Antimicrobial resistance (AMR) is identified by the European Commission together with the Member States, as one of the top three priority health threats in the EU (1). Due to AMR, a growing number of infections that used to be effectively treated with antibiotics are becoming harder to treat as antibiotics become less effective (1). It was estimated that more than 35.000 people in the European Union (2) and more than 1.27 million people globally (3) died in 2020 and 2019, respectively, as a direct consequence of an infection due to bacteria resistant to antibiotics. The health impact of AMR in the European Union is comparable to that of influenza, tuberculosis, and HIV/AIDS combined (4). Currently, antibiotic resistance leads to high costs in the European Union, including € 1.5 billion annually for the healthcare systems (1). However, the costs of AMR will be borne most severely by future generations, impacting economic stability and increasing severe poverty as early as 2030 (5, 6) and causing an expected 10 million deaths per year due to drug-resistant infections by 2050 (7, 8).

Antimicrobial resistance is mainly caused by the inappropriate use of antibiotics. The majority of antibiotics are prescribed in the community. For many years, ecological studies have shown a link between the use of antimicrobials and the frequency of AMR. In 2021, the European Union population-weighted mean consumption of antibacterials for systemic use was 15.0 defined daily doses (DDD) per 1,000 inhabitants per day in the community and 1.4 DDD per 1,000 inhabitants per day in the hospital sector (9). The majority of antibiotics in primary care are prescribed for respiratory tract infections (RTIs), urinary tract infections, and skin infections, with RTIs accounting for more than half of these prescriptions. In addition, in many regions around the world, antibiotics are accessible without prescriptions. Despite being more prevalent in low-income countries, even in some high- and middle-income countries, non-prescription antibiotic provision is a concern (10). A large part of the antibiotics consumed in the community is estimated to be used inappropriately as the majority are self-limiting (viral or bacterial) infections (11, 12). Hence, antibiotic stewardship, which implies a more rational prescription and use of antibiotics, is crucial. A recent point-prevalence audit survey about doctors’ management of nearly 5,000 patients presenting with symptoms of an RTI in 18 European countries found that antibiotic prescribing rates varied considerably. In four countries, less than 20% of consultations led to antibiotic prescribing, and in six countries, more than 40% of consultations resulted in antibiotic prescribing (13). An Italian study including nearly 2000 patients with acute RTIs found antibiotic prescriptions were given in 67.3% of the consultations. Two-third of prescriptions were not according to the guidelines (14). Similarly, substantial over-prescribing has been found in United Kingdom primary care: in 41% of all acute cough consultations when experts advocated for it in only 10% and in 82% for bronchitis vs. the ideal of 13% (15). In primary care in Ireland, approximately 60% of patients with RTIs get antibiotics (16).

Antimicrobial stewardship is urgently needed to safeguard antibiotic effectiveness for future generations. Healthcare regulators and payers play a key role in enabling behaviors that ensure antibiotics are used effectively and sparingly (11, 17–20). C-reactive protein (CRP) point-of-care testing (POCT) is a simple, validated, and cost-effective tool supporting antibiotic prescribing decisions. When CRP POCT is combined with clear guidance, communication skills training, delayed prescribing, and other safety netting procedures, primary care management of RTIs will be improved. CRP can be used to assess the severity of inflammation and predict whether an infection is likely to be self-limiting or severe. Self-limiting infections, whether viral or bacterial, usually resolve on their own without further treatment. Interpreted together with signs and symptoms in the majority of patients with RTIs, the need for antibiotics can be ruled out. Guidance on the use of CRP POCT and complementary strategies to improve antibiotic prescribing has been recently published (21–23) (Table 1).

Antibiotic overuse in humans and AMR are strongly correlated (35–37), and the total global economic direct and indirect costs of AMR due to resistance in five pathogens are already very high, estimated at USD 2.9 billion in the United States alone (38). According to the OECD, the cost of treating complications due to AMR can exceed USD 28.9 billion every year adjusting for purchasing power parity across 34 OECD and EU/EEA countries (39). This is calculated for a period from 2021 up to 2050. Most of the costs are caused by longer hospitalizations. Thus, the costs caused by AMR are high. On the other hand, every USD 1 invested in a mixed policy package across the health and food sectors generates returns equivalent to USD 5 in economic benefits achieved through reductions in health expenditure and increased productivity at work (39).

