Phenomena: Exploring the "natural history" of disease
Idea: Detailed observation (like a naturalist) or detective work--albeit informed by theoretical ideas--may be needed before we can characterize what the phenomenon is we are studying, what questions we need to ask, and what categories we need for subsequent data collection and analysis.
(Make annotations in alphabetical order by author's name and at the end of your entry add your initials and the year you're taking the course.
Guidelines)
Notes and annotations from 2007 course
Initial notes from PT on Natural history
(See also
glossary guidelines )
1. The initial motivation for this class was to highlight that epidemiology does not
necessarily begin with data sets to analyze. There may be exploratory, investigative, detective, anthropological, and naturalist inquiries before phenomena are even noticed, categories are defined, questions are framed. Good examples of this seemed to be provided by John Snow's work on cholera, Barker's (1971) research in Uganda, and on "clues from geography" of infant mortality and heart disease (1998), and the three Lancashire towns, and Oxford's account of the conditions that provided a source for a global pandemic of the 1918 flu. (40 million died from flu, while 8.5m died from war.) Even Barker's (1999) speculation about anomalous French cardiovascular disease rates looks like someone who is able to connect dots of diverse kinds and that are spread out in time.
2. Brody's paper, in addition to drawing attention to the role of maps in this exploratory research, makes the Snow story more complicated and interesting. Snow had clear hypotheses that guided his mapping and his advocacy of stopping the water supply from the Broad St pump -- he was certainly not simply noticing patterns in the data and hypothesizing about the causes. This account opens up broader questions in philosophy of science. E.g., where do hypotheses that get assessed by research come from in the first place?
(Barker is a recently retired but still active epidemiologist whose reputation is linked to the "Barker hypothesis" that chronic diseases of later life are associated with fetal or early-life conditions. We'll address this hypothesis within the frame of "life course epidemiology" in week 10.)
Annotations on common readings
The two primary common readings for this week shared some common themes. Both the Spanish Influenza and Buruli Disease articles consisted of case study approaches to determine origin, transmission, and distribution of disease. Both conditions involve purulent matter, i.e. pus, albeit one is respiratory based while the other involves skin lesions. The presence of each appears to be correlated with issues of overcrowding and sanitation problems. The significant role of geography and population movement in epidemiology is evident, in terms of hindering or facilitating infection. Such migratory and travel patterns are exacerbated by modern air travel and reading the Oxford article reminded me of chapters from The Coming Plague (L.Garrett) which highlight the risk of such rapid transmission.
In the case of influenza, the cause of the disease and mode of transmission are widely understood and accepted. The first article seeks to understand the conditions that foster transmission. In contrast, the conditions under which Buruli flourishes appear clearer to researchers than does the cause or mode of transmission of this elusive disease.
Spanish Influenza (Oxford et al 2005)
This article outlined the influenza pandemic of 1918-1919 that coincided with the end of World War I. The ‘flu’ is an upper respiratory infection, found worldwide, that is spread from person to person and among people and animals, by the influenza virus. Contributing conditions to the spread of infection at the time included; overcrowding in unsanitary conditions among young soldiers on the warfront in Europe, close proximity to pigs, geese, ducks, and chickens which fostered cross-species viral mutation and transmission, and exposure to mutagenic gases (i.e. toxic) that compromised respiratory systems making individuals more vulnerable to the virus.
According to the authors, war travel restrictions hindered the virus in the short-term but then served to incubate more virulent strains, which soldiers carried back to their countries when they returned home at war’s end. It is also noted that the virus may have ‘seeded’ itself in multiple geographic locations prior to the pandemic, however the reference to similar symptoms by different terminology may have delayed recognition of the onset of the epidemic.
Concentrated, complicated vocabulary is used to describe the symptoms of influenza, employing clinical as compared to laymen’s terms. This article was fairly dismissive of alternate hypotheses, such as the 1918 flu may have originated in the U.S., or China, rather than Spain. Current available technology is described, in that researchers have used DNA strains from preserved organs of deceased flu victims, taken over stages of the epidemic from 1918-1919 to identify evolving strains of the flu virus. This research was conducted with the aim to identify where the next flu epidemic may originate from geographically and thus how to effectively prevent it from becoming widespread. The authors acknowledge the limitations of this approach in that this technology has provided limited clues.
In conclusion, Oxford et al posit that contributing conditions are more relevant and predictive of the next pandemic. They recommend that instead of focusing on geographic origin and distribution of flu strains, it would be more instructive to focus on locations where the aforementioned combination of conditions may be found and increase flue surveillance in such places. (Amy H)
Buruli Disease (Barker 1971)
Buruli is an endemic skin disease characterized by open lesions, that is found in tropical countries. This article is compromised of a case-study with a quantitative component, measuring distribution of disease across a sample group. However, the small sample size (N=82) limited the ability of the researcher to establish patterns in disease distribution, whether by geography, age, or gender. Barker did identify that Buruli is more common among females and children, with lesions located on limbs, but this slight variation that was largely unattributable to specific causes. Moreover, with a sample this small, the use of percentages can provide false leads in that 55% may seem to signal a trend until one realizes it is merely referring to 2 of 4 male children, ages five and younger.
