Assessing the Impact of a Training Intervention on Venous Compression Bandaging Application: A Factorial Design Randomised Trial

ABSTRACT

Venous leg ulcers (VLU) are a major complication of chronic venous disease, with compression therapy as the gold‐standard treatment. This 2 × 2 factorial randomised open‐label trial assessed the effect and three‐month sustainability of a one‐hour training intervention on compression bandaging quality among 50 advanced practice nursing students (APNS) in France, using two different compression bandage types (system A or system B) on a mannequin leg. The primary outcome was the Control Score of compression bandaging (CCB score, 0–7). Secondary outcomes included installation conformity, application time, and satisfaction. At baseline, mean CCB score was 3.78, with 52% installation conformity. The intervention did not significantly improve the CCB score compared to controls (adjusted difference: −0.43; 95% CI: −1.01 to 0.15). A significant interaction between bandage type and intervention was found for installation conformity: a tendency to improve with system B and decrease with system A. Improvements observed at 1 month were not sustained at 3 months. Application time decreased over follow‐up, and system B was applied more efficiently than system A. Brief, behaviourist‐based training was insufficient to achieve lasting skill improvement. Findings highlight the need for repeated group practice, feedback, and integration into relevant clinical contexts to enhance compression bandaging proficiency.

KeyPoints

• Effectively training nurses to apply venous compression is an educational concern.

• The aim of the study was to assess the effect and sustainability at three months of a training intervention on the compression bandaging quality on a mannequin leg, for two different compression bandage.

• We conducted a two‐factorial randomized trial for advanced practice nursing students (APNS).

• Skills of APNS in in applying compression bandages are suboptimal.

• Short educational intervention does not significantly improve compression efficiency.

1. Introduction

Venous leg ulcers (VLU) are the main complication of chronic venous disease. The prevalence of VLU is estimated in Europe to be between 0.1% and 2.5% [1, 2, 3], increases each year [4] and costs about 1% of the western countries 2022 health budgets [5]. Consequences for the patients are serious: pain, limited mobility, and decline in the quality of life. Internationally recommended treatment for VLU is compression therapy with multi‐layer or short‐stretch bandages [4, 6, 7, 8, 9, 10, 11].

Nurses currently provide compression therapy. In practice, their knowledge and skill about efficient compression bandaging were suboptimal, regardless of their experience [12, 13, 14, 15]. They could be improved with specific training, although its impact decreased after 3 and 6 months [12, 16]. In order to assess the compression bandaging quality, previous studies have used the Control Score of compression bandaging (CCB score) on a 7‐point scale (7 for the best compression bandaging efficacy) [16] and a pressure monitor allowing nurses to get some feedback on the applied pressure [16, 17]. A multi‐layer bandaging is recommended over short‐stretch bandaging [18, 19]. The objective of the study was to assess the effect and sustainability at 3 months of a training intervention on the compression bandaging quality on a mannequin leg, for two different compression bandage types (system A or system B) [20]. The hypothesis was that the advanced practice nurse's student (APNS) compression bandaging quality was suboptimal and could be improved with a training intervention, regardless of the type of bandage used. The primary outcome was the CCB score. The secondary outcomes were the installation conformity of bandage, the application time and APNS satisfaction.

2. Method

2.1. Study Design

This study involved a 2 × 2 factorial randomised open‐label trial. Intervention was the first factorial and bandage type was the second (Figure 1). Randomization was centralised and stratified on the training year and balanced on the two factors of the factorial design.

FIGURE 1.

Flowchart of the study.

The study targeted all first‐ and second‐year APNSs, in a french university during the 2023–2024 academic year. APNS had worked for at least 3 years as nurse before being able to enrol at university.

2.2. Intervention

The training intervention lasted one hour and was carried out by an expert physiotherapist (CR).

It aimed to teach the principles of venous compression and to adapt compression bandage to the patient's condition and morphology. It included 30 min of theory on venous insufficiency, VLU, types of compression device (short, long, multi‐layer), the pressures to be applied and contraindications, followed by a 20 min practical bandaging demonstration on a mannequin of each compression type and 10 min of discussion with the APNSs. They did not evaluate the quality of the intervention afterwards.

