print logo

Controlling diarrhea and dengue diseases in rural primary schools in Colombia

Diarrheal diseases and dengue fever are major health problems in Colombia. The Norwegian University of Life Sciences in collaboration with researchers from the UK and Colombia are looking at the combined risk of getting diarrhea infections and dengue fever, and whether integrated intervention measures can improve the health situation for school children at primary schools in Colombia.

Frøydis Kvaløy




In Colombia both dengue fever and diarrheal diseases are serious public health problems. Colombia has one of the highest incidences of dengue in Latin America. Diarrheal diseases represent a leading cause of morbidity and one of the 10 most important in terms of mortality. One reason for these public health problems are poor water supply and sanitation. Contamination of stored water is a common source of diarrheal disease, but stored water also acts as possible breeding place for Aedes aegypti, the primary mosquito species that transmit dengue.

– Little is known about the role stored water plays as a common risk of diarrhea and dengue globally, says Hans Jørgen Overgaard, researcher at the Department of Mathematical Sciences and Technology, Norwegian University of Life Sciences. – How water is stored at the household level and where it is collected plays a role in the transmission of both these disease complexes. Therefore it could be possible to develop joint integrated control and management strategies for both.

Hans Jørgen Overgaard

Overgaard is leading a Norwegian-Colombian research team together with Professor Thor-Axel Stenström (former professor at the Norwegian University of Life Sciences, currently based at Durban University of Technology in South Africa). The Colombian partner is the Institute of Health and Environment at the Universidad El Bosque in Bogotá. The London School of Hygiene and Tropical Medicine in London and Liverpool School of Tropical Medicine are also partners.

Overgaard has his main experience from malaria and dengue vector biology and ecology in Southeast Asia, while Stenström has a global overview and work experience in relation to diarrheal disease linked with the World Health Organization (WHO).

The project is carried out by an interdisciplinary group consisting of professionals from various disciplines such as medicine, entomology, environmental engineering, and anthropology.

The objective of the study in Colombia is to investigate whether specific interventions in selected rural primary schools in Colombia could reduce diarrheal disease in school children and concurrently reduce various entomological risk factors. The study area was selected based on previous experience and is located in the Eastern Cordillera in the department of Cundinamarca about 3 hours west of Bogotá.


Study Design

Thirty-four primary rural schools in the municipalities of La Mesa and Anapoima in Cundinamarca province were included in the study. The schools were randomized into one of four groups:

  1. Diarrhea interventions (DIA),
  2. Dengue interventions (DEN),
  3. Combined diarrhea and dengue interventions (DIADEN), and
  4. Control, i.e. no intervention (CONTROL).

The diarrhea interventions (DIA) consist of a combination of filters for clean drinking water, hand washing campaign, clean water jars, lids on water tank and clean toilets. The dengue interventions (DEN) consist of insecticide impregnated curtains, adding pyriproxyfen (a growth regulator that inhibits growth of mosquito larvae) into water breeding containers, garbage cleaning campaigns and mosquito nets and lids on all water tanks.

There is also an educational component (including both a teachers and student manual) where children learn about both diseases and how they can avoid getting sick.

Approximately 800 students from grades 0-5 in 34 schools were recruited for the study, with 8 schools in the DIA group, 9 in the DEN group, 8 in the DIADEN group and 9 in the control group.


Class room with insecticide impregnated curtains

Data collections

All physical interventions were installed in January-May 2012. The first data collection was conducted in June-July 2012, the second in November 2012, the third in June-July 2013, and the fourth in November 2013.

– We are carrying out four different types of data collections to assess the effect of interventions on different outcome measures, explains Overgaard. – Firstly, we record student school absence by disease; secondly, we collect data on adult and larval mosquito abundance; thirdly, we collect water samples to assess fecal contamination in drinking water, and finally, we assess the knowledge, attitudes and practices of school children.


Anticipated outcomes

– The primary outcome we hope to attain by our research is both that the incidence rate of diarrhea among school children and the density of adult female Aedes aegypti mosquitoes per school will be reduced as a result of the interventions, says Overgaard.

– A range of secondary outcomes will also be measured and observed, for example, reduction in school absence in general, improvement of drinking water quality and hygiene practices, reduction in the number larvae or pupae of Aedes aegypti mosquitoes; and improvement of the knowledge and practices of teachers and pupils regarding these diseases.


No clear indications on diarrhea disease

According to the researchers the data collections carried out until the beginning of 2013 show no clear indication that the interventions have had an effect on diarrhea incidence. There were no statistically significantly differences between the four study groups in terms of diarrhea incidence of diarrhea among the school children.

– The reason for this could be that it is too early to see an effect of the interventions, says Overgaard.

– It could be a random result due to the study design or a generally low number of cases. After all there were only 38 cases of diarrhea during the observation period. Furthermore, it is possible that the main effect occurs in the home environment or could be explained by variations in socio-economic conditions. Perhaps there are variables that we have not taken into consideration that can explain these results. This is something we will look into more closely in the future.


Reduced number of mosquitoes

– We are happy to see that the interventions targeting the immature mosquito populations seem to be effective, says Overgaard. – We found that, on average, the number of Aedes aegypti positive containers were the lowest in schools that included the DEN intervention. In fact, during the third data collection we did not find any larvae at all in the DEN and DIADEN schools, whereas the CONTROL and DIA schools had about 40-60 positive containers per 100 schools. Thus, keeping containers covered and treating them with pyriproxyfen seem to have the intended effect that we would expect.

