Skip to contents

Estimating Empirical Coverage with WISCA

Source: vignettes/WISCA.Rmd
WISCA.Rmd

Introduction

Clinical guidelines for empirical antimicrobial therapy require probabilistic reasoning: what is the chance that a regimen will cover the likely infecting organisms, before culture results are available?

This is the purpose of WISCA, or:

Weighted-Incidence Syndromic Combination Antibiogram

WISCA is a Bayesian approach that integrates: - Pathogen prevalence (how often each species causes the syndrome), - Regimen susceptibility (how often a regimen works if the pathogen is known),

to estimate the overall empirical coverage of antimicrobial regimens — with quantified uncertainty.

This vignette explains how WISCA works, why it is useful, and how to apply it in AMR.

Why traditional antibiograms fall short

A standard antibiogram gives you:

``` Species → Antibiotic → Susceptibility %

But clinicians don’t know the species a priori. They need to choose a regimen that covers the likely pathogens — without knowing which one is present.

Traditional antibiograms: - Fragment information by organism, - Do not weight by real-world prevalence, - Do not account for combination therapy or sample size, - Do not provide uncertainty.

The idea of WISCA

WISCA asks:

“What is the probability that this regimen will cover the pathogen, given the syndrome?”

This means combining two things: - Incidence of each pathogen in the syndrome, - Susceptibility of each pathogen to the regimen.

We can write this as:

``` coverage = ∑ (pathogen incidence × susceptibility)

For example, suppose: - E. coli causes 60% of cases, and 90% of E. coli are susceptible to a drug. - Klebsiella causes 40% of cases, and 70% of Klebsiella are susceptible.

Then:

``` coverage = (0.6 × 0.9) + (0.4 × 0.7) = 0.82

But in real data, incidence and susceptibility are estimated from samples — so they carry uncertainty. WISCA models this probabilistically, using conjugate Bayesian distributions.

The Bayesian engine behind WISCA

Pathogen incidence

Let: - K be the number of pathogens, - α = (1, 1, ..., 1) be a **Dirichlet** prior (uniform), - n = (n₁, …, nₖ) be the observed counts per species.

Then the posterior incidence follows:

``` incidence ∼ Dirichlet(α + n)

In simulations, we draw from this posterior using:

``` xᵢ ∼ Gamma(αᵢ + nᵢ, 1)

``` incidenceᵢ = xᵢ / ∑ xⱼ

Susceptibility

Each pathogen–regimen pair has: - prior: Beta(1, 1) - data: S susceptible out of N tested

Then:

``` susceptibility ∼ Beta(1 + S, 1 + (N - S))

In each simulation, we draw random susceptibility per species from this Beta distribution.

Final coverage estimate

Putting it together:

``` For each simulation: - Draw incidence ∼ Dirichlet - Draw susceptibility ∼ Beta - Multiply → coverage estimate

We repeat this (e.g. 1000×) and summarise: - Mean: expected coverage - Quantiles: credible interval (default 95%)

Practical use in AMR

Simulate a synthetic syndrome 

library(AMR)
data <- example_isolates

# Add a fake syndrome column for stratification
data$syndrome <- ifelse(data$mo %like% "coli", "UTI", "Other")

Basic WISCA antibiogram 

antibiogram(data,
            wisca = TRUE)
Amikacin Amoxicillin Amoxicillin/clavulanic acid Ampicillin Azithromycin Benzylpenicillin Cefazolin Cefepime Cefotaxime Cefoxitin Ceftazidime Ceftriaxone Cefuroxime Chloramphenicol Ciprofloxacin Clindamycin Colistin Doxycycline Erythromycin Flucloxacillin Fosfomycin Gentamicin Imipenem Kanamycin Linezolid Meropenem Metronidazole Moxifloxacin Mupirocin Nitrofurantoin Oxacillin Piperacillin/tazobactam Rifampicin Teicoplanin Tetracycline Tigecycline Tobramycin Trimethoprim Trimethoprim/sulfamethoxazole Vancomycin
41.7% (37.2-47.5%) 35.7% (33.3-38.2%) 73.7% (71.7-75.8%) 35.8% (33.6-38.1%) 43.8% (41.5-46%) 28.2% (25.8-30.8%) 69.3% (64.9-73.8%) 74.8% (70.5-78.8%) 73.3% (69.2-77.3%) 69.6% (65.5-73.7%) 35.9% (33.6-38.2%) 73.3% (68.9-77.2%) 71.9% (69.8-74%) 64.9% (51.7-78.5%) 77% (74.5-79.6%) 47.3% (44.7-49.6%) 33% (30.8-35.1%) 63.6% (52.1-74.9%) 43.7% (41.6-46%) 59.3% (47-71%) 60.5% (55.5-65.8%) 72.7% (70.7-74.8%) 78.2% (74-82.2%) 25.6% (13.5-37.7%) 54.9% (50.4-59%) 77.1% (72.8-81.2%) 56.1% (39.5-70.7%) 49.6% (43.6-55.6%) 65.2% (52.7-78.1%) 76.5% (69.4-82.3%) 57.8% (45.4-69.6%) 69.4% (64.2-74.2%) 52.4% (47.6-56.8%) 48.1% (43.4-52.9%) 61.4% (53.6-70.5%) 81.9% (78.1-85.5%) 60.7% (57.8-63.5%) 61% (58.8-63.5%) 76.5% (74.5-78.5%) 61.9% (59.8-64.2%)

