Andrew Cliff

Civil war and the spread of AIDS in Central Africa.

Using ordinary least squares regression techniques this paper demonstrates, for the first time, that the classic association of war and disease substantially accounts for the presently observed geographical distribution of reported clinical AIDS cases in Uganda. Both the spread of HIV 1 infection in the 1980s, and the subsequent development of AIDS to its 1990 spatial pattern, are shown to be significantly and positively correlated with ethnic patterns of recruitment into the Ugandan National Liberation Army (UNLA) after the overthrow of Idi Amin some 10 years earlier in 1979. This correlation reflects the estimated mean incubation period of 8-10 years for HIV 1 and underlines the need for cognizance of historical factors which may have influenced current patterns of AIDS seen in Central Africa. The findings may have important implications for AIDS forecasting and control in African countries which have recently experienced war. The results are compared with parallel analyses of other HIV hypotheses advanced to account for the reported geographical distribution of AIDS in Uganda.

Evaluating the friction of distance parameter in gravity models

Cliff A. D., Martin R. L. and Ord J. K. (1974) Evaluating the friction of distance parameter in gravity models, Reg. Studies 8, 281–286. In this paper we consider whether map pattern (spatial autocorrelation) among the population values in constrained and unconstrained gravity models calibrated by least squares regression hinders interpretation of the coefficient of the distance term. It is shown that, except in certain cases which are most likely to occur in intra-, rather than inter-, urban models, no real problem of interpretation should arise. This finding differs to some extent from those of Curry (1972) and Johnston (1973). The conclusions are supported by both theoretical and empirical results.

Avian influenza A (H5N1) age distribution in humans.

To the Editor: A total of 229 confirmed human cases of avian influenza A (H5N1) were reported to the World Health Organization (WHO) from 10 countries of Africa, Asia, and Europe in the 30 months leading up to July 4, 2006 (1). WHO has highlighted the skewed age distribution of these confirmed cases toward children and young adults, with relatively few cases in older age categories (2). An explanation for this age bias is currently lacking, although a range of behavioral, biological, demographic, and data-related factors may account for the observed pattern (2,3).

The Spatial Dynamics of Poliomyelitis in the United States: From Epidemic Emergence to Vaccine-Induced Retreat, 1910-1971.

Abstract This article seeks to advance an understanding of the spatial dynamics of one of the great emergent viral diseases of the twentieth century—poliomyelitis. From an apparently rare clinical condition occurring only sporadically or in small outbreaks before the late nineteenth century, poliomyelitis had, by the early 1950s, developed into a globally distributed epidemic disease. But, from 1955, continued growth was suddenly and dramatically reversed by the mass administration of inactivated (killed) and live (attenuated) poliovirus vaccines. After almost half a century of vaccine control, the world now stands on the brink of the global eradication of the disease. Against this background, the article draws upon information included in the U.S. Public Health Services Public Health Reports and the U.S. Centers for Disease Control and Preventions Morbidity and Mortality Weekly Report to examine the spatial dynamics of poliomyelitis during the phases of epidemic emergence (1910–1955) and vaccine-induced retreat (1955–1971) in the United States. It is shown that epidemic emergence was accompanied by shifts in the spatial center of activity from early diffusion poles in the northeastern states, to the western seaboard, and then finally to cover all the states of the Union. This was accompanied by accelerating epidemic propagation. The introduction of mass vaccination from the mid-1950s realigned spatial transmission of the disease, producing increased spatial volatility in the geographical center of activity and heightened dependence of epidemic outbreaks upon endemic reservoirs in the most populous states. Finally, the empirical results are generalized to suggest that the emergence and reemergence of many infectious diseases is a distinctively geographical process.

Epidemic Diffusion Processes in a System of U.S. Military Camps: Transfer Diffusion and the Spread of Typhoid Fever in the Spanish-American War, 1898

This article examines the geographical transmission of an epidemic disease in the makeshift encampments of a mobilized army. The choice of location (the southern and eastern United States), the time (an eight-month period of mobilization in the U.S. war with Spain, May to December 1898) and the epidemic disease (typhoid fever) are conditioned by the reports collated by three eminent figures in the history of medicine–Major Walter Reed, Major Victor C. Vaughan, and Major Edward O. Shakespeare–and published in the historic 1904 Report on the Origin and Spread of Typhoid Fever in U.S. Military Camps During the Spanish War of 1898. The Report includes information on the daily occurrence of some 19,000 cases of typhoid fever in 89 volunteer regiments of the U.S. Army. When geo-coded to the locations of army camps, this information is used (i) to reconstruct the spread of typhoid fever with the campwise movements of infected regiments and (ii) to model the diffusion process that drove the spread of the disease. To handle the high degree of regimental mobility in the camp system, a novel process referred to as transfer diffusion is introduced. It is shown that, for the entire system of camps, the spread of typhoid was underpinned by a temporally ordered sequence of diffusion processes in which each process was associated with a discrete stage of the epidemic cycle. The processes reflected the spatial transfer of regiments (epidemic buildup), the geographical proximity of camps (epidemic peak), and the position of camps in the transient population size hierarchy (epidemic fadeout). Importantly, however, the strength and timing of the processes varied by corps. While the findings underscore the singular impact of military mobilization on the spatial dynamics of epidemic diseases, it is suggested that the particular approach adopted in this article may be extended to the study of epidemic transmission in analogous camp settings.

