Leishmaniasis is a parasitic disease caused by different species of protozoan parasites. Cutaneous leishmaniasis (CL) is still a great public health problem in Iran, especially in Isfahan Province. Distribution and abundance of vectors and reservoirs of this disease is affected by different factors such as climatic, socioeconomic and cultural. This study aimed to identify the hotspot areas for CL in Isfahan and assess the relations between the climatic and topographic factors with CL incidence using spatial analysis. We collected data on the total number of CL cases, population at risk, vegetation coverage, altitude and climatic data for each district of the province from 2011 to 2015. Global Moran’s Index was used to map clustering of CL cases across districts and the Getis-Ord (Gi*) statistics was used to determine hotspots areas of the disease in Isfahan. We applied overlay analysis to assess the correlation between the climatic and topographic factors with CL incidence. We found the CL distribution significantly clustered (Moran’s Index=0.17, P<0.001) with the Ardestan and Aran va Bidgol (P<0.01) districts along with the Naein and Natanz districts (P<0.05) to be strong hotspot areas. Overlay analysis revealed a high incidence of CL in areas with relative humidity of 27-30%, mean temperature of 15-19°C, mean precipitation of 5-20 mm, maximum wind speed about 12-16 m/s and an altitude of 600-1,800 m. Our study showed that spatial analysis is a feasible approach for identifying spatial disease pattern and detecting hotspots of this infectious disease.
Leishmaniasis is a parasitic disease caused by several different species of protozoan parasites (Holakouie-Naieni
The least fatal form of the disease, cutaneous leishmaniasis (CL), is caused by
Recent advances in geographic information systems (GIS) and remote sensing (RS) have promoted the study of spatial epidemiology and environmental factors affecting the vector-borne diseases (Kassem
The aims of the present study were to 1) identify the geographical distribution of CL in Isfahan Province; 2) search for hotspot areas; and 3) assess the relations between the climatic and topographic factors with CL incidence in Isfahan Province using the spatial analysis during the period 2011 to 2015.
Isfahan Province is located between 31° 43′ to 34° 22′ N and 49° 38′ to 55° 31′ E and lies in the central parts of the Iranian plateau covering an area of 107,027 km2. Iran is a mountainous country mainly situated ≥1000 m above the mean sea level. The provincial capital is the historic city of Isfahan. According to the census of 2015, it consists of 23 districts with about 4,600,000 inhabitants. The province has a moderate and dry climate on the whole, and is a wellknown endemic area of leishmaniasis (
In order to conduct this study, information on the total number of CL cases, population at risk, climatic data, vegetation coverage and altitude were gathered for each district of the province.
Climatic data including mean precipitation, mean temperature and mean humidity were obtained from Tehran Meteorological Center, collected from synoptic stations in Isfahan and neighbouring provinces, including Lorestan, Kohgiloyeh VA Boyerahmad, Semnan, Chaharmahal & Bakhtiyari, Yazd, Fars, Qom and Markazi. These data were entered in ArcGIS, version 10.3 (ESRI Inc, Redlands, CA, USA) and dealt with in several steps: first, the yearly average of climatic data was calculated for 34 stations from April 2011 to March 2016; a point layer was created for 34 synoptic stations in a second step followed by ordinary Kriging was carried out for interpolation and calculation for all districts to derive a predicted value for unmeasured locations. Weights were based on the distance between the measured points, the prediction locations, and the overall spatial arrangement among the measured points (
Information on vegetation status of study area was obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS) consisting of 16-day composites with 250-meter spatial resolution. These data were used to calculate the NDVI, one of the most commonly used measures to of landscape ecology and useful for the study of the epidemiology of vector-borne disease (Bavia
where NIR stands for near infrared light and RED for red light (Jackson and Huete,
A digital elevation model (DEM) is an ordered array of numbers that represents the elevation over a specified segment of the landscape (Meijerink
Population data for all 23 districts of Isfahan was obtained from National Bureau of Statistics of Iran.
We obtained confirmed CL cases for Isfahan Province from the Isfahan University of medical sciences and Department of Communicable Disease Control (CDC) of the Iranian Ministry of Health in Tehran. The CL incidence rate was calculated as follows:
The total population at risk was defined as population for each district or province, which was obtained for each year.
