Correction: Knowledge, attitudes and practices of smallholder dairy farmers on antimicrobial use in selected districts of Zambia: implications for antimicrobial stewardship

Abstract

Antimicrobial resistance (AMR) is widely regarded as one of the most pressing global health threats of the 21st century. The World Health Organization (WHO) describes it as a "silent pandemic" threatening human health, food safety, and economic stability (1). In 2019 alone, AMR was associated with an estimated 4.95 million deaths globally, with the highest burden recorded in sub-Saharan Africa and other low-and middle-income countries (LMICs) (2). This burden is projected to rise to 10 million annual deaths by 2050 if no effective action is taken (1).Extensive antimicrobial use (AMU) in food-producing animals is recognized as the main driver of AMR in humans (3,4). The livestock sector including dairy farming has come under increasing scrutiny for its role in promoting the development and spread of antibiotic residues (ARs) and resistant pathogens through the food chain and environment (5). In both high-income countries and LMICs, antibiotics are used therapeutically, prophylactically, and at times non-therapeutically to promote growth (6). While such use can increase productivity, the improper or excessive use of antibiotics raises significant concerns for food safety and public health. Common misuse practices include; using antibiotics to treat viral infections, prevent noninfectious disease in healthy animals, promote growth, and administering incorrect dosages or durations of treatment (5,7). Additionally, failure to comply with withdrawal periods and inappropriate prescribing practices contribute to the persistence of residues in milk and other animal products, increasing the risk of resistant bacterial strains entering the human food chain (1,8).In LMICs, particularly across Africa, misuse of antibiotics in livestock production is widely reported (9,10).In many regions, antibiotics are readily accessible without veterinary prescription, often obtained through informal channels (2). KAP studies in various regions of Africa have demonstrated critical gaps. These include low awareness of AMR, reliance on antibiotics as first-line treatment even for non-bacterial infections and poor milk disposal practices when antibiotics are used, self-prescription, incorrect dosages, and noncompliance with withdrawal periods as persistent challenges (11)(12)(13). Poor KAP among farmers is worsened by systemic factors, such as limited veterinary access, weak or lack of regulations, limited diagnostic capacity, and economic constraints (11,14). These factors contribute to the accumulation of ARs in milk and the emergence of resistant strains. Several studies across Africa have reported the presence of ARs and antibiotic-resistant pathogens in raw milk, raising serious concerns about food safety and public health (15,16). Residues of commonly used antibiotics such as tetracyclines, sulfonamides, and penicillins have been detected in milk samples from countries including Zimbabwe, Tanzania, Kenya, and Niger, often exceeding acceptable limits due to non-compliance with withdrawal periods (12,(17)(18)(19). In addition, multidrug-resistant bacteria such as Escherichia coli, Staphylococcus aureus, and Salmonella spp. have been isolated from raw milk, some carrying resistance genes against beta-lactams and other critical antimicrobials (20)(21)(22). These findings highlight the need to promote antimicrobial stewardship (AMS) at the farm level. AMS refers to a set of coordinated strategies aimed at optimizing antimicrobial use, minimizing resistance, and reducing harm to human and animal health (8). AMS educational campaigns have not been widely developed in the agricultural sector of LMICs and have mainly focused on human medicine (23). Effective AMS in livestock systems requires not only policy enforcement but also behavioral change guided by a sound understanding of farmers' knowledge, attitudes, and practices (KAP).In Zambia, Smallholder dairy farmers (SHDFs) who contribute a substantial proportion of milk production, often exceeding 80% of total milk supply take their milk to milk collection centers (MCCs) (24,25).Local studies have confirmed the presence of ARs and resistant genes in raw milk supplied by SHDFs.Muma and others (26) reported a 30.12% prevalence of antibiotic residues in milk samples from Lusaka Province. More recently, Mwasinga and others (27) identified multidrug-resistant Escherichia coli in milk collected from MCCs in Namwala District of Southern province. Despite this evidence, a lack of comprehensive data on the KAP of SHDFs in Zambia remains. Understanding the behavioral and systemic drivers of AMU is essential for designing AMS interventions to reduce the cases of AMR and improve overall food safety of animal-derived foods. The KAP model helps identify gaps in KAP that can inform targeted, context-specific interventions. Therefore, this study aimed to document KAP related to AMU and AMR among SHDFs in selected districts of Zambia. In addition, this study will contribute to evidence-based recommendations for promoting AMS among SHDFs in Zambia.A cross-sectional study was undertaken in the Southern and Lusaka provinces of Zambia from August to September 2025 (Figure 1). In the Southern province, we purposively sampled four districts: Choma, Monze, Namwala, and Zimba. In the Lusaka province, we purposively selected two districts: Chilanga and Chongwe.These districts are known for having significant small-scale milk production in their respective provinces.The target population consisted of smallholder dairy farmers who are members of their local Dairy Cooperatives, both male and female, aged 18 years and above. Smallholder dairy farmers were considered small-scale producers who depend on family labour and produce milk for both home consumption and market sale, with an average herd size of 4 (25,28). SHDFs were selected as they contribute a