Title: Estimate the Burden of Healthcare-Associated Infections
Version: 0.99.2
Author: Benedikt Zacher [aut, cre]
Maintainer: Benedikt Zacher <ZacherB@rki.de>
Description: Provides an approach which is based on the methodology of the Burden of Communicable Diseases in Europe (BCoDE) and can be used for large and small samples such as individual countries. The Burden of Healthcare-Associated Infections (BHAI) is estimated in disability-adjusted life years, number of infections as well as number of deaths per year. Results can be visualized with various plotting functions and exported into tables.
Depends: R (≥ 3.6.0)
License: GPL-3
LazyData: true
RoxygenNote: 6.1.1
Imports: prevtoinc, MCMCpack, plotrix, graphics, grDevices, stats, methods
NeedsCompilation: no
Packaged: 2019-10-01 13:58:13 UTC; zacherb
Repository: CRAN
Date/Publication: 2019-10-06 10:20:02 UTC

BHAI:

Description

The BHAI package

BHAI functions

bhai:

Create a PPS object

Description

This function creates a PPS object.

Usage

PPS(num_hai_patients = NULL, num_survey_patients = NULL,
  length_of_stay = NULL, loi_pps = NULL, hospital_discharges = NULL,
  num_hai_patients_by_stratum = NULL,
  num_hai_patients_by_stratum_prior = NULL, mccabe_scores_distr = NULL,
  mccabe_by_stratum_prior = NULL, mccabe_life_exp = NULL,
  num_survey_patients_by_stratum = NULL, population = NULL,
  country = "")

Arguments

num_hai_patients

Named numeric containing patients having healthcare-associated infections.

num_survey_patients

Number of patients in point prevalence survey.

length_of_stay

Length of stay of all patients in hospitals. This is need for the prevalence to incidence conversion with the Rhame-Sudderth formula.

loi_pps

A list containing length of infections from all patients in the PPS. The length of infection of all healthcare-associated infections. In PPS this is usually approximated as the time from infection onset until the date of the survey.

hospital_discharges

The number of hospital discharges.

num_hai_patients_by_stratum

A list containing for each infection the number of patients in each age and gender stratum.

num_hai_patients_by_stratum_prior

The prior weight (counts) for each infection, age and gender stratum. This is used for smooting the age and gender distribution when small numbers are observed.

mccabe_scores_distr

The observed McCabe scores (counts) for each infection, age and gender stratum from the PPS.

mccabe_by_stratum_prior

The prior weight (counts) for each infection, McCabe score, age and gender stratum. This is used for smooting the age and gender distribution when small numbers are observed.

mccabe_life_exp

Named list containing remaining life expectancies for each McCabe score (NONFATAL, ULTFATAL, RAPFATAL).

num_survey_patients_by_stratum

Number of survey patients stratified by infection, age and gender. If this parameter is provided the methodology described in Cassini et al. (2016) <doi:https://doi.org/10.1371/journal.pmed.1002150> is applied.

population

Population size.

country

Name of the country.

Value

A PPS class object.

See Also

PPS

Examples


data(german_pps_2011_repr)
german_pps_repr = PPS(num_hai_patients = num_hai_patients,
    num_hai_patients_by_stratum = num_hai_patients_by_stratum,
    num_hai_patients_by_stratum_prior = num_hai_patients_by_stratum_prior,
    num_survey_patients = num_survey_patients,
    length_of_stay = length_of_stay,
    loi_pps = loi_pps,
    mccabe_scores_distr = mccabe_scores_distr,
    mccabe_life_exp = mccabe_life_exp,
    hospital_discharges = hospital_discharges,
    population = population,
    country="Germany (representative sample)")
german_pps_repr

This class is a generic container for PPS data sets.

Description

This class is a generic container for PPS data sets.

