A study of three techniques for time‐space clustering in hodgkin's disease

Chen, Rina; Mantel, Nathan; Klingberg, Marcus A.

doi:10.1002/sim.4780030210

Cited by 53 publications

(21 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We are unable to support this conclusion for the following reasons: (1) it is unlikely that MS did not exist in the Faroes before 1940; (2) the apparent excess of cases whose dis ease onset occurred either during or shortly after the occupation by the British troops is: (a) irrelevant, since the date of acquisition of MS (age [5][6][7][8][9][10][11][12][13][14] is the date of importance in determining whether or not an epidemic took place; (b) probably inaccurate, since disease onset is frequently difficult to evalu ate; (c) of borderline statistical significance only when 1 questionable case of MS is accepted, as determined by three different statistical techniques and overlapping 95% confidence intervals of incidence rates of MS by either date of onset or acquisition; (3) it is unclear how exposure to the British troops is defined; the importance of the occupation is difficult to evaluate without a clear under standing of this concept.…”

Section: Discussionmentioning

confidence: 73%

“…the relation ship between known cases). Cluster analysis could not be pursued with the Faroese data for two reasons: first, there are inadequate details on the geographic location of the cases at the time of either disease acquisition or onset, and no information on MS cases subsequently diagnosed amongst the occupy ing British troops; second, the validity of any results calculated using such a small sample size would be questionable, as demonstrated by a recent study of clustering of Hodgkin's disease cases, in which the sensitivity of these methods to detect clusters in samples of 600 or less was low [7], Efficient methods [8][9][10][11][12][13][14][15] have been de veloped for detecting nonrandom clusters of points on a line. These methods have been used to evaluate apparent temporal clusters of disease, to determine whether the cluster might represent an epidemic or whether it might have occurred by chance.…”

Section: Temporal Analysismentioning

confidence: 98%

See 1 more Smart Citation

Analysis of the ‘Epidemic' of Multiple Sclerosis in the Faroe Islands

Hibberd²

1988

Neuroepidemiology

View full text Add to dashboard Cite

A series of biostatistical tests were applied to the data upon which the Faroe epidemic hypothesis was constructed. Temporal, cluster analysis using three commonly applied methods were unsuitable because of the very small size of the sample (32 cases of MS). Methods to detect non-random clusters of disease identified a cluster of 16 cases with date of onset between 1941–1950 (p < 0.05). However, when 1 questionable case of MS was excluded, the resulting cluster did not support the epidemic hypothesis. The overlapping 95% confidence limits of 5- and 10-year based incidence rates of MS, by date of onset (Poisson distribution) argues against the presence of an epidemic. Analysis of the exposure data, i.e. contacts between the Faroese and the British troops, yielded only borderline statistical significance but changing a single case of MS from the exposed to the unexposed category failed to support the exposure theory. All calculations were also carried out using the probable date of acquisition of the disease, between the ages of 5 and 14 years; these also failed to substantiate the hypotheses of epidemic and transmissibility. Since the validity of statistical analysis to test for the presence of epidemics with small sample sizes is not well established, the lack of concordance between the results of the various approaches leads to the conclusion that there was no epidemic of MS in the Faroes.

show abstract

Section: Discussionmentioning

confidence: 73%

Section: Temporal Analysismentioning

confidence: 98%

Analysis of the ‘Epidemic' of Multiple Sclerosis in the Faroe Islands

Hibberd²

1988

Neuroepidemiology

View full text Add to dashboard Cite

show abstract

“…This applies particularly when location is taken as residence at diagnosis as in this study. It also makes space-time interaction methods inappropriate (Chen et al, 1984). For this reason our intention was to investigate spatial clustering only, although the limited time period available for analysis will cause the study to have elements of space-time interaction and hence lack power.…”

Section: Resultsmentioning

confidence: 99%

A specialist leukaemia/lymphoma registry in the UK. Part 2: clustering of Hodgkin's disease

