A Simulated Client Study on Non-Prescription Antibiotic Sales and Counselling Practices in Iraqi Nurse-Led Clinics: Implications for Antimicrobial Resistance

Introduction

Antimicrobial resistance (AMR) is a critical global health challenge, projected to cause up to 10 million deaths annually by 2050 if left unaddressed.¹ Its impact is particularly severe in low- and middle-income countries (LMICs), where regulatory oversight, diagnostic infrastructure, and public awareness are often limited.^2–6 The misuse of antibiotics, especially the dispensing of antibiotics without a prescription, remains widespread in LMICs and is a key driver of AMR.^4,6–12, along with high levels of inappropriate antibiotic prescribing among LMICs^13–15 Studies show that a large proportion of antibiotics are prescribed and obtained without prescriptions for self-limiting viral illnesses such as upper respiratory tract infections (URTIs), where antibiotics offer no clinical benefit.^11,18–21

In Iraq, AMR has become a leading cause of mortality, with 3400 deaths attributable and 2,400 deaths associated with resistant infections in 2019 alone.²² Despite existing regulations, non-prescription dispensing of antibiotics remains widespread, particularly in community pharmacies, where studies report 100% availability of antibiotics without a prescription.^23–26 Inappropriate antibiotic use also extends to hospitals and primary healthcare (PHC) settings, where empirical treatment without microbiological testing is common.^27,28 These trends reflect wider regional challenges exacerbated by post-conflict healthcare fragility and under-resourced stewardship systems.²⁴

The Iraqi healthcare system currently faces a number of challenges with improving future antibiotic use. Current challenges include inadequate enforcement of regulations surrounding the dispensing of antibiotics without a prescription, concerns with public awareness about antibiotics and AMR, especially regarding administering antibiotics without a prescription, as well as insufficient implementation of antimicrobial stewardship (AMS) and antimicrobial stewardship programs (ASPs).^{23,25,26,29,30} There are also concerns that patients or parents stop antibiotics once symptoms improve due to economic reasons.²³ While the Iraqi National Action Plan (NAP) on AMR (2018–2022) emphasizes the need for stricter enforcement of antibiotic regulations and enhanced surveillance,³¹ compliance remains limited, particularly in PHC settings. This is a concern going forward, which is similar to other LMICs where there are financial and personnel concerns with implementing NAPs in practice.^32–34 AMR is not only a national issue but a global public health threat, with resistant genes spreading internationally via travel, trade, and environmental routes.^35,36 It is increasingly viewed as the next pandemic unless urgent action is taken.³⁷ Addressing antibiotic misuse, particularly in LMICs including Iraq, is essential to meet the United Nation’s Goal for AMR reduction.³⁸ In Iraq, reducing inappropriate antibiotic use across all healthcare settings is critical to national health security and achieving the objectives of NAP for AMR, which prioritises surveillance, stewardship, and public awareness in both formal and informal care sectors.³¹

In Iraq, alongside community pharmacies, nurse-led clinics are widely distributed. They were originally established as satellite facilities adjunct to physicians’ clinics, primarily to provide nursing services including wound care and post-procedural management. However, due to prevailing public perceptions and cultural beliefs, these clinics are often regarded as the first point of contact for medical care, particularly in rural and underserved areas in the country. As a result, extending their scope beyond initial concepts. Consequently, they now play a key role in Iraq in PHC delivery with currently widespread reliance on such clinics for ambulatory care in many parts of Iraq, including peri-urban and underserved communities.

Nurse-based clinics are legally permitted in Iraq to dispense over-the-counter (OTC) medications to cover a broad range of ailments. These include medicines for pain management and fever relief, ie, paracetamol, aspirin, ibuprofen, as well as for gastrointestinal disorders including dyspepsia, diarrhea, and constipation, incorporating antacids, loperamide, oral rehydration salts and laxatives, antispasmodics for gastrointestinal discomfort, ie hyoscine butylbromide, as well as guaifenesin and ipecacuanha for acute respiratory conditions including coughs and colds, and loratadine for allergy relief. Moreover, unlike community pharmacies, these clinics are legally permitted to dispense two antibiotics — amoxicillin and erythromycin — without specifying the indications or the rationale. This regulatory flexibility grants nurses the discretion to dispense these two antibiotics based on their clinical judgment. However, this can raise concerns regarding the potential misuse of antibiotics and AMR. This especially given the high prevalence of patients presenting with acute respiratory infections (ARIs) in primary care in LMICs, including Iraq, alongside the considerable expectations among patients, or parents of children, for a prescription for antibiotics for essentially self-limiting conditions.^13,15,39 In addition, concerns regarding the knowledge of antibiotics, AMR and ASPs, among nurse prescribers in LMICs.^15,40 These stewardship challenges are compounded by cultural norms and behavioural drivers common across many LMICs, where antibiotics are often seen by patients as essential to rapid recovery regardless of clinical indication.^6,13,15 Studies from across LMICs have reported that providers face strong pressure to meet patient expectations regarding the prescribing of antibiotics, particularly in settings with limited diagnostic capacity and poor health literacy.^6,13,41 Regulatory enforcement of prescription-only policies remains weak in Iraq, especially in non-pharmacy outlets such as nurse-led clinics. These trends were exacerbated during the COVID-19 pandemic, which disrupted oversight mechanisms and led to a surge in inappropriate antibiotic use across multiple LMICs.^42–45 While our study period (October 2023–April 2024) does not overlap with the height of the pandemic, we include this context to reflect previously observed patterns rather than direct findings from our dataset. Iraq’s situation mirrors broader global patterns in fragile healthcare systems, where unregulated antibiotic access and workforce training gaps pose persistent barriers to AMS activities.^6,13 Nurse-led clinics, in particular, are often excluded from national training frameworks, leading to variability in dispensing behaviour and limited awareness of AMR threats.^13,46