2.3. Compression Bandaging

The system A was URGO K2 kit (Urgo, Chenöve, France), a multitype dual‐band compression system consisting of a KTech short‐stretch bandage and a KPress long‐stretch bandage. The system B was ROSIDAL K bandage (Lohmann‐Rauscher, St Gallen, Switzerland), a short‐stretch bandage requiring an underlay of Cellona orthopaedic padding (Lohman & Rauscher, St Gallen, Switzerland).

2.4. Outcomes

The CCB score [16] was the primary outcome. It included five criteria: installation of the underlay, beginning along the toes, pressure point A between 22 and 27 mmHg, heel included, bandage rolled in the direction of the heart over one point, and pressure point B1 between 50 and 60 mmHg over two points. Point A was located in the space between the big toe and the second toe and point B1 10 cm above the inside of the ankle (Figure 2). The secondary outcomes were installation conformity, based on the four qualitative criteria of the CCB score (Table 1), APNS satisfaction with application and application time. Initial study design defined the primary outcome as pressure at B1 ranging from 37 to 60 mmHg. However, as the evaluation was carried out on a mannequin, the pressure measurement alone cannot be the main outcome, as it cannot be perfectly transposed to a measurement on a patient [13].

FIGURE 2.

Diagram of the positioning of pressure measurement sensors on the mannequin legs. (A) Point B1 position and pressure measurement instrument Picopress, (B) Point A position and mannequin leg characteristics.

TABLE 1.

Control score of compression bandaging and installation conformity criterion.

Control score of compression bandaging/seven points
Installation of underlay a	1
Beginning along the toes a	1
A pressure between 22 and 27 mmHg	1
Heel included a	1
Bandage rolled in direction of the heart a	1
Pressure at B1 between 50 and 60 mmHg	2

Note: CCB score ranges from 0 to 7, higher scores indicate better performance in applying venous compression. The installation conformity corresponds to meeting the four quality criteria of the CCB score.

^a Installation conformity (%): conformity of four criteria.

2.5. Pressure Measurement Equipment

Two 5 cm diameter PICOPRESS probes (Microlab Electronica SAS, Padova, Italia) were used to measure pressures at points A and B1.

2.6. Study Process

Each APNS individually carried out four venous compression evaluations on a mannequin under the supervision of an evaluator, on university premises. The first evaluation involved all advanced practice nurses at inclusion (Figure 1). The pressure value indicated by the pressure measurement equipment was transmitted to the APNS without comment. The training intervention was provided to the intervention group after the first evaluation and to the control group after the second evaluation. Two further evaluations were held after 1 and 3 months for both groups.

Each APNS was equipped with system A or system B with a strip of underlayer padding, depending on their randomization group. They could consult the device instructions if they needed to. At the end of each application, APNS completed a self‐satisfaction questionnaire with compression application, using a four‐modality Likert scale.

2.7. Ethical Considerations

A favourable opinion of the university's ethics committee and written consent of the participant were obtained.

2.8. Statistical Analysis

Analysis was performed on an intention‐to‐treat basis [21, 22]. Continuous outcomes were analysed using linear regression, and categorical outcomes using logistic regression. A first‐order interaction between the training intervention and the type of compression bandage was systematically tested. Missing data were handled using the last observation carried forward (LOCF) method [23]. Only one value was missing in the intervention group at the 3‐month follow‐up. Statistical analyses were performed using Stata SE version 16 (StataCorp, College Station, TX, USA).

3. Results

Out of 58 eligible APNSs, 50 were included in the study, with 25 in the control group and 25 in the intervention group. Within each group, 13 APNSs were in the system A group and 12 in the system B group (Figure 1). The median age of the participants was 40 years, with a mean of 14 years of experience in care. The majority worked in hospitals (52%) or in chronic pathologies (76%). A mere 10% of the APNS cohort had received specific training in venous compression. They were evenly distributed between the control and intervention groups. Groups were homogeneous overall, despite some differences in age and professional experience in certain sub‐branches (data are available on request).

