“Point education” and “rotation communication” on antimicrobial policy implementation: an interrupted time series analysis of declining cephalosporin skin test rates

Abstract

Background

Following China’s 2021 national policy canceling routine cephalosporin skin testing (CST), substantial implementation barriers persist in some hospitals due to traditional clinical practices and fears about liability. For example, prescribers often order CST out of habit, citing concerns about potential allergic reactions and the risk of adverse events being attributed to them without a documented skin test. A similar challenge exists in the nephrology department of our hospital. Three months following the release of the national policy, the rate of routine CST remained persistently high. This highlights an urgent need for clinical pharmacists to implement targeted interventions to facilitate policy adoption and improve practice adherence. This study compares two different strategies: pharmacist-led Collaborative Learning in Practice(CLiP)-aligned “point education” (concentrated, personalized training) vs. “rotation communication” (without pharmacist-led point education, indirect knowledge diffusion through physician rotations and informal information sharing). We analyze their differential effectiveness in reducing skin test rates in a real-world clinical setting.

Methods

A retrospective observational study collected data on hospitalized nephrology patients planned for cephalosporin use between January 2021 and April 2023. Using the April 2022 pharmacist intervention in one treatment team as the node, interrupted time series analysis compared pre-/post-level and trend changes of skin test rates among wards (first nephrology group, second nephrology group, Nephrology Intensive Care Unit (NICU)) and physicians.

Results

The overall skin test rate decreased dramatically from 96.02% pre-intervention to 25.19% post-intervention. The second nephrology ward receiving point education showed a significant immediate drop post-intervention (-71.081%, P < 0.001) with a sustained downward trend (slope: -0.121, P < 0.001). The first nephrology group and NICU, without direct intervention, demonstrated no significant immediate effects but significant continuous declining trends (slope: -0.282, P = 0.009 and − 0.326, P = 0.026, respectively). Physician-level analysis paralleled ward-level findings, with directly educated physicians showing more rapid declines. Notably, after positive skin tests, 43.1% of patients were switched to quinolone antibiotics, indicating continued clinical caution and antibiotic stewardship considerations. These findings confirm our hypotheses: point education showed an immediate effect, while rotation communication showed a sustained trend.

Conclusion

The cephalosporin skin test rate in the Second Nephrology Ward, where the clinical pharmacist was based, showed an immediate decline, which may be associated with the pharmacist-led “point education”. On the basis of this initial effect, “rotation communication” achieved slower but sustained policy implementation across a broader range of clinical settings through peer-to-peer exchanges during physician rotations. These two mechanisms may complement each other to support effective policy implementation in complex clinical settings. These findings provide empirical evidence for optimizing clinical pharmacy service models and developing policy dissemination strategies for antimicrobial stewardship.

Background

Cephalosporins are important medicines in anti-infective therapy due to their broad antimicrobial spectrum, potent bactericidal activity, and good safety profiles [1]. The routine use of skin testing before cephalosporin use has been a long-standing clinical practice in many healthcare settings, particularly in China [2]. However, increasing evidence has challenged the predictive value and clinical utility of this practice for cephalosporins [3, 4]. Concerns regarding the lack of specificity, poor positive predictive value, and absence of correlation between skin test results and true allergic reactions have prompted international guidelines to recommend against routine cephalosporin skin testing [5]. The National Health and Health Commission of China published the Principle for Skin Testing(ST) of β-Lactam Antibacterial Drugs in 2021 [6], which pointed out that there is no need for a skin test before cephalosporin use.

