This is because the amount of adherence your patients demonstrate to your program has a direct correlation to the religious background of the patient, his or her upbringing and family beliefs about health and the role of the physicians, and the official religious concept that your faith system holds belief-wise, in terms of how much to belief in the teachings of your faith, the meaning of the lessons contained within its historical background and reliance upon spiritual and “higher” forms of healing, and how the physician is placed in the paradigm that a someone’s religious leaders follow when it comes to medical school and allied health or experiential medical training beliefs and attitudes.

Reviewing a part of the east coast religious background, for nearly thirty years I knew I was researching a region of the U.S. with quite a long and varied religious background, history, philosophy and combined spiritual and extremely positivist (scientific proofs only are accepted) practices and belief systems.  In the past I taught classes and produced essays and maps demonstrating that the heart of religious diversity is New York.  This is because the other big urban center developing during the first decades were guided by one particular culture–for Philadelphia and Boston it was the United Kingdom and British  philosophy for governing and leadership, with Quakerism contributing to the establishment of the colony in general for Philadelphia, but defining very little of the ruling theocracy for that time, and for Boston, the British and later Irish influences that prevailed, in part to keep the French forces both governmental and of a church-defined nature away from the ruling classes of the British societies forming in the New Britain, New Scotland (Nova Scotia), New England and upper New York parts of these pre-United States colonies.

Between these two British New World countries were the combined belongings of Sweden and the United Netherlands, what we refer to today as the temporary New Sweden settlement followed by the more successful rulership dominated by the Netherlands Dutch families, with their own unique interpretations of the role of the church in a very multi-ethnic, non-Anglican dominated society.  It is the Dutch who allowed unique minority groups with their own religions and religio-medical beliefs to dominate this narrow part of the New World.  With its multiethnic neighborhoods and multiethnic rulers located on the adjacent lands, the ability of this region to remain diverse in its people and belief systems is what made New York , at first, the site where many new religions and medical beliefs could be generated and propagated like the pages of a new version of the Bible.  Even with the diminishment of this alternative thinking in health, religion and medicine, through repeated attempts to define medicine as a predominantly allopathic field throughout the nineteenth and early twentieth centuries, religious groups remained diverse, and due to the diversity of these groups, within the multiethnic society that the Dutch colonists helped to form, we find the most ethnically diverse forms of public health related medical practices evident throughout this part of the United States.


From one of my several pages reviewing GIS, medicine, surveillance, and/or Foreign Born Disease patterns.

The database I am working with to understand this unique multiethnic population, with its  unique racially and ethnically diverse population, where plain “white” or causasians anglo-saxon borne protestants do not constitute the majority.  This population consists of 7.5M patients for the region being analyzed.  Their history and EMR stretch over a 20 to 25 year time frame.  Naturally, the richness of religious data indicating familiar and individualistic religious background provided for mo one of the first opportunities to evaluate the value of religious data in EMRs, for any problems or possible errors that might exist in these data.

The following are my preliminary conclusions to what a 7.5 year dataset of 160 religions tells us, and can be used to design the method for studying religion and health as it pertains to the health planning and administrative process, not the value of religion in eliciting the cure, assisting the patient, or creating the miracle that is needed.

Researchers need to understand the following when trying to understand and map the religion of their area.  The following findings that were made have a significant impact on how I will deal with a dataset when evaluating its people for religious background, from this point on, regardless of community size and type.

First, the amount of data obtained for a patient’s or patient’s family’s religious commitment can be scarce.

It would be no surprise to find that some databases provide as little as 30% of the patients’ data on religious practices and commitment. To substitute for EMR void of religious data, it is necessary to develop a survey to fill in for this missing data, to set guidelines or write up policies that make it essential than we maintain better records about the religion of our patients, providing them with the opportunity to refuse to answer this question if so inclined.  The cultural make up of the community, in particular its churches and religious program that define the area, play an important role in obtaining further insight into your institutional potential and actual patient loads.


Noma, from my page at

For a study I did 5 years ago on religion and unique foreign born forms of disease in parts of the U.S., I found that missionary organizations played a role in introducing cases of foreign or rare diseases to the U.S. EMR databases, but not necessarily in any attempt to result in further propagation or spread.  The severe malnutrition case discussed on another of my web pages for another blog, Noma, provides and excellent example of how a cluster of extremely severe cases may be generated and discovered through foreign borne disease mapping. [See ]

Second, there is an exceptionally large variety of types of religions that can be entered into an EMR, on the old fashioned paper copies of EMRs, documented as part of the first visit, or enrollment application for a new  program.  The lack of data is a problem with older datasets, and is one of the toughest issues to resolve when researching religion and health based upon EMR data.  Most of the time, a patient doesn’t require the institution to know or act upon the institution’s interpretation of religion.  And twenty years ago, many institutions had a history of supporting a particular religious group or two, but many are now in compliance with the federal standards stating that no preferred social group devoted to religion should be favorably promoted by a facility, at the most basic level of providing medical support for all patients.


A large number of different types of religions may now be found in EMRs developed by some of the more successful programs engaged in detailed record keeping.  Since the point of this study is to focus on how the interaction of religious faith, its influence on our personal lifestyle, and the way cultural beliefs influence our reaction to practitioners’ decision making processes, it is important to be able to relate the various faiths to the most basic health or public health related decision that are made.  There are very specific cultural beliefs and practices that could influence the care that is provided, or the likelihood a person may develop and illness.  These include specific dietary practices, the approval or lack of approval for certain child healthcare procedures, or what the mores and taboos about sex and relationships might be in terms of educating or consultation experiences involving a child.  Religion may influence  the way in which family medicine doctors cater to the need of a maturing adolescent, or provide medical information for a young adult pondering marriage and child raising plans.  The needs of the spouse and older parents regarding health care, the type and amount of support we or the practitioner should provide to elders and possible end-of-life stage patients, are also influenced by cultural beliefs and expectations.


Third, religion is a personal attribute that may or may not have the support of patients for sharing this part of the personal history information to the facility they are visiting.  In general, when patients are in the hospital, for one or more overnight stays, or for a surgical procedure, the community and social part of a patient’s life experience often provides that patient with the contacts he/she needs to get that important care of others for being in the hospital.  Similarly, non-denominational religious supporters are common to many hospital settings.  They are there to provide the support a patient needs before the operation or important procedure is to commence.  They play an important role in the mental health of the patient before such a procedure, and reinforce cultural beliefs regarding the interactions between family and patients, sometimes critical to the quality of a patients subsequent experience and/or outcome.


Source: Crane, J. K. (2016). Stem Cell Research and Judaism. Religions: A Scholarly Journal, (2014), 13.  Accessed at

Fourth, the option of filling out the religious background in you medical record can lead to diverse entries, many more than expected by an institution.  Generally speaking, we think of just a few major groups of religions when thinking about how the population’s different faiths distribute across the numbers at hand.  Knowing the practices common to a patient’s community may help with some patients, but provided a very small amount of critical information when this data is due because of its usefulness.  We need to know specific patients and their attachments to specific religious, because that religious faith they admit to and document in their records, provided the medical profession with very valuable insights, assuming the institution’s social scientists, ethnologists, social workers and psychologists/psychiatrists, are familiar with this important piece of demographic information, and its usefulness.


Recently, I had to construct a way to interpret a population of 7.5 million patients, who laying claim to practicing more than 160 different religions.    The response rate for this groups was about 50%, meaning about 50% had a well defined religious group contained in their EMR data.

Even though this data represented about 50% of the total patient population over the past two plus decades, it still provides the insights needed to determine how to analyze religion as a contributor to certain aspects about the quality of care provided to patients within this heavily populated setting.  The types of care they ask for or are against being provided are determined by their background, but especially their race, ethnicity, culture, and religious background.  Of these four possible demographic metrics for any given patient or population, we tend to just pay attention to the first two.

Religion also impacts the special requests people might have about particular parts of the healthcare services normally provided.  It is not unusual for patients to be against certain forms of care, treatment or preventive health measures.

Also playing an important role are the unique types of.ICDs that each population might have entered into the EMR, including ICDs for culturally-bound disease and behavior patterns, culturally linked conditions with some sort of physical science related entity, such as genetics or developmental related illness causes and patterns, and documented race-culture linked traditional ICDs focused on particular medical conditions and health behaviors, such as dietary, domestic living style, and stress management behaviors.

On another page posted at another site, I noted the 10 religious groups or classes I developed for interpreting religion in a population.   Each group presumably shared important features in the average lifestyle that are determined by the social and cultural expectations of that individual and family, based on the individual’s family-related and personal religious background.

The first six of these religious groups make sense:

  1. Catholic
  2. Christian
  3. Christian-derived or related
  4. Judaic
  5. Islamic
  6. Unique Cult or Sect (various titles have been given to this group)

These six have culturally defined church practice activities, church related social pressures put upon families and individuals by their place of worship, and certain practices–such as child upbringing expectations, child health education procedures or programs, recreational allowances, work and social commitment requirements, foodways and drug and alcohol consumption expectations, and beliefs about violence and criminal behaviors in the community setting.  Each of these has particular ICDs, V-codes and E-codes attached to them, which can be assessed in the EMR.


A Pentecostal “healing”, September 1954

The remaining groups are tougher to define, but two of them stand out, and surprising have similar belief-behavior relationships enabling hem to be categorized together.

Atheism and agnosticism both share the general lack in belief in a God or Creator concept as they are defined by the above groups.  Some of the traditional classes do have sections of followers that are essentially focused on the social setting and behave atheist or agnostic as well.  Modern Judaic followers who do not pay attention to the Torah per se, most often fit into this particular group.

Whereas Atheists believe in no God at all, essentially living live as a non-theological, purely nihilistic experience (nothing more matters once you’v departed), agnostics have that uncertainty in their philosophy, enough to “believe in a god, perhaps” but not so certain as to follow a particular faith or belief system due to this ideology.

