epi.about: The library epiR: summary information

In epiR: Tools for the Analysis of Epidemiological Data

The library epiR: summary information

Description

Tools for the analysis of epidemiological data.

Usage

epi.about()

Details

More information about the epiR package can be found at https://mvs.unimelb.edu.au/research/groups/veterinary-epidemiology-melbourne and https://www.ausvet.com.au/.

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 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 epi.sssimpleestb. If two-stage cluster design sampling has been used see epi.ssclus2estb.

How do I work out diagnostic test sensitivity and specificity from a set of test results? See epi.tests.

How do I work out the true prevalence of disease based on results from an imperfect diagnostic test? See 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 epi.2by2.

How do I calculate a confidence interval for an incidence risk estimate? See 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 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 rsu.sep.

How do I calculate the standardised mortality ratio for a single study area? See the epi.smr function. Example 1 in 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 epi.conf function using ctype = "smr".

FUNCTIONS AND DATASETS

The following is a summary of the main functions and datasets in the epiR package. An alphabetical list of all functions and datasets is available by typing library(help = epiR).

For further information on any of these functions, type help(name) or ?name where name is the name of the function or dataset.

For details on how to use epiR for routine epidemiological work start R, type help.start() to open the help browser and navigate to Packages > epiR > Vignettes.

CONTENTS:

The functions in 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:

I. EPIDEMIOLOGY

1. Descriptive statistics

epi.conf	Confidence intervals.
epi.descriptives	Descriptive statistics.

2. Measures of health and measures of association

epi.2by2	Measures of association from data presented in a 2 by 2 table.
epi.directadj	Directly adjusted incidence rate estimates.
epi.edr	Compute estimated dissemination ratios from outbreak event data.
epi.empbayes	Empirical Bayes estimates of observed event counts.
epi.indirectadj	Indirectly adjusted incidence risk estimates.
epi.insthaz	Instantaneous hazard estimates based on Kaplan-Meier survival estimates.
epi.smr	Confidence intervals and tests of statistical significance of the standardised mortality [morbidity] ratio.

3. Diagnostic tests

epi.betabuster	An R version of Wes Johnson and Chun-Lung Su's Betabuster.
epi.blcm.paras	Number of parameters to be inferred and number of informative priors for a BLCA.
epi.fpos	Calculate the expected number of false positive and false negative test results.
epi.herdtest	Estimate the characteristics of diagnostic tests applied at the herd (group) level.
epi.nomogram	Compute the post-test probability of disease given characteristics of a diagnostic test.
epi.pooled	Estimate herd test characteristics when samples are pooled.
epi.prev	Compute the true prevalence of a disease in a population on the basis of an imperfect test.
epi.tests	Sensitivity, specificity and predictive value of a diagnostic test.

4. Meta-analysis

epi.dsl	Mixed-effects meta-analysis of binary outcome data using the DerSimonian and Laird method.
epi.iv	Fixed-effects meta-analysis of binary outcome data using the inverse variance method.
epi.mh	Fixed-effects meta-analysis of binary outcome data using the Mantel-Haenszel method.
epi.smd	Fixed-effects meta-analysis of continuous outcome data using the standardised mean difference method.

5. Regression analysis tools

epi.cp	Extract unique covariate patterns from a data set.
epi.cpresids	Compute covariate pattern residuals from a logistic regression model.
epi.interaction	Relative excess risk due to interaction in a case-control study.

6. Data manipulation tools

epi.asc	Write matrix to an ASCII raster file.
epi.convgrid	Convert British National Grid georeferences to easting and northing coordinates.
epi.dms	Convert decimal degrees to degrees, minutes and seconds and vice versa.
epi.ltd	Calculate lactation to date and standard lactation (that is, 305 or 270 day) milk yields.
epi.offset	Create an offset vector based on a list suitable for WinBUGS.
epi.RtoBUGS	Write data from an R list to a text file in WinBUGS-compatible format.

7. Sample size calculations

The naming convention for the sample size functions in epiR is: epi.ss (sample size) + an abbreviation to represent the sampling design (e.g., simple, strata, clus1, clus2) + an abbreviation of the objectives of the study (est when you want to estimate a population parameter or comp when you want to compare two groups) + a single letter defining the outcome variable type (b for binary, c for continuous and s for survival data).

