Visualizing Antimicrobial Resistance Trends

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-27

Quick Answer

Choose chart types by matching the data structure: MIC distributions use histograms or strip plots, resistance percentages use bar charts, and temporal trends use line charts.
Build dashboards with R Shiny or Tableau using a tidy data format where each row is one isolate and each column is one variable.
Interpret visualizations only within the context of your testing method, breakpoint version, and isolate source, since these factors change what the chart means.

At a Glance

Visualization Goal	Recommended Chart Type	Key Data Requirement	Common Pitfall
Show MIC distribution for one organism-drug pair	Histogram or strip plot with log2 scale	MIC values in doubling dilutions	Plotting raw MIC values on a linear axis obscures the modal peak
Compare resistance percentages across organisms	Grouped bar chart or dot plot	Percent susceptible, intermediate, resistant per organism	Omitting sample size, which makes small-isolate percentages misleading
Show temporal resistance trends	Line chart with confidence interval or loess smoothing	Year or month of isolation, resistance outcome	Connecting sparse time points with straight lines that imply certainty
Display geographic variation	Choropleth map or faceted bar chart	Region or facility identifier per isolate	Using raw counts instead of rates when isolate numbers differ by region
Explore multidrug resistance patterns	UpSet plot or heatmap with clustering	Resistance phenotype per isolate across multiple drugs	Using a Venn diagram for more than three drugs
Present a dashboard overview	Combination of KPI cards, trend line, and MIC histogram	Pre-aggregated summary table plus raw data	Overloading the first screen with too many panels

The Problem with Raw AMR Data

Antimicrobial resistance surveillance produces data that is inherently difficult to interpret without deliberate visualization choices. A typical dataset contains one row per bacterial isolate, with columns for the organism species, the specimen source, the collection date, the testing method, and the MIC value for each antimicrobial tested. The MIC value itself is not a continuous measurement. It is a doubling dilution series, often expressed as 0.25, 0.5, 1, 2, 4, 8, 16, and 32 micrograms per milliliter. The difference between 1 and 2 is the same biological step as the difference between 16 and 32, but a linear axis makes the first gap look small and the second gap look large.

The Clinical and Laboratory Standards Institute publishes the standard methods for antimicrobial susceptibility testing, and those methods define how the MIC is determined and how the result is interpreted. The interpretation categories, susceptible, intermediate, and resistant, depend on the breakpoints in the current standard. Breakpoints change over time. A visualization that labels an isolate as resistant using one breakpoint version may label the same isolate differently under a newer version. Any chart that does not record the breakpoint version is incomplete.

The World Health Organization describes antimicrobial resistance as a global public health concern that requires coordinated surveillance. The purpose of visualizing AMR data is to turn raw laboratory results into information that supports decisions. Those decisions include choosing empiric therapy, updating local treatment guidelines, detecting emerging resistance mechanisms, and evaluating the effect of stewardship interventions. A chart that is technically correct but visually misleading can lead to the wrong decision.

Core Principles for AMR Data Visualization

Match the Chart to the Question

The first decision is not which software to use. The first decision is which question the visualization must answer. Different questions require different chart types.

A question about the distribution of MIC values for one organism and one drug requires a histogram or a strip plot. The histogram shows how many isolates fall at each dilution step. The strip plot shows each isolate as a point, which is useful when the number of isolates is small. Both charts should use a logarithmic scale for the MIC axis because the dilution series is inherently logarithmic.

A question about the percentage of isolates that are resistant to a drug requires a bar chart or a dot plot. The bar chart is appropriate when the number of organisms is small. The dot plot is better when the number of organisms is large, because the dots reduce visual clutter. The chart should show the percentage resistant, and it should also show the number of isolates tested. A percentage based on 10 isolates is not comparable to a percentage based on 1,000 isolates.

A question about how resistance changes over time requires a line chart. The x-axis is the time period, usually the year or quarter of isolation. The y-axis is the percentage resistant. The line chart should include a measure of uncertainty, such as a confidence interval band or a loess curve, when the number of isolates per time period varies.