Slots

infections

Character vector storing names of infections in PPS

num_hai_patients

Named numeric containing patients having healthcare-associated infections.

num_survey_patients

Number of patients in point prevalence survey.

length_of_stay

Length of stay of all patients in hospitals. This is need for the prevalence to incidence conversion with the Rhame-Sudderth formula.

loi_pps

A list containing length of infections from all patients in the PPS. In PPS this is usually calculated as the time from infection onset until the date of the survey.

hospital_discharges

The number of hospital discharges.

num_hai_patients_by_stratum

A list containing for each infection the number of patients in each age and gender stratum.

num_hai_patients_by_stratum_prior

The prior weight (counts) for each infection, age and gender stratum. This is used for smooting the age and gender distribution when small numbers are observed.

mccabe_scores_distr

The observed McCabe scores (counts) for each infection, age and gender stratum from the PPS.

mccabe_by_stratum_prior

The prior weight (counts) for each infection, McCabe score, age and gender stratum. This is used for smooting the age and gender distribution when small numbers are observed.

mccabe_life_exp

Named list containing remaining life expectancies for each McCabe score (NONFATAL, ULTFATAL, RAPFATAL).

num_survey_patients_by_stratum

Number of survey patients stratified by infection, age and gender. If this parameter is provided the methodology described in Cassini et al. (2016) <doi:https://doi.org/10.1371/journal.pmed.1002150> is applied.

population

Population size

country

Name of the country in which PPS was conducted

bhai_options

Options with which bhai was run. If bhai was not run yet, this is an empty list.

bhai_summary

Summary statistics of bhai. If bhai was not run yet, this is an empty list.

Main function of the package to estimation of the burden of healthcare-associated infections

Description

Estimation of the burden of healthcare-associated infections

Usage

bhai(pps, nsim = 1000, pop.sampling = TRUE,
  sample_distr = "rbetamix", estimate_loi_fun = bootstrap_mean_gren,
  stratified_sampling = FALSE, summarize_strata = TRUE,
  use_prior = TRUE)

## S4 method for signature 'PPS'
bhai(pps, nsim = 1000, pop.sampling = TRUE,
  sample_distr = "rbetamix", estimate_loi_fun = bootstrap_mean_gren,
  stratified_sampling = FALSE, summarize_strata = TRUE,
  use_prior = TRUE)

Arguments

pps

The PPS object containing the data.

nsim

Number of Monte Carlo simulations, default: 1000.

pop.sampling

Specifying whether parameters of the disease outcome trees should be sampled on population level, default: TRUE.

sample_distr

Distribution used for prevalence sampling, default: "rbetamix".

estimate_loi_fun

Function used for estimation of the length of infection, default: bootstrap_mean_gren (recommended!).

stratified_sampling

Specifying whether stratified sampling should be done.

summarize_strata

Specifying whether stratum-specific summary statistics should be computed.

use_prior

Specifying whether Prior distributions should be used for computations.

Value

A PPS class object.

See Also

PPS

Examples


data(german_pps_2011_repr)
german_pps_repr = PPS(num_hai_patients = num_hai_patients,
    num_hai_patients_by_stratum = num_hai_patients_by_stratum,
    num_hai_patients_by_stratum_prior = num_hai_patients_by_stratum_prior,
    num_survey_patients = num_survey_patients,
    length_of_stay = length_of_stay,
    loi_pps = loi_pps,
    mccabe_scores_distr = mccabe_scores_distr,
    mccabe_life_exp = mccabe_life_exp,
    hospital_discharges = hospital_discharges,
    population = population,
    country="Germany (representative sample)")
german_pps_repr

set.seed(3)
# The following example is run only for illustratory reasons
# Note that you should never run the function with only 10 Monte-Carlo simulations in practice!
bhai(german_pps_repr, nsim=10)

Barplot of cases, deaths and DALYs.

Description

Barplot of cases, deaths and DALYs.

Usage

bhai.barplot(..., what, infections=NULL, cols1=NULL, cols2=NULL, ylab=NULL, ylim=NULL, 
legend_labs=NULL, main="", names.inf=TRUE, cex.names=1, border=par("fg"), lwd.errors=2)

Arguments

...

Further plotting arguments

what

One of c("Cases", "Deaths", "DALY")

infections

If sepcified only a subset of infections in bhai_summary is plotted.

cols1

Color used to fill the bars.

cols2

Specifies colors of YLDs when plotting DALYs.

ylab

Y-axis labels.

ylim

Limits of y-axis.

legend_labs

Labels of legend.

main

Title of plot

names.inf

Specifying whether names of infections should be plotted.

cex.names

Font size of labels.

border

The color to be used for the border of the bars, default: par("fg").

lwd.errors

Line width of error bars.