A choropleth map was produced to show distribution of CL cases at the district level using a range of colours. Spatial data often are clustered, which means that stronger relationships may be present between proximate observations (Fotheringham and Brunsdon,
Moran’s Index is a global statistic and does not show the local structure of spatial autocorrelation (Chen,
To assess the correlation between DEM, NDVI, humidity, temperature, precipitation and maximum wind speed with CL incidence, we applied overlay analysis. At first, we produced three layers of information including a geographical map the study area (Isfahan), six layers representing each of the environmental factors interpolated with the Kriging method, and a layer indicating the CL distribution. The different layers were overlaid in various combinations, including a final step of all layers together.
Our study showed that incidence of CL in the Isfahan at the district level was significantly clustered (Moran’s Index=0.17, z=4.02, P<0.001). A total of 13,790 confirmed CL cases were reported in the Isfahan Province during 2011-2015 period, with a maximum incidence rate of 62 per 100,000 in 2014 (
The results of the Getis-Ord (Gi*) statistic, used to identify hotspots areas of CL during 2011-2015, showed that both hotspots and coldspots existed at the district level with the former seen in Ardestan and Aran va Bidgol (P<0.01) and Naein and Natanz (P<0.05), while Chadgan was identified as a coldspot area with a CL incidence of 0.05<P<0.1 (
The present study investigated spatial patterns of CL incidence in an endemic area of Iran using the GIS and RS, during 2011-2015. The assessment of spatial characteristics of the CL cases by Moran’s Index and derived Z-scores indicated that CL cases, as a whole, were clustered in the study area. We identified hotspots as well as coldspots for CL incidence, which were clustered in a specific area. The fact that the highest and lowest CL incidences were found in 2011 and 2015, respectively, could be due to interventional programmes, such as reinforced training, health education, disease surveillance and strengthened vector/reservoir control interventions, which were performed these years.
As can be seen in
NDVI is commonly used to separate three types of land cover: surfaces with sparse vegetation (NDVI<0.2), surfaces partially covered by vegetation (0.2≤NDVI≤0.5) and surfaces fully covered by vegetation (NDVI>0.5) (Momeni and Saradjian,
We also found hotspot areas in semi-arid regions with moderate levels of humidity (
Annual variation between moderately high wind speed and CL distribution showed a positive relationship in our study. A recent study has shown that
There are some limitations for this study. The CL surveillance system in Iran is a passive system, so underreporting is a strong possibility, especially in the rural areas. Furthermore, some errors may occur in the surveillance system, such as unreliable diagnosis and notification, or cases acquired in areas other than where they were diagnosed and reported. In addition, climate is only one of many groups of factors influencing vector distribution, while other factors such as vector ecology and socio-economic factors vary from one area to the other and should also be considered in the study of vector ecology. However, we assessed only climatic factors, while we fully understand that comprehensive research needs to consider also other factors, such as cultural, socioeconomic, immigration, demographic, sanitation and vector diversity.
CL is a public health problem in Isfahan. Several hotspot areas were identified using spatial analysis performed by GIS and RS. Overlay analysis revealed a relationship between several climatic factors and incidence of CL in these hotspot-prone areas, the majority of which were located in semi-arid regions with low vegetation coverage. We also found fewer hotspots in lower-altitude regions with higher temperatures and less rain. In addition, a positive correlation between wind speed and hotspot areas was found. The results of the present study indicate that GIS is a feasible approach for identifying spatial disease patterns and detecting hotspots of particular infectious diseases.
The authors would like to thank the CDC Department of Isfahan University and CDC, Ministry of Health and Medical Education in Tehran, Iran.
Study area, Isfahan Province and its districts, centre of Iran.
Incidence rate of cutaneous leishmaniasis in Isfahan Province in the period 2011-2015.
Choropleth map of incidence of cutaneous leishmaniasis in Isfahan Province in the period 2011-2015.
Hotspot map of cutaneous leishmaniasis in Isfahan Province in the period 2011-2015.
Digital elevation model plotted on the study area map and overlaid with cutaneous leishmaniasis incidence (2011-2015).
Normalised difference vegetation index plotted on the study area map and overlaid with cutaneous leishmaniasis incidence (2011-2015).
Mean humidity plotted on the study area map and overlaid with cutaneous leishmaniasis incidence (2011-2015).
Mean temperature plotted on the study area map and overlaid with cutaneous leishmaniasis incidence (2011-2015).
Mean precipitation plotted on the study area map and overlaid with cutaneous leishmaniasis incidence (2011-2015).
Mean of maximum wind speed plotted on the study area map and overlaid with cutaneous leishmaniasis incidence (2011-2015).