Importantly, despite their critical role in healthcare delivery, the extent and appropriateness of antibiotic dispensing practices in nurse-led clinics currently remains largely unexplored in Iraq. As a result, there is a notable paucity of data regarding whether these clinics dispense the two legally permitted antibiotics appropriately in accordance with clinical guidelines. Alternatively, whether they prescribe and dispense these inappropriately alongside restricted antibiotics beyond current regulations. Addressing possible widespread inappropriate prescribing and dispensing of antibiotics in nurse-led clinics in Iraq is important for the country to achieve the goals in its NAP. This is critical given concerns with existing AMR rates in the country, alongside the recent targets and goals from the United Nations Global Assembly (UN GA) to reduce AMR.^31,38

Given the global and local urgency with addressing AMR across LMICs including Iraq, understanding the scale and nature of this issue among nurse-led clinics is essential for informing targeted interventions in Iraq aimed at enhancing future responsible antibiotic use. The findings can inform future policy interventions as part of Iraq’s NAP to instigate targeted AMS activities in this setting, building on recent research with patients and community pharmacists.^23,25,26[ Consequently, this study sought to assess the prevalence and appropriateness of antibiotic prescribing and dispensing practices in nurse-led clinics in Iraq. Specifically, to evaluate the types of antibiotics prescribed, including those from the WHO Watch list, the legality of their dispensing, and the quality of clinical questioning and guidance during these encounters. To the best of our knowledge, no previous studies have examined antibiotic supply or prescribing practices specifically within nurse-led clinics in Iraq. This study therefore fills a critical evidence gap by offering the first empirical insight into this under-researched but widely utilised sector of the Iraqi primary healthcare system.

Method

Study Design and Setting

This study employed a cross-sectional design utilizing the simulated client methodology (SCM) to assess antibiotic dispensing practices among nurse-led clinics within Sulaymaniyah, the largest city in the Kurdistan Region of Iraq (KRI).^47,48 Sulaymaniyah was chosen for this research as it accounts for approximately 41% (2.1 million) of the total population under the jurisdiction of the Kurdistan Regional Government (KRG), and is the third-largest city in Iraq.²⁸ Consequently, this setting provides a representative context for evaluating antibiotic dispensing patterns within PHC facilities in this region and across Iraq.

Study Population and Sample Size

The study population comprised all registered and licensed nurse-led clinics in Sulaymaniyah, as recognised by the Kurdistan Health Cadres Syndicate, the regulatory authority overseeing allied HCPs in the region. HCPs include nurses with a university degree, nurse assistants, physician assistants and pharmacy technicians.

The sample size was estimated at 235 nurse-led clinics, calculated based on a 95% confidence level (Z-score = 1.96), an assumed 0.5 (50%) prevalence of inappropriate antibiotic dispensing (as no prior data were available in the literature to inform a more precise estimate), and a 5% margin of error.⁴⁹ A prevalence of 0.5 was chosen following statistical recommendations and guidelines, as it yields the largest required sample size for survey-based studies.^50,51 Because the study was designed to estimate the prevalence of inappropriate antibiotic dispensing rather than test a hypothesis, a formal power calculation was not required. However, the selected sample size (n=235) ensures sufficient precision to estimate prevalence within ±6.4%, assuming a 50% proportion, which is standard when no prior data exist. A random sampling approach was employed to select the 235 clinics from a comprehensive list of 599 registered nurse-led clinics, obtained from the Kurdistan Health Cadres Syndicate, which served as the sampling frame. While the sample size was calculated assuming simple random sampling and a 50% prevalence estimate to ensure maximum variability, we did not apply a design effect or adjust for clustering by clinic type or area. Since all clinics were located within the same urban region and followed similar operational models, we anticipated minimal intra-cluster correlation; however, the potential impact of clustering on the precision of prevalence estimates should be acknowledged when interpreting confidence intervals.

Study Outcomes: Simulated Client Scenario and Data Collection

The primary outcomes of this study were the extent and practice of dispensing antibiotics with or without a prescription in nurse-led clinics. Data were collected through SCM (Figure 1), with visits conducted between October 2023 and April 2024. The SCM involved trained individuals enacting predefined scenarios to evaluate specific behaviours, providing a more accurate representation of real-world practices while minimizing social desirability bias.⁴⁷ This method has been widely used across LMICs to assess the appropriateness of antibiotics prescribed or dispensed.^48,52

Figure 1 Flowchart summarising the Simulated Client Methodology (SCM) used to assess antibiotic dispensing practices in nurse-led clinics in Iraq. Notes: NB: The figure outlines clinic selection, scenario deployment, demand level escalation, response capture, and analysis.