See Also

PPS

Examples


data(german_pps_2011_repr)
german_pps_repr = PPS(num_hai_patients = num_hai_patients,
    num_hai_patients_by_stratum = num_hai_patients_by_stratum,
    num_hai_patients_by_stratum_prior = num_hai_patients_by_stratum_prior,
    num_survey_patients = num_survey_patients,
    length_of_stay = length_of_stay,
    loi_pps = loi_pps,
    mccabe_scores_distr = mccabe_scores_distr,
    mccabe_life_exp = mccabe_life_exp,
    hospital_discharges = hospital_discharges,
    population = population,
    country="Germany (representative sample)")
german_pps_repr

set.seed(3)
# The following example is run only for illustratory reasons
# Note that you should never run the function with only 10 Monte-Carlo simulations in practice!
result_ger = bhai(german_pps_repr, nsim=10)

bhai.barplot(result_ger, what="Cases")

Summary plot of number of infections, deaths and DALYs

Description

Summary plot of number of infections, deaths and DALYs

Usage

bhai.circleplot(pps, infections=NULL, main="", xlim=NULL, ylim=NULL)

Arguments

pps

The PPS object containing the data.

infections

Infections to be plotted.

main

Title of plot.

xlim

Limits of x-axis.

ylim

Limits of y-axis.

See Also

PPS

Examples


data(german_pps_2011_repr)
german_pps_repr = PPS(num_hai_patients = num_hai_patients,
    num_hai_patients_by_stratum = num_hai_patients_by_stratum,
    num_hai_patients_by_stratum_prior = num_hai_patients_by_stratum_prior,
    num_survey_patients = num_survey_patients,
    length_of_stay = length_of_stay,
    loi_pps = loi_pps,
    mccabe_scores_distr = mccabe_scores_distr,
    mccabe_life_exp = mccabe_life_exp,
    hospital_discharges = hospital_discharges,
    population = population,
    country="Germany (representative sample)")
german_pps_repr

set.seed(3)
# The following example is run only for illustratory reasons
# Note that you should never run the function with only 10 Monte-Carlo simulations in practice!
result = bhai(german_pps_repr, nsim=10)
bhai.circleplot(pps=result)

Create summary table

Description

Create BHAI summary table

Usage

bhai.prettyTable(pps, pop_norm=FALSE, conf.int=TRUE)

Arguments

pps

The PPS object containing the data.

pop_norm

Indicating whether statistics should be computed per 100,000 population, default: TRUE.

conf.int

Specifying whether confidence intervals should be computed, default: TRUE.

Value

A data.frame containing the summarised results.

See Also

PPS

Examples


data(german_pps_2011_repr)
german_pps_repr = PPS(num_hai_patients = num_hai_patients,
    num_hai_patients_by_stratum = num_hai_patients_by_stratum,
    num_hai_patients_by_stratum_prior = num_hai_patients_by_stratum_prior,
    num_survey_patients = num_survey_patients,
    length_of_stay = length_of_stay,
    loi_pps = loi_pps,
    mccabe_scores_distr = mccabe_scores_distr,
    mccabe_life_exp = mccabe_life_exp,
    hospital_discharges = hospital_discharges,
    population = population,
    country="Germany (representative sample)")
german_pps_repr

set.seed(3)
# The following example is run only for illustratory reasons
# Note that you should never run the function with only 10 Monte-Carlo simulations in practice!
result = bhai(german_pps_repr, nsim=10)
bhai.prettyTable(result)

Stratified barplot of cases, deaths and DALYs.

Description

Stratified barplot of cases, deaths and DALYs.

Usage

bhai.strataplot(pps, infection, what, col=NULL, errors=TRUE, lwd.errors=2, xlab=NULL, ...)

Arguments

pps

The PPS object containing the data.

infection

Infection to be plotted.

what

One of c("Cases", "Deaths", "DALY")

col

Color used to fill the bars.

errors

Specifying whether error bars should be plotted, default: TRUE.

lwd.errors

Line width of error bars.

xlab

X-axis labels.

...

