Schistosomiasis control in China : strategy of control and rapid assessment of schistosomiasis risk by remote sensing (RS)and geographic information system (GIS)

Human schistosomiasis remains one of the most prevalent parasitic infections
in the tropics and subtropics. The disease currently is endemic in 76 countries and
territories and continues to be a major public health concern, especially in the
developing world. It is estimated that 650 million people are at risk of infection.
Among the 200 million people actually infected, 120 million are symptomatic and 20
million suffer severe disease. Although morbidity control – in line with
recommendations put forth by the World Health Organization – has been carried out
in China for more than 20 years, it is estimated that 90 million people still live in areas
where they are at risk of infection, and 820,000 people are infected with the parasite,
i.e. Schistosoma japonicum. The estimated area of intermediate host snail habitats
comprise 3,436 km2, concentrated in the 5 lake regions along the Yangtze River that
include the provinces of Anhui, Jiangsu, Jiangxi, Hubei and Hunan. The marshlands
of the Poyang Lake region represent some of the strongholds for the transmission of
S. japonicum. In these settings, for example, the percentages of acute cases and
intermediate host snail habitats represent 79.5% and 96.4%, respectively. With the
World Bank Loan Project (WBLP) to control schistosomiasis in China, the overall
prevalence of S. japonicum was significantly reduced, but in highly endemic areas
the re-infection rates are high.
In the first part of the present thesis, I summarize the 50-year history of China’s
experience and expertise in schistosomiasis control. Particular emphasis is placed on
morbidity control and achievements made by the WBLP carried out between 1992
and 2001. Reviewing this body of literature reveals that morbidity control of
schistosomiasis in China has been successful, and hence this strategy will continue
to form the backbone of protecting people’s health. However, total expenditures have
been considerable, and with the termination of the WBLP there is concern that
schistosomiasis might re-emerge. In the second part of this thesis, I describe the
successful development of a novel compound model to identify the habitats of
Oncomelania hupensis, the intermediate host snail of S. japonicum, and hence the
identification of high-risk areas of disease transmission. There are three findings that
warrant particular notion. First, visual land use classification on multi-temporal
Landsat images was performed for preliminary prediction of O. hupensis habitats.
Second, extraction of the normalized difference vegetation index and the tasseled
cap transformation greenness index were used for improved snail habitat prediction.
Third, buffer zones with distances of 600 and 1,200 m were made around the
predicted snail habitats to differentiate between high (>15%), moderate (3-15%) and
low risk of S. japonicum infection prevalence (< 3%). Preliminary validation of the
compound model against ground-based snail surveys in the Poyang Lake region
revealed that the model had an excellent predictive ability. The model therefore holds
promise for rapid and inexpensive identification of high-risk areas, and can guide
subsequent control interventions, such as whether mass or selective chemotherapy
should be employed. The model can also be used for diseases surveillance in
general and the monitoring of ecological transformations on the transmission
dynamics of S. japonicum, for example in the Three Gorges Dam area.