Rather than depending exclusively on skin testing, effective prevention of hypersensitivity reactions requires comprehensive allergy history screening, careful monitoring during drug administration, readily available emergency medications and equipment, and sufficient training for medical staff in managing severe allergic reactions. Despite these recommendations, de-implementation of established practices faces significant barriers, including ingrained clinical habits, medicolegal concerns, and knowledge gaps among healthcare providers [7–9]. This is particularly evident in nephrology, where patients’ altered pharmacokinetics and heightened susceptibility amplify clinician apprehension [10]. The label of antibiotic allergies is one of the barriers to effective antibiotic treatments, as they may be associated with increased prescriptions of alternative broad-spectrum antibiotics, and consequently, with a high risk of antibiotic resistance [11]. Therefore, a growing body of research has been dedicated to exploring strategies for removing allergen labeling from penicillin and cephalosporin agents, as well as eliminating the requirement for routine skin testing prior to their administration [12, 13]. As they supported, direct oral drug challenge and systematic allergy delabeling represent safer, more effective alternatives for evaluating β-lactam hypersensitivity without universal skin testing. Shang et al. developed and validated a pharmacist-led β-lactam allergy risk stratification tool tailored for Chinese clinical practice. The tool was established based on systematic review, grounded theory, and text co-occurrence analysis, allowing accurate risk classification for patients with reported β-lactam allergy [14].

Since 2005, China has launched a program for training clinical pharmacists. Up to now, more than 30,000 clinical pharmacists have obtained the necessary qualifications and started working [15]. However, the overall number is still very insufficient.“Clinical pharmacists are uniquely positioned to drive antimicrobial stewardship through direct education and collaborative communication [16]. Clinical pharmacists delabel patients with low-risk allergies through structured risk stratification and allergy reassessment, optimizing antimicrobial use in line with national β-lactam guidelines. Their dual role as both knowledge experts and network brokers enables them to bridge evidence-practice gaps [17].

The Second Hospital of Shanxi Medical University, a 2500-bed tertiary teaching hospital in North China, issued a notice on standardizing cephalosporin skin testing in December 2021. However, the Department of Nephrology continued to perform routine cephalosporin skin tests in the following three months. This case catalyzed the implementation of targeted “point education” interventions in the Second Nephrology Ward. Notably, no active or systematic education was provided to the First Nephrology Ward or Nephrology ICU. Instead, policy-related knowledge was disseminated through the department’s existing physician rotation system and informal exchanges during daily work and case discussions (i.e. “rotation communication”). This naturally occurring “intervention-control” scenario offered a unique opportunity to evaluate the impact of different knowledge dissemination models on policy implementation.

This study evaluates two distinct implementation strategies: “point education”—targeted, intensive education delivered directly by clinical pharmacists to one treatment team—and “rotation communication”—indirect knowledge diffusion through routine physician rotations and informal information sharing to non-targeted wards. We hypothesized that point education would produce a rapid, substantial reduction in skin testing rates within the target team (strong immediate effect), while rotation communication would generate a more gradual but sustained decline across the broader department through knowledge spillover and normative shifts. This study aims to quantify and compare the trajectories of practice change resulting from these two implementation models, providing empirical evidence to guide policy promotion strategies and clinical pharmacy service optimization in the context of national antimicrobial stewardship initiatives.

Materials and methods

Study setting and population

This retrospective observational study was conducted at the Second Hospital of Shanxi Medical University. The Department of Nephrology has three wards, namely Nephrology Ward 1, Nephrology Ward 2, and the Nephrology Intensive Care Unit (NICU), with a total of 100 beds. Physicians across the three nephrology wards were assigned on a regular rotation basis, whereas nursing staff were stationed permanently in their respective wards. Clinical pharmacists provide pharmaceutical care exclusively in the Nephrology Ward 2.

In April 2022, clinical pharmacists implemented Collaborative Learning in Practice(CLiP)-aligned point education interventions aimed at discontinuing routine CST, including: (1) collaborative knowledge translation through in-depth communication with attending physicians regarding the core principles and evidence base of the national guideline; (2) Supportive competency development for nurses integrates knowledge training and psychological support, which consists of reassuring clinical staff, addressing fears of allergic reactions and medicolegal risks, and enhancing confidence in guideline-adherent practice to alleviate anxiety about discontinuing routine skin testing; (3) contextual team-based learning via case-centered discussions to strengthen shared understanding and drive sustainable practice change. In contrast, the First Nephrology Ward and Nephrology ICU received no direct education; policy knowledge was instead disseminated through peer-to-peer exchanges among rotating physicians via rotation communication, a peer-driven learning mechanism based on physician rotations and daily clinical interactions. Physicians who had received point education from clinical pharmacists in the Second Nephrology Ward likely served as the key source of this policy-related knowledge.