Barker and his fellow researchers tried to control for confounding factors by limiting their sample according to geography (in terms of location, range, and characteristics), hospital, and doctor. They also focused closely on the effects of the Nile river and water in general, following a chain of theories from water to tsetse flies to crops and finally to exposure of skin to rough grass that may puncture skin, thus making it vulnerable. Speculations and assumptions were left unanswered for the reader. Conditions also come into effect in this article, with occurrence of Buruli at overcrowded unsanitary refugee camps in tropical countries noted. In addition to geography and environmental conditions, Barker tried to establish parameters concerning the recent nature of this disease and the accelerated nature of incidence among the population to understand the epidemiological patterns.
It was also interesting to note in the acknowledgements that the majority of ‘experts’ on this research appeared to be foreign born, as compared to native Ugandans or Rwandans. It would be interesting to know if a subsequent increase in native born physicians and researchers has helped to uncover the source and transmission of Buruli. (Amy H)
Annotated additions by students
Mothers, Babies, and Health in Late Life: Clues from Geography (Barker, 1998)
Does poverty directly affect birth weight, or does it hinder the ability of caregivers to provide their fetus or young child with nutritious food? Do all poverty stricken/low income persons eat unhealthy foods and/or engage in unhealthy behaviors such as smoking, drinking, etc.?
Does the lack of education of poor people play a role in them developing of heart disease? Do they have the knowledge necessary to learn how to eat healthy? Do they have access to quality medical care? Isn’t it true that those who do not have adequate health care options delay (or forgo) seeing a doctor when they become ill?
Should programs such as the ones in some U.S. states (i.e., food stamps, nutrition/education ad campaigns, etc.) be implemented to help lower class individuals gain access to quality foods and healthcare?
In terms of why studies on heart disease are focused primarily on adults: Could that be because it is easier to establish that the disease has occurred (instead of projecting that it will in an infant)? Is it reasonable to then trace that person’s family history to determine cause and origins? Especially if their siblings have poor health as well.
Marmot et al: he says that those who are shorter in stature are assumed to possibly have had a “worse environment in their earlier life.” (ischaemic heart disease). How does he come to this conclusion? Are there medical determinants of susceptibility based on the height of an individual? What kind of environment? In the womb? In the household? In the city or town that they live in?
(CH, 9/16/09)
Mothers, Babies and Health in Later Life (Barker, 1998)
Understanding how human organism works, why some people get sick and others do not, why some get one disease and others get different diseases, or why some get sick at early age and others in later life are the questions that puzzle scientists in different fields, particularly medicine, epidemiology, and public health.
In addition to standard ways of looking at these problems, Barker offers a somewhat different perspective. While his main hypothesis sees nutrition before birth and in infancy as an important predictor of health and diseases developed in later life, Barker also suggests looking into clues from geography. By examining data of standard mortality ratios (SMR) for coronary heart disease in England, Barker observed concentration of high mortality in the northern parts. In support of the thesis that geography has influence on coronary heart disease, Barker showed that even after public policies have resulted in changes in diet, especially lowering fat and sugar intake and increasing fiber intake, heart diseases continued to be on the rise.
As another example in support of using geographic clues Barker used migration. He reported findings that showed that the risk of diseases “moved” with the migrants, and that their place of birth was better predictor of their health than the place of death.
However, there is no clean cut in interpreting geographic clues, because economic conditions of certain areas often follow the mortality rates. Finally, Barker concludes that “mothers are the dominant environmental influence” and that in search of clues the focus should be on babies and their mothers, where and how they lived in the early stages of baby’s life. (DBJ, 09)
Mothers, Babies, and Health in Late Life: Clues from Geography (Barker, 1998)
Barker explores the reasons behind getting sick. Why do some people get sick while others do not? Why do some get diseases and others are immune to falling prey to these same illnesses. Does poverty play a role in low birth weight? Do the eating habits of poor people affect fetal development? Researchers in health related sciences all try to understand the reasoning behind these issues. Barker attempts to understand this with a fresh point of view. Barker acknowledges that nutrition in the fetal stage as well as infancy is important, but he asks readers to go a step further and look into geography. After all, according to Barker, geography studies were some of the first to point out the need to look at the fetal stage for clues to health related events (p. 2).
One example of the need to study geographical settings is Barker’s research into data for coronary death rates in England and Wales. He points to geography as a clue for disease research, as the northern industrial towns of England have the highest numbers of coronary deaths. In comparison, death rates in the eastern and southern towns are low. Barker gives compelling evidence for expanding the view of disease investigation. This must force researchers into thinking about broader ways of seeking knowledge. (LJB, '09)