Stratify by syndrome 

antibiogram(data,
            syndromic_group = "syndrome",
            wisca = TRUE)
Syndromic Group Amikacin Amoxicillin Amoxicillin/clavulanic acid Ampicillin Azithromycin Benzylpenicillin Cefazolin Cefepime Cefotaxime Cefoxitin Ceftazidime Ceftriaxone Cefuroxime Chloramphenicol Ciprofloxacin Clindamycin Colistin Doxycycline Erythromycin Flucloxacillin Fosfomycin Gentamicin Imipenem Kanamycin Linezolid Meropenem Metronidazole Moxifloxacin Mupirocin Nitrofurantoin Oxacillin Piperacillin/tazobactam Rifampicin Teicoplanin Tetracycline Tigecycline Tobramycin Trimethoprim Trimethoprim/sulfamethoxazole Vancomycin
Other 25% (20.2-31.7%) 31.6% (28.7-34%) 70.1% (67.7-72.4%) 31.6% (29.1-34.1%) 56.4% (53.8-58.8%) 36.3% (33.1-39.4%) 61.5% (55.7-66.5%) 68.5% (63.4-73.8%) 66.7% (61.4-71.9%) 63% (57.7-68.6%) 18.3% (15.9-20.8%) 66.6% (61.4-71.5%) 65.5% (62.7-68%) 69.6% (60-77.2%) 74% (70.8-77.2%) 60.9% (58.1-63.6%) 13.9% (11.8-15.8%) 67.4% (63.7-70.9%) 56.4% (54-58.9%) 61.4% (56-67.6%) 49.6% (43.2-56.3%) 65.6% (62.8-68.1%) 71.8% (66.7-77%) 18.6% (13.1-25.9%) 70.8% (65.1-75.8%) 70.6% (65.1-75.7%) 49.8% (34.2-66.6%) 63.3% (56.2-70.3%) 69.8% (62.6-76.4%) 70.5% (61.2-77.5%) 60% (54.4-65.4%) 62.4% (56.4-68.6%) 67.6% (61.9-73.2%) 61.9% (55.4-67.6%) 67.8% (64.8-70.6%) 77% (72.3-81.8%) 50.1% (46.7-53.6%) 61.1% (58.4-64%) 78.8% (76.4-80.9%) 79.6% (77.4-81.8%)
UTI 91.5% (88.8-93.5%) 50% (45.5-54.6%) 80.9% (77.8-84%) 49.9% (45.6-54.3%) 8.2% (6.4-10.5%) 8.2% (6.3-10.3%) 88.9% (84.2-92.3%) 89.4% (86.5-91.8%) 89.9% (87.4-92.1%) 86.1% (82.9-88.9%) 89.8% (87.2-91.9%) 89.8% (87.1-92.1%) 87.4% (84.5-89.8%) NA 81.4% (78.3-84.3%) 8.2% (6.3-10.4%) 91.7% (89.6-93.8%) NA 8.1% (6.3-10.4%) NA 90.6% (86.5-93.3%) 90.2% (87.9-92.2%) 91.8% (89.7-93.8%) NA 8.1% (6.1-10.2%) 91.8% (89.6-93.8%) 71.4% (31.8-91.6%) 9.3% (6.7-13.3%) NA 89.4% (86.9-91.7%) NA 87.2% (84.4-89.6%) 8.2% (6.3-10.4%) 8.2% (6.3-10.3%) 41.2% (14.3-74.4%) 90.9% (87.7-93.3%) 89.6% (87.1-91.8%) 59.1% (54.7-63.4%) 65.3% (61.3-69.2%) 8.2% (6.2-10.3%)

Use combination regimens

The antibiogram() function supports combination regimens:

antibiogram(data,
            antimicrobials = c("AMC", "GEN", "AMC + GEN", "CIP"),
            wisca = TRUE)
Amoxicillin/clavulanic acid Amoxicillin/clavulanic acid + Gentamicin Ciprofloxacin Gentamicin
73.8% (71.7-75.8%) 89.7% (88.2-91.2%) 77% (74.3-79.6%) 72.8% (70.6-74.9%)

Interpretation

Suppose you get this output:

Regimen Coverage Lower_CI Upper_CI
AMC 0.72 0.65 0.78
AMC + GEN 0.88 0.83 0.93

Interpretation:

“AMC + GEN covers 88% of expected pathogens for this syndrome, with 95% certainty that the true coverage lies between 83% and 93%.”

Regimens with few tested isolates will show wider intervals.

Sensible defaults, but you can customise

  • minimum = 30: exclude regimens with <30 isolates tested.
  • simulations = 1000: number of Monte Carlo samples.
  • conf_interval = 0.95: coverage interval width.
  • combine_SI = TRUE: count “I”/“SDD” as susceptible.

Limitations

  • WISCA does not model time trends or temporal resistance shifts.
  • It assumes data are representative of current clinical practice.
  • It does not account for patient-level covariates (yet).
  • Species-specific data are abstracted into syndrome-level estimates.

Reference

Bielicki JA et al. (2016).
Weighted-incidence syndromic combination antibiograms to guide empiric treatment in pediatric bloodstream infections.
J Antimicrob Chemother, 71(2):529–536. doi:10.1093/jac/dkv397

Conclusion

WISCA shifts empirical therapy from simple percent susceptible toward probabilistic, syndrome-based decision support. It is a statistically principled, clinically intuitive method to guide regimen selection — and easy to use via the antibiogram() function in the AMR package.

For antimicrobial stewardship teams, it enables disease-specific, reproducible, and data-driven guidance — even in the face of sparse data.