1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
|
\name{epi.about}
\alias{epi.about}
\title{The library epiR: summary information}
\description{
Tools for the analysis of epidemiological data.
}
\usage{
epi.about()
}
\details{
More information about the \code{epiR} package can be found at \code{https://mvs.unimelb.edu.au/research/groups/veterinary-epidemiology-melbourne} and \code{https://www.ausvet.com.au/}.
}
\section{FREQUENTLY ASKED QUESTIONS}{
I have some diagnostic test results. How do I work out the expected number of true positive and false positive results? See \code{\link{epi.fpos}}: Compute the number of true positives, false positives, true negatives and false negatives given the number of individuals tested, the design prevalence and diagnostic test sensitivity and specificity.
How do I work out a sample size to estimate the prevalence of disease? If simple random sampling has been used see \code{\link{epi.sssimpleestb}}. If two-stage cluster design sampling has been used see \code{\link{epi.ssclus2estb}}.
How do I work out diagnostic test sensitivity and specificity from a set of test results? See \code{\link{epi.tests}}.
How do I work out the true prevalence of disease based on results from an imperfect diagnostic test? See \code{\link{epi.prev}}.
How do I calculate an odds ratio and incidence risk ratio on the basis of data presented in a 2 times 2 table? See \code{\link{epi.2by2}}.
How do I calculate a confidence interval for an incidence risk estimate? See \code{\link{epi.conf}}.
I've collected samples from a population and have tested them --- all returning a negative result. What is the maximum prevalence of disease in this population if disease is actually present? See \code{\link{rsu.pstar}}.
I've collected samples from a population and have tested them --- all returning a negative result. What is the probability that the prevalence of disease is less than or equal to a specified design prevalence? See \code{\link{rsu.sep}}.
How do I calculate the standardised mortality ratio for a single study area? See the \code{\link{epi.smr}} function. Example 1 in \code{\link{epi.bohning}} shows you how to calculate an expected value. To calculate SMRs and their confidence intervals for a series of study areas (instead of one) see the \code{\link{epi.conf}} function using \code{ctype = "smr"}.
}
\section{FUNCTIONS AND DATASETS}{
The following is a summary of the main functions and datasets in the \pkg{epiR} package. An alphabetical list of all functions and datasets is available by typing \code{library(help = epiR)}.
For further information on any of these functions, type \code{help(name)} or \code{?name} where \code{name} is the name of the function or dataset.
For details on how to use \pkg{epiR} for routine epidemiological work start R, type \code{help.start()} to open the help browser and navigate to \code{Packages > epiR > Vignettes}.
}
\section{CONTENTS:}{
The functions in \pkg{epiR} can be categorised into two main groups: tools for epidemiological analysis and tools for the analysis of surveillance data. A summary of the package functions is as follows:
}
\section{I. EPIDEMIOLOGY}{
\subsection{1. Descriptive statistics}{
\tabular{ll}{
\code{\link{epi.conf}} \tab Confidence intervals. \cr
\code{\link{epi.descriptives}} \tab Descriptive statistics. \cr
}
}
\subsection{2. Measures of health and measures of association}{
\tabular{ll}{
\code{\link{epi.2by2}} \tab Measures of association from data presented in a 2 by 2 table. \cr
\code{\link{epi.directadj}} \tab Directly adjusted incidence rate estimates. \cr
