Not really having a direct experince in the use use of Gis for epidemilogical purposes, but I just want to put a general comment about what I came across during my research on schistosomiasis control using local methods.

I do not really belive that Gis can offer substantial thelp to understand how schistomiasis behaves. GIs is there basically to give visual (static) impression of the data characteristics from a far distance. It lacks closeness to the real situation on the ground. Many public health dicisions or policies made based on GIS are hard to implement on the ground. However, I find GIS a wonderful tool for preventive or precautious decisions/policies to travellers/tourists. E.G. The GIS images can clearly pick out all potential schistomiasis transmission sites around tourism water bodies and thus make them aware and avoid the disease. You'll see in most blogs, magazines, website for tourists there is no clear/GIS aided images of potential sites of the disease transmission---they just warn about a certain area or its southern part e.t.c. No certainity at all, but gis can do better

John Snow and his feat in the events of cholera in London, 1854, had already demonstrated the importance of space in understanding the behavior of the disease. Today with advances in technology, GIS provides possibilities of prediction of diseases and disorders through spatial statistics, geostatistics and other tools, thus being of great value in understanding the behavior, however it is reductionist. In my region, the sectors in the management of local public health services use in the routine of the geo and often get good answers when it comes to understanding these behaviors. I understand your position, but disagree on you saying that the GIS does not bring substantial contributions to the understanding, in this case, schistosomiasis and its control. What methods do you use for schistosomiasis control in your locality?

Well, coming from both worlds (urban development and epidemiology) I must say, that in a way I agree with Moir. But just in a way. GIS means GeoInformationSystem. It is simply a database where mostly informations of the builded world were saved e.g. streets, houses etc. To this geo information you can add more information about a specific point or polygon. e.g. this house is a bakery. The way GIS is mostly used in geo science is to measure distances e.g. shortest path, cluster areas based on geographical information, planing the amount of specific infrastructures in an area, measure form, size and structure of an area etc. Using this understanding of GIS, epidemiology can't benefit from GIS. But using GIS as a database of epidemiological information with a geographical connection, we can use this data to understand and manage diseases. But we do this with a statistical program like R not with ArcGIS. The advantage of GIS is to plot the information on a map and store the results with a geographical link. In this case GIS i would say epidemiology benefit from a GIS but in a way of a database based map.

But I would also like to come up with an idea, that might be really beneficial for you. At Northwestern University we worked on a model for the swine flu outbreak that used the travel behavior of people. We know, that these behavior follow a power law distribution. Using this information, it might be possible to predict an airborne disease outbreak more accurately using a shortest path tree of public transportation of an area. But it would probably be more beneficial to prepare for an outbreak than handle one. A multi-agent based simulation based on geographical information, which I want to develop in the future, could be also an interesting model. The question is more, who can spend the money to develop such models. My department at the health ministry in Brandenburg don't want to do that... :-/

Schistosomiasis is both quantitative and qualitative in nature. GIS records, stores, and analyzes info about the features that make up the earth's surface. If we relate it to epidemiology, GIS will assist understanding of the 'incidence and distribution' part, but it will take more than GIS (biostatistics/social e.t.c) to understand the 'severity and conditions influencing the spread of the disease'. Therefore the understanding of the behavioral part may need GIS plus. I meant understanding the 'behaviour' and not control or existence. So I agree advanced GiS can assist.

Almost every good research in a given goegraphical area will need a GIS aided site map. In my country the major snails control methods are (1) Chemotherapy anti-schistosomal drugs chiefly prazquantel (2) Health education to pple in endemic areas (3) Snail control with natural and sythetic moluuscicides/ bilogical control using native fish in pond aquaculture--- the least practised.

Despite the evident endemicity of the disease and the increase in water projects and potential risk of transmission, most control measures come in very late (especially when there is a crisis) by mostly donors and NGOs. Policy wise, schistosomiasis is neglected as most of the attention goes to malaria, Hiv Aids and Soil helminths diseases. Even in aquaculture technology transfer sytems/IAA to local smallholder farmers you can scarcely found the word schistosomiasis. This is just to show you how far we are from using advanced GIS techniques for policy making.

Dominik, I agree in part with what you said. When I said "GIS" I referred to Geoprocesessing which is a set of techniques that work with such information spatially where the GIS is part of it. Perhaps I used the wrong word, but I still think and there are several examples here of how the Geoprocessing brings notes to understand the epidemiological settings subsidizing information. Regarding the use of statistics and spatial statistics in this case, not only R but also provides ArcGIS such analysis, such as Geographically Weighted Regression. This tool creates situations that if we have the variables and a robust model, we can use them to predict the epidemiological settings.

By the way, I was curious about his work with Agent Based Model. You already have some publication or other material that I can access to learn more about this method?

Moir, we must agree that the use of GIS or any other procedure varies greatly with the infrastructure and social policy where the actions are developed. I understand that in some countries, it may be the case, a challenge is the implementation of tools like this for the simple fact that there are other priorities in health management, which currently constitute the most feasible to implement measures against schistosomiasis. But here in Brazil, for example, some measures you mentioned are no longer suitable for combat schistosomiasis, as molluscicides and biological control. Also the Integrated Program of the Schistosomiasis, Oswaldo Cruz Foundation (an important government institution of public health in Brazil), does not agree with the mass treatment of schistosomiasis in our region. This all just shows that according to the reality of each country, region, city, a number of factors need to be mounted, being different for each situation.

