The burden of antibiotic resistance

There is a global consensus that antibiotic resistance poses a profound threat to human health. Yet crucial surveillance and burden data are lacking in large parts of the world, which makes it difficult to get a full overview of the situation. Besides, burden is a broad concept that could mean different things. It is commonly thought of as the mortality caused by a disease or condition of interest but could also be measured in, for example, excess morbidity or additional economic costs.

The economic burden

Antibiotic resistance is indeed very costly – for individuals, health care, and society at large. A study from India found that the median overall extra cost of treating bacteremia caused by a resistant bacterium equals the salary a rural male worker in the country earns after 442 days of work. The impact resistance has on global economies is also substantial. According to the World Bank, inaction on antimicrobial resistance (i.e. resistance to antibiotics, antivirals, antifungals and parasitic agents) could cause a fall in global GDP of between 1.1% and 3.8% by 2050. The World Bank also projects that antimicrobial resistance could push up to an additional 24 million people into extreme poverty by 2030.

The health burden of antibiotic resistance

The health burden of antibiotic resistance is commonly thought of as the number of deaths specifically caused by infections with resistant bacteria. Straightforward as it may seem, there are different ways to measure and calculate this, which complicates burden estimations and data comparisons between studies, regions, and countries. Do we mean the number of people that die while suffering from any antibiotic-resistant infection (irrespective of the presence of other diseases or conditions), the excess number of deaths caused by specific resistant infections, or something else? And what do we compare that to?

Some important factors that affect the generation of mortality data is the diagnostic capacity of health systems and the procedures for determining and recording deaths. For instance, if a patient with terminal cancer gets infected by a resistant bacterium causing sepsis and eventually dies –how is the death recorded? Is it first and foremost a death due cancer? Sepsis? Antibiotic-resistant infection? If the patient lives in an area with limited capacity for laboratory diagnostics, it is not certain that the cause of infection and antibiotic susceptibility can be determined. Antibiotic resistance burden data is particularly scarce in low- and middle-income countries, although it is well-recognized that these countries generally are more afflicted. Work to strengthen health systems and ensure universal health coverage is therefore crucial also for data generation purposes.

Antibiotic resistance – is here – now

Antibiotic resistance could be viewed both as a slow-moving pandemic and a silent tsunami. The burden is increasing, although lots of people already die from resistant infections every day.

Global data from 2019 shows that:

• 1.27 million deaths were a direct result of antibiotic-resistant bacterial infections.
• 1 in 5 deaths from resistant bacteria occurs in children under 5.
• The burden of antibiotic resistance falls disproportionately on low- and middle-income countries.

Global surveillance data

GLASS

To aid the implementation of routine surveillance on a national and global level, the Global Antimicrobial Resistance Surveillance System (GLASS) was introduced by WHO in 2015. GLASS collects harmonized data on the occurrence of antibiotic resistance in eight bacterial pathogens from four types of clinical specimens, studies to estimate AMR burden and related drivers, and the new Sustainable Development Goals (SDGs) AMR indicators. You can read more about GLASS in the infographic at the bottom of this page.

In June 2020, a master protocol for estimating attributable mortality of antimicrobial-resistant bloodstream infections was published as part of the GLASS framework. The protocol focuses, at first hand, on attributable in-hospital mortality of bloodstream infections caused by extended-spectrum beta-lactamase-producing (ESBL) E. coli and methicillin-resistant S. aureus (MRSA). Both these are relatively common pathogens that are resistant to clinically important antibiotics.

Although the initiative is very welcome, it is important to be aware of the limitations. The data will only be as good as the circumstances allow and will be influenced by everything from health-care seeking behavior to availability of clinical diagnostics, laboratories, and trained personnel. The prerequisites for generating useful, representative data therefore varies substantially between countries.

Challenges of measuring global burden of diseases

However, making these estimations is far from straight-forward. Factors complicating burden estimations of antibiotic resistance are described in Dunachie et al. 2020 and include:

• The unique characteristics of antibiotic resistance (for example, that it is not a disease in itself and that bacteria may develop resistance to multiple antibiotics)
• The lack of good microbiological data linked to individual patient data
• The methodology of choice to measure the disease burden

The importance of reliable global burden estimates

Why then is it important to have reliable burden estimates of antibiotic resistance? Isn’t it enough to simply know that antibiotic resistance is a profound problem? Probably not. There is a difference between knowing that something is an issue and being informed about the extent of the problem.

One could make an analogy to, for example, traffic. When driving a car on a road where many car accidents happen, the driver is hopefully paying more attention. However, if the driver knows that on average one traffic accident with a deadly outcome happens every week on that specific road, the response is likely different. Maybe she or he cancels the trip or chooses an alternative, statistically safer, route. Besides, it is probably likelier that authorities and decision-makers prioritize finding solutions to reduce the problem and make the road safer.

The same applies to antibiotic resistance. Knowing the extent of the problem in countries as well as globally would help communicate the urgency of the crisis. It could help inform policies, target interventions and increase general awareness, which all are important components of the antibiotic resistance response. With other words – lack of antibiotic resistance burden data may slow down national and global efforts, which is why it is so important to address. We must look more into how we can overcome the data gap. This includes continuing to strengthen health systems capability and capacity to detect and record the outcome of resistant infections, but also to complement data collection with for example point prevalence studies. While continuing our efforts to overcome the data gap, we must recognize that we have enough knowledge and insight to take action here and now. We have no time to lose.

© Uppsala University