The scientists are directly related to a belief system referred to as positivism–the proof must exist for something to be consider a proven statement or claim.  The counter believe to positive is relativism, which claims that relationships exist, but not necessary indisputable proofs of the cause for these relationships–also referred to as post-positivists, existential proof may be all that is needed.  The existence of something is proof enough of its value or worth.

This loosely linked, fairly uncoordinated and socially detached groups of followers (no major friends or support system for their exact cause exists), result in a pair of believe systems, that can be lumped together as a combined Modern/Postmodern group of patients.

Between the agnostics and the transcendentalists and/or experientialists is another group, which I refer to as the “natural philosophers”.  This group includes those who see a creator or God entity as existing but as part of the universe in general.


Three “healers” from the Hudson Valley, N.Y., ca. 1820 +/- 15 yrs.  Andrew Jackson Davis (left) was a self-proclaimed mystic who developed a world follow, as a product of Quimbyism and Mary Baker Eddyism travelling the New York Hudson Valley Circuit during the Transcendental movement; he was an early experientialist or possibilist.  First or Second generation Mahican descendant Mannessah (either Metis through the missions or Algonkin-Mahican) was a “Christian Indian” who revitalized the Native American healing philosophy and faith around 1790-1840, continuing it decades later in Ohio (left); we could consider he a natural theologian with religious minded experientialist and possibilist influences.  Samuel Mitchell was a NY Congressman and Benjamin Rush’s adversary in New York; he was a realist, actualist, and/or positivist who was broadly trained and highly respected for this scientific approach to interpreting most of disease and medicine, including the geographic perspective that he made famous.

The Quakers for example believe in a “Universal God” that is essentially an energetic thing, a”being”, or causative factor  for what we can experience. The happenstance that defines someone’s life is a result of the existence of this “power”, but its recognition may be missed or not noticed, based upon personal reaction and experience.

The Shakers communicated with this energy when it was founded, and viewed the people who were most gifted in performing these communications as the “most gifted” of worshippers, typically considering them valuable social leaders of “the faith.”  Mary Ann, the founder of Shakers, is certainly one of the better example of these gifted leaders.

The Aesculapians are natural healing, multigod energy worshippers, and view nature and energy/power as important to their natural philosophy faith.

Many Native American “faiths” may fall into this category as well.  The “Manitou” concept is akin to the Shaker’s form of God.

Likewise, the Moravians, actually best considered a Christian religious sect offshoot, are of this type, but due to their much stronger discipline and following as defined by Christian writings, have parts of their faiths overlap with indigenous belief systems, but at the expense of remaining traditional Christian in nature (Group 2 or 3 above, depending upon how you define each).

The Mormons and Seventh Day Adventists are certainly followers of the the Group 3, Christian derived faiths.  They have leaders who are gifted, but still retain some of the older traditions focused on previous gifted leaders, namely the Christ, or Moses, or Mohammed.

Mary Baker Eddy’s Christian Science group places more of the responsibility of utilizing their universal energy on the individual.  They are detached more from the God concept that the Quakers (who aren’t really detached), and almost as detached as energy energy believers who approach atheism or agnosticism, leaving open the option of communicating with that higher entity, energy or being in some way, shape or form.  Native American Manitou-related shamanism may be more natural theological in nature than the faith of a “Christian Scientist.”



This leaves us with a group of religions that are superficial and mostly personally and socially driven agnostic, atheism and purely physical world focused.  Their followers play a role as a social support group.  They have unique energy concepts or interpretations quite often. Examples include Scientology and the non-denominational churches that serve mostly as social gathering for the young to middle age.

The last group is the unstated or unknown religious group.  Those who provided no answer to the clinician when asked this question.

The remaining groups are as follows:

  1. Cultural (Buddhist, Hindu, Sikh, Baha’i, etc.)
  2. Natural Theological or Natural Philosophy
  3. “Contemporary” (Scientologists, Unity, etc.) [aka Modern/Postmodern]
  4. Other, Unknown

(More on this to come.)

Other classification systems paid attention to the natural philosophy aspects of the tradition, enabling some of the groups devoted to mysticism to have their own “transcendentalist” group.

The Encyclopedia Britannica website at offers several historical interpretations of how scholar classified religion during the past two centuries.  The most influential interpreters of the last, with philosophical concepts relative to today’s allopathic interpretation of religious faith and medicine, with a focus on the works of Cornelius Petrus Tiele and William James.

This slideshow requires JavaScript.

Epitaph to a German Lutheran Gravestone, ca. 1802 mortality, probably due to “consumption” or tuberculosis, or alternatively, rheumatism.  


Why Map Congenital?

With EMR we are now able to analyze about anything the medical data present to us.

We are currently struggling, it seems, to get our systems operating such that we can perform according to the standards we are being evaluated, as part of the new Obamacare plan.

However, if we continue to move at the rate this “meaningful use” plan intends for us to show performance, we will never reach the point where we evaluate our systems and document their fullest extent of use.  By relying upon a few dozen elements to define our system, we are focusing only upon the grade we receive for our work, and not the meaningfulness it is mean to provide its most important beneficiaries–the patient population.

I have identified about a dozen to dozen or so “real uses” for EMR data.  These dozen or so “uses” include meaningful use as on of its components.  It also includes a number of evaluations that are obviously meaningful, which many of us performed in the past, but never see the light of in today’s desire to meet mostly the requirements.  It also includes some very unique but essential uses of the EMR data.

One of these socially important uses is the identification of local congenital, development and pregnancy-birth related problems that might strike a given area.  This is one of the more important uses for an EMR, with the help of applying spatial analysis techniques to the data and its processing steps.

The following is a listing of congenital diseases I like to run regularly, derived from ICD9 data (those interested in ICD10 will have to translate).  These are the conditions that are usually rare, or scarce, and which we pay heavily for due to the lack of much effort to monitor and attempt to better understand why these events are happening.  Until we know the cause for  these problems, we are left treating them, their patients, in a reactive manner instead of a preventive proactive manner.

The following list can be cut and past into a surveillance system using ICD9 data for the identification of diagnoses and diseases.

WHEN ICD IN (‘271.0′,’271.1′,’271.2′,’271.3′,’271.4′,’271.8′,’271.9′,’272.0′,’272.1′,’272.2′,’272.3’,
‘275.40’,’275.41′,’275.42′,’275.49′) THEN ‘Endocrine’

WHEN ICD IN (‘277.0′,’277.1′,’277.2′,’277.3′,’277.9′,’277.1′,’277.2′,’277.30′,’277.31′,’277.39′,’277.4’,
‘277.5’,’277.6′,’277.7′,’277.81′,’277.82′,’277.83′,’277.84′,’277.85′,’277.86′,’277.87′,’277.88′,’277.89′) THEN ‘Metabolic’

WHEN ICD IN (‘279.4′,’279.9′,’279.10′,’279.11′,’279.12′,’279.13′,’279.19′,’279.2′,’279.3′,’279.41′,’279.49’) THEN ‘Immunological’

WHEN ICD IN (‘282.0′,’282.1′,’282.2′,’282.3′,’282.40′,’282.41′,’282.42′,’282.43′,’282.44′,’282.45′,’282.46′,’282.47′,’282.49’,
‘282.5’,’282.60′,’282.61′,’282.62′,’282.63′,’282.64′,’282.68′,’282.69′,’282.7′,’282.8′,’282.9′) THEN ‘Hematic’

WHEN ICD IN (‘286.0′,’286.1′,’286.2′,’286.3′,’286.4′,’286.52′,’286.53′,’286.59′,’286.6’) THEN ‘Hematic’

WHEN ICD IN (‘287.33′,’288.1′,’288.1′,’288.2’) THEN ‘Hematic’

WHEN ICD IN (‘330.0′,’330.1′,’330.2′,’330.3′,’330.8′,’330.9′,’331.0′,’331.11′,’331.19′,’331.2′,’331.3′,’331.4′,’331.5′,’331.6′,’331.7′,’331.81′,’331.82′,’331.83′,’331.89′,’331.9′,’332.0′,’332.1′,’333.0′,’333.1′,’333.2′,’333.3′,’333.4′,’333.5′,’333.6′,’333.71′,’333.72′,’333.79′,’333.81′,’333.82′,’333.83’) THEN ‘Neuro-developmental’

WHEN ICD IN (‘334.0′,’334.1′,’334.2′,’334.3′,’334.4′,’334.8′,’334.9′,’335.0′,’335.10′,’335.11′,’335.19′,’335.20′,’335.21′,’335.22′,’335.23′,’335.24′,’335.29′,’335.8′,’335.9′,’336.0′,’336.1’) THEN

WHEN ICD IN (‘340.’,’341.0′,’341.1′,’341.8′,’341.9′,’342.0′,’342.1′,’342.2′,’342.10′,’342.11′,’342.12′,’342.80′,’342.81′,’342.82′,’342.90′,’342.91′,’342.92′,’343.0′,’343.1′,’343.2′,’343.3′,’343.4′,’343.8′,’343.9′) THEN ‘Neurological’

WHEN ICD IN (‘356.0′,’356.1′,’356.2′,’356.3′,’356.4′,’356.8′,’356.9’) THEN ‘Neurological’

WHEN ICD IN (‘359.0′,’359.1′,’359.21′,’359.22′,’359.23’) THEN ‘Neurological’

WHEN ICD IN (‘362.70′,’362.71′,’362.72′,’362.73′,’362.74′,’362.75′,’362.76′,’362.77’) THEN ‘Ocular’

WHEN ICD IN (‘371.50′,’371.51′,’371.52′,’371.53′,’371.54′,’371.55′,’371.56′,’371.57′,’371.58’) THEN ‘Ocular’

WHEN ICD IN (‘377.16′,’377.17’) THEN ‘Ocular’

