Disclosures: Oldenkamp reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
February 08, 2021
2 min read
Save

Study identifies regions in need of antimicrobial resistance surveillance

Disclosures: Oldenkamp reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

The Middle East, sub-Saharan Africa and the Pacific Islands should be prioritized for antimicrobial resistance surveillance, researchers argued in a recent study, saying surveillance in these areas is either needed or could be improved.

“Clinical resistance to antibiotics forms an increasing threat to global health. To help inform treatment and actions to tackle this threat, structural surveillance of clinical resistance is essential,” Rik Oldenkamp, PhD, postdoctoral fellow at the Amsterdam Institute for Global Health and Development and the department of global health at Amsterdam University Medical Centers, told Healio.

Antimicrobial surveillance map
Source: Oldenkamp R, et al. Proc Natl Acad Sci. 2021;doi:10.1073/pnas.2013515118.

“However, setting up and maintaining surveillance is very costly, especially for low- and middle-income countries, and large knowledge gaps still remain in these countries,” Oldenkamp said. “At the same time, these countries are disproportionally burdened by infectious diseases. Therefore, we developed a data-driven method to estimate resistance levels globally and, with that help, prioritize surveillance efforts.”

Using models for nine pathogens resistant to 19 classes of antibiotics, Oldenkamp and colleagues aimed to capture the statistical relationship between antimicrobial resistance (AMR) prevalence and socioeconomic characteristics. According to the study, the researchers combined prevalence data from ResistanceMap, a repository of AMR trends, with socioeconomic profiles constructed from 5,595 World Bank indicators, and used cross-validated models to estimate clinical AMR prevalence and trends for countries that lacked data.

Overall, the study demonstrated high predictive accuracy for six out of nine pathogens. They found that predictive accuracy was lower for Pseudomonas aeruginosa (q2 = 0.58) and Staphylococcus aureus (q2 = 0.56) and the predictive performance for Streptococcus pneumoniae was substantially less (q2 = 0.27). Prediction errors were largest for Enterobacter aerogenes/cloacae and Enterococcus faecium, as well as for Escherichia coli estimates in Iran and Indonesia specifically. E. aerogenes/cloacae and E. faecium are represented in ResistanceMap by a relatively small number of 13 and 46 countries, respectively.

The researchers said that although accurately predicting clinical AMR prevalence in similar countries, performance is lower for target countries with divergent characteristics.

According to Oldenkamp, they were able to identify three regions of interest for prioritization of surveillance based on different data-driven criteria ⎼⎼ countries in the Middle East where resistance levels were highest, sub-Saharan Africa where there was the highest relative resistance increase over the past 20 years and various small island states such as the Pacific Islands that could benefit from the improved coverage.

The researchers acknowledged, however, that their models have “a slight tendency to overestimate resistance prevalence” at the upper limit of the spectrum. However, they said that AMR surveillance efforts should be prioritized in countries where AMR is estimated to increase rapidly.

“We estimate clinical resistance at the level of individual countries. These estimates can support the formulation of (supra-)national surveillance strategies, and possibly national treatment guidelines, but local clinical inferences are more difficult to draw, especially in large and highly heterogeneous countries,” Oldenkamp said. “I think that would require a more local approach tailored to the specific local conditions.”