Inclusion criteria: (1)Adult inpatients admitted to the Department of Nephrology (including general wards and NICU) between January 2021 and April 2023. (2)Patients who were planned for cephalosporin-based anti-infective therapy(Patients with billing records of cephalosporin agents and/or cephalosporin skin test fees during hospitalization). (3) Patients with complete data on prescription records and basic clinical information.Exclusion criteria: (1)Patients with a documented history of severe immediate hypersensitivity to β-lactam antibiotics. (2)Cephalosporin use for surgical prophylaxis only without documented infection.

The study period was divided into pre-intervention (January 2021–April 2022) and post-intervention (May 2022–April 2023) phases, with April 2022 as the cutoff point, when the targeted “point education” intervention was initiated in the Second Nephrology Ward. Based on the intervention model, study participants were categorized into two groups:

Direct intervention group: Patients admitted by doctors who received direct and systematic advice from clinical pharmacists in the Second Nephrology Ward.

Indirect influence group: Patients in the First Nephrology Ward and Nephrology ICU whose doctors did not receive direct systematic advice with policy knowledge primarily obtained through physician rotation and daily communication.

The study was approved by the Research Ethics Committee of the Second Hospital of Shanxi Medical University (Approval No.: 2025 − 251).

Data collection

Data were extracted from the Hospital Information System (HIS) and included:

Demographic information (gender, age); Clinical details (cephalosporin type, administration duration, performance and results of CST, occurrence of adverse drug reactions [ADRs] during therapy, alternative treatment regimens after positive CST); Prescriber information.

Outcomes

Primary outcome: CST rate (%) = (Number of medication episodes with ST performed / Total number of cephalosporin medication episodes in the same period) × 100%.Secondary outcomes: CST positive rate (%) = (Number of positive ST results / Total number of ST performed) × 100%. Adjustment of treatment regimens following positive CST.

Statistical analysis

Continuous data were described as median and interquartile range (IQR), with group comparisons using the Mann-Whitney U test. Categorical data were presented as frequencies and percentages, with comparisons using the χ² test.

Interrupted Time Series Analysis(ITSA) was used to evaluate the impact of interventions on CST rates. A segmented linear regression model was constructed as follows: Y = β₀ + β₁ × T + β₂ × X + β₃ × XT + ε. The definitions of variables and coefficients are detailed in Table 1.

Table 1.

Variable and coefficient definitions and explanations

Variable / Coefficient	Explanation
Y	Observed CST rate at month t
T	Time from the start of the observation period to the last time point t in the series, measured in months
X	Intervention indicator (0 = pre-intervention, 1 = post-intervention)
β 0	Baseline CST rate
β 1	Pre-intervention trend of CST rate
β 2	Immediate effect of intervention (change in intercept)
β 3	Sustained effect of intervention (change in slope)
ε	Random error term.

Analyses were performed at three levels: overall department, individual sub-wards, and individual physicians (who prescribed cephalosporins ≥ 50 times pre- and post-intervention). Model fit was assessed using adjusted R² (R²adj) and Durbin-Watson (DW) statistics. All analyses were conducted using R software (version 4.2.2), with a two-tailed P-value < 0.05 considered statistically significant.

