Antibiotic treatment begins in the consulting room, but its outcome often becomes visible later in the laboratory. The problem grows when prescription records, test results, and resistance information fail to meet. The new measure of healthcare quality is not simply how much medicine was used, but whether the right medicine was selected for the right patient at the right time.

* The goal of antibiotic stewardship is appropriate use, not simply reduced use. Patients who need antibiotics should receive them promptly, and treatment should be adjusted when laboratory results identify a better option.

* When prescriptions and laboratory results remain separate, a critical opportunity to improve treatment can be missed. Data must be linked by patient, time, organism, and antibiotic to evaluate whether prescribing was appropriate.

* Differences in national antibiotic use reflect both overuse and inadequate access. Low consumption alone does not demonstrate good stewardship; the type of antibiotic and the patient’s need for treatment must also be considered.

* Antibiotic data are becoming a new measure of hospital care quality. Hospitals should assess specimen collection, prescription changes after test results, treatment duration, and patient outcomes together.

* Staying ahead of resistant bacteria requires faster connections among data that already exist. When prescriptions, laboratory tests, resistance patterns, and outcomes form one continuous flow, healthcare systems can detect and respond to warning signs earlier.

***

A patient arrives at the emergency department with a high fever and dangerously low blood pressure. Blood tests point to a severe infection, but the bacterium responsible remains unknown. The physician cannot wait a day or two for the laboratory results because any delay could cause the patient’s condition to deteriorate rapidly. Blood and urine samples are collected, and a broad-spectrum antibiotic capable of targeting many types of bacteria is administered immediately.

This is not a mistake. When the identity of the bacterium is unknown, broad protection is a rational emergency measure. The problem begins later. A culture identifies the bacterium and reveals which antibiotics are effective and which ones encounter resistance. The results may also show that a narrower-spectrum antibiotic would be sufficient. Yet the original broad-spectrum treatment sometimes continues unchanged.

The clinician may not have seen the result promptly. Information may have been lost when the patient was transferred to another ward, or the prescribing and laboratory systems may have displayed the information on separate screens. There may also have been too few staff members available at night or over the weekend to review the prescription. The prescription was recorded and the test result was stored, but nothing happened between the two pieces of information.

The central challenge of antibiotic treatment is not to identify the perfect drug at the first attempt. In an emergency, treatment begins broadly and is narrowed when laboratory evidence becomes available. This process, known as antibiotic de-escalation, reduces unnecessary side effects, protects normal bacteria, and lowers the likelihood that resistant organisms will be selected. When prescribing and laboratory data remain disconnected, this critical opportunity can be missed.

Antibiotic susceptibility testing compares a bacterium with several antibiotics to determine which ones remain effective. Results commonly classify the bacterium as susceptible, intermediate, or resistant to each drug. Clinicians then decide whether to continue or change the treatment. In practice, however, the moment a result becomes available does not always coincide with the moment a prescribing decision is made.

The situation becomes more complicated when several departments are involved. The emergency department initiates treatment, the inpatient ward continues it, an infectious-disease specialist provides advice, and the pharmacy monitors use. Meanwhile, the microbiology laboratory produces the test results. Each department can perform its own task correctly while the overall appropriateness of treatment remains unclear because the data are not connected across the patient’s timeline.

Suppose a hospital calculates only the total quantity of antibiotics used during a month. A decline from the previous month may appear to represent an improvement. Yet if antibiotic use fell because patients with serious infections received necessary treatment too late, the result would not indicate better care. Conversely, an increase may not be undesirable if the number of infection cases rose and patients received rapid treatment that matched their laboratory results.

The important question therefore does not end with how much was used. Hospitals must also ask whether a patient who needed an antibiotic received it promptly, whether appropriate specimens were collected before treatment, whether therapy was narrowed after laboratory results became available, and whether treatment continued longer than recommended. Volume is only one piece of the data. The reason for prescribing, the laboratory result, the timing of a change, and the patient’s outcome must be connected before the quality of care can be understood.

Antibiotic treatment for viral illnesses such as the common cold can be examined through the same framework. Antibiotics target bacteria and therefore do not work against viruses. Yet preventive prescriptions may still be issued because of patient expectations, diagnostic uncertainty, or limited consultation time. A usage total alone cannot explain why a drug was prescribed. The diagnosis, testing status, and type of antibiotic must be considered together to distinguish unnecessary use.

The World Health Organization (WHO) released two reports in 2025 examining antibiotic use and resistance. The Global Antimicrobial Resistance and Use Surveillance System (GLASS) report published in April analyzed 2022 antibiotic-use data submitted by 60 countries, territories, and areas. The Global Antibiotic Resistance Surveillance Report released in October examined the scale and direction of resistance using more than 23 million bacteriologically confirmed cases reported by 104 countries in 2023.

