Abstract
Introduction
Bacterial pathogens exhibiting antimicrobial resistance (AMR) are a health threat for humans, companion animals and livestock. Surveillance underpins appropriate AMR responses, but can be biased or expensive. Surveillance using Lot Quality Assurance Sampling (LQAS) has shown promise in human health settings; more evidence on its applicability and acceptability in livestock populations is needed. Drawing on in-depth interviews, this article examines stakeholder perspectives on LQAS for AMR surveillance in livestock in Germany.
Methods
Twenty-five stakeholders were interviewed. They included employees of German universities, research institutes, Federal animal health services / veterinary laboratories, animal producer associations, and veterinarians. Detailed notes of online interviews were analyzed using a framework approach.
Results
Respondents were concerned about AMR in livestock and also about restrictions on antibiotic treatment options. They described the multifaceted, legally prescribed data gathering for farmers to monitor antibiotic consumption and the widespread use of antibiograms to guide treatment in Germany. Respondents saw potential benefits of LQAS for AMR surveillance, in terms of reducing the sample sizes and the need for antibiotic susceptibility tests, but there were questions about surveilling commensal bacteria, with concerns about it leading to further restrictions on antibiotic consumption and driving food production overseas.
Conclusion
An LQAS approach to AMR surveillance requires locally responsive guidance to alleviate concerns about further restriction of treatment options (and about animal welfare). Given existing data collection, recording and reporting burden for farmers and veterinarians, early engagement is needed to agree the rationale and benefits of LQAS, particularly if surveilling resistance in commensal bacteria is considered.
Keywords: AMR surveillance, Livestock, Germany, Qualitative, Stakeholders, Lot quality assurance sampling
Highlights
-
•
Farmers and vets reported arduous AMR surveillance data reporting.
-
•
Respondents were concerned about further restrictions on treatment options.
-
•
Attitudes to LQAS for AMR surveillance in livestock in Germany were mixed.
-
•
Early engagement is needed to agree the rationale and benefits of LQAS.
-
•
This is particularly important if surveilling resistance in commensal bacteria.
1. Introduction
The emergence of bacterial pathogens exhibiting antimicrobial resistance (AMR) is a critical global health threat for humans, companion animals and livestock. Although recent data from Europe suggest decreasing levels of AMR [9], estimates based on more limited data from low- and middle-income countries (LMICs) suggest a more serious situation [27]. Nonetheless, estimating the global burden of AMR in livestock is complicated by the absence of an equivalent to the WHO's Global Antimicrobial Resistance Surveillance System, which collects AMR data for selected indicator bacteria across [19].
For livestock, the direct implications of AMR are clear: more frequent treatment failures, increased severity of infections, and, with fewer treatment options, negative impacts on animal welfare [2]. Indirect consequences for humans include financial losses for producers, and risk to food safety and security [2,17]. With the transfer of antibiotic resistance determinants from bacteria in other animals to bacteria in humans established, AMR in livestock also brings risks for human health [13]. However, more data are needed to understand the complex systems involved in transfer of AMR and quantify this risk [29].
Action is needed to ensure the effectiveness of antimicrobials in human and veterinary medicine, to treat disease and sustain food production. Surveillance is a key component to generate an evidence base for appropriate responses to AMR: it enables researchers and program managers to follow trends in AMR across space and time, can inform empirical treatment guidelines and helps to monitor the impact of interventions. Laboratory-based approaches for AMR surveillance in animals often involve routine culture and susceptibility testing on samples taken in clinical disease management. This approach assumes that samples are routinely submitted in cases of suspected infection. If this assumption is unsound, estimates may be heavily biased. Another approach to surveillance is through population-based sampling of feed, foods and healthy animal species. Such an approach, which is generally designed to produce precise estimates of carriage, is time-intensive and costly. Resultant sample sizes are often inadequate [18,24]. A rapid, feasible and affordable surveillance strategy that can reliably inform prescribing guidelines and identify priority settings for interventions would be a valuable tool.
Lot Quality Assurance Sampling (LQAS) [6] is one surveillance approach to identify local variations in the prevalence of drug-resistance [12,20]. A survey method based on classification, LQAS does not seek a precise estimate for the level of resistance but rather to answer questions such as, “Is the level of AMR above a threshold that requires action?” Using a random sample from a well-defined homogeneous population (“lot”), the lot is classified as having either a high or a low prevalence of a defined outcome based on an a priori decision rule (Fig. 1). Crossing the threshold (e.g. such as a pre-defined prevalence of a bacterium with a specific AMR trait) would initiate action (e.g. such as changing prescription guidance for the specific antimicrobials). Using a classification approach, instead of a conventional estimation of AMR-prevalence, markedly reduces the required survey sample. Given this advantage, LQAS has been applied in monitoring and evaluation of diverse health programmes, including to assess immunization coverage [14,23,25].
