Study design and setting

This was a prospective, pre-post interventional study conducted at Saint Paul General Hospital, a 800-bed tertiary care hospital located in Hanoi, Vietnam. Established as one of the leading surgical centers in the capital with nearly 200 beds for surgery departments, the hospital serves as a regional referral institution for surgical cases. The study aimed to evaluate the impact of a clinical pharmacist-led A&F intervention on compliance to the hospital’s SAP guidelines.

Study population

The study population included patients undergoing elective (scheduled) surgical procedures in the hospital’s surgical departments. Only surgeries that were listed in the hospital’s official SAP guideline and classified as clean or clean-contaminated (as determined by the operating surgeon) were eligible for inclusion. This restriction was based on the scope of the hospital’s SAP guideline, which—at the time of the study—only provided specific recommendations for clean and clean-contaminated surgeries. Therefore, other categories of surgical wounds (e.g., contaminated or dirty) were excluded from the study.

Patients were also excluded if they underwent emergency surgery or had incomplete medical records related to antibiotic use.

Ethics approval

This study was granted ethics approvals by the Ethics Committee and AMS committee of Saint Paul General Hospital. The guidelines for SAP applied for the study was approved by AMS committee of the hospital. A&F program by pharmacists in the study was also approved by AMS committee of the hospital. Consent, which was waived by the Ethics Committee was not obtained because data were analyzed anonymously. To ensure ethical standards, the authors had no access to information that could identify individual participants during or after data collection.

Study procedures

The study procedure is illustrated in Fig 1. Detailed information on the intervention is provided in S1 Table.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Procedure of the study.

https://doi.org/10.1371/journal.pone.0351068.g001

Pre-intervention (initial assessment)

During the pre-intervention phase, prophylactic antibiotics were prescribed independently by surgeons based on the hospital’s internal SAP guideline. This guideline had been developed and approved by the hospital’s AMS committee based on international recommendations and included clear criteria for indications, antibiotic selection, dosing, timing of administration, intraoperative redosing, and duration of use for commonly performed surgerie

From the time the guideline was issued, no intervention, or evaluation regarding its implementation had been conducted. No data was available on actual compliance with the guideline in clinical practice prior to this study.

Therefore, as part of the study design, the research team conducted an initial retrospective assessment of SAP prescribing practices to establish a baseline compliance rate. This assessment served as the reference point (T0) for comparison with post-intervention outcomes. The data collection period was from October 1, 2023, to November 30, 2023 (T0).

Intervention: Audit and Feedback Program

The A&F intervention was implemented in multiple phases and coordinated by a multidisciplinary stewardship team led by clinical pharmacists. The team comprised clinical pharmacists, infectious disease specialists, and surgical representatives. The data collection period was from December 1, 2023, to April 30, 2024 (T1).

• Phase 1 – Initial Retrospective Feedback: Baseline compliance data from the pre-intervention phase were analyzed. Clinical pharmacists presented the results to each surgical department, discussed common non-compliance patterns, and recorded surgeons’ feedback to tailor future interventions.
• Phase 2 – Prospective Audit and Feedback: During the post-intervention phase, the stewardship team conducted regularly prospective audits of patients receiving SAP. For each case, antibiotic selection, dose, timing of first dose, and duration of use were assessed. Individualized case-based feedback when needed was provided by clinical pharmacist and weekly summaries of findings were discussed with surgeons at the department level (Fig 1).
• A total of 156 A&F activities (9 weekly department discussion and 147 individual feedback for each cases) were conducted over the 4 months, covering 147 surgical cases.

Study outcomes.

The primary outcome was the proportion of patients who received fully guideline-compliance of SAP (i.e., appropriate across all five key criteria) before and after the intervention.

Secondary outcomes included individual compliance proportion for:

• Antibiotic selection
• Dose administration
• Timing of administration
• Duration of use

These were compared between the pre-intervention (T0) and post-intervention (T1) phases.

• Sample size calculation

The sample size was calculated based on the primary outcome: the proportion of patients receiving fully appropriate SAP [15]. Assuming a baseline compliance rate of 50% and an anticipated 30% absolute improvement post-intervention, with a two-sided alpha of 0.05 and 80% power, the required sample size was 130 patients per phase.

Data collection and assessment

To minimize assessor bias, data collection was conducted independently from the intervention team during pre-intervention and post-intervention phases. Patient data from electronic medical records and prescriptions were collected by the principal investigator (TS Nguyen) and a postgraduate pharmacy student, both of whom were not involved in the implementation of the audit and feedback intervention.