Unfortunately, the costs associated with AMR are rarely included in economic evaluations of antimicrobials and interventions such as diagnostics and vaccines that affect their consumption (38, 40). The cumulated economic costs of a single course of antibiotic treatment including the costs of AMR, however, are difficult to quantify precisely, and relatively few attempts to do so have been made (41, 42). Cumulative costs will vary depending on the specific context. Models for calculation are available, for example, Shrestha et al. calculated that a standard course of broad-spectrum penicillin has a cumulative economic cost of USD 9.30 in the United States and USD 10.40 in Thailand (38).

Reduced antibiotic prescribing has ecological benefits and leads to reduced selection of resistant strains. Furthermore, healthcare economics have demonstrated that implementing CRP POCT for RTIs is cost-effective:

Van der Pol et al. developed a model for general practice practices in the Netherlands comparing the use of a hypothetical diagnostic strategy to continuing the current standard of care for all patients with suspected acute RTIs. They concluded that primary care costs on consultations will be raised by 9 and 19% when consultation is priced 5 and 10€, respectively, while antibiotic consumption and AMR will be lowered (45). The price per consultation includes the costs of a test analyzer itself and materials used for the test and also costs related to the depreciation and quality assurance related to the use of the hypothetical diagnostic strategy. The efficiency of the CRP POCT device will be higher than accounted for in those studies since many CRP devices have the option to test more parameters for different illnesses on the same device.

Wubishet et al. (20) included 12 studies in their systematic review of economic evaluations of interventions aimed at reducing inappropriate antimicrobial prescribing in primary care. They concluded that there were significant variations in the cost-effectiveness of antimicrobial stewardship interventions across studies and depending on the inclusion of cost components such as the cost of AMR. However, communication skills training and CRP POCT were frequently found to be cost-effective or cost-beneficial for reducing inappropriate antimicrobial prescribing (20).

An effective implementation strategy needs to be chosen depending on the specific context within a country or region. There are a few operating models for CRP POCT implementations that have been successful. However, the strategies may differ from country to country, and there may be more feasible options.

Depending on the chosen model, various practical aspects need to be in place to ensure a successful and sustainable implementation:

Several Nordic countries, Switzerland, and the Netherlands are examples where CRP POCT has been successfully implemented in primary care.

In Norway, the use of CRP POCT started with the launch of a semi-quantitative test, Nycocard^TM CRP Visual, in 1989 (47) followed in 2000 by the launch of the Nycocard^TM CRP Single Test for the NycoCard Reader giving quantitative results (48). Around 1992, the costs attached to CRP POCT became reimbursed. Approximately 60% of all RTI-related consultations at primary care involve a CRP POCT, and in out-of-hour services, CRP POCT is performed in 70% of RTI-related consultations (49). A national guideline “antibiotics in primary care” gives instructions about the use of CRP and the interpretation of values when antibiotic prescribing should be considered for suspected lower RTI (50). Since 2012 there has been a marked steady decline in total antibiotic use with a reduction of 33% (51) and resistance rates are stable and low.

In Sweden, CRP POCT has also been used for 30 years in managing patients with RTIs (52). It is widely used in nearly two-thirds of RTI consultations (53). National guidelines advocate the use of the CRP POCT in primary care RTI consultations primarily for unclear lower RTIs (54, 55). In Sweden, primary care is often provided in group practices, called primary healthcare centers, with an average of four general practitioners (GPs) (56, 57). The primary healthcare centers, not the GP, are charged for the costs (58) when using diagnostic tests. The Swedish Strategic Program Against Antibiotic Resistance (STRAMA) illuminated the problem of resistant bacteria and increased awareness of antibiotic resistance (59). Levels of antibiotic use and resistance in Sweden are among the lowest among the European Union (EU) countries. Between 1992 and 2016, the number of antibiotic prescriptions in outpatient care, including primary healthcare, decreased by 43% (59).