WHEN ICD IN (‘378.0′,’378.1’) THEN ‘Ocular’

WHEN ICD IN (‘378.30′,’378.31′,’378.32′,’378.33′,’378.34′,’378.35′,’378.50’) THEN ‘Optic’

WHEN ICD IN (‘425.11′,’425.2’) THEN ‘Cardiac’

WHEN ICD IN (‘429.83’) THEN ‘Cardiac’

WHEN ICD IN (‘437.5’) THEN ‘Cardiac’

WHEN ICD IN (‘446.7’) THEN ‘Cardiac’

WHEN ICD IN (‘516.61′,’516.62′,’516.63’) THEN ‘Pulmonary’

WHEN ICD IN (‘520.0′,’520.1′,’520.2′,’520.3′,’520.4′,’520.5′,’520.6′,’520.7′,’520.8’) THEN ‘Dental’

WHEN ICD IN (‘524.0′,’524.1′,’524.2′,’524.3′,’524.4′,’524.5′,’524.6′,’524.7′,’524.9′,’524.10′,’524.11’,
‘524.73’,’524.74′,’524.75′,’524.76′,’524.79′,’524.81′,’524.82′,’524.89′) THEN ‘Dental’

ELSE ‘Non-Congenital’

The Formula’s Secret

Quite recently, I began posting my formulas on another site.  These are the “magical” formulas I used to develop the maps that I had for national disease patterns, evaluating population pyramids at the one year age range level for statistically significant differences, using plain old SAS, not even SAS-GIS, to produce 3D images of your research areas, and designing health and disease monitoring and prediction models at the managed care level.

One of the funniest things about human behavior is how much age influences what you’ll do in your field.  Older participants in the field tend to have that stereotyping thrown at them claiming they get behind in the technology and fall behind in learning about anything new.  Not sure if that’s an accurate statement or not.

What I do see however is that the best followers of novelty and innovation are people of the college student age, probably in college.  This means that if and when they learn the new technology I throw at them in terms of statistical formulas, that these formulas cannot be validated or approved by their overseers and upper level managers. Such “old timers” are not in tune with the new ideas, new measures, new ways of reviewing numbers.  More importantly, as time passes once you’ve completed your training in numbers related work, you lose sense of the philosophy, knowledge and understanding that are needed to know why I argue that one theory from one set of processes is easily transferable to another due to underlying parallels in the logic, methodology and equations used to produce the final numbers, p-values or not.

The statistics world is undergoing hybridization of sorts.  Qualitative methods have tended to evolve from quantitative ideologies.  Fro the T-test came the U-test for example, and from the ANOVAs came the non-parametric equivalents such as Kruskal-Wallis and Friedman.  The 2×2 and 2×3 Chi square work best for numbers between 50 and 2500, or did we forget that, and try to apply it to big data, with numbers amounting to tens of thousands on up?

Manager, leaders who forget the theory, and abide by the book, and what they learned eons ago, limit progress for their companies.  This is why students are always smarter in the long run than managers for the most part.  And why managers refuse to accept the fact that they don’t really know a lot usually, unless they themselves just graduates of invented something completely new to work with.

There is this idea that developing innovation is important to the new corporate setting.  But how varied the definitions of innovations can be, makes many of them just as ridiculous to claim as all the “miracles” we hear about occurring daily.  Is that a daily miracle, just experienced by somebody new, or a true miracle that cannot be predicted or explained.  Is that method you developed as a statistician a “new” method that is unheard of, in the literature, or in the field, or by you closest colleagues or international knowledge base equivalent?

Innovations are 90% all on paper–like the ones professed to by nearly all insurance companies out there trying to analyze health.  I have yet to speak with a company with a formula I haven’t yet cracked the code for.  Some of these discoveries are fourth generation, new to Company D, because it stole it from Company C and modified, not knowing that Company C had pulled it years before from Company B, which is a split off of Company A, the original developer of this “black box” routinely used to estimate people’s health risk scores.  Yes, companies like insurance companies numbers 2, 3 and 4, “Optimum health” claimers, PBMs, corporate financial overseers, insurance company #1, have all made such claims of “discovering” the same algorithm.  A claim seen in each of their advertisements for work positions, on screen, in writing, and on the video about the company.  These “innovations” are innovations because the produce believes them to be, thereby revealing the limited knowledge base these developers, producers and managers are already working with.

Some formulas get recognition for their success.  Those that no one knew about, neither have the experience or the past of relying upon them.  And that’s where the most resistance comes from.  That is a good sign your formulas is clean and knew.  If it makes them uncomfortable, and unwilling to learn or listen, your innovation is taking the right path.

Therefore, a formula’s secret to success is developing it undercover, without support or feedback from too many peers, in a way that puts it a generation or two ahead of their work.  This way, when you share that first level with them, and the bickering begins, you already know you are right, because you are two more generations of use and productivity ahead of what their trying to understand.  The worse the management, the more behind companies can be.  This is why.

Research Bibliography

The following are some resources to check out for developing a full fledged population health monitoring program.  They are under review for application to the surveillance program being developed.


Part 1.  Techniques – Machine Learning


Machine learning is using your Big Data/EMR and software tools to develop routine methods for analyzing results.  Most recently, I produced a method for evaluating the top 20 ICDs (by groups I defined) for each of the ethnic groups in the EMR.  These lists were combined in a final table, with ethnics group listed side by side, in descending order for each, for the ‘hot diagnoses’, followed by rank, n and percent.  That SAS took about 1000 lines of programming and a minimum of 53 processes (according to the SAS info that is displayed during a run).

Machine Learning uses two methods–a supervised classification process and an unsupervised classification process.  Supervised classification is where you the researcher manually define the different groupings. For example, my ICD lists and sets are defined based upon personal impressions of ICDs that need to stand out more than they do with the predefined ICD groups defined at CMS.  My first group of ICDs is 135, my second 303.  These groups are to some degree subjective, and based upon clinical observations regarding priority healthcare issues.

Unsupervised classification is where the SAS itself analyzes the data and determines where clusters appear, with the clinical variables (parametric and non-parametric) used to define these clusters.  Although these are fairly easy to produce, they are not always logical and may link one outlier ICD in Group A to the cluster in group B, making the outcomes generated questionable.  There are hundreds of classifications to be tested in terms of ICD and ICD comorbidity relationships.  The following are few examples of these.

Most of these are available in full text form on the internet (sorry, no downloadable pdf copies for the moment, due to copyright concerns.)


Afzal, Z., Engelkes, M., Verhamme, K., Janssens, H. M., Sturkenboom, M. C., Kors, J. A., & Schuemie, M. J. (2013). Automatic generation of case‐detection algorithms to identify children with asthma from large electronic health record databases. Pharmacoepidemiology and Drug Safety, 22(8), 826-833. doi:10.1002/pds.3438

Afzal, Z., Schuemie, M. J., van Blijderveen, J. C., Sen, E. F., Sturkenboom, M. C., & Kors, J. A. (2013). Improving sensitivity of machine learning methods for automated case identification from free-text electronic medical records. BMC medical informatics and decision making, 13(1), 30. doi:10.1186/1472-6947-13-30

Boland, M. R., Tatonetti, N. P., & Hripcsak, G. (2014). CAESAR: a Classification Approach for Extracting Severity Au-tomatically from Electronic Health Records.

Boxwala, A. A., Kim, J., Grillo, J. M., & Ohno-Machado, L. (2011). Using statistical and machine learning to help institutions detect suspicious access to electronic health records. Journal of the American Medical Informatics Association, 18(4), 498-505. doi:10.1136/amiajnl-2011-000217

Caballero Barajas, K. L., & Akella, R. (2015, August). Dynamically Modeling Patient’s Health State from Electronic Medical Records: A Time Series Approach. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 69-78). ACM. doi:10.1145/2783258.2783289

Dua, S., Acharya, U. R., & Dua, P. (2014). Machine learning in healthcare informatics. Springer Berlin Heidelberg.  [Fuzzy logic, supervised and unsupervised classifications, rule learning, black box, predictions, longitudinal data, fraud, imagery.]

FitzHenry, F., Murff, H. J., Matheny, M. E., Gentry, N., Fielstein, E. M., Brown, S. H., … & Speroff, T. (2013). Exploring the Frontier of Electronic Health Record Surveillance: The Case of Post-Operative Complications. Medical Care, 51(6), 509. doi:10.1097/MLR.0b013e31828d1210

Gupta, S., Tran, T., Luo, W., Phung, D., Kennedy, R. L., Broad, A., … & Matheson, L. (2014). Machine-learning prediction of cancer survival: a retrospective study using electronic administrative records and a cancer registry. BMJ open, 4(3), e004007. doi:10.1136/bmjopen-2013-004007

Hoogendoorn, M., Moons, L. M., Numans, M. E., & Sips, R. J. (2014). Utilizing Data Mining for Predictive Modeling of Colorectal Cancer Using Electronic Medical Records. In Dominik Ślȩzak, Ah-Hwee Tan, James F. Peters, Lars Schwabe (Eds.), Brain Informatics and Health (pp. 132-141). Springer International Publishing. doi:10.1007/978-3-319-09891-3_13

Jensen, P. B., Jensen, L. J., & Brunak, S. (2012). Mining electronic health records: towards better research applications and clinical care. Nature Reviews Genetics, 13(6), 395-405. doi:10.1038/nrg3208

Liu, H., Bielinski, S. J., Sohn, S., Murphy, S., Wagholikar, K. B., Jonnalagadda, S. R., … Chute, C. G. (2013). An Information Extraction Framework for Cohort Identification Using Electronic Health Records . AMIA Summits on Translational Science Proceedings, 2013, 149–153.

