Neighborhood antibiotic use tied to risk of resistant bacteria

Jerusalem market
Jerusalem market

Brian Holsclaw / Flickr cc

A large new study by researchers in Israel indicates that increased use of antibiotics in certain neighborhoods is associated with increased risk of acquiring antibiotic-resistant bacteria, even in individuals without prior antibiotic use.

The study, published in The Lancet Infectious Diseases, specifically looked at fluoroquinolone-resistant Escherichia coli found in the urine cultures of Israelis in more than 1,700 neighborhoods across the country, and any association with personal and community use of fluoroquinolones.

The results showed that while higher personal use of fluoroquinolones was associated with increased risk of finding fluoroquinolone-resistant E coli, so was higher neighborhood use. And that risk remained even when the analysis looked at people who had not previously consumed fluoroquinolones.

The authors of the study say the findings suggest that personal consumption of antibiotics isn't the only mechanism driving resistance at a population level.

"Our study may supply some evidence regarding a selective mechanism [for antibiotic resistance] at the community level," lead study author Marcelo Low, MPH, told CIDRAP News in an email.

Large-scale analysis

The population-based case-control study involved data on 2.4 million adult members of Clalit Health Services (CHS), the largest of Israel's four state-mandated health services, living in 1,733 geographic statistical areas (GSAs) in Israel. The research team collected electronic medical record and demographic data on the patients, including antibiotic consumption, and analyzed almost 5 million urine-culture specimens sent to CHS laboratories from 2010 through 2014, focusing on samples with E coli growth and those that were sterile.

In addition to categorizing individuals by the amount of fluoroquinolones they consumed, the researchers categorized each GSA by the amount of fluoroquinolone consumption, which was calculated as the total amount of fluoroquinolones purchased by CHS members in the neighborhood in the prior year.

Low, an epidemiologist with CHS, said that while personal antibiotic use is a well-established risk factor for antibiotic resistance, he and his colleagues were interested in looking at the impact of community-level consumption of antibiotics on antibiotic resistance because of increasing anecdotal reports from Israeli clinicians.

"We've been hearing [of] more and more clinicians in the community treating patients with antibiotic resistant bacteria who never used antibiotics," he said. "The aim was to research the environmental determinants together with the personal ones."  

Of the more than 2.5 million urine cultures analyzed by Low and his colleagues, 300,105 with E coli and 1,899,168 with no pathogen growth were included in the final analysis. The proportion of women with fluoroquinolone-resistant E coli in their urine was 16.8% (45,427 out of 270,190 women), while 29.5% of men (8,835 out of 29,915) had fluoroquinolone-resistant E coli. The fluoroquinolone resistance rate by GSA was 18%, and the mean fluoroquinolone consumption among the entire population by GSAs was 1.51 defined daily doses (DDD) per 1,000 people per day.

Unsurprisingly, the multivariate analysis indicated that personal fluoroquinolone use was associated with the risk of finding fluoroquinolone-resistant E coli in urine compared with fluoroquinolone-susceptible E coli and sterile urine cultures, and the risk increased with higher fluoroquinolone consumption. This association was found among men and women.

The analysis also found that increasing consumption of fluoroquinolones in a given GSA was associated with increased risk of fluoroquinolone-resistant versus fluoroquinolone-susceptible urinary E coli, with similar associations observed among men and women.

Among women, the odds ratios (ORs) for the association between increased neighborhood fluoroquinolone use and fluoroquinolone-resistant E coli were, for the four quintiles with higher neighborhood consumption, 1.15 (95% confidence interval [CI], 1.06 to 1.24), 1.31 (95% CI, 1.20 to 1.43), 1.41 (95% CI, 1.29 to 1.54), and 1.51 (95% CI, 1.38 to 1.65) when compared with the GSAs in the lowest quintile of consumption.

Among men, the ORs were, by quintile, 1.17 (95% CI, 1.02 to 1.35), 1.24 (95% CI, 1.06 to 1.45), 1.35 (95% CI, 1.15 to 1.59), and 1.50 (95% CI, 1.26 to 1.77).

In both women and men, that association remained when the researchers conducted a post-hoc subgroup analysis restricted to patients who hadn't taken fluoroquinolones in the year prior to their urine culture. The association was also independent of other risk factors for antibiotic resistance, including previous hospitalization, older age, and residence in a nursing home.

Personal antibiotic use still remained the most important factor, however. According to the analysis, personal fluoroquinolone consumption accounted for 46% of the cases of fluoroquinolone-resistant E coli in women, while neighborhood fluoroquinolone consumption above 0.87 DDD/1,000 people per day accounted for 25%.

"Our results suggest that increased GSA-level use of antibiotics might pose a small, but real, risk for other individuals in the community in terms of acquiring resistant bacteria," Low and his co-authors write. "This effect might be quantitatively small for an individual, but its overall contribution to the spread of antibiotic resistance in the population is important."

Cross-transmission, environmental contamination

How might higher antibiotic use in a neighborhood increase the risk of finding antibiotic-resistant bacteria in an individual, even in those who haven't been consuming antibiotics? Low and his colleagues hypothesize that more crowded living conditions, and increased physical contact between people in crowded neighborhoods, could be a factor. They note that this theory is supported by the fact that patients with fluoroquinolone-resistant E coli lived in higher-density neighborhoods than those with fluoroquinolone-susceptible E coli and those with sterile urine cultures.

Low explained that more crowded conditions could lead to more opportunities for cross-transmission of resistant bacteria in the gut. "This may occur when a person consumes antibiotics that kill the sensitive bacteria in the gut, allowing [the] resistant bacteria to multiply and spread, also among the people living close," he said.

This kind of transmission through close contact, the study notes, has been observed between mothers colonized with fluoroquinolone-resistant Enterobacteriaceae and children in the same household.

Another possibility is that antibiotics that aren't completely digested are being excreted and "transmitted into the environment in a number of ways, increasing the bacterial resistance problem," Low said.

Environmental transmission of resistance is an area that health officials and researchers are examining more closerly. A 2017 report from the United Nations Environment Programme suggested that up to 80% of antibiotics consumed by people are excreted un-metabolized through urine and feces into sewage systems, and because most treatment plants aren't designed to fully remove antibiotics and other pharmaceuticals from wastewater, drug residues end up being released into surface water. In addition, studies have found that even trace amounts of antibiotics in streams and rivers that receive wastewater can select for clinically important resistance genes.

Although more research is needed to confirm the findings, Low said the study further highlights the need for patients and clinicians to avoid unnecessary antibiotic use.

"Antimicrobial resistance is a public health issue that is not limited to antimicrobial use in the individual patient, [which] reinforces the need for responsible use," he said.

See also:

Mar 4 Lancet Infect Dis abstract

This week's top reads

Our underwriters