Fig. 1.
A graphic representation of LQAS. Only a sample of items from a lot is submitted to assessment. If the number of positive results (coloured in red) equals or exceeds the predefined decision rule (in this case, the threshold is set at five), the entire lot is classified as high (i.e. exceeding the threshold).
To date, LQAS has been applied in human medicine for population-based surveillance of AMR in urinary tract infections (UTIs) [10,28] and in the field of tuberculosis in Western Kenya [16]. Through this approach, locally relevant differences in AMR prevalence classifications were identified across public and private primary care centers and hospital wards [10]. The principles of the LQAS approach apply in other domains and it offers great promise for surveilling AMR in livestock. However, more evidence is needed on its effectiveness and applicability of LQAS-based AMR surveillance in other settings. Investigation of LQAS in diverse contexts and domains, including livestock (surveillance of resistant commensal E. coli in pigs or broiler chickens in Germany) is ongoing [15].
As with any health intervention, effectiveness is contingent on policy translation and implementation. This is particularly the case for LQAS-based AMR surveillance because it yields a classification of lots rather than estimates of prevalence. Policy making and implementation occurs in an environment of diverse stakeholders with varied, sometimes conflicting interests and perspectives [5]. Several stakeholder groups are relevant to policy and implementation of AMR surveillance for livestock in Germany. These include industry representatives, infectious disease experts, microbiologists, veterinarians and farmers. Drawing on interviews with members of these groups, and focusing on pork and chicken production, this article examines potential opportunities for and barriers to policy translation and implementation of LQAS-based AMR surveillance in livestock in Germany.
2. Methods and setting
This exploratory qualitative study used in-depth interviews with stakeholders involved in AMR surveillance in pigs and broiler chickens in Germany.
2.1. Setting
In Germany, two laboratory-based approaches for AMR surveillance in animals are established. One involves susceptibility testing on isolates provided by 25 cooperating national laboratories and obtained from clinical samples by routine culturing [3]. This is termed GERM-Vet and conducted according to the “Act on Veterinary Medicinal Products and on Implementing the European Regulation on Veterinary Medicinal Products in Germany” by the Federal Office for Consumer Protection and Food Safety (BVL). The assessment of AMR characteristics is carried out with clinical breakpoints to evaluate therapeutic options. In case veterinary-specific breakpoints are not available, MIC90 values allow for an estimation of the antimicrobial susceptibility and therapeutic efficacy. The data are evaluated at national level and are published annually by BVL.
The second is population-based AMR surveillance, focusing on zoonotic and commensal bacteria and based on the “AVV Zoonosen Lebensmittelkette”, following EU guidelines (Directive 2003/99/EC; Implementing Decision (EU) 2020/1729). Samples and/or strains from feed, food and clinically healthy animals are examined by the Federal Institute of Risk Assessment (BfR) in accordance with the annual nationwide zoonoses sampling plan. The sample takes regional animal densities into account or, in the case of slaughterhouse samples, the number of slaughtered animals. It provides national prevalence estimates in the livestock population which are essential for risk assessment in the food chain (Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL) [4]).
2.2. Methods
2.2.1. Respondent selection and recruitment
To identify and recruit stakeholders involved in AMR surveillance in livestock, purposive diversity sampling was used. Stakeholders were identified by researchers engaged in fundamental and applied research on AMR-surveillance in animals in Germany. Potential respondents in several stakeholder groups were identified based on their positions in: antibiotic stewardship/guideline committees and state/legal bodies involved in AMR surveillance; laboratories involved in AMR surveillance; veterinary associations; veterinary practice; federal veterinary health services; and associations of farmers/meat producers. Potential respondents from all of these categories were invited to an on-line kick-off meeting during which the LQAS approach to AMR surveillance was explained. The introduction was recorded and made available to potential respondents. These meetings sensitized potential respondents to the LQAS surveillance approach (with which they were likely unfamiliar) helping them to weigh up potential benefits and challenges in an informed and considered manner.
In October and November 2022, potential stakeholders were then invited to an interview via email. Recruitment continued until a point of theoretical saturation was reached, i.e., when no further relevant and novel information was forthcoming from subsequent interviews.
2.2.2. Data collection tool
The interview guide addressed four areas: respondent background (qualifications and current role in AMR surveillance); AMR in general and in relation to their role and their organization; existing AMR surveillance programs; LQAS, its possible use in surveilling AMR in livestock in Germany, plus benefits and disadvantages.