SAP compliance for each case was assessed using a standardized checklist based on the hospital’s SAP guideline. Additional patient information was collected, including age, sex, comorbidities, surgical procedure, treating department, and infection risk factors.

Statistical analysis

All data were entered and analyzed using R version 4.2.1. Continuous variables were expressed as mean ± standard deviation (SD) for normally distributed data or median with interquartile range (IQR) for non-normally distributed data. Categorical and ordinal variables were summarized as frequencies and percentages.

To compare variables between the pre- and post-intervention phases:

• Student’s t-test or Mann–Whitney U test was used for continuous variables,
• Chi-square test or Fisher’s exact test was used for categorical variables, as appropriate.

To assess the effect of the intervention on SAP compliance, univariate logistic regression was performed to estimate odds ratios (OR) and 95% confidence intervals (CI). Multivariable logistic regression was then conducted to calculate adjusted ORs, accounting for potential confounding factors such as patient demographics and surgical characteristics. Prior to regression analysis, multicollinearity among variables was assessed using variance inflation factor (VIF), with a VIF greater than 2 considered indicative of multicollinearity. A p-value < 0.05 was considered statistically significant.

Patient characteristics

The demographic and clinical characteristics between the two patient groups T0 (N = 162) and T1 (N = 147) showed no statistically significant differences in terms of gender, age, BMI, comorbidities, and number of surgical site infection (SSI) risk factors (p > 0.05). No cases of antibiotic allergy were recorded in either group. Overall, the differences between groups were not statistically significant (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Patient characteristics in the study.

https://doi.org/10.1371/journal.pone.0351068.t001

Surgical characteristics of patients

The surgical characteristics between the two patient groups T0 and T1 also showed no statistically significant differences across all criteria (p > 0.05) (S2 Table). Clean surgeries predominated in both groups (T0: 55.56%, T1: 57.82%), and laparoscopic surgeries accounted for nearly half of the procedures (T0: 40.74%, T1: 45.58%). Blood loss during surgery was below 1500 mL in all patients across both groups. The rate of artificial device implantation was 8.25% in T0 and 4.08% in T1, but the difference was not statistically significant (p = 0.150). The median surgery duration remained stable at 50 minutes in both groups. The majority of procedures lasted less than 60 minutes, while surgeries ≥180 minutes were rare (T0: 3.70%, T1: 2.04%).

Antibiotic prophylaxis usage

Details of SAP usage during T0 and T1 are summarized in Table 2. The most frequently prescribed prophylactic antibiotic was cefazolin in both phases, accounting for 61.5% at T0 and increasing to 82.7% at T1. The timing of initial antibiotic administration was 30 minutes before surgical incision for all patients, which also represented the shortest preoperative interval observed. The proportion of cases in which SAP was extended beyond 24 hours postoperatively showed a marked reduction after the intervention, declining from 27.16% in T0 to 11.56% in T1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Antibiotic prophylaxis usage in the study.

https://doi.org/10.1371/journal.pone.0351068.t002

Outcomes

Primary outcomes.

Table 3 presents the results of the univariate and multivariate logistic regression analyses for the risk factors associated with full compliance with the SAP guideline (i.e., the primary outcome). A significant improvement was observed after the interventions. In the pre-intervention phase (T0), 60.49% of patients were classified as fully compliant with SAP guidelines, compared with 76.87% in the post-intervention phase (T1) (adjusted OR 2.97, 95% CI: 1.71–5.28, p < 0.001).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Compliance with SAP Guidelines – Primary Outcomes.

https://doi.org/10.1371/journal.pone.0351068.t003

Secondary outcomes.

The compliance rate with SAP guidelines in T1 showed a significant improvement compared to T0 across multiple criteria (Table 4). Specifically, the “Antibiotic selection” criterion increased from 74.69% to 88.44% (p = 0.002, OR = 2.57, 95% CI: 1.40–4.89), and “Duration of use” increased from 72.84% to 88.44% (p < 0.001, OR = 2.82, 95% CI: 1.55–5.36). The criteria “Timing of administration” and “Additional dosing” maintained 100% compliance, demonstrating stable compliance to these aspects.