For this study, a simulated scenario was developed based on previous literature (Supplementary File 1),^48,52,53 in which the simulated client presented symptoms of an URTI for their 20-year-old sibling at home. This scenario was selected due to the high prevalence of URTIs in ambulatory care settings, where most cases are viral in nature and do not require antibiotic treatment, making them a common source of inappropriate antibiotic use in the community.^6,13,54 To assess the threshold at which antibiotics were dispensed, three levels of demand were applied, based on prior research.⁵⁵ Consistent with previous SCM studies,^47,52 the question “Are you a nurse?” was included at the end of each visit to prompt the provider to self-identify their professional cadre. Responses were recorded as stated by the provider (eg, “bachelor nurse”, “diploma nurse”, “medical assistant”, or “pharmacy assistant”) and documented immediately in the structured SC form (Supplementary File 1). These responses were subsequently categorised into four qualification levels according to the formal educational classifications defined by the Kurdistan Health Cadres Syndicate: (1) bachelor-level nurse (BSc Nursing), (2) diploma nurse, (3) medical assistant (with two-year associate degree), and (4) pharmacy assistant (with pharmacy technician training). This variable was then included as a categorical predictor in the multivariable logistic regression model to examine its association with inappropriate antibiotic dispensing. This approach allowed for an assessment of whether the level of education (as a proxy for knowledge) was associated with the extent and practice of dispensing antibiotics without a prescription. The simulated clients were pharmacy students in their final year who underwent two days of standardised training by a clinical pharmacist and the study supervisor. Training included scenario role-play, instruction on neutral presentation techniques, consistent use of the symptom script, and post-visit data recording procedures. Additionally, to ensure data accuracy including simulated clients’ response post-visit, the simulated clients completed the structured questionnaire immediately after each visit, and responses were reviewed by a trained field coordinator who conducted regular debriefings and provided feedback. While audio recordings were not used to preserve discretion, these validation steps helped ensure consistency and protocol adherence across all visits.

Following each simulated client visit, the details of the encounter were systematically recorded using a structured data collection tool, designed to capture study outcomes and relevant study variables (Supplementary File 1). The scenario and data collection tool underwent an initial pilot phase in 15 nurse-led clinics to assess its feasibility, clarity, and completeness. Modifications would have been made if the simulated client had difficulty consistently delivering the scenario, if the clinical staff frequently misinterpreted the presenting symptoms, or if the structured form failed to capture key variables. However, no such issues were observed during the pilot, and the study proceeded without the need for major revisions of the scenario and data collection tool.

To assess the quality of the antibiotics supplied/sold during the study, these were categorized according to the World Health Organization (WHO) AWaRe classification system, which is being increasingly used across countries to assess antibiotic utilization patterns.^56,57 This evidence-based framework stratifies antibiotics based on their resistance potential.^56,57 Access antibiotics are recommended for first-line treatment due to their lower resistance potential, whereas Watch antibiotics are reserved for more severe infections with a higher resistance risk and Reserve antibiotics are intended for last-line treatment of multidrug-resistant infections. This approach is seen as increasingly important given the UN GA target goal that at least 70% of antibiotic utilization in given sectors should be Access antibiotics.³⁸ Alongside this, we also classified antibiotic utilization patterns under the adapted UK AWaRe classification, which classifies amoxycillin and clavulanic acid as a Watch rather than an Access antibiotic.^58,59 As part of our analysis plan, we included all three AWaRe categories, namely Access, Watch, and Reserve.

To evaluate counselling practices following the dispensing of an antibiotic, data were collected on whether HCPs offered advice regarding the appropriate use of the antibiotic, including the dosage, route of administration, and duration of treatment during the consultation. Additionally, the analysis captured whether providers asked any supplementary questions prior to the antibiotics being dispensed including inquiries about additional symptoms, medical history, allergy status, or other relevant clinical information. In this study, “illegal dispensing” was operationally defined as the supply of any antibiotic without a valid prescription, except for amoxicillin and erythromycin, which are legally permitted to be supplied by nurse-led clinics in Iraq. This definition reflects current national legislation, although we acknowledge that enforcement in practice may be inconsistent. Finally, future guidance to all key stakeholder groups in Iraq is based on the considerable experience of the co-authors in similar circumstances across multiple LMICs.^12,13,60

Data Analysis

Descriptive statistics were utilized to summarize the study variables, with frequencies and proportions reported for categorical variables, while continuous variables were presented as mean (standard deviation, SD) or median (interquartile range, IQR), as appropriate. The overall prevalence of antibiotic dispensing was calculated and further stratified by the study covariates. A multivariable logistic regression model was used to assess the independent effects of each factor and examine the association between a single binary outcome—whether antibiotics were dispensed or not (dependent variable, outcome)—and key predictor variables, including provider qualification, gender, and area-level deprivation. Only one dependent variable was included in the model. This allowed for the ability to ascertain any potential confounding in the relationship between provider characteristics and antibiotic dispensing practices. Prior to model estimation, we assessed multicollinearity among the independent variables (qualification level, gender, and area deprivation) using variance inflation factors (VIFs), which were all below the threshold of concern, indicating no evidence of multicollinearity. Area-level deprivation was classified into three categories—least deprived, moderately deprived, and most deprived—based on neighbourhood-level socioeconomic data provided by the Sulaymaniyah Directorate of Statistics. This included indicators such as average household income, infrastructure access, and population density. Each clinic was mapped to its corresponding area to assign deprivation status. All statistical analyses in this study were performed using R version 4.4.1.