A question about the relationship between resistance to multiple drugs requires a heatmap or an UpSet plot. The heatmap shows the resistance phenotype for each isolate across all drugs. The UpSet plot shows the intersection of resistance phenotypes, such as the number of isolates that are resistant to both ampicillin and ciprofloxacin but susceptible to gentamicin.

The MICRO Scale Is Logarithmic

The MICRO is a dilution series. Each step in the series is a doubling of the previous step. The difference between 1 and 2 is the same biological step as the difference between 16 and 32. A linear axis distorts this relationship. The visual gap between 1 and 2 is tiny, while the gap between 16 and 32 is large, even though both gaps represent one dilution step.

The solution is to plot the MICRO on a logarithmic scale. In R, this is done with the scale_x_log10() function in ggplot2. In Tableau, this is done by changing the axis scale to logarithmic. The log scale makes the dilution steps evenly spaced, so the histogram shows the true modal MICRO and the spread of the distribution.

The modal MICRO is the most common MIC value in the distribution. It is the peak of the histogram. The modal MICRO is a useful summary because it describes the typical susceptibility of the population. The spread of the distribution describes the variability. A narrow distribution with a single peak suggests a homogeneous population. A wide distribution or a distribution with two peaks suggests a heterogeneous population, which may indicate the presence of a resistant subpopulation.

The Breakpoint Problem

The interpretation of an MIC value depends on the breakpoint. The breakpoint is the MIC value that separates the susceptible category from the intermediate category and the intermediate category from the resistant category. The Clinical and Laboratory Standards publishes breakpoints in its performance standards documents. These documents are updated periodically, and the breakpoints can change.

A visualization that shows the percentage resistant must state which breakpoint version was used. If the breakpoint version is not stated, the chart is not reproducible. If the breakpoint version changes between two time periods, the percentage resistant may change even if the MIC distribution is identical. This is a common source of false trends in AMR surveillance.

The solution is to record the breakpoint version in the data table and to include it in the chart title or the chart subtitle. The chart should also include the date of the breakpoint publication. This allows the reader to understand whether a change in the percentage resistant is a real change in the MIC distribution or an artifact of the breakpoint change.

The Denominator Matters

The percentage resistant is a proportion. The denominator is the number of isolates tested. A percentage based on a small denominator is unstable. A single isolate that changes from susceptible to resistant can change the percentage from 0 percent to 10 percent when the denominator is 10.

The visualization should always show the denominator. A bar chart can show the number of isolates above each bar. A dot plot can show the number of isolates in the tooltip. A line chart can show the number of isolates per time point as a secondary axis or as a label.

The American Type Culture Collection provides authenticated biological materials that are used as quality controls in susceptibility testing. The quality control strains have known MIC ranges for each drug. The results from the quality control strains are used to verify that the testing method is working correctly. A visualization that includes quality control results can show whether the testing method was stable over time. A shift in the quality control MIC distribution suggests a problem with the testing method, not a change in the clinical isolates.

Practical Workflow for Building an AMR Dashboard

Step 1: Define the Question and the Audience

The dashboard must answer a specific question for a specific audience. The audience for an AMR dashboard is usually one of three groups. The first group is the laboratory staff, who want to monitor the quality of the testing method. The second group is the infection control team, who want to monitor the resistance patterns in the hospital. The third group is the public health authority, who want to monitor the resistance patterns in the community.

The question determines the content of the dashboard. A dashboard for the laboratory staff should focus on quality control results and the distribution of MIC values. A dashboard for the infection control team should focus on the percentage resistant by organism and by specimen source. A dashboard for the public health authority should focus on the temporal trends and the geographic distribution.

Step 2: Prepare the Data

The data must be in a tidy format. Each row is one isolate. Each column is one variable. The variables include the isolate identifier, the organism, the specimen source, the collection date, the testing facility, and the MIC value for each drug.

The MIC values should be stored as numbers. The interpretation categories should be stored as separate columns. The breakpoint version should be stored as a separate column. The data should be checked for missing values and for duplicate entries.

