Environmental risk factors for Lyme disease identified with geographic information systems

Geographical information systems and bootstrap aggregation (bagging) of tree-based classifiers for Lyme disease risk prediction in Trentino, Italian Alps

The risk of exposure to Lyme disease in the province of Trento, Italian Alps, was predicted through the analysis of the distribution of Ixodes ricinus (L.) nymphs infected with Borrelia burgdorferi s.l. with a model based on bootstrap aggregation (bagging) of tree-based classifiers within a geographical information system (GIS). Data on L ricinus density assessed by dragging the vegetation in 438 sites during 1996 were cross-correlated with the digital cartography of a GIS, which included the variables altitude, exposure and slope, substratum, vegetation type and roe deer density. Ticks were more abundant at altitudes below 1,300 m a.s.l., in the presence of limestone and vegetation cover with thermophile deciduous forests and high densities of roe deer. A bootstrap aggregation procedure (bagging) was used to produce a model for the prediction of tick occurrence, the accuracy of which was tested on actual tick counts assessed by a further dragging campaign carried out during 1997 to determine infection prevalence and resulted in average 77%. Other tests of the model were made on additional and independent data sets. The prevalence of infection with Borrelia burgdorferi s.l, determined by polymerase chain reaction on 2,208 nymphs collected by random dragging in 245 transects selected within eight areas where the model predicted the occurrence of I. ricinus during 1997, was 17.5% and was positively correlated to tick abundance and roe deer density. These findings were used to relate the output of the bagged model (probability of tick occurrence) to the density of infected nymphs through a stepwise model selection procedure and thus to produce a GIS digital map of the probability distribution of infected nymphs in the Province of Trento at high resolution scale (50 by 50-m cell resolution). The application of the bagging procedure increased the accuracy of the prediction made by a single classification tree, a well-known classification method for the analysis of epidemiological data.

Mapping Lyme disease incidence for diagnostic and preventive decisions, Maryland

To support diagnostic and preventive decision making, we analyzed incidence of Lyme disease in Maryland on the zip code level. Areas of high incidence were identified on the Upper Eastern Shore of the Chesapeake Bay and in counties north and east of Baltimore City. These latter foci, especially, are not visible when mapping Lyme disease on the county level.

New perspectives on the use of Geographical Information Systems (GIS) in environmental health sciences

At first glance, the domain of health is no typical area to applicate Geographical Information Systems (GIS). Nevertheless, the recent development clearly shows that also within the domains of environmental health, disease ecology and public health GIS have become an indispensable tool for processing, analysing and visualising spatial data. In the field of geographical epidemiology, GIS are used for drawing up disease maps and for ecological analysis. The striking advantages of GIS for the disease mapping process are the considerably simplified generation and variation of maps as well as a broader variety in terms of determining a real units. In the frame of ecological analysis, GIS can significantly assist with the assessment of the distribution of health-relevant environmental factors via interpolation and modelling. On the other hand, the GIS-supported methods for the detection of striking spatial patterns of disease distribution need to be much improved. An important topic in this respect is the integration of the time dimension. The increasing use of remote sensing as well as the integration into internet functionalities will stimulate the application of GIS in the field of Environmental Health Sciences (EHS). In future, the integration and analysis of health-relevant data in one single data system will open up many new research opportunities.

Predicting the risk of Lyme disease: habitat suitability for Ixodes scapularis in the north central United States

The distribution and abundance of Ixodes scapularis were studied in Wisconsin, northern Illinois, and portions of the Upper Peninsula of Michigan by inspecting small mammals for ticks and by collecting questing ticks at 138 locations in state parks and natural areas. Environmental data were gathered at a local level (i.e., micro and meso levels), and a geographic information system (GIS) was used with several digitized coverages of environmental data to create a habitat profile for each site and a grid map for Wisconsin and Illinois. Results showed that the presence and abundance of I. scapularis varied, even when the host population was adequate. Tick presence was positively associated with deciduous, dry to mesic forests and alfisol-type soils of sandy or loam-sand textures overlying sedimentary rock. Tick absence was associated with grasslands, conifer forests, wet to wet/mesic forests, acidic soils of low fertility and a clay soil texture, and Precambrian bedrock. We performed a discriminant analysis to determine environmental differences between positive and negative tick sites and derived a regression equation to examine the probability of I. scapularis presence per grid. Both analyses indicated that soil order and land cover were the dominant contributors to tick presence. We then constructed a risk map indicating suitable habitats within areas where I. scapularis is already established. The risk map also shows areas of high probability the tick will become established if introduced. Thus, this risk analysis has both explanatory power and predictive capability.