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Compliance with SAP Guidelines – Secondary Outcomes.

https://doi.org/10.1371/journal.pone.0351068.t004

Compliance with SAP Guidelines by department

Fig 2 shows compliance rates with SAP guidelines increased in most departments after the intervention, with particularly notable improvements in some specialties. Specifically, the Gastrointestinal Surgery department increased from 18.18% (2/11) to 71.43% (15/21), with p = 0.008, and the Urology department increased from 10.00% (1/10) to 73.33% (11/15), with p = 0.004. These represent statistically significant improvements. Additionally, the Plastic Surgery department increased from 0.00% (0/6) to 50.00% (2/4), although this change was not statistically significant (p = 0.133). These improvements are particularly important in gastrointestinal and urologic surgeries, where the risk of surgical site infections is highest, and where inappropriate prolonged prophylaxis has been a persistent challenge.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Compliance with SAP guideline by department at pre and post-intervention.

https://doi.org/10.1371/journal.pone.0351068.g002

Discussion on the audit–feedback strategy

Audit–feedback is considered a key strategy within AMS [8,13]. This strategy has been shown to be effective not only in antibiotic stewardship but also in other medication groups for modifying healthcare workers’ practices [16]. In the context of antibiotic stewardship, audit feedback has been extensively implemented in various studies as a primary intervention, especially in low- and middle-income countries, to influence prescribers [17]. Additionally, the Ministry of Health of Vietnam has identified audit feedback as a crucial component of the national antimicrobial stewardship guidelines [13]. The hospital pharmacists provided direct feedback to clinical departments and prescribers in both phases, aiming to modify behaviors and practices to align with SAP guidelines. In developing countries like Vietnam, this strategy is cost-effective, requiring minimal infrastructure investment compared to other interventions [8]. Moreover, it serves as an educational tool, particularly for junior physicians, enhancing their accountability and evidence-based prescribing practices. The impact of this strategy can be directly observed through increased rates of appropriate antibiotic use, improved patient outcomes, reduced hospital costs, and enhanced multidisciplinary collaboration within the AMS program [8]. This strategy has also been applied and shown effective in other antibiotic stewardship studies in Vietnam [14].

In this present study, the audit and feedback intervention was deliberately structured in multiple phases to address different objectives and maximize the sustainability of practice change. Phase 1 (Initial Retrospective Feedback) aimed to build awareness and foster consensus by presenting baseline compliance data to surgeons, highlighting common non-compliance patterns, and collecting feedback to tailor subsequent strategies. Phase 2 (Prospective Audit and Feedback) focused on reinforcing behavior change and maintaining compliance over time, through regular department-level discussions of audit findings and individualized case-based feedback when necessary. This multi-phase approach allowed the stewardship team to sequentially (i) raise awareness of current practice gaps, (ii) sustain continuous improvement through regular reinforcement, and (iii) adapt feedback messages to evolving clinical practice, thereby enhancing both the short-term impact and long-term sustainability of the intervention [8,17].

Baseline comparability of study populations across pre and post intervention

Given the pre–post intervention study design, ensuring baseline comparability of the study populations across the two phases is essential. The analysis showed no statistically significant differences in demographic, clinical, or surgical characteristics between the T0 and T1 groups (p > 0.05). Importantly, the groups were also similar regarding surgical site infection (SSI) risk factors, such as surgery type and implant use. This comparability eliminates confounding from baseline characteristics, thereby strengthening the internal validity and attribution of the improvement in SAP compliance to the audit–feedback intervention.

Magnitude of improvement in SAP guideline compliance

The improvement observed in this study aligns with and extends findings from previous research on audit–feedback interventions for surgical prophylaxis [12,17]. In terms of magnitude of effect, our intervention produced a 19% absolute increase in overall compliance (from 60.49% to 76.87%) within a four-month period. This improvement is comparable to other studies in Vietnam, the post-intervention compliance rate in this study was significantly higher than in Huu Nghi Hospital (24.3%) [18], Thong Nhat Hospital (59.3%) [19], 108 Military Hospital (65.93%) [20]. While cross-hospital comparisons must consider institutional differences, it is noteworthy that these facilities are high-tier public hospitals —suggesting that this intervention may be especially impactful when adapted to provincial hospital contexts.

When compared to the landmark multicenter study by van Kasteren et al. (2003) in the Netherlands, which reported an increase in SAP compliance from 28% to 68% over 2 years, our 16.4% increase within 4 months is notable [11]. A 2017 Cochrane review by Davey et al. also found that audit–feedback programs reduced inappropriate prescribing by 20% on average [12]. Thus, the improvement observed here aligns with global averages and demonstrates the replicability of audit–feedback results in the Vietnamese public hospital setting—even with limited resources.