Results

Prevalence and Characteristics of Nurse-Led Clinics Dispensing Antibiotics Without a Prescription

Of the 235 registered nurse-led clinics visited using the simulated client methodology, all of which were licensed facilities under the Kurdistan Health Cadres Syndicate and did not include any community pharmacies, antibiotics were dispensed without a prescription in 60.0% (95% CI: 53.7%–66.3%) of the encounters (n=141) (Table 1). Of the 141 antibiotics supplied during the simulated visits, 56 (39.7%) were amoxicillin, one of only two antibiotics, along with erythromycin, that is legally permitted for use in nurse-led clinics under current Iraqi regulations. The remaining 85 antibiotics (60.3%) were not legally permitted to be dispensed by nurse-led clinics under current Iraqi regulations. This equates to an overall illegal antibiotic dispensing rate of 36.1% (95% CI: 30.0%–42.4%) (85 out of 235 total encounters).

Table 1 Baseline Characteristics of the 235 Nurse-Led Clinics, Stratified by the Status of Dispensing Antibiotics Without a Prescription

Clinics were predominantly staffed by male personnel (84.3%, n=198), and there were no statistically significant differences in dispensing patterns based on gender (p > 0.05) (Table 1). Similarly, no significant associations were found between dispensing proportions and area-level deprivation. Clinics located in moderately deprived areas accounted for the highest proportion of total antibiotic dispensing (45.8%), followed by those in least deprived (31.0%) and most deprived (22.5%) areas. However, a significant difference in the percentage of dispensing an antibiotic with a prescription was observed across staff qualification levels (p < 0.05) (Table 1). Clinics managed by bachelor-level nurses supplied antibiotics in only 7.8% (n=11) of cases, compared to a higher proportion observed among diploma nurses (43.3%, n=61) and pharmacy assistants (15.6%, n=22) (Table 1).

Table 2 Characteristics of 141 Antibiotics Dispensed Among the 235 Nurse-Led Clinics

Patterns and Quality of Nurse-Led Clinics Dispensing Antibiotics: Types, Duration, and Classification

Among the 141 antibiotics supplied, amoxicillin was the most common (39.7%, n=56), followed by amoxicillin/clavulanic acid (34.0%), azithromycin (21.3%), and smaller proportions of ciprofloxacin, cefixime, and ampicillin (Table 2). Notably, except for amoxicillin, all other agents were not permitted to be prescribed or dispensed by nurse-led clinics, and their dispensing represents regulatory non-compliance. Prescribing and dispensing of antibiotics without a prescription was highly responsive to the level of patient demand. Even in the absence of a specific antibiotic request (Demand Level 1), antibiotics were supplied in 79.4% (n=112) of visits (Table 2). There was no prescribing or dispensing of erythromycin for the SPs with URTIs.

According to the WHO AWaRe framework, 74.5% (n=105) of antibiotics supplied were from the Access group and 25.5% (n=36) from the Watch group. However, classification under the UK AWaRe list revealed a higher proportion of Watch antibiotics (59.6%, n=84) (Table 2). The duration of therapy was also generally sub-optimal. Most antibiotics (65.2%, n=92) were supplied for three days or less, and 31.2% (n=44) were recommended for four to seven days. Only 1.4% (n = 2) of clients were advised to take antibiotics for more than one week. An exploratory univariate analysis using a chi-square test indicated that short-duration antibiotic supply (≤3 days) was most frequently observed with azithromycin (93.3%), followed by amoxicillin (62.5%) and amoxicillin/clavulanic acid (54.2%). In contrast, longer durations (4–7 days) were more commonly observed for cefixime (66.7%) and ciprofloxacin (66.7%). Although these patterns suggest variation in dispensing practices by antibiotic type, the overall association was not statistically significant (χ²(15) = 21.77, p = 0.114), potentially due to small sample sizes within some antibiotic categories limiting statistical power. Oral formulations were predominantly recommended (98.6%, n=139), with only two instances of intravenous antibiotic dispensing (Table 2).

Table 3 Counselling Practices and Inquiry Patterns for Antibiotics Supplied/Sold Among the 235 Nurse-Led Clinics (Key Takeaway Message/ While Dosage Instructions Were Commonly Provided, Less Than 8% of Providers Screened for Allergy, Posing a Notable Patient Safety Concern)

Counselling and Inquiry Practices in Nurse-Led Clinics Dispensing Antibiotics

Counselling regarding antibiotic use was provided in 90.1% (n =127) of antibiotic dispensing cases (Table 3). Advice regarding dosage, frequency, or duration of treatment was provided in 90.1% (n =127) of antibiotic dispensing cases, while 8.9% (n=14) of encounters involved no counselling.