2.2.3. Data collection procedures
All interviews were conducted via online video conferencing software at the respondent's preferred time. Interviews were conducted in German (by RW, a native German speaker and biotechnologist with experience of conducting in-depth interviews with stakeholders). A researcher (one of the co-authors) with expertise in LQAS was present to assist in responding to any questions about LQAS.
2.2.4. Data processing and analysis
With the consent of respondents, interviews were audio recorded. In case the respondent declined audio recording, detailed notes were taken during the interview. Audio-recordings were reviewed and detailed notes were taken in English to summarize the discussion on a question-by-question basis. Identifying information was removed from the detailed notes. The detailed notes were coded using a codebook based on the research question and the structure of the interview guide. The coding of the detailed notes was undertaken deductively in MS Excel by two members of the research team (CP and RW). In the first instance, coding was conducted independently and differences in the coding discussed and resolved in subsequent analysis meetings. Selected excerpts from the audio recordings were transcribed verbatim and translated to English (by RW) to illustrate specific findings.
2.2.5. Ethical approvals
After consultation with the ethical review board of the Justus Liebig University of Giessen, Germany and sharing of the research protocol, the study was deemed exempt from ethical approval (AZ 20/23). Participants were sent “Declaration of consent” and “Data protection declaration of consent according to Art. 4 no. 11, 7 EU GDPR” forms for signature two weeks before interview. Interviews were conducted if signed forms were returned beforehand.
3. Results
Twenty-five stakeholder interviews were carried out. Interviews lasted between 30 and 60 min. All interviews were conducted individually apart from one. In the single group interview respondents from three departments of the same state agency were present and answered the questions, depending on their expertise. All but one interviewee had a background in veterinary science. Many had further board-certified specialization in poultry and swine diseases or microbiology or epidemiology. Some had a double specialization. The interviewed stakeholders were employed by universities or research institutes, animal health services (pig or poultry) or veterinary laboratories of Federal States. Two respondents were responsible for veterinary services in large livestock producers. Two were partners in large private veterinary practices. One was the veterinarian in an association of animal producers. A further respondent was responsible for data on antibiotic consumption in a private quality assurance company. One led a national ministry's AMR monitoring program. Ten were directly involved in the treatment of animals with antibiotics or advising other veterinaries on the use of antibiotics.
From the interviews with stakeholders, the following key themes were identified as relevant to the potential implementation of LQAS for AMR surveillance in livestock in Germany: attitudes towards AMR and antibiotics in livestock; perspectives on current AMR surveillance and data gathering; and attitudes towards LQAS for AMR surveillance in livestock.
3.1. Attitudes towards the use of antibiotics and AMR in livestock
Respondents expressed nuanced attitudes to the risks posed by the use of antibiotics and AMR in the livestock industry for human medicine. They pointed to other factors that were more pertinent to the development of AMR in human pathogens and highlighted dangers to animal welfare regarding greater restriction on the use of antibiotics in livestock. Nonetheless some also conceded that antibiotics were not always used appropriately in the livestock industry.
All respondents mentioned the possible transfer of antibiotic-resistant bacteria (or resistance determinants) in livestock to humans (or bacteria relevant to human pathologies). Many questioned, however, whether this transfer is a major issue. Respondents directly involved in administering antibiotics were not convinced that this transfer plays a major role in the development of resistance in human-relevant pathogens and the subsequent failures in human therapy. They pointed to excessive and improper use of antibiotics in humans as the major cause of AMR in human medicine. Respondents also highlighted the need for nuance with regard to trends in AMR in non-human animals. Concern about AMR in animals varied across livestock classes, household pets and wild species. Household pets were seen as an overlooked area, where the use of antibiotics is less closely controlled than in livestock.
[…] it is not yet entirely comprehensible to me how the resistance problem in human medicine is related to that in veterinary medicine. Well, I have the impression that they have created their own problems and we have created our problems – independently of each other.”
A veterinarian/microbiologist at a federal laboratory (Interview 13).
Most respondents, including those not directly treating animals, were concerned that the armamentarium of antibiotics for the treatment of livestock will be further restricted by regulators. Respondents considered colistin to be the most threatened antibiotic: they highlighted how, in several other European countries, it is already prohibited for veterinary use. This reflected concern about being left without therapeutic alternatives for sick animals.
We have always had to struggle with the fact that many substances have simply been taken away from us. That's actually more of a problem…that colistin is being taken away from us …
A partner in a large veterinarian practice for pigs (I8).