Mo, H., Thompson, W. K., Rasmussen, L. V., Pacheco, J. A., Jiang, G., Kiefer, R., … & Lingren, T. (2015). Desiderata for computable representations of electronic health records-driven phenotype algorithms. Journal of the American Medical Informatics Association, 22(6), 1220-1230. doi:10.1093/jamia/ocv112

Kononenko, I. (2001). Machine learning for medical diagnosis: history, state of the art and perspective. Artificial Intelligence in Medicine, 23(1), 89-109.

Koutsojannis, C., Nabil, E., Tsimara, M., & Hatzilygeroudis, I. (2009, November). Using machine learning techniques to improve the behaviour of a medical decision support system for prostate diseases. In Intelligent Systems Design and Applications, 2009. ISDA’09. Ninth International Conference on (pp. 341-346). IEEE. 10.1109/ISDA.2009.110

Kreuzthaler, M., Schulz, S., & Berghold, A. (2015). Secondary use of electronic health records for building cohort studies through top-down information extraction. Journal of biomedical informatics, 53, 188-195. doi:10.1016/j.jbi.2014.10.010 [COHORTS]

Lin C, Karlson EW, Canhao H, Miller TA, Dligach D, Chen PJ, et al. (2013) Automatic Prediction of Rheumatoid Arthritis Disease Activity from the Electronic Medical Records. PLoS ONE 8(8): e69932. doi:10.1371/journal.pone.0069932

Martin-Sanchez, F., Iakovidis, I., Nørager, S., Maojo, V., de Groen, P., Van der Lei, J., … & Baud, R. (2004). Synergy between medical informatics and bioinformatics: facilitating genomic medicine for future health care. Journal of biomedical informatics, 37(1), 30-42.

Meystre, S. M., Friedlin, F. J., South, B. R., Shen, S., & Samore, M. H. (2010). Automatic de-identification of textual documents in the electronic health record: a review of recent research. BMC medical research methodology, 10(1), 70. doi:10.1186/1471-2288-10-70

Patel, V. L., Shortliffe, E. H., Stefanelli, M., Szolovits, P., Berthold, M. R., Bellazzi, R., & Abu-Hanna, A. (2009). The coming of age of artificial intelligence in medicine. Artificial intelligence in medicine, 46(1), 5-17. doi:10.1016/j.artmed.2008.07.017

Pathak, J., Bailey, K. R., Beebe, C. E., Bethard, S., Carrell, D. S., Chen, P. J., … & Huff, S. M. (2013). Normalization and standardization of electronic health records for high-throughput phenotyping: the SHARPn consortium. Journal of the American Medical Informatics Association, 20(e2), e341-e348. doi:10.1136/amiajnl-2013-001939 [MU]

Pineda, A. L., Ye, Y., Visweswaran, S., Cooper, G. F., Wagner, M. M., & Tsui, F. R. (2015). Comparison of machine learning classifiers for influenza detection from emergency department free-text reports. Journal of Biomedical Informatics, 58, 60-69.. doi:10.1016/j.jbi.2015.08.019

Prather, J. C., Lobach, D. F., Goodwin, L. K., Hales, J. W., Hage, M. L., & Hammond, W. E. (1996, December). Medical data mining: knowledge discovery in a clinical data warehouse. In Proceedings: a conference of the American Medical Informatics Association/… AMIA Annual Fall Symposium. AMIA Fall Symposium (pp. 101-105).

Sada, Y., Hou, J., Richardson, P., El-Serag, H., & Davila, J. (2013). Validation of case finding algorithms for hepatocellular cancer from administrative data and electronic health records using natural language processing. Medical care, 54(2), e9–e14. doi:10.1097/MLR.0b013e3182a30373

Skeppstedt, M., Kvist, M., Nilsson, G. H., & Dalianis, H. (2014). Automatic recognition of disorders, findings, pharmaceuticals and body structures from clinical text: An annotation and machine learning study. Journal of Biomedical Informatics, 49, 148-158. doi:10.1016/j.jbi.2014.01.012

Szarvas, G., Farkas, R., & Busa-Fekete, R. (2007). State-of-the-art anonymization of medical records using an iterative machine learning framework. Journal of the American Medical Informatics Association, 14(5), 574-580. doi:10.1197/j.jamia.M2441

Wang, Z., Shah, A. D., Tate, A. R., Denaxas, S., Shawe-Taylor, J., & Hemingway, H. (2012). Extracting diagnoses and investigation results from unstructured text in electronic health records by semi-supervised machine learning. PLoS One, 7(1), e30412. doi:10.1371/journal.pone.0030412

Wiens, J., Campbell, W. N., Franklin, E. S., Guttag, J. V., & Horvitz, E. (2014, September). Learning Data-Driven Patient Risk Stratification Models for Clostridium difficile. In Open Forum Infectious Diseases (Vol. 1, No. 2, p. ofu045). Oxford University Press. doi: 10.1093/ofid/ofu045

Weiss, J. C., Natarajan, S., Peissig, P. L., McCarty, C. A., & Page, D. (2012). Machine learning for personalized medicine: predicting primary myocardial infarction from electronic health records. AI Magazine, 33(4), 33. doi:10.1609/aimag.v33i4.2438

Wolfson, J., Bandyopadhyay, S., Elidrisi, M., Vazquez-Benitez, G., Musgrove, D., Adomavicius, G., … & O’Connor, P. (2013). A Naive Bayes machine learning approach to risk prediction using censored, time-to-event electronic health record data. [Draft of presentation/publication; not completed.]

Wu, J., Roy, J., & Stewart, W. F. (2010). Prediction modeling using EHR data: challenges, strategies, and a comparison of machine learning approaches. Medical care, 48(6), S106-S113. doi:10.1097/MLR.0b013e3181de9e17


Observational Studies=Data Mining


I most often incorporate GIS into my work by using the raw data provided by EMR, reclassifiying it as need be, and adding longitude-latitude data whenever possible. This use of GIS may be considered an extension of increasing popular “Observations Studies” term and techniques now found in the literature.  A GIS study of the raw or freshly mined and slightly modified data may also be labelled an “ecological study.”

Grimes, D. A., & Schulz, K. F. (2002). Bias and causal associations in observational research. The Lancet, 359(9302), 248-252. doi:10.1016/S0140-6736(02)07451-2

Hansen, R. A., Gray, M. D., Fox, B. I., Hollingsworth, J. C., Gao, J., & Zeng, P. (2013). How well do various health outcome definitions identify appropriate cases in observational studies? Drug Safety, 36(1), 27-32. doi:10.1007/s40264-013-0104-0

Madigan, D., Stang, P. E., Berlin, J. A., Schuemie, M., Overhage, J. M., Suchard, M. A., … & Ryan, P. B. (2014). A systematic statistical approach to evaluating evidence from observational studies. Annual Review of Statistics and Its Application, 1, 11-39. doi:10.1146/annurev-statistics-022513-115645

Nagisetty, N., Huang, E. Y., Wade, G., & Viangteeravat, T. (2014). Building a knowledge base to assist clinical decision-making using the Pediatric Research Database (PRD) and machine learning: a case study on pediatric asthma patients. BMC Bioinformatics, 15(Suppl 10), P17. doi:10.1186/1471-2105-15-S1-S10

Roche, J. J. W., Wenn, R. T., Sahota, O., & Moran, C. G. (2005). Effect of comorbidities and postoperative complications on mortality after hip fracture in elderly people: prospective observational cohort study. BMJ, 331(7529), 1374. doi:10.1136/bmj.38643.663843.55

Schuemie, M. J., Ryan, P. B., DuMouchel, W., Suchard, M. A., & Madigan, D. (2014). Interpreting observational studies: why empirical calibration is needed to correct p‐values. Statistics in medicine, 33(2), 209-218. doi:10.1002/sim.5925

Shiomi, H., Nakagawa, Y., Morimoto, T., Furukawa, Y., Nakano, A., Shirai, S., … & Mitsuoka, H. (2012). Association of onset to balloon and door to balloon time with long term clinical outcome in patients with ST elevation acute myocardial infarction having primary percutaneous coronary intervention: observational study. BMJ, 344, e3257. doi: 10.1136/bmj.e3257

Tannen, R. L., Weiner, M. G., & Xie, D. (2009). Use of primary care electronic medical record database in drug efficacy research on cardiovascular outcomes: comparison of database and randomised controlled trial findings. BMJ, 338. doi:10.1136/bmj.b81

Twisk, J. W. (1997). Different statistical models to analyze epidemiological observational longitudinal data: an example from the Amsterdam Growth and Health Study. International Journal of Sports Medicine, 18, S216-24.

Yost, N. P., Bloom, S. L., McIntire, D. D., & Leveno, K. J. (2005). A prospective observational study of domestic violence during pregnancy. Obstetrics & gynecology, 106(1), 61-65. doi:10.1097/01.AOG.0000164468.06070.2a


Comparative Effectiveness Research

CER is when treatment programs for several programs or facilities are contrasted and compared statistically.  This refers to database settings where the data source is several places, and in order to retain HIPAA compliance, the data is cleaned of the personal identifiers and other data, as specified by some program and/or HIPAA guidelines.  These guidelines are followed as much as possible by the researchers, but realize, full compliance is difficult when the restricted data is essential to the study process itself, such as 5digit zip code identification and even street and house number data.  CER involves institutions cross-comparing their healthcare results and performance.  These measures are often implemented as part of the meaningful use program as well.


Hersh, W. R., Weiner, M. G., Embi, P. J., Logan, J. R., Payne, P. R., Bernstam, E. V., … & Saltz, J. H. (2013). Caveats for the use of operational electronic health record data in comparative effectiveness research. Medical Care, 51(8 Suppl 3), S30-7. doi:10.1097/MLR.0b013e31829b1dbd

Holve, E., Segal, C., Lopez, M. H., Rein, A., & Johnson, B. H. (2012). The Electronic Data Methods (EDM) forum for comparative effectiveness research (CER). Medical care, 50, S7-S10. doi:10.1097/MLR.0b013e318257a66b

Kudyakov, R., Bowen, J., Ewen, E., West, S. L., Daoud, Y., Fleming, N., & Masica, A. (2012). Electronic health record use to classify patients with newly diagnosed versus preexisting type 2 diabetes: infrastructure for comparative effectiveness research and population health management. Population Health Management, 15(1), 3-11. doi:10.1089/pop.2010.0084.