Results

Overall CST rates and CST positive rate

A total of 2022 cephalosporin medications were included, cephalosporin consumption remained stable before and after the intervention. The overall CST rate decreased significantly from 96.02% (916/954) pre-intervention to 25.19% (269/1068) post-intervention, representing an absolute reduction of 70.83%. The overall CST positive rate slightly increased from 5.13% (47/916) pre-intervention to 6.69% (18/269) post-intervention, with no statistically significant difference (P = 0.325). CST rates for all individual cephalosporin varieties decreased substantially post-intervention in Table 2. Cephalosporin Utilization by Prescriber were showed in Table 3. It shows that all physicians rotated across different wards throughout the study period, providing a natural setting for cross-ward intervention diffusion. All physicians had cephalosporin prescribing records. Physicians No. 16, 7, 18, and 20, with > 50 prescriptions in both pre- and post-intervention phases, were selected as the main subjects for subsequent individual-level ITSA.

Table 2.

Cephalosporin use and skin testing in the nephrology department

Drug Category	Pre-intervention			Post-intervention			Reduction rate(CST%)	P-value (CST%)	P- value (Positive Rate)
	Administrations	CST n (%)	Positive Rate n (%)	Administrations	CST n (%)	Positive Rate n (%)
Total	954	916 (96.02%)	47 (5.13%)	1068	269 (25.19%)	18 (6.69%)	-70.83%	< 0.001	0.325
Ceftriaxone	535	503 (94.02%)	25 (4.97%)	568	146 (25.70%)	9 (6.16%)	-68.32%
Cefuroxime	242	241 (99.59%)	15 (6.22%)	284	70 (24.65%)	5 (97.14%)	-74.94%
Cefoperazone/Sulbactam	95	93 (97.89%)	1 (1.08%)	129	22 (17.05%)	1 (4.55%)	-80.84%
Cefoxitin	42	42 (100.00%)	3 (7.14%)	0	0	0 /	/
Ceftazidime	35	35 (100.00%)	3 (8.57%)	61	23 (37.70%)	0 (0.00%)	-62.30%
Cefaclor	2	0	0	3	0(0.00%)	0 /	/
Cefixime	1	0	0 /	0	0 /	0 /	/
Cefazolin	1	1 (100.00%)	0 (0.00%)	19	7 (36.84%)	2 (28.57%)	-63.16%
Cefmetazole	1	1 (100.00%)	0 (0.00%)	1	0 (0.00%)	0 /	/
Cephalexin	0	0	0	2	0 (0.00%)	0 /	/
Ceftazidime/Avibactam	0	0	0	1	1 (100.00%)	1 (100.00%)	/

Note: Data are presented as n or percentage (%). Nephrology Department: Nephrology ICU, Nephrology Ward 1, Nephrology Ward 2

Table 3.

Cephalosporin utilization by prescriber

Doctor Information		Doctor’s Ward		Cephalosporin Prescription Status(n)			Total
Grading	No	Pre-intervention	Post-intervention	Pre-intervention		Post-intervention
Resident Physician
	1	NW-1	NW-1;NW-2	13		60	73
	10	NW-2	NW-2;NICU	56		45	101
	13	NW-1	NW-2;NICU	1		43	44
	17	NW-2;NICU	NW-1;NW-2	67		43	110
Attending Physician
	2	NW-1	NICU; NW-1	16		70	86
	9	NW-2;NICU	/	2		0	2
	12	NW-2;NICU	NICU; NW-1༛NW-2	18		3	21
	14	NW-1;NW-2༛NICU	NW-1;NICU	35		66	101
	16	NW-1	NW-2;NICU༛NW-1	115		68	183
Associate Chief Physician
	4	NW-1	NW-1;NICU	46		50	96
	6	NW-2;NICU	NICU; NW-1༛NW-2	45		76	121
	7	NW-2;NICU	NW-2;NICU	83		67	150
	11	NW-1	NW-1;NICU	27		36	63
	15	NW-2;NICU	NW-2;NICU༛NW-1	68		44	112
	18	NW-2;NICU	NW-1;NICU༛NW-2	150		88	238
	19	NW-2	NW-2;NICU	35		43	78
	20	NW-2	NW-2;NICU	50		83	133
	21	NW-2	NW-2;NICU	5		69	74
Chief Physician
	3	NW-1;NICU	NW-1;NICU	49		26	75
	5	NW-1;NICU	NW-2;NICU	27		14	41
	8	NW-1;NW-2༛NICU	NW-1;NW-2༛NICU	38		65	103
	22	NW-2;NICU	NW-1;NICU	8		9	17
Total				954		1068	2022