Stakeholders from the chicken industry were particularly frustrated to see a shrinkage of approved antibiotics yet also a lack of market authorizations for alternative therapies, which will not interfere with human medicine, such as the use of bacteriophages or competitive exclusion cultures. The regulation of antibiotic use in the poultry industry was described as driven not only by regulatory authorities but also by consumers. A couple of respondents from the poultry sector confirmed that very large customers (e.g. fast food chains) prohibit the use of certain legally available antibiotics in animals from which their meat products are produced. The lack of therapeutic alternatives was a dilemma for veterinarians and farmers: aware of the welfare needs of livestock, they do not want to deprive animals of necessary treatment and do not want to cull them as a last resort.
Nonetheless, there were also references to overuse of antibiotics as a cost-effective quick fix. Some respondents called for greater support for improved animal husbandry and a more holistic approach to livestock health for infection prevention and animal welfare. In addition to vaccination programmes (as well as vaccine development) and improved biosecurity, this must incorporate antibiotics, which were seen in some cases as the most effective way to ensure animal health, particularly given the cost pressures and competition that farmers face. Respondents also highlighted veterinarians' familiarity with the effectiveness of antibiotics on particular farms and the value of this experiential knowledge for successfully treating infections in animals.
3.2. Perspectives on current AMR surveillance and data gathering
Current efforts to address AMR were described, with emphasis on the collection of data on the antibiotics administered and the use of antibiograms in the treatment of sick animals. There were varied reports regarding whether antibiotic susceptibility tests were obligatory in the treatment of all sick animals, with one respondent suggesting that they could be used in some cases and later undertaken if treatment was observed to fail. Respondents also described financial reasons for the treating veterinarians to undertake antibiotic susceptibility tests alongside treatment i.e. they could charge an additional fee for them. In general, antibiograms were described as widely used and available, although there were some questions about laboratory practices and data quality. There were also references to how in-vitro resistance tests should be considered together with the experience of veterinarians in the field with regard to clinical management of disease. And with data on resistance from antibiograms, respondents described being able to monitor trends on a farm-by-farm basis.
The data collected depended on the livestock species and the life-stage of the animal. For example, the veterinarians responsible for broiler chickens had extensive information on bacterial infections and resistances in pathogens. This screening was described as conducted on the initiative of the producer to address any potential complaints from partner farms about infections in the chicks they produce. In such cases, the chick producer therefore has evidence that the bacterium was not present in the freshly hatched chicks.
Farms reported collecting data on antibiotic consumption, with data collection sometimes outsourced to private companies. As a legal requirement, these data are reported to the Federal State authority (and then at European level), who establishes quartiles in terms of antibiotic treatment frequency for specific species and age groups. Farms in the upper quartile of consumption frequency are required to develop and implement an action plan for improved animal husbandry (e.g., greater ventilation and more space for animals). One respondent described how, because of overall low levels of consumption in calves, any treatment meant writing an action plan. Paradoxically, reporting zero administration of antibiotics over six months could be judged by the ‘authorities’ as implausible and prompt an investigation into false reporting or not meeting animal health requirements. Another respondent referred to the unrealistic reports of antibiotics used in pigs, described as being to avoid ‘accountability’. One respondent described how the burden of data gathering and the resultant pressure to reduce antibiotic consumption in Germany could however be sideline by producers' moving operations outside of Germany. Furthermore, a partner in a large veterinarian practice for pigs highlighted how moving production to other European and non-European countries could have an overall detrimental impact on AMR globally because antibiotic stewardship is much less stringent in other settings.
3.3. Attitudes towards LQAS for AMR surveillance in livestock
When asked directly about the potential for an LQAS approach to AMR surveillance in livestock, its advantages and disadvantages, the respondents highlighted the following: reduced sampled size; the need for homogenous lots; surveilling commensal resistance in commensal bacteria; the potential to reduce antibiotic susceptibility tests; and the risk of additional data and analysis leading to further restrictions on antibiotics.
Respondents commonly cited the benefits of a reduced sample size. Benefits and concerns about obtaining homogenous lots were also mentioned. Obtaining homogenous lots was described as potentially difficult to achieve in some instances because of differences in age cohorts on the same farm e.g. piglets and fattening pigs or broiler chickens of different age groups in parallel on the same farm. Hence, in some farms, this might not lead to a reduction in the necessary resources because of the level at which the lot needed to be selected.
But I also find it interesting as an advantage [of the LQAS method] that you are forced to define this lot very precisely and to think very carefully about what is the group for which I am making a statement. This forces you to limit the statements to very specific groups, for now. And not to be able to generalize too much.
A veterinarian/epidemiologist (I12).
Interest in understanding AMR levels in commensal as well as pathogenic bacteria was mixed. Some respondents were uncomfortable about surveillance in commensal bacteria potentially being used to determine the use of antibiotics for pathogenic bacteria when there was uncertainty about the level of transfer between strains. Questions were also raised about which bacteria to monitor and where samples would be collected. Sample location was highlighted because there was the potential for resistant strains to originate from beyond the farm e.g. through the water supply.