\code{\link{epi.edr}} \tab Compute estimated dissemination ratios from outbreak event data. \cr
\code{\link{epi.empbayes}} \tab Empirical Bayes estimates of observed event counts. \cr
\code{\link{epi.indirectadj}} \tab Indirectly adjusted incidence risk estimates. \cr
\code{\link{epi.insthaz}} \tab Instantaneous hazard estimates based on Kaplan-Meier survival estimates. \cr
\code{\link{epi.smr}} \tab Confidence intervals and tests of statistical significance of the standardised mortality [morbidity] ratio. \cr
}
}
\subsection{3. Diagnostic tests}{
\tabular{ll}{
\code{\link{epi.betabuster}} \tab An R version of Wes Johnson and Chun-Lung Su's Betabuster. \cr
\code{\link{epi.blcm.paras}} \tab Number of parameters to be inferred and number of informative priors for a BLCA. \cr
\code{\link{epi.fpos}} \tab Calculate the expected number of false positive and false negative test results. \cr
\code{\link{epi.herdtest}} \tab Estimate the characteristics of diagnostic tests applied at the herd (group) level. \cr
\code{\link{epi.nomogram}} \tab Compute the post-test probability of disease given characteristics of a diagnostic test. \cr
\code{\link{epi.pooled}} \tab Estimate herd test characteristics when samples are pooled. \cr
\code{\link{epi.prev}} \tab Compute the true prevalence of a disease in a population on the basis of an imperfect test. \cr
\code{\link{epi.tests}} \tab Sensitivity, specificity and predictive value of a diagnostic test. \cr
}
}
\subsection{4. Meta-analysis}{
\tabular{ll}{
\code{\link{epi.dsl}} \tab Mixed-effects meta-analysis of binary outcome data using the DerSimonian and Laird method. \cr
\code{\link{epi.iv}} \tab Fixed-effects meta-analysis of binary outcome data using the inverse variance method. \cr
\code{\link{epi.mh}} \tab Fixed-effects meta-analysis of binary outcome data using the Mantel-Haenszel method. \cr
\code{\link{epi.smd}} \tab Fixed-effects meta-analysis of continuous outcome data using the standardised mean difference method. \cr
}
}
\subsection{5. Regression analysis tools}{
\tabular{ll}{
\code{\link{epi.cp}} \tab Extract unique covariate patterns from a data set. \cr
\code{\link{epi.cpresids}} \tab Compute covariate pattern residuals from a logistic regression model. \cr
\code{\link{epi.interaction}} \tab Relative excess risk due to interaction in a case-control study. \cr
}
}
\subsection{6. Data manipulation tools}{
\tabular{ll}{
\code{\link{epi.asc}} \tab Write matrix to an ASCII raster file. \cr
\code{\link{epi.convgrid}} \tab Convert British National Grid georeferences to easting and northing coordinates. \cr
\code{\link{epi.dms}} \tab Convert decimal degrees to degrees, minutes and seconds and vice versa. \cr
\code{\link{epi.ltd}} \tab Calculate lactation to date and standard lactation (that is, 305 or 270 day) milk yields. \cr
\code{\link{epi.offset}} \tab Create an offset vector based on a list suitable for WinBUGS. \cr
\code{\link{epi.RtoBUGS}} \tab Write data from an R list to a text file in WinBUGS-compatible format. \cr
}
}
\subsection{7. Sample size calculations}{
The naming convention for the sample size functions in \pkg{epiR} is: \code{epi.ss} (sample size) + an abbreviation to represent the sampling design (e.g., \code{simple}, \code{strata}, \code{clus1}, \code{clus2}) + an abbreviation of the objectives of the study (\code{est} when you want to estimate a population parameter or \code{comp} when you want to compare two groups) + a single letter defining the outcome variable type (\code{b} for binary, \code{c} for continuous and \code{s} for survival data).
