rflexscan {rflexscan}

Detect spatial disease clusters using the flexible/circular scan statistic

Description

This function analyzes spatial count data using the flexible spatial scan statistic developed by Tango and Takahashi (2005) or Kulldorff's circular spatial scan statistic (1997), and detect spatial disease clusters.

Usage

rflexscan(
  x,
  y,
  lat,
  lon,
  name,
  observed,
  expected,
  population,
  nb,
  clustersize = 15,
  radius = 6370,
  stattype = "ORIGINAL",
  scanmethod = "FLEXIBLE",
  ralpha = 0.2,
  simcount = 999,
  rantype = "MULTINOMIAL",
  comments = "",
  verbose = FALSE,
  secondary = NULL,
  clustertype = "HOT"
)

Arguments

x

A vector of X-coordinates.

y

A vector of Y-coordinates.

lat

(DEPRECATED) A vector of latitude.

lon

(DEPRECATED) A vector of longitude.

name

A vector of names of each area.

observed

A vector with the observed number of disease cases.

expected

A vector with the expected number of disease cases under the null hypothesis. This is used on "Poisson" model.

population

A vector with the background population at risk in each area. This is used on "Binomial" model.

nb

A neighbors list or an adjacency matrix.

clustersize

The number of maximum spatial cluster size to scan, i.e., the maximum number of regions included in the detected cluster

radius

Radius of Earth to calculate a distance between two sets of latitude and longitude. It is approximately 6370 km in Japan. This parameter is used when lat and lon are specified. This is DEPRECATED. The distance calculated using this parameter is not accurate. This feature is implemented to maintain compatibility with FleXScan. It is recommended to transform latitude and longitude onto the Cartesian coordinate system beforehand and use the x and y parameters that are projected coordinates.

stattype

Statistic type to be used (case-insensitive).

"ORIGINAL"

the likelihood ratio statistic by Kulldorff and Nagarwalla (1995)

"RESTRICTED"

the restricted likelihood ratio statistic by Tango (2008), with a preset parameter ralpha for restriction

scanmethod

Scanning method to be used (case-insensitive).

"FLEXIBLE"

flexible scan statistic by Tango and Takahashi (2005)

"CIRCULAR"

circular scan statistic by Kulldorff (1997)

ralpha

Threshold parameter of the middle p-value for the restricted likelihood ratio statistic.

simcount

The number of Monte Carlo replications to calculate a p-value for statistical test.

rantype

The type of random number for Monte Carlo simulation (case-insensitive).

"MULTINOMIAL"

Total number of cases in whole area is fixed. It can be chosen in either Poisson or Binomial model.

"POISSON"

Total number of cases is not fixed. It can be chosen in Poisson model.

comments

Comments for the analysis which will be written in summary.

verbose

Print progress messages.

secondary

The number of secondary clusters to be enumerated. If NULL is specified (default), the search for secondary clusters is stopped when the Monte Carlo p-value reaches 1.

clustertype

Type of cluster to be scanned.

"HOT"

Hot-spot clusters with elevated risk.

"COLD"

Cold-spot clusters with reduced risk.

"BOTH"

Hot- and cold-spot clusters simultaneously.

Details

Centroid coordinates for each region should be specified EITHER by Cartesian coordinates using arguments x and y or by latitudes and longitudes using arguments lat and lon. Note that lat and lon are DEPRECATED due to accuracy issues. This feature is implemented to maintain compatibility with FleXScan software. We recommend to transform latitude and longitude onto the Cartesian coordinate system beforehand (using spTransform function in sp package, for example) and use the x and y parameters that are projected coordinates.

Value

An rflexscan object which contains analysis results and specified parameters.

References

Otani T. and Takahashi K. (2021). Flexible scan statistics for detecting spatial disease clusters: The rflexscan R package, Journal of Statistical Software 99:13.

Tango T. and Takahashi K. (2005). A flexibly shaped spatial scan statistic for detecting clusters, International Journal of Health Geographics 4:11.

Kulldorff M. and Nagarwalla N. (1995). Spatial disease clusters: Detection and Inference. Statistics in Medicine 14:799-810.

Kulldorff M. (1997). A spatial scan statistic. Communications in Statistics: Theory and Methods, 26:1481-1496.

Tango T. (2008). A spatial scan statistic with a restricted likelihood ratio. Japanese Journal of Biometrics 29(2):75-95.

See Also

summary.rflexscan, plot.rflexscan, choropleth

Examples

# load sample data (North Carolina SIDS data)
library(spdep)
data("nc.sids")

# calculate the expected numbers of cases
expected <- nc.sids$BIR74 * sum(nc.sids$SID74) / sum(nc.sids$BIR74)

# run FleXScan
fls <- rflexscan(x = nc.sids$x, y = nc.sids$y,
                 observed = nc.sids$SID74,
                 expected = expected,
                 name = rownames(nc.sids),
                 clustersize = 10,
                 nb = ncCR85.nb)

# print rflexscan object
print(fls)

# print properties of the most likely cluster
print(fls$cluster[[1]])

# print summary to the terminal
summary(fls)

# plot graph
plot(fls, col = palette())
labs <- 1:length(fls$cluster)
legend("bottomleft", legend = labs, col = palette(), lty = 1)