The Bacteriological Analytical Manual from the U.S. Food and Drug Administration describes the official methods for the detection of foodborne pathogens. These methods include the isolation and the identification of the organisms. The susceptibility testing is performed after the organism is identified. The data from the susceptibility testing should be linked to the data from the isolation and the identification.

Step 3: Choose the Software

Two common software options are R Shiny and Tableau. R Shiny is a web application framework for the R programming language. It requires programming skills, but it offers the most flexibility. Tableau is a visual analytics platform that does not require programming. It is easier to learn, but it is less flexible.

The choice between the two depends on the skills of the team and the requirements of the dashboard. A team that already uses R for data analysis will find it easier to build a Shiny dashboard. A team that does not use R will find it easier to build a Tableau dashboard.

Step 4: Build the Dashboard Layout

The dashboard layout should follow a logical flow. The top of the dashboard should have the key performance indicators, such as the total number of isolates, the percentage resistant to a specific drug, and the number of organisms tested. The middle of the dashboard should have the main charts, such as the MIC distribution and the temporal trend. The bottom of the dashboard should have the detailed tables and the filters.

The filters are the most important part of the dashboard. The user should be able to filter by the organism, the specimen source, the time period, and the testing facility. The filters should be linked to all the charts on the dashboard. When the user selects a filter, all the charts should update.

Step 5: Validate the Dashboard

The dashboard must be validated before it is shared. The validation includes checking the data against the source records, checking the calculations against the known values, and checking the charts against the expected patterns.

The validation should be done by a second person who was not involved in the building of the dashboard. The second person should check the data for errors, the calculations for the correct formulas, and the charts for the correct interpretation.

Options and Tradeoffs in Chart Selection

Histogram versus Strip Plot for MIC Distributions

The histogram is the default choice for displaying a MIC distribution. The histogram groups the MIC values into bins and shows the count of isolates in each bin. The bin width should be one dilution step. The histogram is easy to read and works well with a large number of isolates.

The strip plot is an alternative to the histogram. The strip plot shows each isolate as a point on the axis. The strip plot is useful when the number of isolates is small, because the histogram bins may be too sparse. The strip plot is also useful when the user wants to see the individual isolates, such as when the isolates are colored by the specimen source.

The strip plot has a limitation. When the number of isolates is large, the points overlap and the plot becomes unreadable. The solution is to add a small amount of random noise to the position of the points, which is called jittering. The jittering makes the points visible, but it also makes the plot less precise.

Bar Chart versus Dot Plot for Resistance Percentages

The bar chart is the default choice for resistance percentages. The bar chart shows the percentage resistant for each organism or each drug. The bar chart is easy to read and the bars are easy to compare.

The dot plot is an alternative to the bar chart. The dot plot shows the percentage resistant as a point on the axis. The dot plot is useful when the number of organisms is large. The dot plot is also useful when the user wants to show the uncertainty in the percentage. The error bars can be added to the dot plot more easily than to the bar chart.

The dot plot has a limitation. The dot plot is less familiar to the general audience than the bar chart. The user may need to explain the dot plot to the audience.

Line Chart versus Loess Curve for Temporal Trends

The line chart is the default choice for temporal trends. The line chart connects the data points with a straight line. The line chart is easy to read and the trend is easy to see.

The loess curve is an alternative to the line chart. The loess curve is a smooth curve that is fitted to the data points. The loess curve is useful when the data is noisy and the user wants to see the underlying trend. The loess curve is also useful when the number of data points is small.

The loess curve has a limitation. The loess curve can overfit the data. The overfitting means that the curve follows the noise in the data instead of the underlying trend. The user should check the loess curve against the line chart to make sure that the curve is not overfitting.

Heatmap versus UpSet Plot for Multidrug Resistance

The heatmap is the default choice for multidrug resistance patterns. The heatmap shows the resistance phenotype for each isolate across each drug. The rows are the isolates and the columns are the drugs. The cells are colored by the resistance phenotype.

The UpSet plot is an alternative to the heatmap. The UpSet plot shows the intersections of the resistance patterns. The UpSet plot is useful when the user wants to see the number of isolates that are resistant to a specific combination of drugs.

