CLSI M39 antibiogram software that builds your cumulative antibiogram by the M39 rules and shows the rules it applied

What the CAP checklist asks for, and where M39 comes in

For a CAP accredited lab the requirement is simple to state. Checklist item MIC.12040, Cumulative Antimicrobial Susceptibility Data, requires microbiology laboratories that provide primary testing for hospitals to maintain and report cumulative susceptibility data to the medical staff at least yearly. CAP TODAY described it in April 2026 as part of the consolidated susceptibility subsection in the checklist edition released in late 2025, and noted that the organisms included in the annual antibiogram are decided at the institutional level.

The checklist does not dictate the method, but the inspector, your pharmacy and therapeutics committee and your stewardship physician will all read the report against M39, because there is nothing else to read it against. A report that counts every isolate, or prints 64 percent on 11 isolates, is hard to defend in that conversation even when the checklist box is ticked.

The same checklist carries the breakpoint rule. MIC.11385 required current breakpoints for MIC and disk diffusion interpretation by January 1, 2024, and requires new breakpoints to be implemented within three years of publication. An antibiogram pooled across a breakpoint change needs that change written on it.

How breakpoint versions are handled in the builder is on the CLSI and EUCAST breakpoints page. Hospitals that also answer to The Joint Commission can see how the report fits the 2026 medication management standards on Joint Commission antimicrobial stewardship.

Where antibiograms usually break M39

Very few labs set out to ignore the guideline. The problems come from the report the lab system produces by default, and from the spreadsheet steps someone added years ago.

If your hospital runs Oracle Health, the duplicate rule question is covered in detail in antibiogram software for Cerner hospitals. Epic and MEDITECH builds have their own versions of the same problem.

• Every isolate counted. A default LIS report often totals all finalized isolates, so a patient cultured nine times in an ICU stay counts nine times and pulls the whole report toward the resistance pattern of the sickest patients
• A different duplicate rule. Some systems remove duplicates by patient, species and susceptibility pattern rather than keeping the first isolate. A published comparison of rules on the same data (White and colleagues, Diagnostic Microbiology and Infectious Disease, 2001) found that the pattern-based rule gave figures that differed from the other methods by minus 11 to plus 25 percent
• De-duplication in the wrong order. A unit or source cut made after de-duplication keeps whichever isolate happened to be first hospital-wide, not the first one in the ICU or the first urine
• Thin cells printed as percentages. Pseudomonas at a small hospital, or anything in a NICU, rarely reaches 30 isolates a year
• Intermediate counted as susceptible, which flatters exactly the agents where the MIC sits close to the breakpoint
• Screening swabs left in the export, which turns a colonization program into an apparent jump in MRSA or CRE
• Cascade suppressed results missing, so a narrow agent looks as if it was barely tested and its percentage is built on the resistant isolates only

CLSI M39 rule by rule, compared

WHONET is free and capable, and many labs keep it. The practical differences, including running it on a Mac, are on the WHONET alternative page.

What changed in the M39 5th edition

The 5th edition, summarized by Simner and colleagues in the Journal of Clinical Microbiology in October 2022, removed some older recommendations and added content on antimicrobial stewardship and on verifying data before it is reported. It also added guidance on extracting data from different sources for antibiogram preparation, on combining rapid diagnostic and resistance marker results with the antibiogram, on antibiograms for long-term care facilities, and on multifacility antibiograms.

Three of those additions change what a buyer should ask of software. Data verification means the tool should show you what it removed and why, not just the final grid. Long-term care guidance means a nursing home antibiogram built from a small count is a recognized product, with its own rules for pooling. Multifacility guidance means a system antibiogram should pool isolates, not average the percentages of member hospitals.

The long-term care version is covered on the nursing home antibiogram page, and pooling several hospitals in regional antibiogram software for health systems.

Does a hospital antibiogram have to follow CLSI M39?

No regulation names M39, but in practice yes. CAP requires the annual cumulative report, CMS requires hospitals to run an antibiotic stewardship program, and M39 is the method every reviewer uses to judge whether the numbers are sound. Following it is what makes last year and this year comparable, and it is what lets a stewardship committee put a percentage into an order set without an argument.

Where a hospital departs from M39, for example by pooling three years for a small unit or by counting intermediate as resistant, the fix is to say so on the report. The rule receipt that prints with every Antimicrobe export exists for exactly that.

How it works

The builder at the top of this page runs the same engine on sample data, so you can switch the first-isolate rule off and watch the percentages move before you buy anything.

• Ask the lab for 12 months of finalized susceptibility results as CSV: pseudonymous patient key, collection date, specimen source, unit, organism, antibiotic and S, I or R. Include the cascade suppressed results if your LIS can export them
• Map free-text sources and units once, and mark screening sources so they stay out of the report
• Build the house-wide report. The first-isolate rule, the 30 isolate threshold and the intermediate setting are applied and shown, with the counts removed at each step
• Add the cuts the committee asks for, urine, blood, ICU, each de-duplicated inside its own cut
• Export the antibiogram with its rule receipt for the medical staff, the P&T committee and the CAP binder, and rebuild it from next year's export with the same settings

Which plan fits

A single hospital lab producing the annual house-wide report and a few source cuts needs the Lab plan: $249 a month for 1 facility, 25,000 isolates a year and 2 seats, with cuts by source and period, CLSI breakpoints and CSV and print exports. It is bought by card and there is no implementation fee.

A stewardship program that also wants unit cuts, syndrome groups, quarterly trends and MDRO counts, or up to three facilities, uses Stewardship at $749 a month. A lab or system that produces antibiograms for up to 12 facilities uses Network at $2,400 a month.

Every limit is on the pricing page. How labs compare this with the instrument data manager and Excel is in antibiogram software for microbiology labs, and the export itself is described on laboratory information system exports.

What this does not do

• It does not read raw MIC values or reinterpret them. It uses the S, I and R your lab reported, so breakpoint changes are made in the lab system
• It does not print confidence intervals on each cell. Cells under the threshold are suppressed and marked instead
• It does not submit to the NHSN AR Option or any public health system
• There is no EHR or HL7 interface on the Lab and Stewardship plans. It works from exported files