epi.sssimpleestb	Sample size to estimate a binary outcome using simple random sampling.
epi.sssimpleestc	Sample size to estimate a continuous outcome using simple random sampling.
epi.ssstrataestb	Sample size to estimate a binary outcome using stratified random sampling.
epi.ssstrataestc	Sample size to estimate a continuous outcome using stratified random sampling.
epi.ssclus1estb	Sample size to estimate a binary outcome using one-stage cluster sampling.
epi.ssclus1estc	Sample size to estimate a continuous outcome using one-stage cluster sampling.
epi.ssclus2estb	Sample size to estimate a binary outcome using two-stage cluster sampling.
epi.ssclus2estc	Sample size to estimate a continuous outcome using two-stage cluster sampling.
epi.ssxsectn	Sample size, power or detectable prevalence ratio for a cross-sectional study.
epi.sscohortc	Sample size, power or detectable risk ratio for a cohort study using count data.
epi.sscohortt	Sample size, power or detectable risk ratio for a cohort study using time at risk data.
epi.sscc	Sample size, power or detectable odds ratio for case-control studies.
epi.sscompb	Sample size, power and detectable risk ratio when comparing binary outcomes.
epi.sscompc	Sample size, power and detectable risk ratio when comparing continuous outcomes.
epi.sscomps	Sample size, power and detectable hazard when comparing time to event.
epi.ssequb	Sample size for a parallel equivalence or equality trial, binary outcome.
epi.ssequc	Sample size for a parallel equivalence or equality trial, continuous outcome.
epi.sssupb	Sample size for a parallel superiority trial, binary outcome.
epi.sssupc	Sample size for a parallel superiority trial, continuous outcome.
epi.ssninfb	Sample size for a non-inferiority trial, binary outcome.
epi.ssninfc	Sample size for a non-inferiority trial, continuous outcome.
epi.ssdetect	Sample size to detect an event.
epi.ssdxsesp	Sample size to estimate the sensitivity or specificity of a diagnostic test.
epi.ssdxtest	Sample size to validate a diagnostic test in the absence of a gold standard.

8. Miscellaneous functions

epi.prcc	Compute partial rank correlation coefficients.
epi.psi	Compute proportional similarity indices.
epi.realrisk	Return absolute risks from odds, incidence risk and hazard ratios.

9. Data sets

epi.epidural	Rates of use of epidural anaesthesia in trials of caregiver support.
epi.incin	Laryngeal and lung cancer cases in Lancashire 1974 - 1983.
epi.SClip	Lip cancer in Scotland 1975 - 1980.

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.

1. Representative sampling — sample size

rsu.sspfree.rs	Defined probability of disease freedom.
rsu.sssep.rs	SSe, perfect test specificity.
rsu.sssep.rs2st	SSe, two stage sampling.
rsu.sssep.rsfreecalc	SSe, imperfect test specificity.
rsu.sssep.rspool	SSe, pooled sampling.

2. Representative sampling — surveillance system sensitivity and specificity

rsu.sep.rs	SSe, representative sampling.
rsu.sep.rs2st	SSe, representative two-stage sampling.
rsu.sep.rsmult	SSe, representative multiple surveillance components.
rsu.sep.rsfreecalc	SSe, imperfect test specificity.
rsu.sep.rspool	SSe, representative pooled sampling.
rsu.sep.rsvarse	SSe, varying surveillance unit sensitivity.
rsu.spp.rs	Surveillance system specificity.

3. Representative sampling — probability of disease freedom

rsu.pfree.rs	Probability of disease freedom for a single or multiple time periods.
rsu.pfree.equ	Equilibrium probability of disease freedom.

4. Risk-based sampling — sample size

rsu.sssep.rbsrg	SSe, single sensitivity for each risk group.
rsu.sssep.rbmrg	SSe, multiple sensitivities within risk groups.
rsu.sssep.rb2st1rf	SSe, 2 stage sampling, 1 risk factor.
rsu.sssep.rb2st2rf	SSe, 2 stage sampling, 2 risk factors.

5. Risk-based sampling — surveillance system sensitivity and specificity

rsu.sep.rb	SSe, risk-based sampling.
rsu.sep.rb1rf	SSe, risk-based sampling, 1 risk factor.
rsu.sep.rb2rf	SSe, risk-based sampling, 2 risk factors.
rsu.sep.rbvarse	SSe, risk-based sampling, varying unit sensitivity.
rsu.sep.rb2st	SSe, 2-stage risk-based sampling.

6. Risk-based sampling — probability of disease freedom

rsu.pfree.equ	Equilibrium probability of disease freedom.

7. Census sampling — surveillance system sensitivity

rsu.sep.cens	SSe, census sampling.

8. Passive surveillance — surveillance system sensitivity

rsu.sep.pass	SSe, passive surveillance.

9. Miscellaneous functions

rsu.adjrisk	Adjusted risk values.
rsu.dxtest	Series and parallel diagnostic test interpretation.
rsu.epinf	Effective probability of disease.
rsu.pstar	Design prevalence back calculation.
rsu.sep	Probability disease is less than specified design prevalence.

Author(s)

Mark Stevenson ([e-mail]), Melbourne Veterinary School, Faculty of Science, The University of Melbourne, Parkville Victoria 3010, Australia.

Evan Sergeant ([e-mail]), 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.

epiR documentation built on April 12, 2025, 1:31 a.m.