Diagnostics are often lifted as a part solution in containing the development and spread of antibiotic resistance. Sometimes, the word is used as if it was a magic wand – if we only had more or better diagnostics all problems would be solved. But how do diagnostics work, and what effects can we expect from implementing a new diagnostic method?
In an almost trivial sense, diagnostics are used to determine what disease a patient is afflicted by, and in the case of infectious diseases what is causing the disease. This information is then used to determine what the appropriate therapy is: for example a specific antibiotic, other medicines or bed rest. In the absence of a proper diagnosis many patients may not get the correct antibiotic treatment when they need it, but another common problem is that patients who do not need antibiotics are given one just in case it is a bacterial infection.
Thus, diagnostics can be said to serve a dual purpose: to reduce morbidity and mortality by giving information about which antibiotic to use and to reduce the total amount of antibiotics consumed. By extension this would reduce the development and spread of resistance. In addition, diagnostics serve as a source of information for surveillance programs and making antibiograms. This information can be used to help choose antibiotic in those cases where antibiotic therapy needs to be started without diagnostic confirmation.
Three categories of diagnostics
In broad terms, three categories of diagnostics can be identified:
1. Clinical algorithms
These are decision-making trees or guides, where clinical characteristics and risk factors are used to give a diagnosis. An example would be ALMANACH, which has shown to reduce antibiotic prescriptions in children by 80% in Tanzania without negative health impacts.
2. Biomarker tests
These determine the concentrations of markers that are released in the body in response to an infection. The tests can be used to determine the presence of an infection and provide information on whether the infection is likely to be caused by a bacterium or a virus. Examples of biomarkers include CRP, PCT, TRAIL and HNL.
3. Species identification and antimicrobial susceptibility tests
A multitude of techniques are available in this category that provide information about the species of the infecting micro-organism or its susceptibility to antimicrobials.
The diagnostic chain
The diagnostic chain can utilize clinical algorithms at the point of care while the physician or health care worker meets the patient. Biomarker tests can, depending on equipment and training level, be performed as an add-on at the point of care or near the point of care, for example in a separate room at a health center, and can give results within an hour of taking the sample. For the identification of the pathogen and its susceptibility, current methods require dedicated laboratories with trained staff and more time. A typical, culture-based diagnostic such as depicted in the image above will require at least three days to provide results, unless rapid diagnostic tests are used to speed up the processes. However, current and most upcoming diagnostic methods target the last day in the chain as they generally require a primary culture or otherwise enriched sample. New advancements in rapid diagnostics should therefore include ways to decrease the time from sampling to the actual diagnostic test.
Reality check new diagnostic methods
It is easy to call for newer and faster diagnostics to solve the problems related to antibiotic misuse and resistance. New rapid diagnostics can certainly have a great impact on stemming improper antibiotic use and improving patient care. But unfortunately the diagnostic of our dreams; cheap, rapid and giving specific treatment guidance, is often just that – a dream. And before getting carried away with a new solution that promises the moon, its always good to do a reality check:
- How does the new method fit into existing healthcare or laboratory systems?
- What capacities are required to be able to implement the method?
- Is the data generated sufficient or superfluous relative to the need?
- What will the total cost and individual test cost be over the expected lifetime?
At the end of the day, the main question to ask is probably this:
Will the test have the desired impact on clinical and prescribing practices – reducing morbidity and mortality or increasing appropriateness of antibiotic use?
You can also read articles
More news and opinion
- Five focus areas at the 2nd Ministerial Conference on AMR hosted by the Netherlands
- Safety concerns of fecal microbiota transplants
- Upcoming ReAct Africa Conference: universal health coverage and antimicrobial resistance in focus
- Mother Earth conference in Argentina – the environment in focus
- Diagnostics – what are we talking about?
- Connecting global to local civil-society-agenda on AMR at CSO convening in Geneva
- ReAct colleagues featured in WHO Bulletin as leading profiles in the work on reacting to antibiotic resistance
- RAN stakeholder at WHO IPC consultation – for standards and guidelines in African Union member states
- WHA conversation on Antibiotic Resistance as a Global Development Problem co-organized by ReAct
- 8 insights from ReAct Asia Pacific project on antibiotic stewardship in secondary level hospitals in India
- Open letter to UN Member States from former IACG members Anthony So and Otto Cars
- ReAct UHC Intervention at UNGA Multi-stakeholder Hearing for High-level Meeting on UHC
- ReAct Latin America honors Earth Day
- Medicines Patent Pool’s view on the role of licenses for antibiotics – World Intellectual Property Day
- Second time for Innovate4AMR competition!
- World Health Day 2019: Universal Health Coverage
- Constraints for successful implementation of diagnostics
- Antibiotic Shortages: magnitude, causes and possible solutions: A new WHO meeting report
- Erry Setyawan, FAO, on Indonesian NAP: We need to work together to make it possible to manage AMR
- ReAct’s new 5-year strategic plan receives funding from Sida
- How infections spread and how to stop them
- Generating data for policy and practice