We faced a similar problem

Article A World Malaria Map: Plasmodium falciparum Endemicity in 2007

Apologies I sent to early. We have faced similar problems and adopted a solution based on an old classification of parasite rate as explained in the attached references. The answer is thus a little dependent on the data you have available.

Article A World Malaria Map: Plasmodium falciparum Endemicity in 2007

Article A new world malaria map: Plasmodium falciparum endemicity in 2010

Hi Paul,

There could be regions of  very high endemicity, high endemicity, moderate endemicity and low endemicity.

Kindly find as attached this 2013 research article:

http://www.malariajournal.com/content/12/1/339

Hope this, in addition to others, helps

Kind regards

Olugbenga Oluwagbemi

I think A world malaria map: Plasmodium falciparum endemicity in 2007 on its way  to  be updated

You should try different cutoffs and choose one that better fits the purpose of your analysis. Malaria is so high in Africa that in most cases the classification proposed by Hay et al may not work to classify regions within a country that has low endemicity. If such is your case I will suggest that you stratify your country regions based on the logarithm of your malaria incidence first and move on from there. I.e. In Peru we are using the annual parasite incidence (malaria cases counts per 1,000 inhabitants per year) to compare malaria endemic districts, so the classification that better fitted our needs was: 0,1/1,000, 1/1,000, 1/10,000, and 1/100,000. I hope it helps.  

I am not really providing an answer to your quest, but it brings to mind a very interesting phenomenon being observed in Haiti where as you know malaria is endemic. In the 1960s, Francois "Papa Doc" Duvalier initiated a program SNEM which stands for Service National pour l' Eradication de la Malaria. It worked beyond expectations reducing the number of active cases by 90% or more.

Since the chaos the country was plunged in after decades of political instability and therefore lack of consistent public health policies, Haitian doctors are seeing spikes in malaria cases followed by inexplicable lulls. The same with Dengue fever cases. Yet no national control measures are in place and there are no obvious correlation between lulls and dry season or spikes and rainy season. Somewhat puzzling.

I am using PfPr data standardized to age limit 2-10 years, historical limits data and gridded population data to categorize a countries' (Guinea) districts into the endemicity classes

Dear  Paul Ouma 

The age itself depend on the endemicity as different age groups are infected

Dear Simon

I am using the similar data to the one your paper proposes, but now trying to adjust the populations at risk at the sub national level (which are the decision making units)

You may contact with Mohammad Shafiul Alam, icddrb. He has conducted mapping malaria endimicity in Bangladesh. He is available in researchgate.

I would recommend use of incidence cases as already advised by Quispe above. if available incidence data has age variable, age standardised incidence would be more precise. normally land terrain would naturally divide any country into different endemic levels with valleys having high and highlands low, fitting a model that incorporates altitude , temperature and rainfall would greatly improve the ability of the same to predict the same.

I would rather go for stratification of malaria risk. Stratification approach uses epidemiological data (prevalence, incidence, API, mortality, vector and parasite species ....what so ever available) as well as operational data (socioeconomic, geography and climatic data, programme status, health system....) to classify districts to strata. Each strata will have specified intervention/s.

More data about the stratification could be found at WHO website and malaria journal and other journals (you can use the term "malaria startification" for searching) 

We developed an automated monitoring system based on an algorithm. The algorithm employs quartile diagrams to classify municipalities according to variations in the incidence of malaria: group 1 (reduced incidence)—municipalities with below the expected incidence rates; group 2 (expected incidence)- 

within the expected incidence rates; group 3 (epidemics)–higher than the expected incidence rates; and group 4 (sporadic case)–a single case during a year.  You can see details at:

http://www.scielosp.org/pdf/rpsp/v35n3/a04v35n3.pdf 

http://www.scielosp.org/pdf/rbepid/v17n3/1415-790X-rbepid-17-03-00615.pdf 

Dear Rui Moreira Braz

Is there any  age factors in these categories?

Dear Ishag

We do not evaluate age factors. We evaluate the number of cases only (crude data)

You can go through;

Public health 127 (2013) 572-578 by Nath et al

Geocarto International, 27:1, 79-88 by Nath et al

Intl. J. Geographical Info. Sci. DOI No. 10.1080/13658816.2011.5 by Yadav et al

Thanks Sunil Dhiman

For these valuable sites

 Test positivity rate (TPR) could be one good method for mapping of malaria endemicity.

I think it desire so

I Think malaria is an important public health issue and needs to be mapped on different aspects,epidemiology,genetics,pharmacological ...

Malaria mapping: understanding the global endemicity of falciparum and vivax malaria

Article Malaria mapping: Understanding the global endemicity of falc...

Thanks for all the tips! This was extremely helpful

Geospatial mapping technique is the answer for you. You can identify the endemic cases of the particular area.