Prevention of tick-borne diseases

Ticks are a part of the landscape where humans live, work, and play. Because ticks carry a wide range of organisms that potentially can cause disease in humans, many studies have focused on ways to reduce risk of these diseases. Ticks have biologically complex interactions with microorganisms and with their vertebrate hosts, on whom they depend for blood meals and survival. To consider ways to reduce the burden of tick-borne diseases in humans, it is necessary to understand the biology and ecology of ticks and their interface with humans. In many areas, changes in land use, reforestation, and patterns of human settlements have led to more abundant tick populations, increasing rates of infections in ticks, and increasing contact with human populations. Warmer winter temperatures in temperate regions may extend the transmission season for some ticks and pathogens. Although much of the discussion in this article has focused on I. scapularis and the Lyme disease spirochete (because they have been studied extensively), other tick-pathogen pairs may differ in risk factors for infection and transmission dynamics. Interventions studied to reduce the burden of tick-borne diseases include changing the environment, controlling vertebrate hosts, killing ticks, altering the behavior of humans, treating tick bites, and trying to protect humans through immunologic means (vaccine). All of these approaches have limitations and drawbacks. From a public health perspective, a plan that employs multiple strategies may be most effective. This article has reviewed what is known about preventive interventions, including the vaccine.

Early effects of climate change: do they include changes in vector-borne disease?

The world's climate appears now to be changing at an unprecedented rate. Shifts in the distribution and behaviour of insect and bird species indicate that biological systems are already responding to this change. It is well established that climate is an important determinant of the spatial and temporal distribution of vectors and pathogens. In theory, a change in climate would be expected to cause changes in the geographical range, seasonality (intra–annual variability), and in the incidence rate (with or without changes in geographical or seasonal patterns). The detection and then attribution of such changes to climate change is an emerging task for scientists. We discuss the evidence required to attribute changes in disease and vectors to the early effects of anthropogenic climate change. The literature to date indicates that there is a lack of strong evidence of the impact of climate change on vector–borne diseases (i.e. malaria, dengue, leishmaniasis, tick–borne diseases). New approaches to monitoring, such as frequent and long–term sampling along transects to monitor the full latitudinal and altitudinal range of specific vector species, are necessary in order to provide convincing direct evidence of climate change effects. There is a need to reassess the appropriate levels of evidence, including dealing with the uncertainties attached to detecting the health impacts of global change.

The shifting landscape of tick-borne zoonoses: tick-borne encephalitis and Lyme borreliosis in Europe

The two major vector-borne diseases of northern temperate regions, tick-borne encephalitis (TBE) and Lyme borreliosis (LB), show very different epidemiological patterns, but both have increased significantly in incidence since the 1980s. Insight into the temporal dynamics of TBE, gained from statistical analysis of spatial patterns integrated with biological explanation, suggests that the recent increases in TBE cases in Central Europe and the Baltic States may have arisen largely from changes in human behaviour that have brought more people into contact with infected ticks. Under forecast climate change scenarios, it is predicted that enzootic cycles of TBE virus may not survive along the southern edge of their present range, e.g. in Slovenia, Croatia and Hungary, where case numbers are indeed decreasing. New foci, however, are predicted and have been observed in Scandinavia. At the same time, human impact on the landscape, increasing both the habitat and wildlife hosts of ticks, has allowed tick populations to multiply significantly. This probably accounts for a genuine emergence of LB, with its high potential transmission rate, in both the USA and Europe, although the rate of emergence has been exaggerated by improved surveillance and diagnosis.

Locational uncertainty in georeferencing public health datasets

The assignment of locational attributes to a study subject in epidemiologic analyses is commonly referred to as georeferencing. When georeferencing study subjects to a point location using their residential street address, most researchers rely on the street centerline data model. This study assessed the potential locational bias introduced using street centerline data. It also evaluated georeferencing effects on a location-dependent, exposure assessment process. For comparison purposes, subjects were georeferenced to the center of their residential parcel of land using digitized parcel maps. A total of 10,026 study subjects residing in Jefferson County, Alabama were georeferenced using both street centerline and residential parcel methods. The mean nondirectional, linear distance between points georeferenced using both methods was 246 ft with a range of 11 to 13,260 ft. Correlation coefficients comparing differences in exposure estimates were generated for all 10,026 subjects. Coefficients increased as the geographic areas of analysis around study subjects increased, indicating the influence of nondifferential exposure misclassification.