Only 24.8% (n=35) of providers asked additional clinical questions prior to dispensing antibiotics. Among those who did, the most frequently asked questions related to respiratory symptoms (34.3%, n= 12), allergy history (31.4%, n=11), and chronic conditions such as hypertension or diabetes (14.3%, n=5) (Table 3). Queries related to lifestyle factors, gastrointestinal symptoms, or surgical history were rare. Whilst 31.4% (n=11/35) of the 35 providers who asked any additional clinical questions inquired specifically about allergy status, this accounted for only 7.8% of all antibiotic dispensing encounters (n=11/141), highlighting a critical gap in routine safety screening prior to dispensing. We observed a trend toward higher screening rates in clinics staffed by more qualified personnel (Bachelor of Nursing (9.1%), Diploma of Nursing (3.3%), Medical Assistant (4.3%), Pharmacy Assistant 0.0%) and in clinics located in higher-income areas (4.6%) vs 3.1% in the most deprived areas. However, these observations were based on univariate analyses with small subgroup sizes and were not adjusted for potential confounders such as patient age, antibiotic type, or demand level. Moreover, identifying predictors of allergy screening was not a primary aim of this study and should be explored further in future research.

Among the 94 clinics that declined to dispense antibiotics for a URTI, despite an explicit request by the simulated client (demand level 3), 38.9% (n=35) cited the need for a prescription, 37.8% (n=34) judged the condition as not requiring antibiotics, and 23.3% (n=21) referred the client to a physician. Furthermore, among the clinics that did not dispense antibiotics (n=94), 25.5% (n=24) offered alternative over-the-counter advice for symptom relief.

Discussion

We believe this is the first study of its kind in Iraq to assess current antibiotic prescribing practices in nurse-led clinics, an important provider of public healthcare in Iraq. Overall, this study provides one of the most detailed explorations to date regarding antibiotic prescribing, dispensing and counselling practices among nurse-led clinics in a LMIC context. The findings demonstrate that 60% of the surveyed nurse-led clinics dispensed non-approved antibiotics without a prescription, which is a concern. This is because URTIs, such as acute bronchitis, are typically viral in origin where first-line treatment should not be an antibiotic but symptomatic treatment as documented for instance in the WHO AWaRe Book.^63,64 Among these cases, 36.1% (n=85/235) of antibiotics supplied were deemed illegal under national legal definitions in Iraq, which is also a concern exacerbating AMR in Iraq. Another concern was that 25.5% of dispensed antibiotics were from the Watch category, which increased to 59.6% in the UK adaptation of the AWaRe classification primarily due to the reclassification of amoxicillin/clavulanic acid as a Watch antibiotic.⁵⁹ This divergence has major implications for stewardship strategies, especially in countries that use any modified AWaRe list to monitor current utilisation practices. Both the WHO and UK AWaRe classifications were applied in this study to provide a comparative framework for assessing antibiotic use. While the WHO list offers a global benchmark, the UK adaptation adopts a more conservative stance, reclassifying certain antibiotics such as amoxicillin/clavulanic acid from Access to Watch. Given Iraq’s high usage rates and increasing resistance to broad-spectrum antibiotics, adopting a similar stewardship perspective may be more appropriate for guiding local policy, even though Iraq does not yet have a national AWaRe adaptation.

Whilst erythromycin is legally permitted for dispensing by nurse-led clinics, it was not supplied in any encounter. This likely reflects its lower demand due to tolerability concerns, particularly gastrointestinal side effects, which are more common with erythromycin compared to other broad-spectrum agents such as amoxicillin/clavulanic acid. Additionally, when a macrolide was considered appropriate, providers appeared to favour azithromycin, a second-generation macrolide, possibly due to its more favourable pharmacokinetic and safety profile.⁶⁵ Azithromycin offers several clinical advantages over erythromycin, including a longer half-life allowing for once-daily dosing, fewer gastrointestinal side effects, reduced drug–drug interactions, and improved overall tolerability.⁶⁵ While both agents are effective protein synthesis inhibitors with overlapping indications, azithromycin is generally preferred in practice, which may explain the absence of erythromycin in our study.⁶⁵

Prescribing or dispensing Watch antibiotics (other than erythromycin) or any antibiotic for self-limiting conditions such as URTIs should be avoided to help curb AMR.^13,54 Instigating prescribing targets based on the WHO AWaRe book should help in this regard going forward, especially among LMICs with no recognized robust guidelines for managing infections typically seen in ambulatory care.^64,66,67 However, this requires suitable IT systems to enable real-time education and monitoring, reducing reliance on time- and resource-intensive paper-based data collection. While most nurse-led clinics in Iraq currently operate without formal IT infrastructure, particularly in informal or peri-urban settings, the development of scalable digital or mobile-based monitoring tools should be considered as part of a phased implementation aligned with Iraq’s National Action Plan on AMR. Pilot programmes using low-cost reporting tools or digital registers in high-volume clinics could offer a pragmatic starting point, which is currently being explored.⁶⁸ In high-income countries, online access to prescribing data, combined with clear guidance and real-time monitoring, has significantly improved adherence to prescribing standards.^69,70 This provides an exemplar for health authority personnel in Iraq to help meet NAP and UN GA targets.