Note: Doctor numbers range from 1 to 22; NICU: Nephrology ICU; NW-1: First Nephrology Ward; NW-2: Second Nephrology Ward

Cephalosporin use and CST rates by sub-ward

As shown in Fig. 1, NICU patients exhibited low cephalosporin skin testing, which may be explained by patient transfer patterns. Some patients were initially admitted to general wards and had already undergone cephalosporin skin testing there, thereby avoiding repeated testing after being transferred to the NICU.Regarding cephalosporin prescribing volume, the intervention of eliminating routine skin testing was associated with increased cephalosporin use across all three nephrology wards. Prescriptions rose from 126 to 163 in the Nephrology ICU (NICU), from 405 to 425 in the Second Nephrology Ward (NW-2), and from 423 to 480 in the First Nephrology Ward (NW-1). These trends indicate that the intervention did not compromise the clinical accessibility of cephalosporins but was instead accompanied by a moderate increase in their appropriate use. In contrast, cephalosporin skin testing rates decreased significantly in all units post-intervention. The number of skin tests dropped from 105 to 49 in the NICU, from 391 to 52 in NW-2, and from 420 to 168 in NW-1. Collectively, these data demonstrate that the intervention successfully reduced unnecessary cephalosporin skin testing across all clinical settings while maintaining, and even slightly increasing, cephalosporin prescription volumes.

Fig. 1.

Cephalosporin prescription and skin testing practices pre- and post-intervention; Pre/Post: Pre/Post: Pre/Post -intervention-abolition of routine cephalosporin skin test; NICU-Nephrology ICU; NW-1: First Nephrology Ward; NW-2: Second Nephrology Ward. N-CST: No Cephalosporin Skin Testing CST: Cephalosporin Skin Testing

ITSA Results

Overall and sub-ward CST rate trends

Overall, ITSA showed that the CST rate decreased significantly by 26.63% immediately post-intervention (P = 0.008), with an additional monthly reduction of 0.215% (P = 0.008). The model demonstrated good fit (R²adj = 0.763, DW = 1.707) (Table 4; Fig. 2A). Second Nephrology Ward (direct intervention group): A highly significant immediate reduction in CST rate (-71.081%, P < 0.001) and a significant sustained declining trend (slope: -0.121, P < 0.001) (Table 4; Fig. 2B). First Nephrology Ward (indirect influence group): No significant immediate effect (-6.899%, P = 0.621), but a significant sustained declining trend (slope: -0.282, P = 0.009).Nephrology ICU (indirect influence group): No significant immediate effect (5.321%, P = 0.761), but a significant sustained declining trend (slope: − 0.326, P = 0.026).

Table 4.