When it comes to the adult mosquito populations the team also found lower mosquito numbers in schools with the DEN intervention, however this difference was not statistically significant.

-The reason for this could be that the main control method targeting adults, i.e. insecticide treated curtains, did not perform well enough, says Overgaard. – There are probably too many openings, e.g. doors, eaves, ventilation holes etc. that could not be covered by the curtains. Furthermore, mosquitoes also breed in habitats that were not targeted in our interventions and adult mosquitoes can fly to the schools from the surrounding areas.


Dengue virus found in mosquitoes

Baseline data collected in 2011 indicate that the mosquito infestation in the study area is close to or above the recognized limit for transmission of arboviruses (groups of viruses that are transmitted by insects) and therefore remains a risk area for dengue transmission. To confirm if the dengue virus is circulating in the mosquito populations in the study area, the researchers performed additional adult mosquito collections in 2012 from several households close to the study schools.

– We were very surprised to find a high proportion of dengue-infected mosquitoes, says Overgaard. The virus that causes dengue fever is in fact a set of four dengue virus serotypes. We were astonished that we were able to detect all four dengue serotypes in just a few preliminary samples that we have analyzed so far.


Mosquito net and lid on water tank

Improved drinking water quality

The quality of water was measured by the presence and levels of the bacteria Escherichia coli (E. coli), which is the internationally used indicator of fecal contamination of water. The data shows that the average E. coli concentration was lower in the study groups that had diarrhea interventions (i.e. the DIA and DIADEN groups) than in the CONTROL and DEN groups.

–This is good news, says Overgaard. – The lowest level was in the DIA group, meaning that drinking water quality seems to be better in schools where the diarrhea interventions have been introduced. It shows that the interventions are effective.

– We also wanted to check the reduction efficiency of some of the interventions, Overgaard continues. – The water filters that we installed in the schools have been shown to be effective in various other projects in Central and South America. The filters are simple clay containers with an improved design that lets the water seep through the container walls and thereby clean the water from contaminants. We analyzed the quality of the water before and after filtering the water. Fifty percent of water samples taken before the filters were installed were E. coli positive, whereas only 10 percent of the water samples taken after the filters were added were E. coli positive. The mean E. coli concentration in the sample taken before the filters was more than 11 times higher than in the samples taken after the filters.

– In addition, we also compared water from the filters with water that had been boiled and stored. We found that about 44 percent of water samples taken from boiled, stored water were contaminated with E. coli, whereas only 6 percent of water samples that were filtered were contaminated.

These results show that the filters are working well and seem to work better than boiling the water.


Difficult to assess school absence

The researchers found that it is difficult to accurately assess school absences. One of the reasons is that absence registration is not common in rural Colombia. Many students live far from the schools and they are difficult to reach. Teachers' reports are not always accurate and trustworthy either.

– To be able to collect data on absence, we have established a system of weekly telephone follow-ups with teachers and parents in order to record all absences and their causes. Preliminary results show that the main reasons for school absence are sickness (27 percent), travel (13 percent), bad weather (13 percent), lack of incentive (13 percent), and other reasons (34 percent). The most common health reasons for school absences were the common flu (44 percent), diarrhea (17 percent), and other reasons (38 percent).


Better knowledge, attitudes and practices

The data collection to assess knowledge, attitudes and practices regarding dengue and diarrhea was based on a questionnaire distributed to 132 students in the 4th and 5th grades in 2012.

– It is interesting to see that integrating teaching about both diseases at the same time is more effective in terms of knowledge about either of the two diseases, than teaching about either disease alone. Students who were taught about both diseases, i.e. in the DIADEN group, knew more about each disease than students who were taught about either disease alone. Although this is a positive results we don’t exactly know why this is so.

– Another positive outcome of the project is that the study material and educational guides for teachers and children that were developed by the Colombian team will be distributed to other larger urban and rural schools in the municipalities. We hope this might have a multiplicative effect.

– The school principals of these schools were very interested in including our educational guides, says Overgaard. – We are of course happy that some of our activities will be adopted by the communities and, thus, ensuring sustainability.


Final seminar

– During 12th - 14th of May 2014 we will arrange a conference in Norway to present the final results of the project, concludes Overgaard. – We will invite international key note speakers. Presentations of the project results and discussions will hopefully lead to recommendations for integrated disease control in general and for dengue and diarrheal disease in particular. We have received NorLARNet support for this conference. We are very grateful for the hard work the Colombian team has put into this project and are therefore very pleased to be able to invite them to Norway.




Dr. Hans Jørgen Overgaard studied forestry at the Norwegian University of Life Sciences (NMBU) but soon changed direction into medical entomology. He did his PhD and postdoc work on malaria mosquito ecology and insecticide resistance in Thailand. He has since worked in the malaria control program of Equatorial Guinea in Central Africa and is currently involved in several malaria related projects in Africa. He is still actively doing research on dengue and diarrhea diseases in Southeast Asia.


Prof. Thor Axel Stenström is currently acting as a South African Research Chair instituted by the National Research Council (NRF) of South Africa at the Durban University of Technology (SARChI). He had a former position as Prof II at the Norwegian University of Life Sciences (NMBU) and has been acting as an Advisor for the World Health Organization in relation to Water, Sanitation and Health.

Tags: Colombia
Published Jan. 6, 2014 10:30 AM - Last modified Jan. 22, 2014 2:44 PM