5.1 Types of surveillance and implications for sampling
In most cases, isolates from patients with infections are tested for resistant pathogens to guide clinical decision-making. That is, people who are unwell, particularly those admitted to hospital, have samples collected for bacterial identification and AST so that the most appropriate antimicrobial to which the infection is susceptible can be used in treatment. Generating surveillance data or monitoring the prevalence of AMR within the population is generally a secondary objective or an additional benefit of this testing. In general, this means that sampling for AMR happens within the context of
Sometimes, people who are healthy are asked to provide specimens (for example, skin or nasal swabs, or urine samples), to find out if resistant bacteria are present in healthy individuals. This is sometimes referred to as ‘asymptomatic carriage’ of resistant bacteria. This type of sampling happens in the context of
You can find out more about the difference between active and passive surveillance in Introducing AMR surveillance systems module.
Sampling in the context of passive surveillance
Usually, people sampled in the context of passive AMR surveillance are unwell. This has both advantages and disadvantages. Unwell people present to hospitals and other medical facilities and are therefore easily accessible. They are often tested to inform their treatment, and the cost of the testing may be covered by their health insurance or government funding. However, unwell people are not representative of the general population.
What are some reasons why unwell people do not represent the general population?
They are sick, whereas most people are healthy most of the time. Another reason is that they might be older.
When only sick hospitalised patients are selected for AMR surveillance, we cannot assume the results are externally valid for the whole human population in that region: hospitalised patients may be older or sicker, may have been exposed to more treatments, or may have easier access to hospitals, etc., than the general population. In contrast, younger or healthier people, or people with less access to hospitals, might get sick with a bacterial infection, but only seek care in their community. This means that the target population in these studeies should be defined as the hospitalised population, not the general population.
The importance of this depends on what you are studying. If you want to know more about resistance in bacteria causing bloodstream infections, you would only be interested in hospitalised patients because healthy people in the community don’t have bacteria in their bloodstream (if they did, they’d be unwell – and likely end up in hospital!) However, if you want to study asymptomatic carriage of methicillin-resistant Staphylococcus aureus (MRSA), or mild UTIs, looking at only hospitalised patients may not tell you very much about how many people in the community have certain types of resistant infections.
Passive surveillance also takes place in the community; for example, patients presenting to general practitioners with UTIs. There are advantages and disadvantages to hospital versus community sampling. Hospital patients are sampled more frequently than community-based patients, and they are a more accessible population in general. However, they also tend to differ more systematically from the general population than community-based patients. It is also difficult to determine if resistant pathogens found in hospital patients were acquired before they entered the hospital or in the hospital itself, especially if they have been in hospital for some time – though some resistant pathogens, such as MRSA, mainly occur in hospitalised patients. Community-based patients are sampled less frequently in most passive surveillance systems, but these patients, and the resistance profiles of the bacteria they carry, may be more representative of the general population and their pathogens. This is because of patients’ individual characteristics mentioned above (age, socio-economic status, etc.) and if resistant pathogens are isolated, it is likely patients were infected in the community rather than in a hospital or other healthcare setting.
Sampling in the context of active surveillance
As you have seen, sampling in the context of passive AMR surveillance means that there are likely to be challenges with bias (see the Fundamentals of data for AMR module). This is particularly because patients included in passive surveillance might not be representative of the source or target population, especially for bacterial diseases that could be found in people who are unwell in the community as well as people who are in hospital. (Remember that for severe bacterial infections such as bloodstream infections, almost everyone who gets sick will go to hospital, so there’s no bias caused by restricting surveillance to hospital settings!)
An alternative is to conduct active AMR surveillance; however, active surveillance programmes based on probability sampling methods as described above are resource-intensive and not always practical. So what can be done to generate reliable and comprehensive data on the frequency of resistant bacteria for common and sometimes self-limiting infections, such as UTIs? Researchers have proposed a cost-effective, practical way to determine the categorical level of resistant bacteria in a particular facility (such as a primary care facility) called
In the context of active AMR surveillance or research in healthy populations, where the topic of interest is carriage of a particular type of resistant bacteria, it can be difficult to identify people to sample because it is challenging to encourage those with no symptoms to undergo testing. Moreover, sampling people purely for surveillance purposes in a representative way is expensive and resource-intensive, and has ethical implications that need considering. This is one reason why environmental sampling strategies are being developed, such as testing sewage (wastewater) for resistant pathogens.
Alternatively, research studies actively seek volunteers to participate in research. This can be a useful way to sample people outside of healthcare settings; however, research participants also differ systematically from the general population. For example, many research studies exclude population groups such as children or pregnant women from participating in research, to protect them from any risk of harms from doing so. (See the Legal and ethical considerations in AMR data module if you are interested in ethical aspects of AMR.) However, this exclusion also means data may not be available for these groups.
5 Putting it all together: sampling for AMR studies and surveillance