Lopez, M. H., Holve, E., Sarkar, I. N., & Segal, C. (2012). Building the informatics infrastructure for comparative effectiveness research (CER): a review of the literature. Medical Care, 50, S38-S48. doi: 10.1097/MLR.0b013e318259becd

Masica, M. D., & Collinsworth, M. P. H. (2012). Leveraging Electronic Health Records in Comparative Effectiveness Research. Prescriptions for Excellence in Health Care Newsletter Supplement, 1(14), 6.

Ogunyemi, O. I., Meeker, D., Kim, H. E., Ashish, N., Farzaneh, S., & Boxwala, A. (2013). Identifying appropriate reference data models for comparative effectiveness research (CER) studies based on data from clinical information systems. Medical Care, 51, S45-S52. doi:10.1097/MLR.0b013e31829b1e0b

Toh, S., Platt, R., Steiner, J. F., & Brown, J. S. (2011). Comparative‐Effectiveness Research in Distributed Health Data Networks. Clinical Pharmacology & Therapeutics, 90(6), 883-887. doi:10.1038/clpt.2011.236

Toh, S., & Platt, R. (2013). Is size the next big thing in epidemiology?. Epidemiology, 24(3), 349-351. doi:10.1097/EDE.0b013e31828ac65e

Data Sharing (iDASH, a HIPAA certified cloud)

Ohno-Machado, L., Bafna, V., Boxwala, A. A., Chapman, B. E., Chapman, W. W., Chaudhuri, K., … & Kim, H. (2012). iDASH: integrating data for analysis, anonymization, and sharing. Journal of the American Medical Informatics Association, 19(2), 196-201. 10.1136/amiajnl-2011-000538

Reisinger, S. J., Ryan, P. B., O’Hara, D. J., Powell, G. E., Painter, J. L., Pattishall, E. N., & Morris, J. A. (2010). Development and evaluation of a common data model enabling active drug safety surveillance using disparate healthcare databases. Journal of the American Medical Informatics Association, 17(6), 652-662. doi:10.1136/jamia.2009.002477

Data Quality Assessment Model

Kahn, M. G., Raebel, M. A., Glanz, J. M., Riedlinger, K., & Steiner, J. F. (2012). A pragmatic framework for single-site and multisite data quality assessment in electronic health record-based clinical research. Medical care, 50. doi:10.1097/MLR.0b013e318257dd67   Accessed at

Brown, J., Kahn, M., & Toh, S. (2013). Data quality assessment for comparative effectiveness research in distributed data networks. Medical care, 51(8 0 3), S22. doi:10.1097/MLR.0b013e31829b1e2c

Dreyer, N. A., Schneeweiss, S., McNeil, B. J., Berger, M. L., Walker, A. M., Ollendorf, D. A., & Gliklich, R. E. (2010). GRACE principles: recognizing high-quality observational studies of comparative effectiveness. The American Journal of Managed Care, 16(6), 467-471.

Data Accuracy

Cipparone, C. W., Withiam-Leitch, M., Kimminau, K. S., Fox, C. H., Singh, R., & Kahn, L. (2015). Inaccuracy of ICD-9 Codes for Chronic Kidney Disease: A Study from Two Practice-based Research Networks (PBRNs). The Journal of the American Board of Family Medicine, 28(5), 678-682. doi:10.3122/jabfm.2015.05.140136


Malin, B. A., El Emam, K., & O’Keefe, C. M. (2013). Biomedical data privacy: problems, perspectives, and recent advances. Journal of the American medical informatics association, 20(1), 2-6. doi:10.1136/amiajnl-2012-001509

Tran, D. T., Halgrim, S., & Carrell, D. (2014). C3-4: An Algorithm to Combine Machine Learning and Structured Data to Automate De-identification of Clinical Text. Clinical Medicine & Research, 12(1-2), 94-95. doi:10.3121/cmr.2014.1250.c3-4



Part 2 – Applications, Methods and Skills

These are examples of how to employ population health analysis procedures.   


Alghwiri, A., Alghadir, A., & Awad, H. (2014). The Arab Risk (ARABRISK): Translation and Validation. Biomedical Research, 25(2), 271-275.

Carroll, R. J., Thompson, W. K., Eyler, A. E., Mandelin, A. M., Cai, T., Zink, R. M., … & Karlson, E. W. (2012). Portability of an algorithm to identify rheumatoid arthritis in electronic health records. Journal of the American Medical Informatics Association, 19(e1), e162-e169. doi:10.1136/amiajnl-2011-000583

Holroyd-Leduc, J. M., Lorenzetti, D., Straus, S. E., Sykes, L., & Quan, H. (2011). The impact of the electronic medical record on structure, process, and outcomes within primary care: a systematic review of the evidence. Journal of the American Medical Informatics Association, 18(6), 732-737. doi:10.1136/amiajnl-2010-000019

Lin, Y. K., Chen, H., Brown, R., Li, S. H., & Yang, H. J. (2014). Time-to-Event Predictive Modeling for Chronic Conditions using Electronic Health Records. Intelligent Systems, IEEE, 29(3), 14-20. doi:10.1109/MIS.2014.18

Sovio, U., Skow, A., Falconer, C., Park, M. H., Viner, R. M., & Kinra, S. (2013). Improving prediction algorithms for cardiometabolic risk in children and adolescents. Journal of obesity, 2013. doi:10.1155/2013/684782


Anderson, A. E., Kerr, W. T., Thames, A., Li, T., Xiao, J., & Cohen, M. S. (2015). Electronic health record phenotyping improves detection and screening of type 2 diabetes in the general United States population: A cross-sectional, unselected, retrospective study. arXiv preprint arXiv:1501.02402.

Boland, M. R., Tatonetti, N. P., & Hripcsak, G. (2015). Development and validation of a classification approach for extracting severity automatically from electronic health records. Journal of Biomedical Semantics, 6(1), 14.  doi:10.1186/s13326-015-0010-8

Carroll, R. J., Eyler, A. E., & Denny, J. C. (2011). Naïve electronic health record phenotype identification for rheumatoid arthritis. In AMIA annual symposium proceedings (Vol. 2011, p. 189). American Medical Informatics Association.

Chen, Y., Carroll, R. J., Hinz, E. R. M., Shah, A., Eyler, A. E., Denny, J. C., & Xu, H. (2013). Applying active learning to high-throughput phenotyping algorithms for electronic health records data. Journal of the American Medical Informatics Association, 20(e2), e253-e259. doi:10.1136/amiajnl-2013-001945

Pecci, A., Klersy, C., Gresele, P., Lee, K. J., De Rocco, D., Bozzi, V., … & Fabris, F. (2014). MYH9‐Related Disease: A Novel Prognostic Model to Predict the Clinical Evolution of the Disease Based on Genotype–Phenotype Correlations. Human Mutation, 35(2), 236-247. doi:10.1002/humu.22476

Hripcsak, G., & Albers, D. J. (2013). Next-generation phenotyping of electronic health records. Journal of the American Medical Informatics Association, 20(1), 117-121. doi:10.1136/amiajnl-2012-001145


Peissig, P. L., Costa, V. S., Caldwell, M. D., Rottscheit, C., Berg, R. L., Mendonca, E. A., & Page, D. (2014). Relational machine learning for electronic health record-driven phenotyping. Journal of biomedical informatics, 52, 260-270. doi:10.1016/j.jbi.2014.07.007

Rasmussen, L. V., Thompson, W. K., Pacheco, J. A., Kho, A. N., Carrell, D. S., Pathak, J., … & Starren, J. B. (2014). Design patterns for the development of electronic health record-driven phenotype extraction algorithms. Journal of Biomedical Informatics, 51, 280-286. doi:10.1016/j.jbi.2014.06.007

Shivade, C., Raghavan, P., Fosler-Lussier, E., Embi, P. J., Elhadad, N., Johnson, S. B., & Lai, A. M. (2014). A review of approaches to identifying patient phenotype cohorts using electronic health records. Journal of the American Medical Informatics Association, 21(2), 221-230. doi:10.1136/amiajnl-2013-001935

Wei, W. Q., Teixeira, P. L., Mo, H., Cronin, R. M., Warner, J. L., & Denny, J. C. (2015). Combining billing codes, clinical notes, and medications from electronic health records provides superior phenotyping performance. Journal of the American Medical Informatics Association, ocv130.