\tabular{ll}{
\code{\link{epi.sssimpleestb}} \tab Sample size to estimate a binary outcome using simple random sampling. \cr
\code{\link{epi.sssimpleestc}} \tab Sample size to estimate a continuous outcome using simple random sampling. \cr
\tab \cr
\code{\link{epi.ssstrataestb}} \tab Sample size to estimate a binary outcome using stratified random sampling. \cr
\code{\link{epi.ssstrataestc}} \tab Sample size to estimate a continuous outcome using stratified random sampling. \cr
\tab \cr
\code{\link{epi.ssclus1estb}} \tab Sample size to estimate a binary outcome using one-stage cluster sampling. \cr
\code{\link{epi.ssclus1estc}} \tab Sample size to estimate a continuous outcome using one-stage cluster sampling. \cr
\tab \cr
\code{\link{epi.ssclus2estb}} \tab Sample size to estimate a binary outcome using two-stage cluster sampling. \cr
\code{\link{epi.ssclus2estc}} \tab Sample size to estimate a continuous outcome using two-stage cluster sampling. \cr
\tab \cr
\code{\link{epi.ssxsectn}} \tab Sample size, power or detectable prevalence ratio for a cross-sectional study. \cr
\code{\link{epi.sscohortc}} \tab Sample size, power or detectable risk ratio for a cohort study using count data. \cr
\code{\link{epi.sscohortt}} \tab Sample size, power or detectable risk ratio for a cohort study using time at risk data. \cr
\code{\link{epi.sscc}} \tab Sample size, power or detectable odds ratio for case-control studies. \cr
\tab \cr
\code{\link{epi.sscompb}} \tab Sample size, power and detectable risk ratio when comparing binary outcomes. \cr
\code{\link{epi.sscompc}} \tab Sample size, power and detectable risk ratio when comparing continuous outcomes. \cr
\code{\link{epi.sscomps}} \tab Sample size, power and detectable hazard when comparing time to event. \cr
\tab \cr
\code{\link{epi.ssequb}} \tab Sample size for a parallel equivalence or equality trial, binary outcome. \cr
\code{\link{epi.ssequc}} \tab Sample size for a parallel equivalence or equality trial, continuous outcome. \cr
\tab \cr
\code{\link{epi.sssupb}} \tab Sample size for a parallel superiority trial, binary outcome. \cr
\code{\link{epi.sssupc}} \tab Sample size for a parallel superiority trial, continuous outcome. \cr
\tab \cr
\code{\link{epi.ssninfb}} \tab Sample size for a non-inferiority trial, binary outcome. \cr
\code{\link{epi.ssninfc}} \tab Sample size for a non-inferiority trial, continuous outcome. \cr
\tab \cr
\code{\link{epi.ssdetect}} \tab Sample size to detect an event. \cr
\code{\link{epi.ssdxsesp}} \tab Sample size to estimate the sensitivity or specificity of a diagnostic test. \cr
\code{\link{epi.ssdxtest}} \tab Sample size to validate a diagnostic test in the absence of a gold standard. \cr
}
}
\subsection{8. Miscellaneous functions}{
\tabular{ll}{
\code{\link{epi.prcc}} \tab Compute partial rank correlation coefficients. \cr
\code{\link{epi.psi}} \tab Compute proportional similarity indices. \cr
\code{\link{epi.realrisk}} \tab Return absolute risks from odds, incidence risk and hazard ratios. \cr
}
}
\subsection{9. Data sets}{
\tabular{ll}{
\code{\link{epi.epidural}} \tab Rates of use of epidural anaesthesia in trials of caregiver support. \cr
\code{\link{epi.incin}} \tab Laryngeal and lung cancer cases in Lancashire 1974 - 1983. \cr
\code{\link{epi.SClip}} \tab Lip cancer in Scotland 1975 - 1980. \cr
}
}
}
\section{II. SURVEILLANCE}{
Below, SSe stands for surveillance system sensitivity. That is, the average probability that a surveillance system (as a whole) will return a positive surveillance outcome, given disease is present in the population at a level equal to or greater than a specified design prevalence.
\subsection{1. Representative sampling --- sample size}{
\tabular{ll}{
\code{\link{rsu.sspfree.rs}} \tab Defined probability of disease freedom.\cr
\code{\link{rsu.sssep.rs}} \tab SSe, perfect test specificity. \cr
\code{\link{rsu.sssep.rs2st}} \tab SSe, two stage sampling. \cr