In a background of frequent use of antibiotic susceptibility tests prior to treating sick animals, LQAS was viewed as having the potential to reduce the number of such tests. This could have cost savings, which was seen as important to producers in such a cost-sensitive and competitive industry. In these circumstances, an LQAS approach was described as a useful additional data source to take action (e.g. influencing treatment), but not the sole determinant.
On the other hand, it was mentioned that the additional data collected, their analysis and interpretation could lead to further restrictions on the use of antibiotics, which were considered already excessive and hence would lead to risks for animal health. This might particularly be a risk if the data are inappropriately interpreted, with misinterpretations leading to unnecessary restrictions on commonly used antibiotics. This was also seen as likely given that it is a new and rather unknown approach.
4. Discussion
Interviews with a diverse group of 25 stakeholders highlighted a range of issues relevant to potential translation of LQAS into policy and its implementation in practice AMR surveillance in livestock, particularly pigs and broiler chickens, in Germany. Respondents were concerned about AMR in livestock but also about further (and generalized) reductions in the antibiotics available for treating sick animals. They described the data gathering activities that farmers in Germany conducted to monitor antibiotic consumption and for AMR surveillance, and the widespread use of antibiotic susceptibility testing to guide treatment. Finally, respondents saw potential benefits in terms of an LQAS approach to surveillance, in terms of reducing the sample size and potentially leading to the need for less antibiotic susceptibility testing. However, they had questions about obtaining ‘lots’, the need to surveil commensal bacteria and the potential for additional data collection leading to further restrictions of antibiotics.
Respondents raised questions about the contribution of AMR in livestock to the problem in human medicine. This is not the first time that stakeholders in the European livestock sector have expressed skepticism about the contribution of antibiotic use in livestock to the development of AMR human pathogens [1,8,11,22,26]. Although there is evidence that transfer occurs [13], given the complex systems involved, more research on the frequency and risk is needed [29]]. This is however beside the threat that AMR poses to animal welfare, and food safety and security [2,17]. Given that these attitudes have been repeatedly described, it seems that attention is needed to how the threat of AMR in the livestock sector is communicated. One approach is to acknowledge potential risk to human health, and to emphasize the impacts of AMR on animal welfare, food production and livelihoods, and acknowledge concerns about reducing treatment options for sick animals. A good understanding of the risk (to humans and other animals) is necessary to facilitate such communication. This underlines the need for improved surveillance and further fundamental research on transfer of resistance traits.
Respondents expressed concerns about animal welfare, when referring to wide-ranging bans on antibiotics. What they saw as broad-brush bans (e.g., for colistin) highlights the need for locally relevant and specific (drug-bug) estimates of AMR. In this case, guidance (rather than prohibition) can be adjusted to maximize treatment success given locally relevant data. This is precisely what an LQAS approach can offer [10,28], in contrast to the current Germany-wide AMR surveillance on selected bacterial pathogens, which is not used for – and unlikely to provide meaningful – estimates at sub-national level (e.g. federal state) given the small sample sizes.
During the kick-off meeting with stakeholders, a current project undertaken by the authors was presented as an example of using LQAS focusing on commensal bacteria. Although it was also emphasized that an LQAS can be applied for commensals or pathogens, respondents repeatedly made skeptical remarks about AMR surveillance in commensal bacteria. These doubts and the concerns about the potential implications were notable given that, as part of the BfR-led surveillance of zoonotic bacteria and AMR following EU guidelines (Directive 2003/99/EC; Implementing Decision (EU) [3]/1729), data on commensal bacteria are already collected. With other qualitative research on the drivers of AMR indicating that education and awareness raising are not sufficient [21], this highlights the important of engaging farmers/producers and veterinarians in discussions on the benefits of surveilling commensal bacteria. Nuanced communication, within the context of public and community engagement [30], regarding the purpose and value of surveilling commensals is needed.
Respondents emphasized the financial pressures that livestock producers face. Some pressures were linked to efforts to surveil and address AMR. The economic impacts of AMR on livestock, in terms of untreatable infections, were not highlighted. The reference to the risk of off-shoring the livestock industry, moving production to territories with a lower burden of AMR-related controls, particularly highlights the need to balance the burden of surveillance with the costs. Elsewhere, producers have been described as stuck between the demands of the market for affordable (and sometimes antibiotic free) meat and the need to maintain standards of animal welfare and biosecurity [1]. LQAS may offer potential cost savings if the approach is accompanied by changes in policies regarding the application of antibiotic susceptibility testing.