The first picture revealed by GLASS was one of enormous inequality. Antibiotic use differed by approximately tenfold between the highest- and lowest-using countries and territories. Low use, however, does not necessarily indicate successful stewardship. Some places prescribe too many unnecessary antibiotics, whereas patients elsewhere cannot obtain antibiotics even when treatment is essential. The same statistical decline may represent reduced overuse in one country and deteriorating access to care in another.

WHO classifies antibiotics into the Access, Watch and Reserve (AWaRe) groups. Access antibiotics are recommended as first-line treatments for common infections and carry a relatively lower risk of resistance. Watch antibiotics have a greater potential to encourage resistance and should be used more selectively. Reserve antibiotics are the final line of defense for severe infections caused by bacteria resistant to multiple treatments.

Access antibiotics accounted for approximately 52.7% of global use in 2022. The international community has set a target for at least 70% of human antibiotic use to come from the Access group by 2030. The choice of antibiotic, rather than only the total amount used, has therefore become an international public-health objective.

The October report presented an even more direct picture of resistance. Approximately one in six laboratory-confirmed common bacterial infections worldwide in 2023 was resistant to antibiotic treatment. Resistance increased in more than 40% of the bacterium-antibiotic combinations monitored between 2018 and 2023, with average annual increases of approximately 5?15%. Choices beginning with a single prescription were accumulating across hospitals and communities, altering the world’s ability to treat infection.

Describing antibiotic stewardship simply as a campaign to reduce use can create a dangerous misunderstanding. In conditions such as bacterial sepsis, where rapid treatment can determine whether a patient survives, the appropriate antibiotic must be administered without delay. Reducing necessary treatment is not stewardship but neglect.

Unnecessarily using a broad-spectrum drug for a mild infection or continuing a powerful antibiotic after laboratory results support a narrower option leaves costs for both the patient and society. Side effects such as diarrhea and allergic reactions can occur, and the normal microbial ecosystem may be damaged. Bacteria that survive gain another opportunity to withstand future treatment.

Antibiotics differ from many other medicines because one person’s use can affect another person’s treatment options. Taking too many painkillers does not cause a neighbor’s painkillers to stop working. When inappropriate antibiotic use increases the number of resistant bacteria, however, those organisms can spread through hospitals and communities. A prescription issued to one patient today can alter the effectiveness of a medicine needed by another patient tomorrow.

An Antimicrobial Stewardship Program (ASP) is therefore not a system designed simply to block prescriptions. It helps clinicians use the necessary medicine at the correct dose, through the correct route, and for an appropriate duration. It reviews broad-spectrum therapy after a defined period, adjusts treatment when culture results become available, and identifies when an intravenous drug can be changed to an oral medicine.

The Centers for Disease Control and Prevention (CDC) identifies accountable leadership, professional expertise, targeted improvement activities, tracking, reporting, and education as core elements of hospital antibiotic stewardship. Measuring prescribing data and providing regular feedback to clinicians are not secondary administrative tasks. They stand at the center of an effective stewardship program.

The role of data changes at this point. Traditional statistics showed how many bottles of antibiotics a hospital used during the previous month. Newer indicators ask how many hours passed before a prescription was adjusted after a laboratory result became available. The emphasis is shifting from measuring consumption to measuring the quality of decisions.

Evaluating antibiotic use as a healthcare quality measure requires several questions to be answered together. Did a patient with a severe infection receive the first antibiotic promptly? Was an appropriate specimen collected before treatment? Did a culture identify the responsible bacterium? Was the medicine changed to reflect the susceptibility result? Did treatment follow the recommended duration? Did the patient recover without readmission?

The information needed to answer these questions is scattered across different systems. Prescription information resides in the electronic medical record, actual administration times appear in nursing records, and culture and susceptibility results are stored in the laboratory information system. The pharmacy retains dispensing volumes and costs, while infection-control teams collect data on resistant bacteria and transmission within the hospital. Readmission and treatment failure generate still more records after discharge.

Even when every system contains abundant data, meaningful questions remain difficult to answer unless the information can be linked by patient, time, organism, and antibiotic. The time a culture result became available must be connected to the time the prescription changed in order to measure responsiveness. The chosen antibiotic must be compared with the organism’s susceptibility to evaluate appropriateness. Ward-level consumption must be viewed alongside resistance trends to detect warning signs.