In Denmark, CRP POCT for RTIs has been implemented for more than 20 years, and GPs have been paid a fee for performing a CRP test (60, 61). The first step was the implementation of CRP POCT in the national guidelines of the Danish College of General Practitioners (62). Danish Regions facilitated quality improvement programs based on audits and feedback on diagnostics and antibiotic use. The quality intervention program was organized by Audit Project Odense (APO). A nationwide retrospective cross-sectional register-based study found that between 2015 and 2017, approximately half of the antibiotic prescriptions for adults in general practice (49.6%) had an RTI stated as the indication, and a CRP test was performed in 45.2% of these cases (61). A 25% reduction in the consumption of antibiotics in primary care during 2011–2020 has been observed (63).

In Switzerland, CRP POCT has also been widely used for many years (64). Already in 2006, CRP has been the most used test for acute RTIs. According to Briel et al. for 42% of patients with acute RTI, at least one test was carried out by GPs in 2004, and in 35%, it was a CRP test (65). In 2022, a survey among GPs found that most of them (92–98%) selected a CRP POC test alone or combined with other diagnostics in the management of RTI (64). For participation in proficiency testing, practices hold a “practice laboratory” certificate. The CRP POCT is reimbursed and used in adults and children with acute infections (64). Switzerland has one of the countries lowest prescribing rates in Europe (66).

In the Netherlands, guidelines from the Dutch College of GPs recommend CRP POCT for the management of adults with acute cough (67, 68), suspected diverticulitis (69), and exacerbations of COPD (70) since 2011. Healthcare solutions recommended in clinical guidelines are considered as usual care and are almost automatically included in the “Basic Package of Care” and thus will get reimbursed. A large majority of the Dutch GPs have contracted a (regional) hospital or primary care lab for collaboration on POCT (CRP and possibly other tests). GPs are outsourced by the lab on purchase, organizational aspects, and logistics and can fully concentrate on patient care without financial incentives. In return, the lab can reimburse the test costs at the insurance company. The GPs who do not collaborate with a laboratory (a minority) have purchased the device themselves and can ask the insurance company for reimbursement of the actual test costs per patient. Since 2011, almost all Dutch GPs have used CRP POCT routinely 24/7 with high satisfaction of users and patients (71). The already low rate of antibiotic prescribing further decreased by 14%. Causality can only be suspected, but the use of CRP POCT has probably contributed to this decrease (72).

The implementation of POCT involves several actors and processes. Dewez et al. (72) evaluated which factors contribute to high vs. low adoption of CRP POCT in the Netherlands and England. They used a comparative qualitative case study approach and collected data through a review of documents and interviews with stakeholders. In both countries, early adopters of the tests advocated for its implementation by generating robust evidence and engaging with all relevant stakeholders. This led to the inclusion of CRP POCT in clinical guidelines in both countries. However, only in the Netherlands did this result in the reimbursement of the test costs. In addition, the better integration of health services enabled operational support from laboratories to general practice practices in the Netherlands. In contrast, in England, the development of a reimbursement program was prevented by the funding constraints of the National Health Service and the prioritization of alternative and less expensive antimicrobial stewardship interventions such as practice visits, needs assessment, peer feedback, and audits (72, 73). In addition, the lack of integration between health services limits the operational support to general practice practices.

Lingervelder et al. (74) compared the implementation of POCT in The Netherlands and Norway, where POCT is more widely implemented, with that in England and Australia, where POCT is not widely implemented. They analyzed the structure of healthcare operations and the transactions between stakeholders and found that the biggest challenge for countries with low POCT uptake was the lack of effective communication between the several organizations involved with POCT as well as the high workload for GPs aiming to implement POCT. They concluded that setting up a single national authority may be an effective step toward realizing the full benefits of POCT.

In June 2023, the Council of the European Union (75) and the European Parliament (76) published recommendations on EU actions to combat AMR. They encourage their member states to support the prudent use of antimicrobial agents, in healthcare settings, including primary healthcare settings and long-term care facilities, and community care in particular. One recommendation is the improvement of the availability, cost-effectiveness, and timeliness of diagnostic tests (76). Specific consideration is given to rapid tests conducted prior to the prescription of antimicrobial treatment, in particular in primary care, to ensure the optimal prescription. Where possible, antibiotic prescription should be restricted to face-to-face consultations. The European Parliament calls on the Commission “to work toward the development of EU guidelines including recommendations to strive to always carry out diagnostic tests, including rapid tests when available, prior to the prescription of antimicrobial treatment, … to ensure their optimal and most prudent use” (76).