The spatial anatomy of an epidemic: influenza in London and the county boroughs of England and Wales, 1918–1919

From uncertain origins in the spring of 1918, an apparently new variant of influenza A virus spread around the world as three distinct diffusion waves, infecting half a billion and probably killing around 40 million people. This paper examines the spatial structure of influenza transmission during the ten–month course of the epidemic in England and Wales, June 1918–April 1919, using the weekly counts of influenza deaths in London and the county boroughs as collated by the General Register Office, London. In addition, a particular case study of the borough of Cambridge is presented. From mid–1916, Cambridge contained, as well as its undergraduate population, a large naval contingent billeted in both the colleges and the town. It therefore affords the opportunity of studying the effect of the epidemic in contiguous groups with widely differing demographic characteristics. Through the application of a range of statistical methods (average lags, correlations and regressions), it is shown that the three waves that comprised the pandemic had fundamentally different spatial and temporal characteristics. The first, moving through a population that was a virgin soil to the new virus strain, was explosive in its north to south progress across the country. The second wave was somewhat slower in its rate of diffusion and displayed a south to north drift. Finally, the third wave reverted more closely to the form of the first. The spread of all three waves, however, was underpinned by a clearly defined process of spatial contagion. The Cambridge study showed the special characteristics of this pandemic in terms of the ages of those attacked: high rates were experienced across the age spectrum, a feature also seen internationally.

Statistical modelling of measles and influenza outbreaks

This paper reviews the application of statistical models to outbreaks of two common respiratory viral diseases, measles and influenza. For each disease, we look first at its epidemiological characteristics and assess the extent to which these either aid or hinder modelling. We then turn to the models that have been developed to simulate geographical spread. For measles, a distinction is drawn between process-based and time series models; for influenza, it is the scale of the communities (from small groups to global populations) which primarily determines modelling style. Applications are provided from work by the authors, largely using Icelandic data. Finally we consider the forecasting potential of the models described.

Acquired Immune Deficiency Syndrome (AIDS): literature, geographical origins and global patterns

The year 1979 was a watershed in the epidemiological history of the planet. Following a concerted campaign by the World Health Organization (WHO), the last naturally occurring case of smallpox had been tracked down in Somalia in October 1977. After a two-year period in which no other cases (apart from laboratory accidents) were recorded, WHO formally announced in December 1979 that the global eradication of smallpox was complete. In the same year, a patient in a New York hospital was in the final stages of a long and mysterious illness characterized by repeated attacks of chance infections which are usually easily repelled by the body. Cross-checks with other US hospitals revealed not only an alarming number of similar cases, but also a steep rise in hospital admissions of patients suffering from illnesses which pointed to collapse of the immune system. Because of this, the new disease was called Acquired Immune Deficiency Syndrome or AIDS. Thus simultaneously with the elimination of one of the historical scourges of mankind, smallpox, the world stood on the edge of another great pandemic.

A swash–backwash model of the single epidemic wave

While there is a large literature on the form of epidemic waves in the time domain, models of their structure and shape in the spatial domain remain poorly developed. This paper concentrates on the changing spatial distribution of an epidemic wave over time and presents a simple method for identifying the leading and trailing edges of the spatial advance and retreat of such waves. Analysis of edge characteristics is used to (a) disaggregate waves into ‘swash’ and ‘backwash’ stages, (b) measure the phase transitions of areas from susceptible, S, through infective, I, to recovered, R, status (S → I → R) as dimensionless integrals and (c) estimate a spatial version of the basic reproduction number, R0. The methods used are illustrated by application to measles waves in Iceland over a 60-year period from 1915 to 1974. Extensions of the methods for use with more complex waves are possible through modifying the threshold values used to define the start and end points of an event.

Computing the Spatial Correspondence between Geographical Patterns

A Sign test for evaluating the goodness-of-fit between maps of area-based data is described. The test is based upon measures of contiguity and is used to determine the degree of correspondence between some maps of observed and simulated patterns of adopters of an innovation given by T. Hagerstrand (I953). The observed and simulated patterns are shown to have systematic spatial differences, reasons for which are given. GEOGRAPHICAL research often requires estimates to be made of the amount of agreement between patterns shown on maps. The patterns may consist of points, lines, areas, intensities, flows, or any combination of these elements (W. Tobler, I965). As examples of the kinds of problem which arise, Tobler states (p. 131) that it might be required to compute the spatial correspondence between the railway and road patterns of the United States; or the percentage agreement between the observed distribution of cities within Europe and Christallers theoretical arrangement of central places. The comparison of an observed pattern with a theoretically derived pattern is of considerable importance (W. W. Bunge, I963), since it is usually undertaken in order to verify how closely the model from which the theoretical pattern was obtained conforms to reality. Several recent studies (H. H. McCarty et al., 1956; A. H. Robinson and R. A. Bryson, I957; R. Bachi, 1962; R. F. Minnick, I964; Tobler, 1965; D. Neft, 1966) have focused upon methods for computing the degree of spatial correspondence between maps of line or point patterns. The problem of comparing maps of area-based or county data has, however, been neglected. The purpose of this paper is to provide a test which measures the spatial correspondence between observed and theoretically derived maps of county data.