The UpSet plot has a limitation. The UpSet plot is more complex than the heatmap. The user may need to explain the UpSet plot to the audience.

Records and Measurements

What to Record

The data for the AMR dashboard comes from the laboratory records. The records include the isolate identifier, the organism, the specimen source, the collection date, the testing facility, and the MIC value for each drug. The records also include the interpretation category for each drug and the breakpoint version.

The records should be stored in a database or a spreadsheet. The database is preferred because it allows for the querying and the filtering of the data. The spreadsheet is acceptable for a small dataset.

The records should be backed up regularly. The backup should be stored in a separate location from the original records. The backup should be tested regularly to make sure that the backup can be restored.

The Quality Control Records

The quality control records are a separate set of records. The quality control records include the results of the quality control strains for each testing run. The quality control records also include the date of the testing run and the name of the technician who performed the testing.

The quality control records should be reviewed regularly. The review should check that the quality control results are within the expected range. The review should also check that the quality control results are consistent over time.

The American Type Culture Collection provides the authenticated biological materials that are used as quality control strains. The quality control strains are used to verify that the testing method is correct. The quality control strains have known MIC values for each drug. The results from the quality control strains are compared to the known values.

The Measurement of the MIC

The MIC is measured using a standard method. The method is described in the Clinical and Laboratory Standards Institute documents. The method involves the preparation of a bacterial suspension, the inoculation of the suspension into a series of wells with different concentrations of the drug, and the incubation of the wells for a specified period.

The MIC is the lowest concentration of the drug that inhibits the visible growth of the bacteria. The MIC is read from the wells after the incubation period. The MIC is recorded as a number in the dilution series.

The MIC measurement is subject to variation. The variation can come from the preparation of the bacterial suspension, the incubation conditions, and the reading of the results. The variation is controlled by the use of the quality control strains.

Common Failure Patterns in AMR Visualization

Failure Pattern 1: Using a Linear Axis for MIC Values

The most common failure pattern is the use of a linear axis for the MIC values. The linear axis distorts the distribution. The MIC values are a dilution series, and the dilution series is logarithmic. The linear axis makes the lower MIC values look more common than they are and the higher MIC values look less common than they are.

The solution is to use a logarithmic axis for the MIC values. The logarithmic axis makes the dilution steps evenly spaced. The distribution is then shown correctly.

Failure Pattern 2: Ignoring the Breakpoint Version

The second common failure pattern is the ignoring of the breakpoint version. The breakpoint version determines the interpretation category. The breakpoint version changes over time. The chart that does not state the breakpoint version is not reproducible.

The solution is to record the breakpoint version in the data and to include it in the chart. The chart should also include the date of the breakpoint publication.

Failure Pattern 3: Showing Percentages without Denominators

The third common failure pattern is the showing of the percentages without the denominators. The percentage is a proportion. The proportion is unstable when the denominator is small. The chart that does not show the denominator is misleading.

The solution is to show the denominator in the chart. The denominator can be shown as a label above the bar or as a tooltip in the interactive chart.

Failure Pattern 4: Overloading the Dashboard

The fourth common failure pattern is the overloading of the dashboard. The dashboard has too many charts. The user cannot find the information that they need. The dashboard is confusing.

The solution is to limit the number of charts on the dashboard. The dashboard should have a clear focus. The dashboard should answer one question, not many questions.

Failure Pattern 5: Using a Venn Diagram for More Than Three Drugs

The fifth common failure pattern is the use of a Venn diagram for more than three drugs. The Venn diagram is difficult to read for more than three sets. The Venn diagram is also difficult to draw for more than three sets.

The solution is to use an UpSet plot for more than three drugs. The UpSet plot shows the intersections of the sets in a clear and readable way.

Biosafety and Containment Context

The visualization of AMR data is a data analysis task, but the data comes from a laboratory that handles biological materials. The laboratory must follow the biosafety guidelines that are described in the Biosafety in Microbiological and Biomedical Laboratories from the CDC and the NIH. The guidelines describe the risk assessment, the containment, and the decontamination procedures.