Alexander

Williams²,

McKinney³

et al. 1989

Br J Cancer

View full text Add to dashboard Cite

Summary Part 1 describes the epidemiology of Hodgkin's disease occurring in those parts of the United Kingdom which are included in the Leukaemia Research Fund data collection survey. A total of 1,023 cases diagnosed between 1984 and 1986 were available for analysis. At county and district levels there was little heterogeneity in the distribution of cases. However, at the electoral ward level there were real differences for the younger age group . In this paper methods of investigation which are not dependent on census boundaries are applied and the presence of localised spatial clustering is confirmed. There is some evidence that the pattern of clustering relates to the nodular sclerosing subtype. These results are related to hypotheses of an infectious aetiology.The distribution of Hodgkin's disease has raised questions of long-standing interest to epidemiologists. The unusual agedistribution was first noted by MacMahon (1966), who suggested that it might result from a combination of two distinct diseases. From the histological appearance, clinical course and anatomical distribution of the lesions as well as the age distribution he postulated an infectious aetiology in those aged less than 35 and an aetiology similar to other lymphomas in cases over 50. Subsequent research is in broad agreement with the two disease hypothesis (Roush, 1987;Gutensohn, 1982). Evidence for a viral aetiology comes from anecdotal reports of micro-clusters and of case linkage (Vianna et al., 1971;Grufferman et al., 1977), from school cohort studies (Vianna & Polan, 1973), formal analyses of space-time interaction (Kryscio et al., 1973;Greenberg et al., 1983), case-control studies of social linkage (Zack et al., 1977;Scherr et al., 1984), and follow-up of cases of infectious mononucleosis (Rosendahl, 1974; Muellar, 1987). Although the evidence is inconclusive and somewhat conflicting there is a consensus that an infectious aetiology is more likely for younger cases.The Leukaemia Research Fund data collection survey has been described (McKinney et al., 1989) and the basic descriptive epidemiology of Hodgkin's disease reported. This showed some evidence of localised aggregations of cases, or 'clustering', for the younger age group. The present study concentrates on spatial clustering using more sophisticated methods of analysis which do not depend on arbitrary administrative boundaries. MethodsIncidence data have been taken from the Leukaemia Research Fund data collection survey (DCS) described earlier (McKinney et al., 1989 al., 1989;Besag, 1989) and by control locations in the second (Cuzick & Edwards, 1989). Both provide global tests of the extent of localised spatial clustering. The NNA method involves consideration of each case in turn. We first identify the two nearest cases and then estimate the population from which these cases derived. If the expected number in this population is significantly small (P<0.05) then the case is the centre of a small local aggregation and is said to be 'clustered'. In a random distribution approxi...

show abstract

“…For space-time clusters, suitable methodology has been available for several years (Knox, 1964). Although appropriate for acute infectious diseases, it has low statistical power for chronic disease with long and variable latent periods (Chen et al, 1984). For these, a study of spatial clustering is more relevant and suitable methodology is now available and validated (Draper, 1991;.…”

mentioning

confidence: 99%

Spatial clustering of childhood leukaemia: summary results from the EUROCLUS project

Alexander

Boyle²,

Carli³

et al. 1998

Br J Cancer

View full text Add to dashboard Cite

Summary The interpretation of reports of clusters of childhood leukaemia is difficult, first because little is known about the causes of the disease, and second because there is insufficient information on whether cases show a generalized tendency to cluster geographically. The EUROCLUS project is a European collaborative study whose primary objective is to determine whether the residence locations of cases at diagnosis show a general tendency towards spatial clustering. The second objective is to interpret any patterns observed and, in particular, to see if clustering can be explained in terms of either infectious agents or environmental hazards as aetiological agents. The spatial distribution of 13 351 cases of childhood leukaemia diagnosed in 17 countries between 1980 and 1989 has been analysed using the Potthoff-Whittinghill method. The overall results show statistically significant evidence of clustering of total childhood leukaemia within small census areas (P= 0.03) but the magnitude of the clustering is small (extra-Poisson component of variance (%) = 1.7 with 90% confidence interval 0.2-3.1). The clustering is most marked in areas that have intermediate population density (150-499 persons km-2). It cannot be attributed to any specific age group at diagnosis or cell type and involves spatial aggregation of cases of different ages and cell types. The results indicate that intense clusters are a rare phenomenon that merit careful investigation, although aetiological insights are more likely to come from investigation of large numbers of cases. We present a method for detecting clustering that is simple and readily available to cancer registries and similar groups.

show abstract

A study of three techniques for time‐space clustering in hodgkin's disease

Cited by 53 publications

References 12 publications

Analysis of the ‘Epidemic' of Multiple Sclerosis in the Faroe Islands

Analysis of the ‘Epidemic' of Multiple Sclerosis in the Faroe Islands

A specialist leukaemia/lymphoma registry in the UK. Part 2: clustering of Hodgkin's disease

Spatial clustering of childhood leukaemia: summary results from the EUROCLUS project

Contact Info

Product

Resources

About