substantial proportion of milk in the country (24,25).Sample size estimation was conducted using Epitools (http:// epitools.ausvet.com.au/). With an estimated 5,408 dairy farmers affiliated with their respective cooperatives in both provinces, the sample size was determined based on an assumption of 50% prevalence of 'good' KAP and 5% desired precision at a 95% confidence level. In the absence of prior data on the proportion of good knowledge in this population, a prevalence of 50 percent was assumed to maximize sample size (1). A sample size of 360 was estimated. The Southern province had 5,091 registered farmers, and Lusaka province had 317, resulting in a proportional distribution of 339 to 21 respondents from each province, respectively. With a small sample size from Lusaka province, statistical power to analyse KAP would have been inadequate. To overcome this limitation, the sample distribution was adjusted to 250 and 110, respectively, maintaining the dominance of the Southern province. This adjustment reflected an adequate representation from Lusaka province. Similar sampling adjustments have been applied in other KAP studies where oversampling of participants was necessary to strengthen statistical analyses (11,28). The number of farmers interviewed was proportionate to the quantity of bulk milk supply per district at the time, ensuring equitable representation. Convenient sampling was employed, and participants were engaged voluntarily at the MCCs with the help of the MCC managers. They were assured of anonymity and allowed to leave at any time during the interview if they wished to.The questionnaire design was informed by previous studies (11,29). The structured questionnaire was administered during face-to-face interviews to all eligible dairy farmers. The questionnaires were originally written in the English language and translated into the local language (Tonga) of the Southern Province for participants who were unable to respond in English and (Nyanja) for those in Lusaka province. Translation accuracy was ensured through forward translation into local languages (Tonga and Nyanja) followed by back-translation into English.A pilot study was conducted among 36 randomly selected farmers to pretest the questionnaire for face and content validity, and the questionnaire was adjusted accordingly. The selection of several farmers for pre-testing was based on 10% of the study's sample size. Internal consistency reliability was assessed using Cronbach's alpha for each KAP domain which gave alpha (>0.9). The findings from the pilot study were not included in the analysis of the main research. A total of 43 questions on KAP were administered. Most of the questions were multiple-choice and were classified as either correct or incorrect. Answers to openended questions were coded into categorical variables. The questionnaire consisted of four sections. Section A (3 questions) focused on the socio-demographic information of the participants; Section B (13 questions) addressed knowledge regarding the use of antibiotics in dairy production; Section C (8 questions) included questions related to the participants' attitudes (perceptions about antibiotic misuse), and Section D (18 questions) covered practices (antibiotic usage and withdrawal periods) concerning the use of antibiotics in their livestock. In Section D, 11 questions were used to assess the practices, while the remaining questions, such as the types of antibiotics used, provided contextual insights into farmer practices in this study area.Therefore, the KAP was assessed based on 32 questions.After collecting the data, data entry was performed using Microsoft Excel version 16.22, and the data were further analyzed using RStudio version 4.4. To determine KAP scores, the method described in (30) was adopted. One mark was awarded for each correct answer, while zero was given for incorrect or uncertain responses. Multiple-response questions were scored using a structured system that awarded one point for exclusively correct or predominantly correct answers, while responses that included incorrect options received no score. This method ensured an accurate reflection of respondents' KAP without inflating scores due to guessing. For KAP studies, Blooms criteria categorize scores of 80%-100% as good, 60%-79% as moderate, and < 60% as poor (2,3). In this study, we used a cutoff of 75%, which was a modification of the Blooms cutoff point. Similar peer-reviewed studies have adopted a dichotomized classification using a 75% threshold to define good scores (4,5). Therefore, a cut-off point of 75% was established to define the KAP scores as "good" or "poor". Participants with ≥75% correct responses in the knowledge, attitude or practice questions were categorized as having good knowledge, good attitude and good practices, respectively. On the other hand, those scoring < 75% were classified as having poor knowledge, poor attitude and poor practices. In addition, the percentage of correct answers for each participant was calculated for all 32 KAP questions. Respondents having ≥75% correct answers in all the knowledge, attitude and practice questions were considered as having good KAP scores and those with <75% were considered as having poor KAP scores.To identify the normal distribution of the KAP score, the Shapiro-Wilk normality test was applied.Since the scores were not normally distributed (W = 0.949, p < 0.001), Spearman's rank correlation was employed to explore the correlation values between the outcomes of Knowledge, Attitude, and Practice. Chisquare tests were used for the association between demographics and KAP levels. Logistic Regression was performed to identify the predictors of good KAP outcomes. Independent including and were included in this The KAP of "good" were the in the The factors that a significant association < during the analysis were used to a analysis to identify the KAP AMU and AMR in the study total of 360 farmers in the Southern province had 250 respondents