C-reactive protein point-of-care testing is infrequently mentioned in national AMR (77–80) plans or by NGOs such as the WHO (81, 82) and the ECDC (83). Rather tests to identify the cause of an infection or susceptibility testing are usually recommended. Several policies promote at least the use of rapid diagnostic tests, for example, the French national AMR plan (79). The German government promotes in DART 2030 (78) the use of rapid diagnostic tests to differentiate between viral and bacterial infections as the initial investigation at the point of care.

An exception is the report from the OECD. The report “Stemming the Superbug Tide: Just A Few Dollars More” (6) gives detailed insight into the evidence and implementation of CRP POCT. The new OECD report “Embracing a One Health Framework to Fight Antimicrobial Resistance” (39) calls for scaling up the use of rapid CRP POCT in addition to other community-based interventions such as delayed antimicrobial prescriptions, introducing financial incentives to optimize antimicrobial use, scaling up mass media campaigns, and scaling up prescriber training. The availability of CRP POCT should be increased in ambulatory care settings in combination with antibiotic treatment guidelines. The modeled increase in the availability of CRP is assumed to reduce immediate antibiotic prescribing by 32% in adults and 46% in children under 18 years of age, and the intervention is assumed to yield immediate effects on antibiotic prescribing behaviors.

The WHO Policy Brief 63 is another exception. It highlights CRP POCT as an example of rapid laboratory testing, which should be made available to guide antibiotic prescription within a few minutes (84).

An overview of available CRP POCT, their use in primary care, and the reimbursement in specific countries can be found in addition to other important information in the review of the European Network for Health Technology Assessment (2019) (26) and more recently in the report of the PHG Foundation’s independent research and analysis for Health Action International (2023) (85).

Having access to a CRP POCT, on its own, is not enough to reduce antibiotic prescribing. Clear guidance to support physicians in their decision-making is needed to achieve safe and meaningful reductions. Several national respiratory guidelines (50, 54, 55, 62, 68, 89–98) from Belgium, Denmark, Germany, Norway, the Netherlands, Spain, Sweden, Switzerland, the United Kingdom, and South Africa have already implemented CRP POCT. Most of them provide cutoff recommendations, usually 20, 40, and/or 100 mg/L (Table 1). A few do not recommend any cutoffs (55, 94, 95). A clear cutoff guidance is essential as outlined before and should include detailed guidance on the interpretation of CRP values and guidance for the selection of appropriate patients, who would benefit from testing as recently provided by the authors. An update of national guidelines is recommended once CRP POCT for the guidance of antibiotic prescribing for RTIs is not yet or insufficiently implemented.

To maximize the safe reduction of antibiotic over-consumption, CRP POCT should be implemented in conjunction with several complementary strategies (31–33, 99, 100). As discussed previously (21–23), maximal impact can be gained when policymakers offer communication skills training to medical students and physicians, enable the application of delayed prescribing techniques, and encourage the use of decision aids during the consultation.

Lack of reimbursement and costs associated with CRP POCT for physicians have been identified as the main reason for non-adherence to guidelines that recommend the use thereof (11, 101). In general, a neutral reimbursement level that balances costs for physicians, health insurance, and the healthcare system is recommended to ensure the proper use of CRP POCT. Neutral reimbursement levels would not incentivize overuse, nor de-incentivize use of the tests when they could be beneficial, thus maximizing the systemic effectiveness of an implementation. Slight over-reimbursement might be a solution in the short term if the healthcare-economics business case permits it, to encourage the use of CRP POCT, especially in an early adoption stage.

Table 3 covers the elements of cost to consider when determining appropriate reimbursement levels. Actual costs will differ per country/region. Adequate reimbursement should go to the party that is performing the action or service and thus carrying the cost, so in the case of collaborations with labs or other third parties, reimbursement should be split.

Though POCT devices are usually designed to be easy to handle also by non-lab-staff, the implementation of POCT may be facilitated for physicians/practices if they are accompanied by specialists from laboratories, national institutes, or non-profit organizations. Those specialists may ensure that the devices are used as intended and according to the local standards. They may support with consultation, giving advice regarding where to place the device best, for example, not next to a heater, or about the cooling logistics for the tests, perform training, and offer troubleshooting. Especially they may keep control of the quality assessment to ensure the establishment of quality standards.