The risk assessment is the first step in the biosafety process. The risk assessment considers the organism, the procedure, and the laboratory facility. The risk assessment determines the biosafety level of the laboratory.

The containment is the second step of the biosafety process. The containment includes the physical containment, such as the biosafety cabinet, and the operational containment, such as the standard operating procedures. The containment is designed to prevent the release of the organism from the laboratory.

The decontamination is the third step of the biosafety process. The decontamination includes the cleaning and the disinfection of the laboratory surfaces and the equipment. The decontamination also includes the disposal of the biological waste.

The data analysis is performed outside of the laboratory. The data analysis does not involve the handling of the biological materials. The data analysis is a safe activity. The data analysis does not require the biosafety level of the laboratory.

The American Society for Microbiology provides the professional education and the laboratory practice context for the microbiology workforce. The ASM also provides the guidelines for the laboratory safety and the quality. The ASM guidelines are a useful reference for the laboratory professional.

Professional Escalation Criteria

The visualization of AMR data is a technical task, but the results can have clinical and public health implications. The analyst should escalate the results to the appropriate professional when the results indicate a problem.

The analyst should escalate the results when the quality control results are out of range. The out-of-range quality control results indicate a problem with the testing method. The problem must be corrected before the clinical results are reported.

The analyst should escalate the results when the resistance percentage for a specific organism-drug pair is unusually high. The high resistance percentage may indicate an outbreak or a change in the local epidemiology. The infection control team should be notified.

The analyst should escalate the results when the temporal trend shows a sudden increase in the resistance percentage. The sudden increase may indicate a change in the testing method or a change in the breakpoint version. The change must be investigated before the trend is reported.

The analyst should escalate the results when the data is incomplete or the data is inconsistent. The incomplete data may lead to the wrong conclusion. The inconsistent data may indicate a problem with the data entry or the data transfer.

The escalation should be documented. The documentation should include the date of the escalation, the reason for the escalation, and the person who was notified. The documentation should be stored with the data.

A Decision Framework for Selecting AMR Visualizations

Choosing the right chart type for AMR surveillance data is not a matter of preference. It is a decision that affects whether the audience sees the true resistance pattern or a distorted version of it. A practical decision framework helps the analyst move from the raw data structure to the appropriate visualization without relying on guesswork. This framework is built around three sequential questions that the analyst must answer before opening any plotting software.

Question 1: What Is the Unit of Analysis?

The first question is whether the visualization will show individual isolates, aggregated groups, or time periods. This question determines the fundamental structure of the chart.

When the unit of analysis is the individual isolate, the chart must preserve the MIC value for each isolate. A histogram or strip plot is appropriate because both show the distribution of MIC values across the isolate population. The histogram bins the isolates by MIC value, while the strip plot shows each isolate as a point. Both require a logarithmic scale on the MIC axis because the dilution series is inherently logarithmic.

When the unit of analysis is the group, such as a species, a specimen source, or a facility, the chart must show a summary statistic for each group. The summary statistic is usually the percentage resistant, the percentage susceptible, or the modal MIC. A bar chart or a dot plot is appropriate. The chart must include the denominator for each group, because a percentage based on 10 isolates is not comparable to a percentage based on 1,000 isolates.

When the unit of analysis is the time period, the chart must show how the summary statistic changes over time. A line chart is appropriate. The x-axis is the time period, and the y-axis is the summary statistic. The line chart should include a measure of uncertainty, such as a confidence interval band, when the number of isolates per time period varies.

Question 2: The Number of Variables Being Compared

The second question is how many variables the chart must compare simultaneously. This question determines the complexity of the chart.

A single variable, such as the MIC distribution for one organism and one drug, requires a simple chart. A histogram or a strip plot is sufficient. Two variables, such as the resistance percentage for two different drugs, require a grouped bar chart or a dot plot with two series. Three or more variables, such as the resistance percentage for multiple drugs across multiple organisms, require a faceted chart or a heatmap.