while Lusaka had Most of the participants were male The which had most participants was years The highest of for most of the participants was was assessed against 11 (Figure A significant proportion of participants were to define an antibiotic only of participants the of antibiotics with to or number of define of respondents demonstrated good scores from to by and district aged and were to have good knowledge to = a with farmers the highest knowledge scores to only among those with no < was also significant < and recorded the highest with good knowledge, while and Chilanga had the were assessed against (Figure A proportion of farmers that antibiotics used for disease that milk or reduce antibiotic residues from the only of farmers demonstrated good attitude scores from to attitudes to = with of respondents had good attitudes among those with no < farmers were also to attitudes = were with Chilanga the highest attitudes to = were assessed against 11 (Figure farmers that they used The most used of antibiotics reported in the study were tetracyclines, and percent of SHDFs reported using antibiotics for disease growth in to a of the study population that they not the regarding or number of The for this poor practice not the the treatment not and the milk production (Figure of respondents reported good practices. practice scores from to was associated with practices = with those at and practices to those without by district < and reported the highest to good practices, Namwala and were identified district and as the predictors of good KAP outcomes to farmers in had of good knowledge and of good practices in Namwala had of good practices 95% was associated with of good attitudes and practices to no KAP scores were obtained by by the total number of questions (30) by the total number of correct answers, as for The overall KAP scores were as of participants scored the cut-off point was not associated with KAP outcomes = no significant association was with = A outcomes was with increasing = were significant in KAP outcomes by district < by province, Southern province had a prevalence of good KAP to Lusaka In Southern district recorded the highest KAP scores and in Lusaka province, it was Chilanga district confirmed that both and district were significant predictors of overall KAP < the of good outcomes by 95% District 95% was most while Namwala 95% was to good The model demonstrated good power = on antimicrobial use was reported by of farmers analysis a significant association between and good KAP = p = who had received in antibiotic use and animal health were as to good KAP to those who had not 95% rank correlation significant among the KAP was with attitudes = p = and practices = p < were also with practices = p < The association was between knowledge and has as one of the threats to and health, with misuse of antibiotics in livestock production identified as a driver (1). Understanding the knowledge, attitudes, and practices of smallholder dairy farmers is not but for designing interventions that promote AMS to food safety and public findings of the study provided some critical insights into the KAP of SHDFs in districts of Zambia and on of respondents antibiotics and an even proportion had of only ARs This between awareness and accurate understanding of ARs is not to Zambia and has been in other For and others reported that of livestock in had of antibiotic their health or the of withdrawal a study by and others that only of farmers in that residues in milk and the to including and resistance only define AMR, and AMR in the dairy These that while awareness is is limited of understanding of the critical AMR These findings with studies in and where poor knowledge of resistance development and public health associated with misuse of antibiotics were knowledge limited knowledge in critical Similar findings in were reported in findings that good study studies in and that is a of knowledge in that and to knowledge and are all farmers the of when animals are A KAP study of dairy farmers in reported of the participants the of the using antibiotics This is a good for interventions. On the other hand, a significant proportion antibiotics were necessary in dairy farming and their use for disease 60% that antibiotics increase milk production in A proportion of that antibiotics can to the emergence of that can disease in human the reported in of SHDFs in this study that milk antibiotic a critical about the stability of and the of food This is particularly as it a of safety among farmers and to the and consumption of milk informal dairy studies have that many antibiotics such as and beta-lactams are and can even or These can AMS by to public for milk, and with withdrawal this through campaigns is particularly in regions where informal milk is and are were with only of respondents farmers and respondents were to The of attitudes among farmers and those of knowledge and risk respectively. A of and this many of sub-Saharan Africa, Zambia, are in livestock and for This they are to to disease and practices This increase awareness of and improve attitudes and use of Additionally, often take on in health for the which to and to using antibiotics and behavioral particularly on male farmers who production but attitudes in to AMS the this improve farmer attitudes by increasing use in dairy farming was in this study, as all the farmers reported using antibiotics in their production A was also in and where use of antibiotics in the dairy sector was The excessive use of antibiotics to farmers' to promote growth, treat various infections such as and milk have also of antibiotic use in dairy farming The of antibiotic use is projected to increase farmers reported that they on veterinary for of in a study that of farmers a when their and in reported that antibiotics were exclusively by a A KAP study of dairy farmers in reported that of the participants the of the using antibiotics reported of the to findings in to previous findings in and most farmers in study