Consider an intensive care unit where use of a particular broad-spectrum antibiotic rises steadily for several months. If bacteria resistant to the same drug increase over the same period, the two trends should be reviewed together. The association alone does not prove causation, but it provides an early signal prompting the infection-control team to investigate prescribing practices, patient characteristics, and possible transmission within the hospital.

Connected data can also work at the patient’s bedside. A system can notify clinicians when an antibiotic that does not match the laboratory result remains active. It can request a review when therapy continues for a long period despite a negative culture or exceeds the recommended duration. Too many warnings, however, create alert fatigue and may cause clinicians to ignore them. A useful data system does not send the greatest number of alerts; it delivers the right information at the clinically important moment.

Context must also be preserved when data become quality indicators. A tertiary hospital treating large numbers of critically ill patients will naturally use more antibiotics. The same standards cannot be applied without adjustment to a long-term care facility, a children’s hospital, and a cancer center. Rankings that ignore disease severity, infection type, and institutional role may penalize hospitals providing necessary treatment.

WHO’s GLASS standardizes national antibiotic-use and resistance data to create a global picture. The European Centre for Disease Prevention and Control (ECDC) also operates the European Surveillance of Antimicrobial Consumption Network (ESAC-Net), which tracks antimicrobial use, and the European Antimicrobial Resistance Surveillance Network (EARS-Net), which monitors resistance.

These surveillance systems are essential for identifying countries and regions where particular forms of resistance are rising and for setting policy priorities. National averages, however, cannot tell a clinician which prescription should be changed in a ward today. A considerable distance remains between a global resistance map and an individual patient’s prescription.

The next challenge is to close this distance. National consumption and resistance rates, regional outbreaks, hospital prescribing, and patient-level test results must be connected as different layers of the same information system. If a global surveillance network acts as a compass, linked hospital data functions more like navigation guiding actual clinical decisions.

Data standards are equally important. Comparisons become difficult when hospitals use different names for organisms, testing methods, antibiotic codes, and prescribing units. One institution may record a medicine by its generic name while another uses a brand name. Automated analysis can also produce errors when susceptibility criteria change or different systems use incompatible time formats.

Interoperability is more than the ability of systems to exchange files. Different institutions must be able to interpret and analyze the same concepts in the same way. This is why linking prescription, testing, and resistance data matters. Data become actionable not because they are abundant, but because they are connected in a form that allows different systems and professionals to understand them consistently.

Antibiotic resistance does not arrive as a sudden disaster. It sends gradual signals: use of a particular drug rises in one ward, susceptibility declines among several bacteria, and treatment durations and hospital stays begin to lengthen. When the information remains in separate systems, each change can appear minor. When it is connected on one screen, it becomes a recognizable pattern.

A strong antibiotic-data system is more than a tool for monitoring prescriptions. It becomes infrastructure that helps clinicians make faster and safer decisions. It supports the selection of the most effective medicine, reduces unnecessary broad-spectrum treatment, and detects early signs that resistant bacteria are spreading. It shows hospital executives the added length of stay and cost associated with infection and gives public-health authorities signals needed for regional action.

Data cannot replace clinical judgment. A negative culture does not always mean that an infection is absent, and the statistically preferred medicine is not suitable for every patient. Allergies, organ function, previous treatment, and the location of infection must all be considered. The purpose of connected data is not to control physicians automatically. It is to assemble scattered clues at the right time so that better decisions become possible.

The hospital using the least antibiotics is not necessarily the best hospital. Better care means treating patients without delay when antibiotics are needed, switching to the most appropriate medicine when laboratory results arrive, and stopping treatment promptly when it is no longer necessary. Measuring this performance requires more than counting the antibiotics leaving the pharmacy.

When prescriptions, laboratory results, resistance information, and treatment outcomes form one continuous flow, antibiotic use changes from a cost item into a measure of healthcare quality. Bacteria continue to change. Prescribing data must move with the discoveries made in the laboratory. Staying one step ahead of resistant organisms does not begin with a stronger antibiotic. It begins by connecting information that already exists more quickly and accurately.

World Health Organization, October 2025, Global Antibiotic Resistance Surveillance Report 2025

World Health Organization, April 2025, Global Antimicrobial Resistance and Use Surveillance System Report: Antibiotic Use Data for 2022

World Health Organization, September 2025, Updated WHO Dashboard Offers New Insights on Antimicrobial Resistance and Use

Centers for Disease Control and Prevention, 2025, Core Elements of Hospital Antibiotic Stewardship Programs

European Centre for Disease Prevention and Control, 2025, European Antimicrobial Resistance Surveillance Network and European Surveillance of Antimicrobial Consumption Network