I am not an expert in epidemiology but find this discussion most interesting. The historical example of cholera in London certainly caught my imagination when I learned of this many years ago. As a Canadian I feel very fortunate that we in more north climes have far less incidence of parasites though in recent years West Nile Virus as well as Lyme's Disease have become of concern. GIS and spatial analysis systems enable analysis of data in ways not possible or difficult in past. I am aware of more use of these systems in addressing a variety of health related issues, but of course the key is being able to collect data efficiently. In my work in integrated pest management in the urban environment this has been one of the most difficult aspects, but with the resurgence of bed bugs has become more common. As in your discipline, this kind of data enables better decisions and realizations, not unlike the historical case of cholera mentioned. Just my reflections on this as a layperson. Most interesting discussion here.. I am in my senior years and the modern access to information through the net as I look back at the infancy of computers and before, gives me awe much as I had as a boy observing nature and learning wonders as watching a butterfly emerge from a chrysalis.

Best wishes to all. Good luck on your endeavours in this area.

Onicio, I didn't disagree with you, I just wanted to make sure, that other without a deeper knowledge understand what the difference is. I completely agree that geoprocessing is more than what I described. I also didn't know, that ArcGIS has implemented such a function, but I'm also just a social scientist, not a geodesist... ;-)

I have some publication about that, but all in german so far. But you can take a look here http://rocs.northwestern.edu/publications/index_assets/belik2011.pdf

http://rocs.northwestern.edu/publications/index_assets/PhysRevX.1.011001.pdf

http://rocs.northwestern.edu/publications/index_assets/brockmann2009schuster.pdf

For the simulation models you could start here:

I. Benenson. Multi-agent simulations of residential dynamics in the city. Computers, Environment an Urban Systems, 22(1):25–42, 1998.

I think GIS has huge potential in epidemiological research. Given the wealth of literature on the method I am sure we can safely say that there are numerous uses for GIS and its output in Health research. Even a simple google scholar serach will being up thousands of useful articles on the subject and I would suggest reading through a few of them to get a feel for the types of work others are currently doing on projects similar to yours.

When you come to thinking about your own research I think it is important to ascertain from the outset the particular perspective you are wanting to take, and how this in turn, will influence your methodology and analysis.

I.e-If you are coming at it from a humanistic, largely qualitative angle then GIS and its applications will probably take a lesser role than if you were doing a geo-stat, mapping or quant project. In my own experiences I have found GIS to be useful in 'setting the scene'/ helping to justify the need for investigation and easily displaying data in a visual way. I have used it laregly as a supporting tool, rather than it being the focus as I am a human geogrpaher/social scientist not a GIS expert :)

Others have brought up an important issue, that GIS (and its associated outputs) is somewhat reductionist, which I think it true, but as with any tool you recognise its limitations. Again it would depend on your method but as Onicio has said, pairing the data with GWR (etc) can be hugely influencial and I think engaging with GIS in a mixed-method study can really add weight to your findings . Almost every project in health and epidemiological research could and would benefit from GIS analysis in someway, so I would certainly suggest investigating further. Good luck.

As the common saying "A picture worth more than 1000 words" GIS has been a very important tool in the area of epidemiology most especially in explaining the impact of disease on a particular geographical area or the spread of disease from a geographical area to another. In the Houston Department of Health we were able to use the tools for the epidemiological description of the outbreaks/clusters of Shigella, West Nile Virus in the community to mention a few.

I have worked on data pertaining to schistosomiasis and its relationship with anaemia using biostatistical tools. I can say that applied GIS to helminth epidemiology is a very interesting tool. Indeed, you will find it is used for finding areas of endemicity to schistosomiasis and other soil-transmitted helminths and mapping prevalence of infection. The decision to go in and administer praziquintel and albendazole in areas heavily affected is all based on information derived from GIS. It is again used to monitor over time whether drug treatment campaigns are working to reduce (or eradicate) endemicity.

I think it is a very useful tool in this area of research. You will find a large body of literature on how its implemented. I suggest reading the following:

(i) Schistosomiasis Control Initiative;

(ii) recommended papers by Brooker et al., (2010) The global atlas of helminth infection: mapping the way forward for NTD control; Brooker et al., (2009) An updated atlas of human helminth infections: the example of East Africa; and Clement et al., (2006) Bayesian spatial analysis and disease mapping: tools to enhance planning and implementation of a schistosomiasis control programme in Tanzania

I would encourage people to make the switch from using GIS to the broader term of geomatics or even spatial intelligence. As has been alluded to, GIS is a system which in the traditional definition is a complement of software, hardware, data and people. Having worked as a spatial epidemiologist (I am an epi by training) it has become clear that simply using the term GIS usually means simply an IT tool to many people and that the value is really in the ability to incorporate the spatial and temporal dimensions in epidemiological or public health practice. To answer the other question posed, I have used Geomatics to look at movement of tuberculosis, delivery of breast cancer screening, location privacy in public health, spread of H1N1 and the distribution of vulnerable populations, threat modeling....it is a powerful ally and is definitely an asset to have - and I also co-develop and co-facilitate various workshops on this for health professionals. After all, place is one of the pillars of descriptive epi...