Prevention of Lyme disease: a review of the evidence

The Healthy People 2010 public health goals targeted a 44% decrease in the incidence of Lyme disease, the most commonly reported tick-borne illness in the United States. To review Lyme disease prevention, clinical trials, epidemiological and experimental studies, and predictive models were evaluated. Geographic distribution of ixodid vectors and local landscape predict Lyme disease risk. Density of infected ticks correlates with incidence and prevalence of Lyme disease, but risk quantitation is made uncertain by tick aggregation and inability to predict tick-human interactions. Outdoor activities are inconsistently or weakly associated with risk, and most infections likely occur in residential areas during routine activities. Tick control (burning or removing vegetation, acaricide use, and deer elimination) reduces Ixodes scapularis populations by up to 94%, and acaricide application to wildlife decreases nymphal I scapularis populations by up to 83%. The effect of these strategies on incidence of Lyme disease in humans is unknown. Studies show that only 40% to 50% of adults take precautions against tick bites even when they are aware of Lyme disease. Effective protection afforded by personal precautions (wearing protective clothing, avoiding ticks, and using insect repellant) has not been shown prospectively. Antimicrobial prophylaxis of tick bites is not warranted. Clinical trials showed vaccines containing recombinant OspA of Borrelia burgdorferi to be efficacious and well tolerated. Currently, vaccination is the only empirically demonstrated method to prevent Lyme disease. The best evidence supports prevention efforts focused on practices that encourage immunization, Lyme disease awareness, and possibly treatment of deer.

Ticks and tick-borne disease systems in space and from space

Analyses within geographical information systems (GISs) indicate that small- and large-scale ranges of hard tick species (Ixodidae) are determined more by climate and vegetation than by host-related factors. Spatial distributions of ticks may therefore be analysed by statistical methods that seek correlations between known tick presence/absence and ground- or remotely-sensed (RS) environmental factors. In this way, local habitats of Amblyomma variegatum in the Caribbean and Ixodes ricinus in Europe have been mapped using Landsat RS imagery, while regional and continental distributions of African and temperate tick species have been predicted using multi-temporal information from the National Oceanic and Atmospheric Administration-Advanced Very High Resolution Radiometer (NOAA-AVHRR) imagery. These studies illustrate ways of maximizing statistical accuracy, whose interpretation is then discussed in a biological framework. Methods such as discriminant analysis are biologically transparent and interpretable, while others, such as logistic regression and tree-based classifications, are less so. Furthermore, the most consistently significant variable for predicting tick distributions, the RS Normalized Difference Vegetation Index (NDVI), has a sound biological basis in that it is related to moisture availability to free-living ticks and correlated with tick mortality rates. The development of biological process-based models for predicting the spatial dynamics of ticks is a top priority, especially as the risk of tick-borne infections is commonly related not simply to the vector's density, but to its seasonal population dynamics. Nevertheless, using statistical pattern-matching, the combination of RS temperature indices and NDVI successfully predicts certain temporal features essential for the transmission of tick-borne encephalitis virus, which translate into a spatial pattern of disease foci on a continental scale.

Spatial statistics and geographical information systems in epidemiology and public health

This chapter surveys the principles behind spatial statistics and geographic information systems (GIS), and their application to epidemiology and public health. Like the other introductory chapters, it is aimed mainly to facilitate understanding in the chapters specific to certain diseases that follow, and to provide a short introduction to the field. A brief overview of spatial statistics and GIS is provided in the introduction. The sections that follow explore the ways in which we can map the distribution of disease, ways in which we can look for spatial patterns in the distribution of disease, and ways in which we can apply spatial statistics and GIS to the problem of identifying the causal factors of observed patterns. In the last section I discuss some of the ways in which these techniques have been applied to assist decision making for disease intervention, and conclude by discussing future developments in the field, and some of the issues surrounding the integration of spatial statistics and GIS.

Risk factors for lyme disease in Chester County, Pennsylvania

OBJECTIVE

To identify factors associated with increased or decreased risk of infection for Lyme disease in Chester County, Pennyslvania.

METHODS

The authors designed an unmatched case-control study involving 294 incident cases reported to the Chester County Health Department in 1998 and 449 controls selected by random digit dialing. All case and control participants were interviewed by telephone.