Chai, K. E., Anthony, S., Coiera, E., & Magrabi, F. (2013). Using statistical text classification to identify health information technology incidents. Journal of the American Medical Informatics Association, 20(5), 980-985. doi:10.1136/amiajnl-2012-001409

Dai, W., Brisimi, T. S., Adams, W. G., Mela, T., Saligrama, V., & Paschalidis, I. C. (2015). Prediction of hospitalization due to heart diseases by supervised learning methods. International journal of medical informatics, 84(3), 189-197. doi:10.1016/j.ijmedinf.2014.10.002

Pak, T. R., & Kasarskis, A. (2015). How next-generation sequencing and multiscale data analysis will transform infectious disease management. Clinical Infectious Diseases, 61(11), 1695-1702.  doi: 10.1093/cid/civ670

Ye, Y., Tsui, F., Wagner, M., Espino, J. U., & Li, Q. (2014). Influenza detection from emergency department reports using natural language processing and Bayesian network classifiers. Journal of the American Medical Informatics Association, 21(5), 815-823. doi:10.1136/amiajnl-2013-001934

QOL & Dx post-Tx

Penson, D. F., Feng, Z., Kuniyuki, A., McClerran, D., Albertsen, P. C., Deapen, D., … & Stanford, J. L. (2003). General quality of life 2 years following treatment for prostate cancer: what influences outcomes? Results from the prostate cancer outcomes study. Journal of Clinical Oncology, 21(6), 1147-1154. doi:10.1200/JCO.2003.07.139


Newhouse, J. P., & McClellan, M. (1998). Econometrics in outcomes research: the use of instrumental variables. Annual Review of Public Health, 19(1), 17-34. doi:10.1146/annurev.publhealth.19.1.17

Comorbidity Scores

Austin, S.R., Wong, Y.N., Uzzo, R.G., Beck, J.R., Egleston, B.L. (2015). Why Summary Comorbidity Measures Such As the Charlson Comorbidity Index and Elixhauser Score Work.  Medical Care, 53(9), e65-72. doi:10.1097/MLR.0b013e318297429c.  Accessed at

Bang, J. H., Hwang, S.-H., Lee, E.-J., & Kim, Y. (2013). The predictability of claim-data-based comorbidity-adjusted models could be improved by using medication data. BMC Medical Informatics and Decision Making, 13, 128.

Chu, Y.-T., Ng, Y.-Y., & Wu, S.-C. (2010). Comparison of different comorbidity measures for use with administrative data in predicting short- and long-term mortality. BMC Health Services Research, 10, 140.

Gutacker, N, Bloor, K, Cookson, R. (2015). Comparing the performance of the Charlson/Deyo and Elixhauser comorbidity measures across five European countries and three conditions.  European Journal of Public Health. 25 Suppl 1, 15-20. doi:10.1093/eurpub/cku221.

Johnson, A. E., Kramer, A. A., & Clifford, G. D. (2013). A new severity of illness scale using a subset of acute physiology and chronic health evaluation data elements shows comparable predictive accuracy*. Critical Care Medicine, 41(7), 1711-1718. doi: 10.1097/CCM.0b013e31828a24fe  [Oxford Acute Severity of Illness Score ; Particle Swarm Optimization]

Menendez, Mariano E. et al. The Elixhauser Comorbidity Method Outperforms the Charlson Index in Predicting Inpatient Death After Orthopaedic Surgery. Clinical Orthopaedics and Related Research 472.9 (2014): 2878–2886. PMC. Web. 24 Jan. 2016.

Schneeweiss, S., Maclure, M.  (2000).  Use of comorbidity scores for control of confounding in studies using administrative databases. International Journal of Epidemiology, 29(5), 891-8. Accessed at

Stausberg J, Hagn S (2015) New Morbidity and Comorbidity Scores based on the Structure of the ICD-10. PLoS ONE 10(12): e0143365. doi:10.1371/journal.pone.0143365  Accessed at

Yang, M., Mehta, H.B., Bali, V., Gupta, P., Wang, X., Johnson, M.L., Aparasu, R. R.
(2015). Which risk-adjustment index performs better in predicting 30-day mortality? A systematic review and meta-analysis. Journal Evaluation Clinical Practice, 21(2), 292-9. doi: 10.1111/jep.12307. [Includes several speciality disease scores]


Castro, V. M., Clements, C. C., Murphy, S. N., Gainer, V. S., Fava, M., Weilburg, J. B., … & Smoller, J. W. (2013). QT interval and antidepressant use: a cross sectional study of electronic health records. BMJ, 346, f288. doi:10.1136/bmj.f288

Costa, F. F. (2014). Big data in biomedicine. Drug discovery today, 19(4), 433-440. doi:10.1016/j.drudis.2013.10.012

Khoury, M. J., Rich, E. C., Randhawa, G., Teutsch, S. M., & Niederhuber, J. (2009). Comparative effectiveness research and genomic medicine: an evolving partnership for 21st century medicine. Genetics in Medicine, 11(10), 707-711. doi:10.1097/GIM.0b013e3181b99b90

Analytics (other)

Schulam, P., Wigley, F., & Saria, S. (2015, February). Clustering Longitudinal Clinical Marker Trajectories from Electronic Health Data: Applications to Phenotyping and Endotype Discovery. In Twenty-Ninth AAAI Conference on Artificial Intelligence.


Part 3 – Risk Analysis

Predicting Risk for Diabetes

Eggleston, E. M., & Klompas, M. (2014). Rational use of electronic health records for diabetes population management. Current Diabetes Reports, 14(4), 1-10. 10.1007/s11892-014-0479-z

Exalto, L. G., Biessels, G. J., Karter, A. J., Huang, E. S., Katon, W. J., Minkoff, J. R., & Whitmer, R. A. (2013). Risk score for prediction of 10 year dementia risk in individuals with type 2 diabetes: a cohort study. The Lancet Diabetes & Endocrinology, 1(3), 183-190. doi:10.1016/S2213-8587(13)70048-2

Herman, W. H. (2009). Predicting risk for diabetes: choosing (or building) the right model. Annals of Internal Medicine, 150(11), 812-814.

Jin, H., & Benyshek, D. C. (2013). The “metabolic syndrome index”: A novel, comprehensive method for evaluating the efficacy of diabetes prevention programs. doi:10.4236/jdm.2013.32014

Lawrence, J. M., Black, M. H., Zhang, J. L., Slezak, J. M., Takhar, H. S., Koebnick, C., … & Reynolds, K. (2013). Validation of pediatric diabetes case identification approaches for diagnosed cases by using information in the electronic health records of a large integrated managed health care organization. American Journal of Epidemiology, kwt230. doi:10.1093/aje/kwt230

Makam, A. N., Nguyen, O. K., Moore, B., Ma, Y., & Amarasingham, R. (2013). Identifying patients with diabetes and the earliest date of diagnosis in real time: an electronic health record case-finding algorithm. BMC medical informatics and decision making, 13(1), 81. doi:10.1186/1472-6947-13-81

Onitilo, A. A., Stankowski, R. V., Berg, R. L., Engel, J. M., Williams, G. M., & Doi, S. A. (2014). A novel method for studying the temporal relationship between type 2 diabetes mellitus and cancer using the electronic medical record. BMC medical informatics and decision making, 14(1), 38. doi:10.1186/1472-6947-14-38

Reed, M., Huang, J., Brand, R., Graetz, I., Neugebauer, R., Fireman, B., … & Hsu, J. (2013). Implementation of an outpatient electronic health record and emergency department visits, hospitalizations, and office visits among patients with diabetes. JAMA, 310(10), 1060-1065. doi:10.1001/jama.2013.276733.

Riaz, M., Basit, A., Hydrie, M. Z. I., Shaheen, F., Hussain, A., Hakeem, R., & Shera, A. S. (2012). Risk assessment of Pakistani individuals for diabetes (RAPID). Primary care diabetes, 6(4), 297-302. doi:10.1016/j.pcd.2012.04.002

Tankova, T., Chakarova, N., Atanassova, I., & Dakovska, L. (2011). Evaluation of the Finnish Diabetes Risk Score as a screening tool for impaired fasting glucose, impaired glucose tolerance and undetected diabetes. Diabetes Research and Clinical Practice, 92(1), 46-52. doi:10.1016/j.diabres.2010.12.020

Wang, H., Liu, T., Qiu, Q., Karp, E., Ding, P., He, Y. H., & Chen, W. Q. (2015). Development and validation of a simple risk score for prevalent undiagnosed type 2 diabetes in Southern Chinese population. International Journal of Diabetes in Developing Countries, 35(3), 1-9. doi:10.1007/s13410-014-0285-9

Prediction, Risk, in General

Bandyopadhyay, S., Wolfson, J., Vock, D. M., Vazquez-Benitez, G., Adomavicius, G., Elidrisi, M., … & O’Connor, P. J. (2014). Data mining for censored time-to-event data: A Bayesian network model for predicting cardiovascular risk from electronic health record data. Data Mining and Knowledge Discovery, 1-37. doi: 10.1007/s10618-014-0386-6

Eggleston, E. M., & Weitzman, E. R. (2014). Innovative uses of electronic health records and social media for public health surveillance. Current Diabetes Reports, 14(3), 1-9. doi:10.1007/s11892-013-0468-7

Fox, K. A., Dabbous, O. H., Goldberg, R. J., Pieper, K. S., Eagle, K. A., Van de Werf, F., … & Granger, C. B. (2006). Prediction of risk of death and myocardial infarction in the six months after presentation with acute coronary syndrome: prospective multinational observational study (GRACE). BMJ, 333(7578), 1091. doi:10.1136/bmj.38985.646481.55

Goldstein, B. A., Chang, T. I., Mitani, A. A., Assimes, T. L., & Winkelmayer, W. C. (2014). Near-term prediction of sudden cardiac death in older hemodialysis patients using electronic health records. Clinical Journal of the American Society of Nephrology, 9(1), 82-91. doi:10.2215/​CJN.03050313

Gultepe, E., Green, J. P., Nguyen, H., Adams, J., Albertson, T., & Tagkopoulos, I. (2014). From vital signs to clinical outcomes for patients with sepsis: a machine learning basis for a clinical decision support system. Journal of the American Medical Informatics Association, 21(2), 315-325. doi:10.1136/amiajnl-2013-001815

Himes, B. E., Dai, Y., Kohane, I. S., Weiss, S. T., & Ramoni, M. F. (2009). Prediction of chronic obstructive pulmonary disease (COPD) in asthma patients using electronic medical records. Journal of the American Medical Informatics Association, 16(3), 371-379. doi:10.1197/jamia.M2846

Hubbard, R. (2014). Statistical methods for misclassified outcomes and exposures in data from electronic medical records. [Report]. Accessed at

Li, D., Simon, G., Chute, C. G., & Pathak, J. (2013). Using Association Rule Mining for Phenotype Extraction from Electronic Health Records . AMIA Summits on Translational Science Proceedings, 2013, 142–146. [ARM Model building]

Mani, S., Ozdas, A., Aliferis, C., Varol, H. A., Chen, Q., Carnevale, R., … & Weitkamp, J. H. (2014). Medical decision support using machine learning for early detection of late-onset neonatal sepsis. Journal of the American Medical Informatics Association, 21(2), 326-336. doi:10.1136/amiajnl-2013-001854

Melton, L. J., Atkinson, E. J., St Sauver, J. L., Achenbach, S. J., Therneau, T. M., Rocca, W. A., & Amin, S. (2014). Predictors of Excess Mortality After Fracture: A Population‐Based Cohort Study. Journal of Bone and Mineral Research, 29(7), 1681-1690. doi:10.1002/jbmr.2193

Murray, R. E., Ryan, P. B., & Reisinger, S. J. (2011). Design and validation of a data simulation model for longitudinal healthcare data. In AMIA Annual Symposium Proceedings (Vol. 2011, p. 1176). American Medical Informatics Association.