reported antibiotics from veterinary The between reliance on and substantial a in LMICs where small farmers for but antibiotics due to limited or of Similar have been in LMICs, where veterinary with informal practices practice is often associated with and antimicrobial from reported in study that has been in studies, often resulting in inappropriate selection and use most are often not animal health which further inappropriate antimicrobial use practices in livestock systems study that included Zambia reported that economic a significant role in antimicrobial use which farmers to of veterinary a lack of for antimicrobial use in the animal health sector was to a to promote and use in the sector This was reported to promote the use of to veterinary and coordinated antimicrobial have been recognized as essential strategies to inappropriate antibiotic use in livestock systems will antibiotic are with good practices at the point of of the study that of farmers that they followed the withdrawal with the findings of and others The that some of the farmers noncompliance with antibiotic withdrawal periods can to the of ARs in dairy In this can the emergence of pathogens with AMR genes a to from or treatment with of constraints as the while they treatment if the In a in 21 reported that they the treatment when A study in this where farmers that they often to animals the antibiotic to by the treatment These findings point to constraints and low risk in studies In and others reported that of participants they the antibiotic and of as as the animals no of They also reported that participants antibiotics if animals to at some point using These findings that farmers commonly take the to change antibiotic use of antibiotics for disease or growth reported by a of antibiotics as a the between and of farmers' KAP in and also reported this practice in a significant proportion is that of the participants to milk of antibiotics to animals, while reported such Similar practices are These practices have for public health and AMR, to antibiotic in which can resistant in the domain of practices that good practices are not a of good knowledge but are in and economic scores were with respondents scoring the threshold of These findings are with previous studies in sub-Saharan Africa which reported low or poor KAP scores among dairy farmers studies have a between KAP and other factors including and of In the study, a significant between KAP and two and was as a of good Participants with or were to good KAP those with or no These findings the role of in risk critical and to public health was significant from district KAP scores to the other Chilanga district of Lusaka province had the highest KAP scores and district of Southern province. districts are in a in to the other districts in their respective This in to veterinary and and further The in KAP outcomes also that interventions context-specific and evidence-based to gaps. This study not data on disease prevalence across in disease burden in antimicrobial use practices and in also KAP outcomes. Participants who had received on antibiotic use and animal health were to have good This the that farmer and can attitudes and studies have established that on dairy farming and antibiotic use and their are an essential for overall KAP among dairy farmers correlation analysis between knowledge, attitudes, and practices. The correlation was between knowledge and practices, followed by attitudes and practices, and knowledge and These findings that while knowledge contribute to practices, the is and by and practices on antibiotic use are by several and factors, which addressed by knowledge alone, such as those by economic For a farmer about the of not withdrawal periods the of antibiotics to most of but not the time due to of as by the of participants who that they milk administering antibiotic that while knowledge is a necessary change requires and For enforcement of and a veterinary system will also necessary to and public findings of this study were in to the Zambia One Health with on antimicrobial resistance food safety, and system With to systems for and the persistence of practices, including to withdrawal periods and informal antibiotic use that limited to and but behavioral at the production level. For the of antibiotics to to and of public health while some of awareness it not into practices, that strategies focused on information is need for interventions that and economic to farmer the of KAP findings with the Zambia One Health the need for a from and interventions. both and the of One Health strategies on AMS and food safety is to study one of the of KAP regarding AMU and AMR among SHDFs in Zambia. This study that while knowledge and practices regarding antimicrobial use were acceptable among SHDFs in Zambia, attitudes a critical weak overall stewardship constraints further This that AMU is not a of but also of and One such would the of to the of antibiotics to farmer with treatment the between knowledge, attitudes, and practices will interventions that are and farmers through to veterinary and AMU into is In AMS in the dairy sector as both a and knowledge with systems and attitudes, Zambia and other LMICs can strengthen food safety, public health, and contribute to the global against of this study was the use of a sample and statistical including and correlation study data and study on responses from This have and participants provided responses that with or acceptable Participants have or their This have to an of To this participants were assured of anonymity and study employed a cross-sectional data at a point in While this provided a of it not assess studies to knowledge, attitudes, and practices particularly in to study not include some such as herd farming and to veterinary which have been used in studies and antimicrobial use were using sampling at the which have farmers or informal milk These farmers have to veterinary and AMU practices. the study findings KAP among the population of small farmers.