RESULTS

Age is a risk factor for Lyme disease for groups aged 10-19 years old and 50 years or older. Sex was not a risk factor. Incidence of Lyme disease in a rural setting was three times the incidence in an urban setting. Increased risk also was associated with living in single family homes, homes with yards or attached land, woods on the land, signs of tick hosts seen on the land, and homes within 100 feet of woodland. Gardening for more than four hours per week was also a risk factor, but most other outdoor activities were not. Twice as many participants took protective measures against tick bites before outdoor employment than those who merely ventured into the yard or land associated with the home. Only checking for ticks during outdoor activity and the use of repellents prior to outdoor activities outside the yard were unequivocally associated with a reduced risk of Lyme disease.

CONCLUSIONS

It is important to increase public awareness about the risk of acquiring Lyme disease from ticks in the immediate environment of the home.

Morphoea is neither associated with features of Borrelia burgdorferi infection, nor is this agent detectable in lesional skin by polymerase chain reaction

BACKGROUND

The aetiology of morphoea is still unknown. Borrelia burgdorferi as a causative agent of morphoea has been discussed since 1985, but the relationship remains uncertain.

OBJECTIVES

We aimed to find evidence for infection with B. burgdorferi by combined evaluation of different clinical and laboratory data in a group of 54 patients with morphoea.

METHODS

In each patient, an evaluation of the case history was performed with regard to infection with B. burgdorferi, using a standardized questionnaire. Questions focused on previous tick bites and skin changes suspicious for erythema migrans (EM). The case history data of 52 patients were compared with those of 104 matched control subjects and of 25 patients with acrodermatitis chronica atrophicans (ACA). Serological examinations were performed in 53 patients with morphoea. Furthermore, lesional skin was examined for borrelial DNA in 33 patients, using nested polymerase chain reaction (PCR) for the ospA and the borrelial rRNA gene.

RESULTS

Results of the questionnaire showed no differences between patients with morphoea and matched controls. In contrast, patients with ACA showed a much higher prevalence of tick bites and skin changes suspicious for EM as compared with patients with morphoea. Serological examination was positive in only one patient with morphoea alone and in two additional patients with coexistent ACA. No borrelial DNA was detected by PCR in lesional skin of 33 patients with morphoea.

CONCLUSIONS

No evidence was found for B. burgdorferi infection in patients with morphoea.

Identifying populations potentially exposed to agricultural pesticides using remote sensing and a Geographic Information System

Pesticides used in agriculture may cause adverse health effects among the population living near agricultural areas. However, identifying the populations most likely to be exposed is difficult. We conducted a feasibility study to determine whether satellite imagery could be used to reconstruct historical crop patterns. We used historical Farm Service Agency records as a source of ground reference data to classify a late summer 1984 satellite image into crop species in a three-county area in south central Nebraska. Residences from a population-based epidemiologic study of non-Hodgkin lymphoma were located on the crop maps using a geographic information system (GIS). Corn, soybeans, sorghum, and alfalfa were the major crops grown in the study area. Eighty-five percent of residences could be located, and of these 22% had one of the four major crops within 500 m of the residence, an intermediate distance for the range of drift effects from pesticides applied in agriculture. We determined the proximity of residences to specific crop species and calculated crop-specific probabilities of pesticide use based on available data. This feasibility study demonstrated that remote sensing data and historical records on crop location can be used to create historical crop maps. The crop pesticides that were likely to have been applied can be estimated when information about crop-specific pesticide use is available. Using a GIS, zones of potential exposure to agricultural pesticides and proximity measures can be determined for residences in a study.

Using remotely sensed data to identify areas at risk for hantavirus pulmonary syndrome

The 1993 U.S. hantavirus pulmonary syndrome (HPS) outbreak was attributed to environmental conditions and increased rodent populations caused by unusual weather in 1991- 92. In a case-control study to test this hypothesis, we estimated precipitation at 28 HPS and 170 control sites during the springs of 1992 and 1993 and compared it with precipitation during the previous 6 years by using rainfall patterns at 196 weather stations. We also used elevation data and Landsat Thematic Mapper satellite imagery collected the year before the outbreak to estimate HPS risk by logistic regression analysis. Rainfall at case sites was not higher during 1992-93 than in previous years. However, elevation, as well as satellite data, showed association between environmental conditions and HPS risk the following year. Repeated analysis using satellite imagery from 1995 showed substantial decrease in medium- to high-risk areas. Only one case of HPS was identified in 1996.

