New study finds 2 out of 5 antibiotic prescriptions could be unnecessary

By Dr. Liji Thomas, MDDec 12 2019

A new study in the British Medical Journal published on December 11, 2019, finds that up to 2 in every 5 prescriptions in the US might not be relevant clinically. This means that clinicians have a long way to go in rational antibiotic use and achieving proper antibiotic stewardship. The characteristics linked with antibiotic prescription without documented indication could help improve practices in this field.

Image Credit: i viewfinder / Shutterstock

Antibiotic stewardship

Irrational antibiotic prescribing has become a globally felt threat to public health, with multiple researchers pointing out several different adverse impacts of poor prescribing practices. One major danger is the rapid rise of antibiotic resistant infections all over the world.

Many studies on appropriate antibiotic prescribing have drawn data from medical records which are supposed to show the medical indication for the antibiotic prescription. However, because this is not a required field, it is often missing, which leads to unrealistically low figures of antibiotic use.

In 2016, antibiotics were prescribed at about 836/1,000 people, which is an extremely high rate. Moreover, during peak flu season, many providers prescribe antibiotics though flu is a viral infection. Overall, research shows that about a third of antibiotic prescriptions in outpatients are inappropriate.

For this reason, the current study looked specifically at how many times antibiotics are prescribed without a documented indication, in ambulatory patient care settings within the US. They also examined the factors that increase or decrease the likelihood of documentation.

The study

They used data drawn from the National Ambulatory Medical Care Survey (NAMCS), based on over 28,000 visits for healthcare at a doctor’s office, that were selected as representative of almost 1,000 million similar visits across America in the year 2015. About 13% of these, or 130 million, were taken for the study as they involved antibiotic prescriptions.

These prescriptions were then examined using the standardized diagnostic codes so that it could be determined if each of them was appropriate or inappropriate or was issued without a documented indication.

The findings

The results showed that more than half of all prescriptions (about 57%) were appropriately issued. However, a quarter were not, while almost a fifth were given without documented indication. Adding up these figures, this could mean that almost 43% of prescriptions were not necessary.

Risk factors for non-documented prescriptions

Demographic data showed that about a fifth of people between 18 and 64 years, and 22% of people over the age of 65 years, were prescribed antibiotics without documented indication, vs only 8% of people under 18 years.

Secondly, longer visits (17 minutes or longer) meant a higher chance of being prescribed an antibiotic without documented indication (15% vs 21% for shorter vs longer visits respectively).

Those with chronic medical illnesses had a 22% risk of antibiotic prescription compared to 14% for those without one. Men were also at higher risk for non-documented antibiotic prescriptions, as was the prescription of a sulfa antibiotic or one for urinary infection.

The lowest rate of antibiotic prescription without a documented indication was among primary care physicians and other healthcare providers, at 12%. On the other hand, specialists who often prescribed antibiotics, and doctors in other specialties had a 24% and 29% risk of prescribing an antibiotic without such indication.

The researchers failed to find a pattern in the type of antibiotics prescribed, the race, day of the week or season when the prescription was issued.

Limitations

The researchers point out that they took into consideration only the first five diagnostic codes in the medical record, and in addition, they could not consider prescriptions given over telephone or other virtual means. This could mean the actual incidence was still higher.

Secondly, the observational nature of the study means that the actual cause cannot be deduced from it.

However, the representative nature of the sample allows the results to be broadly generalized to the US population. In short, it shows that clinicians don’t consistently document why they prescribe antibiotics, leading to a lower than actual estimate of the incidence of unnecessary antibiotic use in outpatient care. It also found that many factors describing the patient, the provider and the visit itself were associated with this practice.

Knowing these factors could help to evolve measures to improve documentation in this area.

An editorial in the same journal by Alastair Hay at the University of Bristol suggests the importance of a comprehensive coding system to achieve good antibiotic stewardship. Every antibiotic prescription should be accompanied by a diagnostic code being recorded, and every infection should also receive a code, whether or not it is treated with antibiotics.

Additional measures could include improved control of infections, immunization, and clearer diagnostic principles. Implementing all of these strategies could help improve prescribing practices for antibiotics in outpatient care, which account for 60% of the total spending on such drugs, and up to 90% of the use of these medications.