A surveillance system that involves an unfamiliar approach to sampling was seen as being potentially misunderstood. This was particularly concerning if the approach was accompanied by prohibitions of antibiotics based on specific thresholds. A participatory approach in the French pig and poultry sectors highlighted the need for consensual standardized monitoring tools that allow farmers and veterinarians to jointly monitor the health, welfare, AMR, and antimicrobial use on farms [7]. This stresses the need for meaningful and long-term engagement with stakeholders regarding any implementation of LQAS. LQAS does not per se imply that antibiotics are prohibited once a threshold is crossed. For example, LQAS could also be used to classify lots under the defined threshold as a quality criterion for producers. Another approach would be to inform veterinarians, farmers and the public (e.g. via labelling) about high prevalence lots, so that they could make informed decisions with regard to the treatment, sale and consumption of these animal products.
4.1. Strengths and limitations
The respondents included stakeholders purposely sampled from a diverse range of institutions: state bodies and private actors, including producers and veterinarians. Because of the specific roles of stakeholders, a purposive approach to sampling and recruitment was essential. This was combined with specific sampling across the various respondent groups to ensure a broad range of perspectives. The findings are however limited by the absence of representatives from different scales of producers (i.e. small or medium scale). Introducing potential respondents to the LQAS approach in a kick-off meeting also meant that respondents were able to contribute informed reflections on its implications and potential barriers. The presence of an expert during the interview, although useful in terms of addressing any points of doubt about the approach, may have contributed to desirability bias, particularly because of previous professional connections with the respondents. We deemed that this was outweighed by the need for presence of the expert to respond to any queries from respondents regarding LQAS. Given the novelty of LQAS as an approach to AMR surveillance, this was particularly important to enable the respondents to consider potential pitfalls and benefits.
5. Conclusion
Lot quality assurance sampling is a novel approach to AMR surveillance, under assessment for use among livestock. Interviews with stakeholders highlighted considerations for its potential translation into policy and implementation in practice for livestock AMR surveillance, particularly in pigs and broiler chickens. There were concerns about AMR in livestock but also about generalized restrictions on antibiotics available for treatment and the burdens of data collection. Respondents saw potential benefits of LQAS for surveillance, by reducing the sample size and potentially leading to fewer antibiotic susceptibility tests. There were questions about the ‘lots’ that could be used, the benefits of surveilling commensal bacteria. A corresponding locally responsive treatment guidance is needed to alleviate concerns about further restriction of treatment options (and about animal welfare). Given the existing data collection burden for farmers and veterinarians, early engagement is needed to agree on the rationale and benefits, particularly of surveilling resistance in commensal bacteria.
CRediT authorship contribution statement
Christopher Pell: Writing – review & editing, Writing – original draft, Supervision, Methodology, Funding acquisition, Formal analysis, Conceptualization. René Wagner: Writing – review & editing, Investigation. Christa Ewers: Writing – review & editing, Funding acquisition, Conceptualization. Constance Schultsz: Writing – review & editing, Supervision, Funding acquisition. Timo Homeier-Bachmann: Writing – review & editing, Supervision, Funding acquisition. Carsten Heydel: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. Frank van Leth: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. Christian Menge: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.
Funding statement
The research on which this article is based was conducted as part of the OASIS project, which is funded in the 9th call (diagnostics and surveillance) of the Joint Programming Initiative on Antimicrobial Resistance (JPIAMR).
Declaration of competing interest
The authors declare no competing interests.
Acknowledgements
The authors would like to thank the respondents for their time and contributions, responding to the questions during the interviews.
Appendix A. In-depth interview guide
Instructions
Use the vignette to explain the different components of the LQAS approach to antibiotic surveillance | |
| Topics | Possible questions and probes |
|
Opening |
Hello, my named is…
|
|
Background information |
|
|
Antibiotic resistance |
|
| Current antibiotic resistance surveillance |
|
|
Introduction to LQAS LQAS is a strategy used in industry to assess the quality of goods in a batch. It is now used in public health to quickly assess the coverage of interventions, including vaccination. For the surveillance of antibiotic resistance, LQAS does not seek a precise estimate for the level of resistance. Instead, LQAS seeks to answer the questions of whether prevalence of resistance to a particular antibiotic in a specific area/district/farm is above or below a specific threshold. We can decide on this threshold beforehand and act if it is reached (or not), for example, through moving to an alternative antibiotic. LQAS does not require population-based surveillance. Instead smaller numbers of samples from specific areas are used to determine whether resistance has reached the specific threshold. The aim is that the results can be used to identify whether it is time to change to an alternative antibiotic. | |
| Questions about this introduction to LQAS |
|
| Implementing LQAS |
|
|
Potential impact of LQAS |
|
| Closing |
|
Data availability
The data that has been used is confidential.