Further plotting arguments

See Also

PPS

Examples

 
data(german_pps_2011_repr)
german_pps_repr = PPS(num_hai_patients = num_hai_patients,
    num_hai_patients_by_stratum = num_hai_patients_by_stratum,
    num_hai_patients_by_stratum_prior = num_hai_patients_by_stratum_prior,
    num_survey_patients = num_survey_patients,
    length_of_stay = length_of_stay,
    loi_pps = loi_pps,
    mccabe_scores_distr = mccabe_scores_distr,
    mccabe_life_exp = mccabe_life_exp,
    hospital_discharges = hospital_discharges,
    population = population,
    country="Germany (representative sample)")
german_pps_repr

set.seed(3)
# The following example is run only for illustratory reasons
# Note that you should never run the function with only 10 Monte-Carlo simulations in practice!
result = bhai(german_pps_repr, nsim=10)
bhai.strataplot(pps=result, infection="HAP", what="Cases")

Aggregated data of the ECDC PPS 2010-2011.

Description

Aggregated data of the ECDC PPS 2010-2011.

Usage

data(eu_pps_2011)

Format

A PPS object.

Aggregated data of the german PPS 2010-2011 (convenience sample).

Description

Aggregated data of the german PPS 2010-2011 (convenience sample).

Usage

data(german_pps_2011_conv)

Format

A PPS object.

Hospital discharges in Germany (2011)

Description

Hospital discharges in Germany (2011)

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Average length of stay of survey patients in german PPS 2011 (representative sample)

Description

Average length of stay of survey patients in german PPS 2011 (representative sample)

Usage

data(german_pps_2011_repr)

Format

A PPS object.

A list containing length of infections from all patients in the german PPS 2011 representative sample.

Description

A list containing length of infections from all patients in the german PPS 2011 representative sample.

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Named list containing remaining life expectancies for each McCabe score (NONFATAL, ULTFATAL, RAPFATAL).

Description

Named list containing remaining life expectancies for each McCabe score (NONFATAL, ULTFATAL, RAPFATAL).

Usage

data(german_pps_2011_repr)

Format

A PPS object.

The observed McCabe scores (counts) for each infection, age and gender stratum from the ECDC PPS 2011-2012.

Description

The observed McCabe scores (counts) for each infection, age and gender stratum from the ECDC PPS 2011-2012.

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Number of cases for each infection in the german PPS 2011 (representative sample)

Description

Number of cases for each infection in the german PPS 2011 (representative sample)

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Stratified number of cases for each infection in the german PPS 2011 (representative sample)

Description

Stratified number of cases for each infection in the german PPS 2011 (representative sample)

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Stratified number of cases for each infection in the german PPS 2011 (convenience sample). This distribution is used as a Prior for the representative sample.

Description

Stratified number of cases for each infection in the german PPS 2011 (convenience sample). This distribution is used as a Prior for the representative sample.

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Number of survey patients in the german PPS 2011 (representative sample).

Description

Number of survey patients in the german PPS 2011 (representative sample).

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Population size of Germany in 2011.

Description

Population size of Germany in 2011.

Usage

data(german_pps_2011_repr)

Format

A PPS object.

Simulate PPS data

Description

Simulate PPS data

Usage

sample.pps(pps_data, num_survey_patients)

Arguments

pps_data

The PPS object containing the data. Parameters for simulations are extracted from this data.

num_survey_patients

Numeric vector indicating sample sizes for simulations.

Value

A simulated PPS object.

See Also

PPS

Examples


# Specify the number of survey patients
sim_survey_patients = 10000
# Subsample data sets from european PPS
sim_pps = sample.pps(eu_pps, num_survey_patients = sim_survey_patients)

Simulated/subsampled data sets from european PPS

Description

Simulated/subsampled data sets from european PPS

Usage

data(simulations)

Format

A PPS object.

BHAI with default options was applied to simulated/subsampled data sets from european PPS

Description

BHAI with default options was applied to simulated/subsampled data sets from european PPS

Usage

data(simulations)

Format

A PPS object.

BHAI with prior was applied to simulated/subsampled data sets from european PPS

Description

BHAI with prior was applied to simulated/subsampled data sets from european PPS

Usage

data(simulations)

Format

A PPS object.

BHAI with stratified sampling was applied to simulated/subsampled data sets from european PPS

Description

BHAI with stratified sampling was applied to simulated/subsampled data sets from european PPS

Usage

data(simulations)

Format

A PPS object.