CDC resistance data · verification framework
CDC’s Antimicrobial Resistance & Patient Safety Portal combines datasets with different sources, measures and coverage. A publishable update must lock the organism, drug phenotype, setting, geography, period and denominator before describing change.
What the portal contains
CDC describes the AR & Patient Safety Portal as an interactive application displaying information from the National Healthcare Safety Network, the Antimicrobial Resistance Laboratory Network and other sources. It includes resistance, antibiotic-use, stewardship and healthcare-associated-infection datasets.
The resistance display includes healthcare facilities that reported at least one healthcare-associated infection to NHSN and specified infection types. CDC also provides methodology and phenotype definitions. Those inclusion rules are part of the finding, not footnotes to omit.
Why percentages and counts answer different questions
A resistance percentage describes the proportion of tested pathogens meeting a named resistance definition. It is not the total burden of infection. CDC illustrates the problem: one area can have a higher percentage yet far fewer infections than another because the denominators differ.
AR Lab Network displays require another caution. CDC states that the network is not a surveillance system that can be generalized to the entire nation; testing and submissions can vary with state priorities and rules. A higher detection frequency does not necessarily mean a higher incidence rate.
Record the dataset, organism or phenotype, measure, denominator, geography and data period. Save the portal retrieval date and methodology version with the exported value.
What remains uncertain
This guide does not assert a future portal trend. Before it can publish as news, the editor must select an exact updated display, compare equivalent periods and preserve any changed methodology or facility coverage. A visual difference without comparable definitions is not a verified trend.
What clinicians and readers can verify next
- Open the “About the Data” methodology for the selected display.
- Check whether the metric is a count, percentage, rate or risk-adjusted ratio.
- Confirm which facilities, infections and years are included.
- Do not use a surveillance display to choose treatment for an individual.