References
- 1.Albernaz-Gonçalves R., Olmos Antillón G., Hötzel M.J. Waiting for markets to change me—high-stakeholders’ views of antibiotic use and antibiotic resistance in pig production in Brazil. Frontiers in Veterinary Science. 2022;9 doi: 10.3389/fvets.2022.980546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bengtsson B., Greko C. Antibiotic resistance—consequences for animal health, welfare, and food production. Ups. J. Med. Sci. 2014;119(2):96–102. doi: 10.3109/03009734.2014.901445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL) Resistenzsituation bei klinisch wichtigen tierpathogenen Bakterien [Resistance situation in clinically important animal pathogenic bacteria] 2020. https://www.bvl.bund.de/SharedDocs/Berichte/07_Resistenzmonitoringstudie/Bericht_Resistenzmonitoring_2020.pdf?__blob=publicationFile&v=5
- 4.Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL) Zusammenfassender Bericht über die Ergebnisse der Prävalenzuntersuchungen im Zoonosen- Monitoring der Jahre 2010–2019. 2022. https://www.bvl.bund.de/SharedDocs/Downloads/01_Lebensmittel/04_Zoonosen_Monitoring/Zoonosen_Monitoring_Bericht_2010-2019.pdf;jsessionid=727CBE8531257ECB8970B05EE1ECA20B.internet011?__blob=publicationFile&v=8
- 5.Cairney P., Oliver K. Evidence-based policymaking is not like evidence-based medicine, so how far should you go to bridge the divide between evidence and policy? Health Research Policy and Systems. 2017;15(1):35. doi: 10.1186/s12961-017-0192-x. s12961-017-0192–x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dodge H.F., Romig H. A method of sampling inspection. Bell Syst. Tech. J. 1929;8(4):613–631. [Google Scholar]
- 7.Ducrot C., Guénin M.-J., Hemonic A., Rousset N., Carre Y., Facon C., Le Coz P., Marguerie J., Petiot J.-M., Jarnoux M., Leblanc-Maridor M., Paul M., Molia S., Belloc C. Towards a better use of antimicrobials on farms: insights from a participatory approach in the French pig and poultry sectors. Antibiotics. 2022;11(10):1370. doi: 10.3390/antibiotics11101370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Etienne J., Chirico S., Gunabalasingham T., Dautzenberg S., Gysen S. EU insights – perceptions on the human health impact of antimicrobial resistance (AMR) and antibiotics use in animals across the EU. EFSA Supporting Publications. 2017;14(3) doi: 10.2903/sp.efsa.2017.EN-1183. [DOI] [Google Scholar]
- 9.European Food Safety Authority (EFSA) & European Centre for Disease Prevention and Control (ECDC) The European Union Summary Report on Antimicrobial Resistance in zoonotic and indicator bacteria from humans, animals and food in 2020/2021. EFSA J. 2023;21(3) doi: 10.2903/j.efsa.2023.7867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ginting F., Sugianli A.K., Bijl G., Saragih R.H., Kusumawati R.L., Parwati I., De Jong M.D., Schultsz C., Van Leth F. Rethinking antimicrobial resistance surveillance: a role for lot quality assurance sampling. Am. J. Epidemiol. 2019;188(4):734–742. doi: 10.1093/aje/kwy276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Golding S.E., Ogden J., Higgins H.M. Shared goals, different barriers: a qualitative study of UK veterinarians’ and farmers’ beliefs about antimicrobial resistance and stewardship. Frontiers in Veterinary Science. 2019;6:132. doi: 10.3389/fvets.2019.00132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hedt B.L., van Leth F., Zignol M., Cobelens F., van Gemert W., Nhung N.V., Lyepshina S., Egwaga S., Cohen T. Multidrug resistance among new tuberculosis cases: detecting local variation through lot quality-assurance sampling. Epidemiology (Cambridge, Mass.) 2012;23(2):293. doi: 10.1097/EDE.0b013e3182459455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hoelzer K., Wong N., Thomas J., Talkington K., Jungman E., Coukell A. Antimicrobial drug use in food-producing animals and associated human health risks: what, and how strong, is the evidence? BMC Vet. Res. 2017;13(1):211. doi: 10.1186/s12917-017-1131-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hora R., Ray A., Longchar I., Rio G.R., Mehra R., Koshal S.S., Kumari A., Quadri S.F., Kaur A., Roy A.D. Lot quality assurance sampling for coverage evaluation of a new vaccine: a pilot study. Vaccine. 2024;X, 21 doi: 10.1016/j.jvacx.2024.100578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.2025. https://www.oasisresist.org/