\code{\link{rsu.sssep.rsfreecalc}} \tab SSe, imperfect test specificity. \cr
\code{\link{rsu.sssep.rspool}} \tab SSe, pooled sampling. \cr
}
}
\subsection{2. Representative sampling --- surveillance system sensitivity and specificity}{
\tabular{ll}{
\code{\link{rsu.sep.rs}} \tab SSe, representative sampling. \cr
\code{\link{rsu.sep.rs2st}} \tab SSe, representative two-stage sampling. \cr
\code{\link{rsu.sep.rsmult}} \tab SSe, representative multiple surveillance components. \cr
\code{\link{rsu.sep.rsfreecalc}} \tab SSe, imperfect test specificity. \cr
\code{\link{rsu.sep.rspool}} \tab SSe, representative pooled sampling. \cr
\code{\link{rsu.sep.rsvarse}} \tab SSe, varying surveillance unit sensitivity. \cr
\code{\link{rsu.spp.rs}} \tab Surveillance system specificity. \cr
}
}
\subsection{3. Representative sampling --- probability of disease freedom}{
\tabular{ll}{
\code{\link{rsu.pfree.rs}} \tab Probability of disease freedom for a single or multiple time periods. \cr
\code{\link{rsu.pfree.equ}} \tab Equilibrium probability of disease freedom. \cr
}
}
\subsection{4. Risk-based sampling --- sample size}{
\tabular{ll}{
\code{\link{rsu.sssep.rbsrg}} \tab SSe, single sensitivity for each risk group. \cr
\code{\link{rsu.sssep.rbmrg}} \tab SSe, multiple sensitivities within risk groups. \cr
\code{\link{rsu.sssep.rb2st1rf}} \tab SSe, 2 stage sampling, 1 risk factor. \cr
\code{\link{rsu.sssep.rb2st2rf}} \tab SSe, 2 stage sampling, 2 risk factors. \cr
}
}
\subsection{5. Risk-based sampling --- surveillance system sensitivity and specificity}{
\tabular{ll}{
\code{\link{rsu.sep.rb}} \tab SSe, risk-based sampling. \cr
\code{\link{rsu.sep.rb1rf}} \tab SSe, risk-based sampling, 1 risk factor. \cr
\code{\link{rsu.sep.rb2rf}} \tab SSe, risk-based sampling, 2 risk factors. \cr
\code{\link{rsu.sep.rbvarse}} \tab SSe, risk-based sampling, varying unit sensitivity. \cr
\code{\link{rsu.sep.rb2st}} \tab SSe, 2-stage risk-based sampling. \cr
}
}
\subsection{6. Risk-based sampling --- probability of disease freedom}{
\tabular{ll}{
\code{\link{rsu.pfree.equ}} \tab Equilibrium probability of disease freedom. \cr
}
}
\subsection{7. Census sampling --- surveillance system sensitivity}{
\tabular{ll}{
\code{\link{rsu.sep.cens}} \tab SSe, census sampling. \cr
}
}
\subsection{8. Passive surveillance --- surveillance system sensitivity}{
\tabular{ll}{
\code{\link{rsu.sep.pass}} \tab SSe, passive surveillance. \cr
}
}
\subsection{9. Miscellaneous functions}{
\tabular{ll}{
\code{\link{rsu.adjrisk}} \tab Adjusted risk values. \cr
\code{\link{rsu.dxtest}} \tab Series and parallel diagnostic test interpretation. \cr
\code{\link{rsu.epinf}} \tab Effective probability of disease. \cr
\code{\link{rsu.pstar}} \tab Design prevalence back calculation. \cr
\code{\link{rsu.sep}} \tab Probability disease is less than specified design prevalence. \cr
}
}
}
\author{
Mark Stevenson (\email{mark.stevenson1@unimelb.edu.au}), Melbourne Veterinary School, Faculty of Science, The University of Melbourne, Parkville Victoria 3010, Australia.
Evan Sergeant (\email{evansergeant@gmail.com}), Ausvet Pty Ltd, Level 1 34 Thynne St, Bruce ACT 2617, Australia.
Simon Firestone, Melbourne Veterinary School, Faculty of Science, The University of Melbourne, Parkville Victoria 3010, Australia.
Telmo Nunes, UISEE/DETSA, Faculdade de Medicina Veterinaria --- UTL, Rua Prof. Cid dos Santos, 1300 - 477 Lisboa Portugal.
Javier Sanchez, Atlantic Veterinary College, University of Prince Edward Island, Charlottetown Prince Edward Island, C1A 4P3, Canada.
Ron Thornton, Ministry for Primary Industries New Zealand, PO Box 2526 Wellington, New Zealand.
With contributions from: Cord Heuer, Jonathon Marshall, Jeno Reiczigel, Jim Robison-Cox, Paola Sebastiani, Peter Solymos, Yoshida Kazuki, Geoff Jones, Sarah Pirikahu, Ryan Kyle, Johann Popp, Methew Jay, Allison Cheung, Nagendra Singanallur, Aniko Szabo and Ahmad Rabiee.
}
\keyword{univar}
|