Results of the segmented regression analysis of CST%

Outcomes	Coefficient	Standard Error	t - Statistic	p - value
Total CST% (DW = 1.707)
Baseline level	94.041	16.869	5.575	< 0.001
Baseline trend	0.001	0.051	0.015	0.989
Level change after intervention	-26.630	9.185	-2.900	0.008
Trend change after intervention	-0.215	0.073	-2.927	0.008
NW1 CST% (DW = 1.629)
Baseline level	96.418	21.955	4.392	< 0.001
Baseline trend	0.003	0.068	0.043	0.966
Level change after intervention	-6.899	13.781	-0.501	0.621
Trend change after intervention	-0.282	0.100	-2.818	0.009
NW2 CST% (DW = 1.640)
Baseline level	93.136	3.568	26.105	< 0.001
Baseline trend	0.016	0.014	1.188	0.247
Level change after intervention	-71.081	5.237	-13.574	< 0.001
Trend change after intervention	-0.121	0.022	-5.565	< 0.001
NICU CST% (DW = 1.812)
Baseline level	70.707	33.102	2.136	0.044
Baseline trend	0.032	0.099	0.319	0.752
Level change after intervention	5.321	17.311	0.307	0.761
Trend change after intervention	-0.326	0.137	-2.385	0.026
Dr.18 CST% (DW = 1.219)
Baseline level	100.000	7.286	13.861	< 0.001
Baseline trend	-0.008	0.028	-0.291	0.774
Level change after intervention	-56.752	10.809	-5.251	< 0.001
Trend change after intervention	-0.140	0.044	-3.191	0.004
Dr.16 CST% (DW = 2.058)
Baseline level	97.507	16.383	5.952	< 0.001
Baseline trend	0.004	0.055	0.077	0.939
Level change after intervention	-1.379	15.354	-0.090	0.929
Trend change after intervention	-0.327	0.082	-3.966	< 0.001
Dr.7 CST% (DW = 2.018)
Baseline level	95.481	6.507	14.674	< 0.001
Baseline trend	0.007	0.026	0.267	0.792
Level change after intervention	-63.160	9.332	-6.768	< 0.001
Trend change after intervention	-0.119	0.037	-3.189	0.005
Dr.20 CST% (DW = 1.714)
Baseline level	99.636	15.765	6.320	< 0.001
Baseline trend	-0.053	0.161	-0.330	0.746
Level change after intervention	-38.219	21.264	-1.797	0.092
Trend change after intervention	-0.137	0.167	0.818	0.426

Note: NICU: Nephrology ICU; NW-1: First Nephrology Ward; NW-2: Second Nephrology Ward. pre-CU: pre-intervention Cephalosporin use; post-CU: Pre-intervention Cephalosporin use; pre-CST: pre-intervention-Cephalosporin Skin Testing; post-CST: post-intervention-Cephalosporin Skin Testing

Fig. 2.

A: Trend of cephalosporin skin test rate pre- and post-intervention (ITSA); B: Trend of Cephalosporin Skin Test Rate by Sub-ward Pre- and Post-intervention (ITSA); C: Trend of cephalosporin skin test rate by physicians. NICU: Nephrology ICU; NW-1: First Nephrology Ward; NW-2: Second Nephrology Ward. Note: The vertical dashed line indicates the intervention start (April 2022)

Individual physician CST rate trends

Analysis of four high-volume prescribers showed consistent results with sub-ward-level findings (Table 4; Fig. 2C): Physicians 18 and 7 (directly associated with the Second Nephrology Ward and recipients of “point education”) exhibited significant immediate reductions (P < 0.001) and sustained declining trends (P < 0.01) in CST rates post-intervention. Physicians 16 and 20 (who obtained policy knowledge primarily through departmental communication) showed no significant immediate effects. However, Physician 16 demonstrated a highly significant sustained declining trend (slope: -0.327, P < 0.001), while Physician 20 showed no significant sustained trend (P = 0.426).

Treatment adjustments after positive CST

Among 65 positive CST episodes, quinolones (levofloxacin, moxifloxacin) were the most common alternative therapy (28, 43.1%). Thirteen episodes (20.0%) were switched to other β-lactam antibiotics (including different cephalosporin varieties, β-lactamase inhibitor combinations, and cephamycins). Two patients with positive cefazolin ST were switched to meropenem (Table 5), reflecting clinical caution regarding cross-reactivity risks among β-lactam antibiotics.

Table 5.