Pearson, J. F., Brownstein, C. A., & Brownstein, J. S. (2011). Potential for electronic health records and online social networking to redefine medical research. Clinical chemistry, 57(2), 196-204. doi:10.1373/clinchem.2010.148668

Ryan, P. B., Schuemie, M. J., Gruber, S., Zorych, I., & Madigan, D. (2013). Empirical performance of a new user cohort method: lessons for developing a risk identification and analysis system. Drug safety, 36(1), 59-72. doi:10.1007/s40264-013-0099-6

Ryan, P. B., Schuemie, M. J. (2013). Evaluating Performance of Risk Identification Methods Through a Large-Scale Simulation of Observational Data. Drug Safety, 36(1), 171-180. doi:10.1007/s40264-013-0110-2

Weiss, J. C., Natarajan, S., Peissig, P. L., McCarty, C. A., & Page, D. (2012, July). Statistical Relational Learning to Predict Primary Myocardial Infarction from Electronic Health Records. In IAAI. Twenty-Fourth IAAI Conference, Toronto, Ontario, Canada, July 22, 2012 – July 26, 2012



Part 4 – Other Applications, Methodology Information



Gobbel, G. T., Reeves, R., Jayaramaraja, S., Giuse, D., Speroff, T., Brown, S. H., … & Matheny, M. E. (2014). Development and evaluation of RapTAT: a machine learning system for concept mapping of phrases from medical narratives. Journal of biomedical informatics, 48, 54-65. doi:10.1016/j.jbi.2013.11.008

Jonnagaddala, J., Dai, H. J., Ray, P., & Liaw, S. T. (2015). A preliminary study on automatic identification of patient smoking status in unstructured electronic health records. ACL-IJCNLP 2015, 147. Accessed at

Poulin, C., Shiner, B., Thompson, P., Vepstas, L., Young-Xu, Y., Goertzel, B., … & McAllister, T. (2014). Predicting the risk of suicide by analyzing the text of clinical notes. PloS one, 9(1). doi:10.1371/journal.pone.0085733

Strauss, J. A., Chao, C. R., Kwan, M. L., Ahmed, S. A., Schottinger, J. E., & Quinn, V. P. (2013). Identifying primary and recurrent cancers using a SAS-based natural language processing algorithm. Journal of the American Medical Informatics Association, 20(2), 349-355. doi:10.1136/amiajnl-2012-000928

Zheng, C., Rashid, N., Wu, Y. L., Koblick, R., Lin, A. T., Levy, G. D., & Cheetham, T. C. (2014). Using natural language processing and machine learning to identify gout flares from electronic clinical notes. Arthritis Care & Research, 66(11), 1740-1748. doi:10.1002/acr.22324


Hripcsak, G., Albers, D. J., & Perotte, A. (2015). Parameterizing time in electronic health record studies. Journal of the American Medical Informatics Association, ocu051. doi:10.1093/jamia/ocu051

Software Other

Mowery, D., Wiebe, J., Ross, M., Vellupillai, S., Mystere, S., Chapman, W. W. Generating Patient Problem Lists from the ShARe Corpus using SNOMED CT/SNOMED CT CORE Problem List  In Proceedings of the 2014 Workshop on Biomedical Natural Language Processing (BioNLP 2014) (pages 54–58). Baltimore, Maryland USA, June 26-27 2014. Accessed at

Ng, K., Ghoting, A., Steinhubl, S. R., Stewart, W. F., Malin, B., & Sun, J. (2014). PARAMO: A PARAllel predictive MOdeling platform for healthcare analytic research using electronic health records. Journal of biomedical informatics, 48, 160-170. doi:10.1016/j.jbi.2013.12.012

Savova, G. K., Masanz, J. J., Ogren, P. V., Zheng, J., Sohn, S., Kipper-Schuler, K. C., & Chute, C. G. (2010). Mayo clinical Text Analysis and Knowledge Extraction System (cTAKES): architecture, component evaluation and applications. Journal of the American Medical Informatics Association, 17(5), 507-513. doi:10.1136/jamia.2009.001560

Software Sentinel

Behrman, R. E., Benner, J. S., Brown, J. S., McClellan, M., Woodcock, J., & Platt, R. (2011). Developing the Sentinel System—a national resource for evidence development. New England Journal of Medicine, 364(6), 498-499. doi:10.1056/NEJMp1014427

Curtis, L. H., Weiner, M. G., Boudreau, D. M., Cooper, W. O., Daniel, G. W., Nair, V. P., … & Brown, J. S. (2012). Design considerations, architecture, and use of the Mini‐Sentinel distributed data system. Pharmacoepidemiology and Drug Safety, 21(S1), 23-31. 10.1002/pds.2336

Madigan, D., & Ryan, P. (2011). Commentary: What Can We Really Learn From Observational Studies?: The Need for Empirical Assessment of Methodology for Active Drug Safety Surveillance and Comparative Effectiveness Research. Epidemiology, 22(5), 629-631. doi:10.1097/EDE.0b013e318228ca1d

Maro, J. C., Platt, R., Holmes, J. H., Strom, B. L., Hennessy, S., Lazarus, R., & Brown, J. S. (2009). Design of a national distributed health data network. Annals of Internal Medicine, 151(5), 341-344. doi:10.7326/0003-4819-151-5-200909010-00139

Overhage, J. M., Ryan, P. B., Reich, C. G., Hartzema, A. G., & Stang, P. E. (2012). Validation of a common data model for active safety surveillance research. Journal of the American Medical Informatics Association, 19(1), 54-60. doi:10.1136/amiajnl-2011-000376 [Values]

Reich, C., Ryan, P. B., Stang, P. E., & Rocca, M. (2012). Evaluation of alternative standardized terminologies for medical conditions within a network of observational healthcare databases. Journal of Biomedical Informatics, 45(4), 689-696. doi:10.1016/j.jbi.2012.05.002

Ryan, P. B., Madigan, D., Stang, P. E., Marc Overhage, J., Racoosin, J. A., & Hartzema, A. G. (2012). Empirical assessment of methods for risk identification in healthcare data: results from the experiments of the Observational Medical Outcomes Partnership. Statistics in Medicine, 31(30), 4401-4415.  doi:10.1002/sim.5620

Stang, P. E., Ryan, P. B., Racoosin, J. A., Overhage, J. M., Hartzema, A. G., Reich, C., … & Woodcock, J. (2010). Advancing the science for active surveillance: rationale and design for the Observational Medical Outcomes Partnership. Annals of internal medicine, 153(9), 600-606. doi:10.7326/0003-4819-153-9-201011020-00010

Stang, P. E., Ryan, P. B., Dusetzina, S. B., Hartzema, A. G., Reich, C., Overhage, J. M., & Racoosin, J. A. (2012). Health outcomes of interest in observational data: issues in identifying definitions in the literature. Health Outcomes Research in Medicine, 3(1), e37-e44. doi:10.1016/j.ehrm.2011.11.003

Coloma, P. M., Trifirò, G., Schuemie, M. J., Gini, R., Herings, R., Hippisley‐Cox, J., … & Lei, J. (2012). Electronic healthcare databases for active drug safety surveillance: is there enough leverage?. Pharmacoepidemiology and Drug Safety, 21(6), 611-621. doi:10.1002/pds.3197


MISC (some nice reads on this topic)

For more on OMOP, see

Brooks, R., & Grotz, C. (2010). Implementation of electronic medical records: How healthcare providers are managing the challenges of going digital. Journal of Business & Economics Research (JBER), 8(6). doi:10.19030/jber.v8i6.736

Hansen, M. M., Miron-Shatz, T., Lau, A. Y. S., & Paton, C. (2014). Big Data in Science and Healthcare: A Review of Recent Literature and Perspectives: Contribution of the IMIA Social Media Working Group. Yearbook of Medical Informatics, 9(1), 21-26. doi:10.15265/IY-2014-0004  [Table provides examples: location use, visualization, assess disease spread, evaluate cause, predict, define social and environmental factors, crisis and disaster management planning, tracking, storing and mining population health data, bring together data from different sources, monitor, cost model]

Herland, M., Khoshgoftaar, T. M., & Wald, R. (2013, December). Survey of Clinical Data Mining Applications on Big Data in Health Informatics. In Machine Learning and Applications (ICMLA), 2013 12th International Conference (Vol. 2, pp. 465-472). IEEE. doi:10.1109/ICMLA.2013.163

Kennedy, E. H., Wiitala, W. L., Hayward, R. A., & Sussman, J. B. (2013). Improved cardiovascular risk prediction using nonparametric regression and electronic health record data. Medical care, 51(3), 251. doi:10.1097/MLR.0b013e31827da594

Kerr, W. T., Lau, E. P., Owens, G. E., & Trefler, A. (2012). The future of medical diagnostics: large digitized databases. The Yale journal of biology and medicine, 85(3), 363.