Computer-generated dot maps as an epidemiologic tool: investigating an outbreak of toxoplasmosis

We used computer-generated dot maps to examine the spatial distribution of 94 Toxoplasma gondii infections associated with an outbreak in British Columbia, Canada. The incidence among patients served by one water distribution system was 3.52 times that of patients served by other sources. Acute T. gondii infection among 3, 812 pregnant women was associated with the incriminated distribution system.

[Maps of epidemiological rates: a Bayesian approach]

This article presents statistical methods recently developed for the analysis of maps of disease rates when the geographic units have small populations at risk. They adopt the Bayesian approach and use intensive computational methods for estimating risk in each area. The objective of the methods is to separate the variability of rates due to differences between regions from the background risk due to pure random fluctuation. Risk estimates have a total mean quadratic error smaller than usual estimates. We apply these new methods to estimate infant mortality risk in the municipalities of the State of Minas Gerais in 1994.

Using geographic information systems to assess spatial patterns of drug use, selection bias and attrition among a sample of injection drug users

This study sought to assess whether frequency and type of drug use are geographically located within the city of Baltimore independent of neighborhood characteristics. The second goal was to assess geographic factors associated with sample selection and attrition. The sample consisted of 597 inner-city injection drug users who were enrolled in a HIV prevention study. The residential locations were plotted using Geographic Information Systems (GIS) software. Three patterns of drug use in the prior 6 months were examined: daily use of injection heroin, daily use of injection cocaine and any use of crack cocaine. Daily use of cocaine and any use of crack were found to be statistically associated with residing in the western portion of the city and distance from the western district sexually transmitted disease clinic. After adjusting for individual level characteristics and neighborhood level variables, as measured by 1990 census tract data, daily use of cocaine was found to be associated with residing in a more southern area of the city and distance from the western district sexually transmitted disease clinic, and any crack use was found to be associated with residing in a more western area of the city and distance from the western district sexually transmitted disease clinic. Men and younger participants were more likely to reside further away from the study clinic as were individuals who dropped out of the intervention condition. The results of this study suggest that type and frequency of drug use is associated with specific geographic areas, independent of neighborhood characteristics. These results have implications for the location of drug prevention, needle exchange and other HIV prevention activities.

The public health risks of Lyme disease in Breckland, U.K.: an investigation of environmental and social factors

This paper considers the public health risks of Lyme disease, a borrelial infection transmitted to humans chiefly by nymphal Ixodes ticks. A study undertaken in the Breckland area of East Anglia, U.K., combined analysis of the spatial and temporal factors affecting tick activity at recreational sites with a survey of current levels of disease awareness among visitors to these locations. Significant relationships were found between densities of questing ticks and vegetation type, relative humidity and temperature. More than two thirds of the general public visiting the sites were aware ticks could carry diseases, but only 13% recognized an unfed nymph, and under half knew that Lyme disease could be contracted from tick bites. Such results need to be taken into account when formulating public health and education measures.

Geographic information systems: their use in environmental epidemiologic research

Advances in geographic information system (GIS) technology, developed by geographers, provide new opportunities for environmental epidemiologists to study associations between environmental exposures and the spatial distribution of disease. A GIS is a powerful computer mapping and analysis technology capable of integrating large quantities of geographic (spatial) data as well as linking geographic with nongeographic data (e.g., demographic information, environmental exposure levels). In this paper we provide an overview of some of the capabilities and limitations of GIS technology; we illustrate, through practical examples, the use of several functions of a GIS including automated address matching, distance functions, buffer analysis, spatial query, and polygon overlay; we discuss methods and limitations of address geocoding, often central to the use of a GIS in environmental epidemiologic research; and we suggest ways to facilitate its use in future studies. Collaborative efforts between epidemiologists, biostatisticians, environmental scientists, GIS specialists, and medical geographers are needed to realize the full potential of GIS technology in environmental health research and may lead to innovative solutions to complex questions.

On epidemiology and geographic information systems: a review and discussion of future directions

Geographic information systems are powerful automated systems for the capture, storage, retrieval, analysis, and display of spatial data. While the systems have been in development for more than 20 years, recent software has made them substantially easier to use for those outside the field. The systems offer new and expanding opportunities for epidemiology because they allow an informed user to choose between options when geographic distributions are part of the problem. Even when used minimally, these systems allow a spatial perspective on disease. Used to their optimum level, as tools for analysis and decision making, they are indeed a new information management vehicle with a rich potential for public health and epidemiology.