Alternative treatment regimens after positive cephalosporin skin test alternative therapy number of episodes percentage (%)

Type	PST(n)	Alternative Anti-infective Drugs
Ceftriaxone	34	Levofloxacin (11), Clindamycin Phosphate (6), Moxifloxacin (8), Moxifloxacin Tablets + Azithromycin Tablets (1), Cefuroxime Sodium (1), Cefoxitin Sodium (1), Unknown (5), Piperacillin/Sulbactam (1)
Cefuroxime	20	Clindamycin Phosphate (5), Levofloxacin (6), Azithromycin (2), Moxifloxacin Tablets (1), Cefoperazone/Sulbactam (1), Ceftriaxone (2), Unknown (3)
Cefoperazone/Sulbactam	1	Ceftriaxone (1)
Cefoxitin	3	Cefoperazone/Sulbactam (2), Levofloxacin (1)
Ceftazidime	2	Cefoperazone/Sulbactam (1), Ceftriaxone (1)
Cefazolin	2	Meropenem (2)
Ceftazidime/Avibactam	1	Azithromycin + Moxifloxacin (1)

Note: PST: Positive Cephalosporin Skin Test

Discussion

“Point education” delivers immediate initiating effects

ITSA results demonstrated that the Second Nephrology Ward, which received systematic “point education” led by clinical pharmacists, showed a statistically significant immediate reduction in CST rates. This observation suggests that targeted, case-based interventions delivered by pharmacy professionals may help rapidly shift long-standing clinical behaviors, and it may contribute to overcoming early implementation barriers for policy adoption [18].

The effectiveness of “point education” appears related to its focused nature: it concentrates on addressing knowledge and cognitive barriers among key stakeholders (physicians, head nurses) and shares successful clinical cases to support team confidence and willingness to follow guidelines. This finding are consistent with previous studies highlighting the value of tailored, evidence-based education in supporting behavioral change in antimicrobial stewardship [19, 20]. They also align with implementation science evidence that intensive, targeted interventions for key team members may be more effective than passive dissemination strategies in changing complex clinical practices. The use of a real-world local case likely enhanced acceptance and credibility, in line with evidence that case-based learning can facilitate knowledge translation in clinical settings [21].

“Rotation communication” drives widespread and sustained improvement

A notable finding was that CST rates showed significant sustained declines in the First Nephrology Ward and Nephrology ICU, even in the absence of direct pharmacist education. This pattern is unlikely to be explained by natural temporal trends alone and maybe largely attributable to the department’s physician rotation system and informal peer communication. Rotating physicians may act as “knowledge carriers,” potentially spreading guideline concepts, practical experience, and successful examples from the Second Nephrology Ward to other sub-wards.

Such diffusion appears gradual and pervastive, reflected as changes in trend slope rather than an immediate level shift. It confirms that within healthcare organizations, effective practices can spread substantially through personnel mobility and daily interactions, even without formal cross-departmental training [22]. Physician-level analysis (e.g. Physician 16) provides further support for this mechanism, and highlights the potential value of internal organizational communication channels for guideline dissemination. This pattern is consistent with social network theory, in which healthcare professionals may serve as ‘knowledge brokers’ within departmental networks [23]. The gradual slope change is typical of innovations diffusion through trusted peer networks, where adoption tends to occur as individuals observe and discuss new practices with colleagues. These observations underscore the potential of existing organizational structures to support policy dissemination.”

Synergistic driving mechanism of “point education” and “rotation communication.”

This study suggests a dual-driving mechanism that may support policy implementation in clinical departments. Point education was designed and delivered as a structured, pharmacist-led intervention aligned with Collaborative Learning in Practice (CLiP). As an active, intensive intervention, it directly targets all three components of the Capability-Opportunity-Motivation-Behavior (COM-B) behavior change system [24]: enhancing capability by updating knowledge and skills, strengthening motivation by alleviating anxiety and medicolegal concerns, and improving opportunity by embedding guideline-adherent practices into daily workflow. It may create local demonstration sites with early observable impact, making it particularly useful during early implementation or for teams with persistent practice gaps, although it requires dedicated personnel resources and may not be broadly feasible across all wards simultaneously.Rotation communication: A passive, low-intensity but networked diffusion model that utilizes existing internal communication pathways to support natural, low-cost knowledge spread. It mainly acts on the opportunity domain of COM-B by expanding peer learning and social influence, gradually reshaping collective norms.However, its effects typically emerge more gradually and may depend on the frequency and quality of peer interactions.