When examining department-level effects, the largest improvements were seen in gastrointestinal and urologic surgery, where compliance rose from 18.2% to 71.4% and 10% to 73.3%, respectively. These specialties are associated with clean-contaminated procedures and higher baseline risk for surgical site infections, underscoring the importance of targeted stewardship interventions in these departments. Such focused improvements demonstrate that audit–feedback can deliver disproportionate benefits in high-risk surgical areas, complementing international evidence that stewardship interventions are most impactful when applied to vulnerable patient populations.

Shifts in antibiotic selection and duration

Another important finding was the positive shift in antibiotic selection and post-operative duration. Prescribers increasingly favored narrow-spectrum agents, especially cefazolin monotherapy, while significantly reducing the use of broad-spectrum antibiotics like third-generation cephalosporins and betalactam/betalactam inhibitors. Overuse of broad-spectrum agents in clean and clean-contaminated surgeries has been documented in prior Vietnamese studies [7,21]. This results in a higher risk of developing AMR, causing adverse events in patients, and incurring unnecessary treatment costs [22]. Meanwhile, the use of cefazolin is also considered a cost-saving measure for patients, as this antibiotic is less expensive than the aforementioned broad-spectrum options.

Post-operative antibiotic duration also improved, with compliance to recommended duration rising from 72.84% to 88.44%. Inappropriate extension of antibiotics until hospital discharge is common in Vietnam’s public hospitals [7,18,23], whereas current guidelines recommend discontinuation within 24 hours for these surgeries [13]. This improvement reflects increased clinician trust in and adoption of evidence-based hospital guidelines.

More importantly, these changes were most evident in gastrointestinal and urologic surgery departments—both associated with clean-contaminated procedures and endogenous microbial risk—highlighting areas where targeted stewardship can have significant impact.

Sustainability and the role of clinical pharmacists

In AMS programs, pharmacists play a vital role in shaping prescribers’ behavior through various interventions. These include education, training, guideline development, direct communication, and audit-feedback [8,17]. At Saint Paul Hospital, the SAP guideline was issued in 2022. However, merely issuing a guideline seems insufficient to ensure good compliance with prophylactic antibiotic use. This has been observed in several hospitals in Vietnam where antimicrobial stewardship programs with appropriate strategies have not yet been implemented [18,24]. This underlines the necessity of periodic audit–feedback interventions to maintain real-world compliance, beyond the passive dissemination of policy documents.

In response, clinical pharmacists initiated the AMS team’s involvement and proposed hospital-wide audit–feedback in all surgical departments. For sustainable impact and broader implementation, deeper understanding of prescriber behaviors is needed—particularly regarding combination therapy, preference for broad-spectrum cephalosporins, and rationale for prolonged prophylaxis. Such insights would inform more refined interventions aimed at optimizing SAP in terms of both clinical efficacy and cost-efficiency.

Sustaining the long-term effectiveness of antimicrobial prophylaxis interventions likely requires multifaceted strategies, including ongoing education, audit and feedback, and system-based approaches. In this context, the integration of digital technologies, such as electronic health record (EHR)-based prompts, may help reinforce and maintain adherence to prophylaxis guidelines over time [8,12]. However, the implementation of such system-level solutions remains limited by infrastructure and resource constraints, and may not yet be widely feasible across healthcare institutions. Therefore, the AMS committee and clinical pharmacy services play a crucial role in sustaining appropriate antimicrobial use through audit and feedback activities. The achieved improvement stemmed from coordinated efforts—systematic intervention guided by hospital leadership, and regular clinical pharmacy activities at the ward level. Everyday, clinical pharmacists and the study team reviewed and analyzed SAP use by department and delivered direct feedback to physicians. This routine not only served as a reminder but also facilitated bidirectional communication, enabling prescribers to better understand non-compliance issues and receive immediate support for correction. Physician engagement was a decisive factor in the success and smooth execution of the audit–feedback process.

Limitations

The study has several limitations. First, post-discharge clinical outcomes were not assessed, making it impossible to link improved compliance with reduced SSI rates. Future studies should incorporate follow-up of SSI incidence to strengthen the causal connection between compliance and patient safety. Second, given the pre–post design, there is potential for unmeasured confounding factors. External influences such as staff turnover, seasonal variation in surgical case mix, or national AMR awareness campaigns could have contributed to changes in prescribing practices. Although baseline characteristics were comparable across phases, causality should be interpreted cautiously. Third, the short follow-up period (four months) limits insights into long-term sustainability. Improvements achieved through audit–feedback may wane without ongoing reinforcement, and periodic monitoring will likely be required to maintain compliance.