At the time of inclusion, the average CCB score for all APNSs was 3.78/7. Installation conformity was achieved for 52%. Median pressures were 17.4 mmHg at point A and 62.5 mmHg at point B1. Fifty‐six per cent of APNSs were satisfied with the application.

Higher levels of conformity installation and longer application time were observed in the groups who used system A.

3.1. Intervention Effect

After intervention, no significant improvement in the CCB score was observed in the intervention group in comparison with the control group (Table 2). The mean score in the intervention group was found to be 0.44 points lower than the control group (95% confidence interval [CI]: −1.03 to 0.15) before adjustment, and the difference remained non‐significant after adjustment for baseline CCB score and bandage (−0.43; 95% CI: −1.01 to 0.15).

TABLE 2.

Changes in CCB score (primary outcome) from baseline to second evaluation in control and intervention groups, including unadjusted and adjusted differences [95% CI].

Outcome	Group	Baseline mean (SE)	Second evaluation mean (SE)	Mean change (SE)	Difference [95% CI]	p
CCB score/7	Control (n = 25)	3.80 (0.28)	4.52 (0.21)	0.72 (0.28)	(…)	(…)
	Intervention (n = 25)	3.76 (0.23)	4.08 (0.21)	0.32 (0.29)	−0.40 [−1.21, 0.41]	0.324
Unadjusted model	(…)	(…)	(…)	(…)	−0.44 [−1.03, 0.15]	0.142
Adjusted model a	(…)	(…)	(…)	(…)	−0.43 [−1.01, 0.15]	0.139

Note: CCB score ranges from 0 to 7, higher scores indicate better performance in applying venous compression. Positive mean change indicates improvement in CCB score between evaluations.

Abbreviations: CI, confidence interval; SE, standard error.

^a Adjusted model controls for baseline CCB score and bandage type. No significant interaction was observed between bandage and intervention group.

A significant interaction between bandage type and intervention was observed for conformity installation (p = 0.007) (Table 3). In the system B group, intervention exhibited a tendency to enhance conformity (adjusted OR = 5.76 [95% CI: 0.92–35.92]). Conversely, within the system A group, there was a tendency for a reduction in conformity (adjusted OR = 0.14 [95% CI: 0.01–1.41]).

TABLE 3.

Difference in secondary outcome, installation conformity and APNS satisfaction, between study groups with unadjusted and adjusted odds ratios [95% CI].

Secondary outcomes	No. students	n	(%)	Unadjusted OR (95% CI)		Adjusted OR (95% CI)
Installation conformity a , b
System A
Control	13	12	(92.3)	1.00	(…)	1.00	(…)
Intervention	13	8	(61.5)	0.13	(0.01–1.36)	0.14	(0.01–1.41)
System B
Control	12	3	(25.0)	1.00	(…)	1.00	(…)
Intervention	12	8	(66.7)	6.00	(1.02–35.37)	5.76	(0.92–35.92)
APNS satisfaction c
Baseline
Control	25	13	(52.0)	1.00	(…)	1.00	(…)
Intervention	25	15	(60.0)	1.38	(0.45–4.25)	1.46	(0.39–5.39)
Second evaluation
Control	25	17	(68.0)	1.00	(…)	1.00	(…)
Intervention	25	16	(64.0)	0.84	(0.26–2.70)	0.80	(0.21–3.11)
30‐days
Control	25	14	(56.0)	1.00	(…)	1.00	(…)
Intervention	25	14	(56.0)	1.00	(0.33–3.06)	1.00	(0.27–3.67)
90‐days d
Control	25	17	(68.0)	1.00	(…)	1.00	(…)
Intervention	25	16	(64.0)	0.84	(0.26–2.70)	0.80	(0.21–3.11)

Note: Installation conformity refers to conformity of four criteria of the CCB score: installation of underlay, beginning along the toes, heel included, bandage rolled in direction of the heart. APNS satisfaction measured at baseline, second evaluation, 30‐day and 90‐day follow‐up.