The number of variables also determines whether the chart can be read at a glance or whether it requires careful study. A chart with more than three variables is difficult to read at a glance. The analyst should consider whether the chart can be split into multiple smaller charts, each showing one or two variables, and arranged in a dashboard.

Question 3: The Comparison Type

The third question is whether the comparison is across groups, across time, or across drugs. This question determines the orientation of the chart.

A comparison across groups, such as the resistance percentage for different organisms, is best shown with a bar chart or a dot plot. The groups are on the x-axis, and the percentage is on the y-axis. A comparison across time is best shown with a line chart. The time is on the x-axis, and the percentage is on the y-axis. A comparison across drugs is best shown with a heatmap. The drugs are on the x-axis, and the isolates are on the y-axis.

The comparison type also determines whether the chart should be sorted. A bar chart that is sorted by the percentage resistant is easier to read than a bar chart that is sorted alphabetically. The sorted chart shows the highest resistance at the top and the lowest resistance at the bottom. The sorted order makes the pattern visible.

The Decision Matrix

The three questions can be combined into a decision matrix. The matrix has the unit of analysis on one axis and the comparison type on the other axis. The intersection of the two axes gives the recommended chart type.

Unit of Analysis	Comparison Across Groups	Comparison Across Time	Comparison Across Drugs
Individual isolate	Histogram or strip plot with log2 scale	Not applicable	Heatmap with isolates as rows
Aggregated group	Bar chart or dot plot with denominator labels	Line chart with confidence interval	Grouped bar chart or heatmap
Time period	Line chart with multiple series	Line chart with confidence interval	Faceted line chart by drug

The matrix is a starting point, not a rigid rule. The analyst should use the matrix to narrow the options and then choose the chart that best communicates the specific message. The matrix is most useful when the analyst is working with a new dataset and does not have a previous chart to reference.

The Decision Record

The decision framework should be documented. The documentation should include the question that the chart answers, the unit of analysis, the comparison type, and the chart type that was chosen. The documentation should also include the reason for the choice and the alternatives that were considered.

The documentation serves two purposes. First, it makes the chart reproducible. A colleague who reads the documentation can recreate the chart from the same data. Second, the documentation makes the chart auditable. A reviewer who reads the documentation can check whether the chart type was appropriate for the data.

The documentation should be stored with the data and the chart. The documentation can be a text file, a spreadsheet, or a section in the laboratory notebook. The documentation should be updated when the chart is updated.

The Validation Step

The decision framework is not complete without a validation step. The validation step checks whether the chart type is appropriate for the data. The validation step is performed after the chart is created and before the chart is shared.

The validation step has three parts. The first part is the data check. The analyst checks that the data is complete, that the MIC values are in the correct format, and that the breakpoint version is recorded. The second part is the chart check. The analyst checks that the chart shows the correct data, that the axis is on the correct scale, and that the chart is not misleading. The third part is the interpretation check. The analyst checks that the chart answers the question that it was designed to answer.

The validation step should be done by a second person who was not involved in the creation of the chart. The second person can see the chart with fresh eyes and can catch errors that the original analyst missed. The second person should be familiar with the data and the chart type.

The Escalation Criteria

The decision framework includes an escalation criteria. The analyst should escalate the chart to a supervisor or a colleague when the chart shows an unexpected pattern. The unexpected pattern may be a sudden increase in the resistance percentage, a shift in the MIC distribution, or a change in the quality control results.

The escalation is not a failure. The escalation is a sign that the analyst is paying attention to the data. The escalation allows the supervisor to investigate the pattern and to determine whether the pattern is real or an artifact of the data.

The Role of the Decision Framework in the Dashboard

The decision framework is also for individual charts. The framework is also used to design the dashboard. The dashboard is a collection of charts that answer a set of questions. The framework is used to choose the chart type for each question.

The dashboard should have a clear focus. The dashboard should answer one question, not many questions. The dashboard should have a limited number of charts. The dashboard should have a logical flow. The top of the dashboard should have the key performance indicators, the middle of the dashboard should have the main charts, and the bottom of the dashboard should have the detailed tables.

The dashboard should be validated as a whole. The validation should check that the charts are consistent with each other. The validation should check that the filters are working correctly. The validation should check that the dashboard is easy to use.