Table 4 Suggested Activities for Key Stakeholders to Reduce Inappropriate Antibiotic Use and AMR in Iraq

Suboptimal counselling and failure to check for allergies raise serious patient safety concerns and may contribute to AMR and inadequate treatment. In addition, in 65.2% of encounters, the recommended duration was ≤3 days, which is often insufficient for complete resolution of common bacterial infections, especially RTIs or UTIs of bacterial origin, thereby contributing to sub-therapeutic exposure and resistance development.^63,66 The findings also revealed that only 31.4% (n=11/35) of the HCPs in the nurse-led clinics who asked follow-up questions before dispensing antibiotics queried the allergy status of the SPs. This equates to under 8% of all antibiotic dispensing encounters. This is also a major concern given the known risks of antibiotic allergies, further highlighting patterns of irrational antibiotic prescribing and dispensing in nurse-led clinics. This finding is particularly troubling given that approximately three-quarters of the antibiotics supplied in the nurse-led clinics were penicillin derivatives including amoxicillin and amoxicillin/clavulanic acid. Failure to routinely ask and assess for allergies, especially to beta-lactam antibiotics poses a serious patient safety risk, including the possibility of severe hypersensitivity reactions such as anaphylaxis. This urgently needs to be addressed going forward. This gap reflects a wider lack of proper clinical assessment in Iraq’s primary healthcare setting. It highlights the urgent need for structured training and standardised protocols. Implementing WHO AWaRe guidance will be key to ensuring safer and more appropriate antibiotic use (Table 4). Furthermore, the limited provision of symptomatic advice may reflect missed opportunities for promoting appropriate self-care and AMS. However, although infrequent, a few providers offered appropriate non-antibiotic symptom management advice, which included recommending paracetamol or advising on rest and hydration, and this could be leveraged in future AMS programs as positive behavioural anchors for wider capacity-building efforts. Enhancing clinical inquiry, especially allergy screening, should be prioritized in any future intervention or ASPs targeting nurse-led clinics in Iraq and similar LMIC contexts. We know ASPs can be more challenging in LMICs due to personnel and resource issue; however, this is now changing with multiple ASPs successfully introduced in LMICs in recent years including among prescribers, dispensers and patients.^12,13,53,76

As seen, multivariate analysis revealed significant associations between inappropriate antibiotic dispensing and the level of professional qualifications among the HCPs. Clinics staffed by those with bachelor-level nursing degrees were less likely to dispense antibiotics inappropriately compared to those staffed by diploma or assistant-level personnel. This is similar to HCPs in other LMICs especially when confronted by patients requesting antibiotics to treat essentially viral infections.^13,54,77 No significant differences though were observed based on clinic location or gender of the provider. While the regression model adjusted for qualification, gender, and area-level deprivation, we were unable to account for other potential confounders such as clinic workload or patient demand level due to data limitations. However, it is notable that the majority of antibiotics were supplied at Demand Level 1, the lowest level of patient insistence, indicating that patient pressure may not have been the primary driver of inappropriate dispensing in this context. Nevertheless, other unmeasured structural factors may still influence dispensing behaviours and should be investigated in future studies using more granular operational data. Overall, these collective behaviours pose an appreciable challenge to improve future antibiotic use, and the containment of AMR, in Iraq. These challenges need to be urgently addressed in order to meet the AMR objectives contained in the Iraq NAP for AMR given current high rates of AMR in the country and the mortality implications.³¹

The high prevalence of inappropriate and, in a number of cases, illegal antibiotic dispensing seen in the nurse-led clinics in our study underscores the need for multiple, coordinated approaches, to improve future antibiotic use in Iraq, similar to other LMICs.^13,54,78 Suggested activities start in universities and continue post-qualification for HCPs. They also include greater monitoring of clinic activities and patient education given concerns with patient knowledge and activities across LMICs including Iraq.^6,13,54 Evidence from LMICs demonstrates that culturally tailored public education campaigns can effectively reduce inappropriate antibiotic use and improve community-level practices.^13,15,74 Notable examples among LMICs include Thailand’s Antibiotic Smart Use programme and adaptations of the UK’s Antibiotic Guardian initiative in South Africa, both of which successfully improved public understanding and reduced misuse.^79,80 Similarly, a multi-country review highlighted how Ghana’s public awareness campaigns—incorporating pharmacy posters, radio broadcasts, and school-based materials—enhanced knowledge of AWaRe antibiotics and lowered demand for unnecessary Watch antibiotics.¹⁵ Comparable community-facing strategies could be adapted to Iraq’s context through collaboration with local health directorates, media outlets, and educational institutions to promote safer antibiotic use and strengthen public alignment with national AMR goals. Subsequently, Table 4 contains a number of suggested activities for all key stakeholder groups going forward in Iraq to improve future antibiotic prescribing, which draws on both our study data as well as relevant global guidance to improve future antibiotic use.^6,13,54 As a result, reduce the levels of AMR in Iraq in line with the goals in the NAP.^22,31 Given the legal constraints on antibiotic supply in nurse-led clinics in Iraq, AMS indicators should prioritise counselling quality and regulatory compliance rather than prescribing metrics. Suggested indicators include: (i) % of encounters with no antibiotic supplied without a prescription, (ii) % with documented allergy screening, (iii) % involving non-antibiotic symptomatic advice, and (iv) % of permitted antibiotic supplies given with full duration and clear instructions. These indicators are pragmatic, measurable, and align with Strategic Objective 3 of the Iraq’s National Action Plan on AMR, which calls for strengthening surveillance and optimising use, even in informal settings.