- 16.Jezmir J., Cohen T., Zignol M., Nyakan E., Hedt-Gauthier B.L., Gardner A., Kamle L., Injera W., Carter E.J. Use of lot quality assurance sampling to ascertain levels of drug resistant tuberculosis in Western Kenya. PloS One. 2016;11(5) doi: 10.1371/journal.pone.0154142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.King T., Cole M., Farber J.M., Eisenbrand G., Zabaras D., Fox E.M., Hill J.P. Food safety for food security: relationship between global megatrends and developments in food safety. Trends Food Sci. Technol. 2017;68:160–175. doi: 10.1016/j.tifs.2017.08.014. [DOI] [Google Scholar]
- 18.Laupland K.B., Ross T., Pitout J.D., Church D.L., Gregson D.B. Investigation of sources of potential bias in laboratory surveillance for anti-microbial resistance. Clin. Invest. Med. 2007:E159–E166. doi: 10.25011/cim.v30i4.1777. [DOI] [PubMed] [Google Scholar]
- 19.Magnusson U., Moodley A., Osbjer K. Antimicrobial resistance at the livestock human interface: Implications for Veterinary Services. Revue Scientifique et Technique de l’OIE. 2021;40(2):511–521. doi: 10.20506/rst.40.2.3241. [DOI] [PubMed] [Google Scholar]
- 20.Mayor A., Ishengoma D.S., Proctor J.L., Verity R. Sampling for malaria molecular surveillance. Trends Parasitol. 2023;39(11):954–968. doi: 10.1016/j.pt.2023.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mitchell J., Cooke P., Ahorlu C., Arjyal A., Baral S., Carter L., Dasgupta R., Fieroze F., Fonseca-Braga M., Huque R., Lewycka S., Kalpana P., Saxena D., Tomley F., Tsekleves E., Quynh Vu Thi, G., & King, R. Community engagement: the key to tackling antimicrobial resistance (AMR) across a one health context? Glob. Public Health. 2022;17(11):2647–2664. doi: 10.1080/17441692.2021.2003839. [DOI] [PubMed] [Google Scholar]
- 22.Padda H., Wemette M., Safi A.G., Beauvais W., Shapiro M.A., Moroni P., Ivanek R. New York state dairy veterinarians’ perceptions of antibiotic use and resistance: a qualitative interview study. Prev. Vet. Med. 2021;194 doi: 10.1016/j.prevetmed.2021.105428. [DOI] [PubMed] [Google Scholar]
- 23.Rath R.S., Solanki H.K. Review of lot quality assurance sampling, methodology and its application in public health. Nepal Journal of Epidemiology. 2019;9(3):781–787. doi: 10.3126/nje.v9i3.24507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Rempel O., Laupland K. Surveillance for antimicrobial resistant organisms: potential sources and magnitude of bias. Epidemiology & Infection. 2009;137(12):1665–1673. doi: 10.1017/S0950268809990100. [DOI] [PubMed] [Google Scholar]
- 25.Robertson S.E., Valadez J.J. Global review of health care surveys using lot quality assurance sampling (LQAS), 1984–2004. Soc. Sci. Med. 2006;63(6):1648–1660. doi: 10.1016/j.socscimed.2006.04.011. [DOI] [PubMed] [Google Scholar]
- 26.Speksnijder D.C., Jaarsma A.D.C., Van Der Gugten A.C., Verheij T.J.M., Wagenaar J.A. Determinants associated with veterinary antimicrobial prescribing in farm animals in the Netherlands: a qualitative study. Zoonoses Public Health. 2015;62:39–51. doi: 10.1111/zph.12168. [DOI] [PubMed] [Google Scholar]
- 27.Van Boeckel T.P., Pires J., Silvester R., Zhao C., Song J., Criscuolo N.G., Gilbert M., Bonhoeffer S., Laxminarayan R. Global trends in antimicrobial resistance in animals in low- and middle-income countries. Science. 2019;365(6459):eaaw1944. doi: 10.1126/science.aaw1944. [DOI] [PubMed] [Google Scholar]
- 28.van Leth F., den Heijer C., Beerepoot M., Stobberingh E., Geerlings S., Schultsz C. Rapid assessment of antimicrobial resistance prevalence using a lot quality assurance sampling approach. Future Microbiol. 2017;12(5):369–377. doi: 10.2217/fmb-2016-0170. [DOI] [PubMed] [Google Scholar]
- 29.Wee B.A., Muloi D.M., van Bunnik B.A.D. Quantifying the transmission of antimicrobial resistance at the human and livestock interface with genomics. Clin. Microbiol. Infect. 2020;26(12):1612–1616. doi: 10.1016/j.cmi.2020.09.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.World Health Organization . World Health Organization; 2019. No Time to Wait: Securing the Future from Drug-Resistant Infections. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that has been used is confidential.