The World Health Organization describes the global surveillance of antimicrobial resistance. The WHO also provides the guidance for the national surveillance systems. The WHO guidance is a useful reference for the public health authority.

The Clinical and Laboratory Standards Institute publishes the standard methods for the antimicrobial susceptibility testing. The standard methods define the MIC measurement and the interpretation categories. The standard methods are the basis for the data that is visualized.

The Biosafety in Microbiological and Biomedical Laboratories from the CDC and the NIH describes the biosafety guidelines for the laboratory. The guidelines describe the risk assessment, the containment, and the decontamination procedures. The guidelines are the reference for the laboratory safety.

Frequently Asked Questions

What is the best chart type for showing MIC distributions?

The best chart type for showing MIC distributions is a histogram with a logarithmic scale on the x-axis. The histogram shows the number of isolates at each dilution step. The logarithmic scale makes the dilution steps evenly spaced. The histogram is easy to read and the modal MIC is easy to identify. A strip plot is a good alternative when the number of isolates is small.

How do I handle breakpoint changes in my visualization?

Record the breakpoint version in the data and include it in the chart title or subtitle. When the breakpoint version changes, the percentage resistant may change even though the MIC distribution is identical. The chart should state the breakpoint version so that the reader can compare the results across time.

Why should I use a logarithmic scale for MIC values?

The MIC values are a dilution series. Each step in the series is a doubling of the previous step. A linear axis distorts the relationship between the steps. A logarithmic axis makes the steps evenly spaced. The log scale is the correct scale for the MIC values.

What is the difference between a histogram and a strip plot?

A histogram shows the frequency of isolates in each bin. A strip plot shows each isolate as a point on the axis. The histogram is better for a large number of isolates. The strip plot is better for a small number of isolates. The strip plot can show the individual isolates values, but the points can overlap.

How do I show the denominator in a resistance percentage chart?

The denominator can be shown as a label above the bar or as a tooltip in an interactive chart. The denominator is the number of isolates tested. The percentage is unstable when the denominator is small. The chart should show the denominator so that the reader can assess the reliability of the percentage.

What is the best way to show multidrug resistance patterns?

The best way is to use a heatmap or an UpSet plot. The heatmap shows the resistance phenotype for each isolate across each drug. The UpSet plot shows the intersections of the resistance patterns. A Venn diagram is difficult to read for more than three drugs.

What software should I use to build an AMR dashboard?

The choice depends on the skills of the team. R Shiny is a programming framework that offers the most flexibility. Tableau is a visual analytics platform that does not require programming. A team that uses R will likely choose Shiny. A team that does not use R will likely choose Tableau.

How do I validate my AMR dashboard?

The validation includes checking the data against the source records, checking the calculations against the known values, and checking the charts against the expected patterns. The validation should be done by a second person who was not involved in the building of the dashboard.

Related Bacteriology Guides

References and Further Reading

Biosafety in Microbiological and Biomedical Laboratories. CDC and NIH.
Microbiology Resources. American Society for Microbiology.
Antimicrobial Resistance. World Health Organization.
Bacteriological Analytical Manual. U.S. Food and Drug Administration.
Antimicrobial Susceptibility Testing. Clinical and Laboratory Standards Institute.
Culture Collections. American Type Culture Collection.
NCBI Pathogen Detection. National Center for Biotechnology Information.
AMRnet: a data visualization platform to interactively explore pathogen variants and antimicrobial resistance.. Nucleic acids research, 2026.
AMRViz enables seamless genomics analysis and visualization of antimicrobial resistance.. BMC bioinformatics, 2024.
Progress made in digitalizing antimicrobial resistance surveillance in a One Health approach in Kenya.. Frontiers in public health, 2024.
Characterizing approaches used to display antimicrobial resistance data in veterinary and human medicine: a scoping review.. Antimicrobial stewardship & healthcare epidemiology : ASHE, 2025.

This article is educational and does not replace validated laboratory procedures, clinical diagnosis, institutional biosafety review, or public-health guidance.