The influence of provider knowledge and training was evident in the observed association between qualification level and antibiotic dispensing behaviour. Clinics managed by bachelor-level nurses were significantly less likely to dispense antibiotics compared to those run by diploma-holding nurses or pharmacy assistants. Although we did not directly assess knowledge, educational qualification served as a useful proxy. This suggests that higher levels of clinical training may equip prescribers with a better understanding of antibiotic indications, resistance risks, and the need for appropriate counselling. This is similar with other studies from LMICs linking knowledge gaps among healthcare workers with poor antibiotic stewardship practices.^13,54,81

Demand level was also a key determinant of dispensing behaviour. Our findings revealed a direct association between the simulated client’s level of demand and the likelihood of receiving an antibiotic. Among the 141 clinics which supplied antibiotics, nearly 80% were supplied at Demand Level 1, ie where no antibiotic was requested. This suggests a low threshold for empirical dispensing of antibiotics. These findings mirror those from global studies that identify patient pressure, whether perceived or actual, as a major driver of inappropriate antibiotic provision.^13,14,77 This pattern underscores the complex interplay between provider knowledge, patient expectations, and the structural incentives that shape dispensing behaviour in PHCs. All these complex factors need to be taken into consideration going forward.^13,53,82 We will continue to monitor the situation as Iraq seeks to reduce AMR rates in line with its NAP.

Beyond provider-level factors, the antibiotic dispensing practices observed in this study must also be understood within Iraq’s broader socio-economic and healthcare context. It is important to recognise that antibiotic dispensing decisions by providers in nurse-led clinics often occur under considerable systemic and socio-economic pressure. Many such clinics operate in regulatory grey zones with limited access to diagnostic tools, sparse oversight, and high patient loads, particularly in underserved or peri-urban regions. In this environment, HCPs may feel compelled to meet patient expectations for antibiotics, which are often rooted in cultural norms that equate antibiotics with effective treatment, regardless of clinical necessity.⁸² These expectations, when coupled with the economic reliance of some clinics on patient satisfaction and repeat visits, can contribute to inappropriate dispensing behaviours even among well-meaning providers.²⁵ Similar pressures have been documented in other LMICs, where prescribers and dispensers navigate complex trade-offs between financial sustainability, limited infrastructure, and public demand.^6,13,54 Acknowledging these real-world constraints is essential for designing ASPs that are not only evidence-based but also context-sensitive and implementable It is important to note that while our study did not directly measure systemic pressures such as patient expectations or financial incentives, these factors were inferred based on observed behaviours. To date, to our knowledge, no published studies have explored these influences in nurse-led clinics in Iraq. However, related research in community pharmacy settings has reported that non-prescription antibiotic supply is frequently driven by financial motives and the desire to maintain customer satisfaction.²⁵ These findings are consistent with our results, particularly as the majority of antibiotics in our study were supplied at Demand Level 1, suggesting that provider-initiated practices, rather than patient pressure, may be the primary driver of inappropriate dispensing in this context. Collectively, it is important, therefore, to note that while the simulated client method allowed for objective observation of dispensing practices and counselling behaviours, these observed actions likely reflect deeper systemic and structural drivers. Factors such as limited enforcement of regulations, economic dependence on medicine sales, lack of structured clinical training for informal providers, and barriers to accessing formal healthcare all contribute to the widespread misuse of antibiotics. These systemic issues cannot be fully captured through behavioural observation alone but are essential to understanding the context in which such practices occur. As such, our findings should be interpreted as a reflection not only of individual provider decision-making but also of broader structural constraints that shape practice norms in under-regulated primary care settings in Iraq.

We believe this study offers a number of unique strengths that contribute valuable insights into a critically understudied are in Iraq, with an appreciable impact on AMR in Iraq. Firstly, to the best of our knowledge, we believe this is the first investigation in Iraq to systematically examine antibiotic prescribing behaviour in nurse-led clinics in Iraq. Given, as mentioned, the widespread reliance on such clinics for primary care in many parts of Iraq, including peri-urban and underserved communities, this study captures a real-world dimension of antibiotic access and use that has been largely overlooked in national surveillance and stewardship efforts to date. Second, the use of simulated client methodology allows for a realistic and standardised assessment of provider behaviour in actual practice, reducing the risk of response bias commonly associated with interviews or self-reported data.³⁸ This approach enabled us to observe dispensing patterns, counselling behaviour, and clinical decision-making in a controlled yet naturalistic context, offering a high degree of ecological validity.