Enable or facilitate the gathering and sharing of data and encourage ongoing peer-to-peer interactions/discussions, small group meetings or focus group discussions (102–104). The data should include comparative numbers between practices from within a region, normalized per patient type. These could compare relative antibiotic prescription behavior, the number of CRP POCTs performed, and the satisfaction of patients, physicians, and supporting staff. Data should be tracked over time, including baseline numbers from before the implementation of CRP POCT and complementary strategies.

Additional positive reinforcement through recognition of best performers may be considered in countries where CRP POCT is not yet established or in the early stages of adoption, and antibiotic prescribing rates are high. This can be an effective way to influence physicians’ behaviors according to the nudge theory, a concept that proposes positive reinforcement and indirect suggestions as ways to influence the behavior and decision-making of groups or individuals (105, 106).

Financial incentives are not recommended, since good performance indicators for this subject are notoriously difficult to define, track, and interpret, and poorly defined and implemented incentivization strategies can encourage wrong behaviors and be counterproductive.

Recently, the Joint Programming Initiative on Antimicrobial Resistance—Primary Care Antibiotic Audit and Feedback Network (JPIAMR-PAAN) developed best practice guidelines for peer comparison audit and feedback for antibiotic prescribing in primary care in high-income countries (107). They defined four categories with 13 best practice recommendations. For example,

As highlighted by the JPIAMR-PAAN, inappropriate antibiotic prescribing behavior is complex (107). On the one hand, there are clinically important factors, such as AMR and the risk of adverse drug reactions. On the other hand, there are emotional factors driving unnecessary prescribing, such as perceived patient expectations, time constraints, habits, and fear of patients worsening. Thus, multifaceted approaches to antibiotic prescribing in primary care to facilitate a change in prescribing are essential. Audit and feedback should be implemented alongside clinical decision support, point-of-care diagnostics, patient and prescriber education, and safety netting procedures.

Despite the existing evidence and seeming consensus regarding the usefulness of CRP POCT and other strategies in reducing antibiotic over-prescription in primary care settings, implementation and adoption are still limited. This is linked, in part, to the lack of context-specific evidence for the effectiveness of CRP POCT use. Countries/regions may need to set up pilot projects or clinical research projects to generate a local, context-specific assessment of efficacy and cost-effectiveness and understand barriers and facilitators to scaling-up.

Communication with the broader public—the patients and parents—will be key (7). In many countries, it may be necessary to build awareness and understanding that the inappropriate use of antibiotics is the main cause of AMR to enable and reinforce the adoption of desired behaviors. It is important that patients and parents understand their key role in reducing antibiotic consumption and that antibiotics are not antipyretics.

A multi-channel communication approach is suggested, deploying messages simultaneously via various target audiences such as parents, adults, and teenagers. Campaigns should be repeated every 3–4 months for continuity and higher impact, and strongest during (possible) periods with the highest prevalence of RTIs.

Healthcare professionals will play a key role in both reducing the number of antibiotic prescriptions and also in dialoguing with patients and parents, spreading the messages related to antibiotic stewardship. Physicians need to understand the basics of AMR, the risks related to the misuse of antibiotics and what constitutes appropriate use, and the clinical benefits of CRP POCT in reducing diagnostic uncertainty and the antibiotic prescription decision-making process.

Physicians need to get the chance to learn of, and experience, the support of CRP POCT in daily practice. Although AMR will be seen as a problem, on a day-to-day basis this is mostly not experienced in the consultation room with a patient with a lower RTI. Instead, diagnostic uncertainty, time pressure, and perceived antibiotic beliefs of patients are important concerns that are nowadays still mostly silenced by prescribing an antibiotic. CRP POCT as a tool and a low CRP test result at hand will encourage both the patient and the physician to change this (high) antibiotic prescribing behavior.

Adapted communication and training programs (19, 26) should be utilized for pilot projects and during the broader roll-out of the CRP POCT to primary care to ensure the tests are deployed effectively and in combination with the strategies that maximize impact. Continued communication is recommended to keep interest in CRP POCT high and to continue convincing physicians of its benefits and increase use.