Kushida, C. A., Nichols, D. A., Jadrnicek, R., Miller, R., Walsh, J. K., & Griffin, K. (2012). Strategies for de-identification and anonymization of electronic health record data for use in multicenter research studies. Medical care, 50, S82-S101. doi:10.1097/MLR.0b013e3182585355


Murdoch, T. B., & Detsky, A. S. (2013). The inevitable application of big data to health care. Jama, 309(13), 1351-1352.

Pathak, J., Kho, A. N., & Denny, J. C. (2013). Electronic health records-driven phenotyping: challenges, recent advances, and perspectives. Journal of the American Medical Informatics Association, 20(e2), e206-e211. doi:10.1136/amiajnl-2013-002428

Savage, N. (2012). Better medicine through machine learning. Communications of the ACM, 55(1), 17-19. doi:10.1145/2063176.2063182

Schneeweiss, S. (2014). Learning from big health care data. New England Journal of Medicine, 370(23), 2161-2163. doi:10.1056/NEJMp1401111

Scruggs, S. B., Watson, K., Su, A. I., Hermjakob, H., Yates, J. R., Lindsey, M. L., & Ping, P. (2015). Harnessing the Heart of Big Data. Circulation Research, 116(7), 1115-1119. doi:10.1161/CIRCRESAHA.115.306013

Sweeney, L. (2002). k-anonymity: A model for protecting privacy. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 10(05), 557-570. doi:10.1142/S0218488502001648

More on the Plan for HIT-GIS

The following are questions and answers regarding the actions taken to develop an HIT-GIS within a managed care system.  This will be either edited or rewritten in more detail as time passes.

Question: What is the Gap?  How do studies performed approach the gap?

Answer: Primary gap is leadership and potentials

  • Lack of management knowledge base and awareness
  • Lack of leadership with specific goals and processes in place proving their stake
  • Lack of preliminary plan, due to lack of understanding of potential and possible end products

Q:  What are the basic rules for constructing a theory for the project?

A: The Rules are

  • Assume that Barriers related to the topic, in turn can be broken down into parts and used to define the hypotheses.
  • Various methods of developing these barriers might exist (need a model, and perhaps a paradigm)
  • Major concepts or variables are to be tested, based upon the model

Q: What is known about these relationships.  Why study the relationship defined for this final PhD project?

A:  Hypotheses define the expectations of researchers

  • The inability of patients to use HIT (versus GIS-HIT), versus practitioners, versus administrators and technicians, versus managers relate to different models—barriers may differ considerably.
  • Highly successful versus smaller institutions can differ, and may in fact experience greater barriers related to employee skills and knowledge base, regional familiarity with the technology, etc.

Q: What about selecting to GIS to use as an option?

A.  The value and reason(s) for GIS:

  • Location/Access improvements; redesigning plans and services
  • Cost savings by redesigning facilities, determining needs
  • Services — quality and adding new services, documenting this
  • Future planning (projection of health and patients and costs; plans/goals)
  • Standardized reporting of valuable QA information; meet ACO requirements
    • Performance Improvement QI scores, documenting and reporting
    • Ad hoc reporting per local needs
  • Increase recognition and Support:
    • Professional reputation
    • Public Support and recognition
    • Press related support and recognition
    • Obtaining other institutional support (npos, clinics, allied health) and recognition
    • Financier support and recognition (improve investments)
    • Allied corporations support
    • Federal or governmental support and recognition

Q: What are the Requirements?

A: The Requirements, Rules and Regulations for implementing a GIS at the Managed Care level needs to address HIPAA possible concerns.  This means, an institution has to have a guidelines book in place for GIS, and/or a Policy defined that is reviewed annually and whenever a major technological change in GIS occurs, that is of interest to the insttitution.  If the institution has a Committee or team directed to technological assessments and decision making regarding use, financing the new technology, etc., they must undergo a regular review of this policy as well.  HIPAA related policy makers or overseers might also need to give a final approval on implementing, updating or modifying the HIT-GIS Policy.

Q:  What steps do we go through to develop this policy?

A:  Policy implementation steps (tentative, as of 10;25/15):

  • Design an institutional GIS-HIT Policy
  • Define metrics levels
  • Define stages and order of development (see metrics list below, choose several projects to initiate with)
  • Define stages and levels of reporting (direct/indirect; next level=clinical/clinical admin; next=upper level admin, public reporting (if any))
  • Define the process of developing an HIT-GIS station; define tool or tools, correlate with projects/spatial analytic steps taken
  • Set Goals for each stage in this process, define dates for those goals.
  • Assign an overall goal of implementing fully functional workstation for certain parts of the HIT-GIS program, testing its performance. [Recommendation: 9 months for Level 1 reporting, 12 months for approval of statement that goal was met.  15 months=produce first report(s) or models for reports.
  • Have a Management Report plan and format in place, in order to administer these reports
  • Define the exact Metrics [34 types, about 1000-2000 metrics, depending upon reports/subgroupings)
  • Initiate use either in sections, or as need requires.
  • Produce chronology or GANTT on this
  • Evaluate rate of development of the HIT-GIS work stations over time.
  • Evaluate rate of documentation of critical events or improvements in HIT-GIS development

Q: What are examples of the metrics we have to consider?

A:  You can design your metrics based upon the quality improvement, meaningful use, projects developed each year for annual reviews.  Add to these some very generic reports pertaining to population age-gender-ethnicity-location information.  Next, define two or more special topics to implement and apply to this program, preferably those reporting on quarterly to biannual rates, such as Emergency line use, case management activities, institutions and patients, smoking cessation calls, complaints lines use and reporting, changes in a standard care performance that could constitute a high risk patient indicator (quarterly blood tests, late visits or skipped appointments).

The following is an initial list of the kinds of metrics to consider:

  • QA – missing data [unk, total ICD? Region, facility?]
  • Standard metrics – population data on ICDs, plus demographics/area [2-16]
  • Standard metrics – IH LOS, OP, LOS [7-10],
  • Standard ICD QOC metrics – Ch Well Visits; CDM Yrly visits; 2MoFU for MH [10-15]
  • Standard Rx Compliance Measures – Refill rates [2-3]
  • Ethnicity measure for overall [2-16]
  • Basic Foreign Language requirement, documentation metric; external use metric? [8,14]
  • Specific Ethnic ICD/place measures (2 levels) – AfrAm or Hisp, other, Asian, AI/NA [5 each, incl W]
  • ACOs [45-60]
  • Other than ACO/QOC-S, simple basic QIA/PIP metrics (defined elsewhere) [15-20]
  • HEDIS Admin metrics focused on facilities [25]
  • Institutional 300+ Disease reporting tool (set for monthly or bimonthly use) [repeats above, 1 report, 300+ metrics]
  • Ethnic-focused ICD reporting tool/report [40-100, x 4-5, x 8 or 11]
  • SES-focused reporting tool, with special topics: poverty groups and areas [ditto]
  • Age-subgroups tools: suicide, epilepsy SUNDS etc. [5-7]
  • V-codes: ch/adult abuse, abandonment, refusal of care for religious reason, immunization refusal [socially meaningful, 12-15]
  • ICD codes: suicide risk, special cultural/region risks, ID risks [age/gender related, 7 x 12 = 94]
  • 65+ yo ICDs report [dupl. above but with different ages, ca. 20-100]
  • Congenital Diseases report [20-45]
  • Genetic Diseases report [1-4, or up to 100+ ICDs]
  • Pregnancy/Child-bearing report (Chlamydia and non-chlamydia, age) [30]
  • Comprehensive Chronic Disease Mgmt report (2-4 levels) [n cd, 4 grps, 8 regions, 11 facilities, 4-5 ethn]
  • Charlson Disease list and scores report (totals, levels 1, 4, 6+) [ditto]
  • Prediction modeling formula/reports (regionally, institutionally, by boroughs; SES-PH, BCBS and OptumHealth “black box” tools) [2 formulas, x 8, 11]]
  • Newborn/Infant Care (3 or 4 years), risk levels (prediction tool) [25-40 icds, x 8,11]
  • Young:Old Childcare (0-8, 9-17) (tic, beh h, ep, pulm, drug, smoking, alc ab, drug ab, suicide, fx, disl, tc.) (prediction tool) 25-40 icds, x 2 age groups, compared x 8, 11]
  • Child Schooling (4-17), (Beh H, Nutr, CD, BMI, Injuries, Fx/Disl, ED use) [1 group, ditto]
  • Mother:Child relationship (families, unmarr. vs married?; 10-45 yo) [1 group, ditto]
  • Childhd:Adult [changing coverage] QOC conversion rates (asthma, CD, MH), 10-30, 12-26, 15-25/27) [25-40 icds, 8, 11]
  • Pre-post 65+/- [changing coverage] QOC relationship (55-64, 65-74) [15-40 icds, 2 groups]
  • SNGs reports: IDs, Autism, EP, MS, PTSD, Specific Drug Abuse hxs, Bullying, Ch Viol. Ch Sx Ab, Age-Fxs, BioT [15-25 icd small groups, 8, 11]
  • Special reports, Risk Areas: Child Abuse indicators combination ICDs and V-codes, Suicide, Adult Abuse, Gender specifics; Infib, Violence indicators 6-12 icd subgroups]
  • Special reports, cultural: culturally-bound indicators, culturally-linked indicators, cultural QOC comparisons (B vs W, H vs W, A vs W) [3 x 40 = 120, starting]
  • Standardized Special reports: Diabetes, asthma, MS, epilepsy [3-7]
  • Design modules, major beneficiaries
  • Define and contact internal Gatekeepers for other options
  • Document reporting requirements
  • Fit requirements into HIT-GIS plan
    • PHI
    • HIPAA
  • Establish IRB engagement process

Note: a fully developed HIT-GIS will be capable of reporting on nearly all of these, and then some.