However, certain limitations should also be acknowledged. Firstly, while this study provides important insights into antibiotic dispensing practices in nurse-led clinics in Sulaymaniyah, some caution is warranted when generalising findings to other regions of Iraq. However, given that the regulatory framework governing nurse-led clinics is issued at the national level, and enforcement mechanisms are similarly limited across provinces, it is likely that similar dispensing behaviours will occur elsewhere in the country. Nonetheless, regional differences in provider density, patient demographics, and informal care networks, may influence the extent and nature of antibiotic misuse. Moreover, the study was conducted in urban and peri-urban nurse-led clinics in Sulaymaniyah, and findings may not be fully generalizable to rural areas of Iraq, where resource constraints, staffing limitations, and regulatory enforcement are often more severe. Rural clinics may exhibit different dispensing behaviours—potentially either more or less frequent inappropriate supply—depending on local demand dynamics, health literacy, and oversight. Future research should explore these settings to assess variation in practices across the urban–rural spectrum. Further research is needed to examine whether similar patterns occur in community pharmacies, or other private-sector providers operating under distinct governance frameworks, in Iraq going forward. Secondly, although qualification levels were used as a proxy for provider knowledge, this does not capture the full complexity of clinical reasoning or the influence of ongoing professional development. Future research should incorporate validated, literature-informed¹⁵ instruments to directly assess healthcare providers’ knowledge, attitudes, and beliefs (KAB) regarding antibiotic use and antimicrobial resistance (AMR). Approaches such as structured questionnaires, in-depth interviews, and standardised KAP surveys can offer more detailed insights into the behavioural and cognitive drivers underlying antibiotic dispensing practices. Thirdly, the simulated client approach, while valuable for standardisation and reducing bias, may not fully capture the complexities of real-world clinical interactions. The study period encompassed both the peak URTI season in Iraq (October to February) and the spring months (March to April), when infection rates typically decline. This although provides a reasonable representation of antibiotic supply behaviours across different seasonal contexts, it may have influenced provider decision-making. Additionally, all simulated clients were standardised young adult males; dispensing practices may differ for other demographic groups, such as children, the elderly, or individuals with comorbidities. Real patients may express their needs differently or present with more complex symptoms, potentially influencing providers’ decision-making. Future research should consider varying both case profiles and study timing to better capture these contextual influences. Nonetheless, the simulation scenarios were designed to reflect common and recognizable presentations for mild URTIs, a condition widely known to be viral and not requiring antibiotics. As a result, offering a valid test of rational prescribing thresholds. Lastly, while the simulated client approach provides a reliable snapshot of behaviour, it cannot capture internal decision-making processes or systemic pressures such as drug availability, patient load, or financial incentives. Qualitative follow-up studies with different prescribers in Iraq may help to unpack these drivers and offer deeper insights into the motivations behind current antibiotic prescribing and dispensing practices. However, despite these limitations, we believe this study delivers a robust and much-needed contribution to the evidence base on antibiotic use in a critical area of primary-care delivery in Iraq.

Conclusion

This study provides the first in-depth, empirical investigation into the patterns, appropriateness, and counselling practices surrounding antibiotic dispensing in nurse-led clinics in Iraq. Our findings reveal widespread inappropriate and, in many cases, illegal dispensing of antibiotics for conditions that do not warrant their use, reflecting substantial gaps in patient assessment and adherence to national regulations. The results also underscore the vulnerability of primary care settings to systemic drivers such as insufficient provider training, weak enforcement, and patient pressure. These findings have broad implications not only for Iraq’s NAP on AMR but also for global efforts to contain AMR. As a result, there is an urgent need for context-specific stewardship policies as well as HCP training and community-level education. In addition, enhanced regulatory monitoring to improve antibiotic use in this important sector in ambulatory care in Iraq. Our findings present an opportunity to inform the ongoing revision and implementation of Iraq’s NAP by incorporating measurable AMS indicators tailored to nurse-led clinics. Potential indicators include an agreed percentage of Access-group antibiotic use, routine documentation of allergy screening, and an agreed adherence target to standardized treatment protocols for URTIs building on the WHO AWaRe guidance. Embedding such indicators into Iraq’s stewardship monitoring frameworks would not only enhance accountability but also support resource allocation and capacity-building efforts in one of the country’s most widespread, yet least regulated, primary care settings. Providers with lower qualifications were more likely to dispense antibiotics inappropriately, suggesting that strengthening clinical education and regulatory literacy is imperative to improve future antibiotic use. The frequent dispensing of Watch antibiotics highlights a misalignment with the WHO AWaRe framework and guidance further underscoring the need for targeted prescribing protocols alongside agreed quality indicators. Furthermore, providers often responded to patient pressure during simulated demand escalation indicating the importance of empowering staff with communication tools to manage inappropriate requests. In the short term, measures such as structured training, public education campaigns, and stewardship indicators specific to primary care could be implemented with minimal disruption. However, policymakers should also anticipate and mitigate unintended consequences. For example, by ensuring that regulatory crackdowns are accompanied by support systems to avoid pushing inappropriate dispensing into informal channels. By addressing these dynamics in a coordinated and context-sensitive manner, Iraq can move closer to realising its AMR containment goals. Aligning our findings with Iraq’s NAP implementation priorities would therefore enhance the country’s ability to meet the AMR containment targets set by the United Nations and the WHO.