In this study, the two approaches appeared to act synergistically. CLip-aligned “point education” in the Second Nephrology Ward established a “successful local example and built practical experience, providing credible content for peer diffusion via “rotation communication”; in turn, “rotation communication” helped extend these localized improvements more broadly across the department, supporting gradual and sustained guideline adherence.

This observations may suggests a practical“benchmark-to-diffusion” strategy for promoting clinical medication management policies: focus resources to develop successful local examples, while supporting internal peer communication channels (e.g. rotations, case discussions, ward handovers) to share effective practices. Such a strategy may help balance implementation speed and coverage.

Reconfirmation of safety and emerging issues in discontinuing routine CST

This studyadds to existing evidence that discontinuing routine CST does not appear to increase apparent allergy risks when accompanied by careful allergy history screening and patient monitoring, as reflected by the non-significant change in overall CST positive rates. This aligns with findings from domestic and international studies [25, 26], providing institutional-level support for broader guideline implementation.

Analysis of treatment adjustments reveals a notable trend of quinolone use after positive CST, continued risk aversion among prescribers but could potentially contribute to unnecessary broader-spectrum antibiotic use and antimicrobial resistance [25]. This pattern—here concerns regarding β‑lactam hypersensitivity may lead to preferential use of alternative agents—could partly offset the goals of antimicrobial stewardship [27]. Therefore, the CST waivering can reduce inappropriate adjustments to treatment regimens caused by false-positive skin test results. In addition, it lowers medical costs, improves clinical workflow efficiency, and is consistent with the national goals of Antimicrobial Stewardship (AMS).

Study limitations

This was a single-center retrospective observational study, which may limit the generalizability of findings. The measurement of ‘rotation communication’ was indirect; future studies could consider social network analysis or qualitative methods to more directly characterize knowledge flow and peer influence within the department.

Conclusion

In the implementation of the policy to discontinue routine cephalosporin skin testing in the nephrology department, clinical pharmacist-led “point education” was associated with rapid early practice change and local benchmarks establishment, demonstrating significant immediate effects. In contrast, “rotation communication” based on physician rotations and daily peer interaction appeared to serve as an important pathway for gradual, system-wide diffusion of guideline knowledge and practical experience, supporting broad and sustained improvement across the department. These two mechanisms—one targeted and intensive, the other network-based and gradual—may complement each other to support effective policy implementation in complex clinical settings.

The findings suggest that under constrained resources, strategy of targeted benchmarking plus internal network diffusion may be effective for promoting rational antibiotic use policies. Clinical pharmacists may serve not only as knowledge educators but also as facilitators connecting practice groups to support experience sharing, potentially maximizing the impact of antimicrobial stewardship initiatives.

Author contributions

Conceptualization: Donghong Yin; Methodology: Zhihong Ren, Donghong Yin; Investigation: Lin Chen, Qian Guo; Visualization: Jinju Duan; Supervision: Donghong Yin, Shuyun Wang; Writing—Donghong Yin, Zhihong Ren; All authors reviewed the manuscript.

Funding

This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

Data are contained within the manuscript.

Declarations

Ethical considerations and consent to participate

The study was conducted in accordance with the Declaration of Helsinki and the General Data Protection Regulation (GDPR). It was a retrospective observational study based on routine clinical practice. No personally identifiable information, including patient and doctors' names or ID numbers, was recorded; instead, only coded identifiers were used. Since the study did not involve any modifications to patients' treatment regimens, it was granted ethical exemption. In compliance with the Ethical Review Measures for Life Science and Medical Research Involving Humans issued by the National Health Commission of the People's Republic of China, written informed consent was waived for all participants. Ethical approval was obtained from the Ethics Committee of the Second Hospital of Shanxi Medical University (Approval No.: [2025] YX-251) on July 14, 2025.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.