Abbreviations: CI, confidence interval; OR, odds ratio.

^a Installation conformity was stratified by bandage (p interaction = 0.007).

^b OR adjusted for baseline CCB installation conformity. OR > 1 indicates higher odds of correct installation in the intervention group compared with control.

^c ORs of satisfaction were estimated using random‐intercept logistic regression with fixed effect of time (four levels), study group, and a first‐order interaction term entered as independent variables for 200 observations nested within 50 students. There was no significant interaction between time and intervention group. OR > 1 indicates higher odds of satisfaction in the intervention group.

^d Missing value was replaced by LOCF imputation (n = 1 in intervention group at 90‐days).

3.2. Evolution Over 1 and 3 Months

After 1 month, the CCB score demonstrated an improvement in all groups, with the exception of the system A control group (data are available on request). Three months later, the CCB score demonstrated stability in both the system A groups. A decline was observed in system B control and intervention groups, with a decrease of 0.23 and 0.25 points out of 7, respectively.

The post‐intervention levels of APNS satisfaction remained comparable across the groups, both at the 1‐month and 3‐month follow‐ups. The application time decreased at 1 and 3 months for all groups. The utilization of system B exhibited a 1.7‐fold increase in efficiency compared to the application of system A.

4. Discussion

Learning and performing the application of multi‐bandage compression is essential in chronic venous leg ulcers. Despite advances in the ease of applying multi‐layer compression, tutorials and QR codes, it has been shown that the quality of application by healthcare professionals is suboptimal [15, 24].

In line with the initial hypothesis, APNSs in our study performed poorly in compression application, installation conformity, and pressure levels. Despite possessing over 14 years of professional experience, few APNSs had received specific training in venous compression, and they did not routinely engage in this practice. Nevertheless, the quality of compression application is contingent upon continuous education [16] and practical experience in compression techniques [2, 9, 25, 26]. Theoretical knowledge alone does not guarantee proficiency in application techniques or effective pressure delivery, resulting in suboptimal clinical practice [14].

4.1. Intervention Impact

The educational intervention did not yield a significant effect on the quality of venous compression installation, contrary to previous findings [13, 14, 16, 17, 24]. Although the mean of CCB score was higher than in Protz's study [16], the improvement was less pronounced, particularly in the intervention group. This result is surprising and disappointing compared to the educational efforts implemented. The training intervention, based on the behaviourist transmission model, aimed to enhance psychomotor skills by enabling APNSs to conceptualise and shape cognitive changes related to compression bandage installation [27, 28, 29]. However, the lack of clinical practice opportunities may have led to the loss of these cognitive changes. Repetition of the second evaluation, independent of training, triggered gestural learning and reactivated appropriate bandaging behaviours among experienced control group participants. Immediate feedback on pressure values at points A and B1 reinforced installation effectiveness, potentially leading to modifications [17, 30]. Providing normative feedback could have been more effective. Gestural repetition enhances performance [27, 28, 30], with studies indicating that nine repetitions are necessary for intact retention after 3 months of non‐practice. However, APNSs in this study did not have the opportunity for multiple practice sessions, as seen in studies by Tidhar, Keller, and Taylor [12, 17, 31]. Collaborative learning environments promote the conceptualization of new ideas, the observation of diverse clinical behaviours, and the critical reflection necessary to challenge pre‐existing beliefs in favour of behavioural change. Conducting compression bandaging in isolation may have constrained learners' ability to refine their practice, due to the absence of peer modelling and shared experiential learning.

4.2. Motivation and Perception of Skill

Training is effective when participants perceive it as meaningful [32] and when it equips them with relevant skills that can be applied in clinical practice. Advanced practice nurses work in consultative roles, with few engaged in vascular medicine or regularly applying compression bandages. The intervention was insufficiently integrated with key aspects of advanced practice nursing, such as medical device prescribing [2], team support, and clinical leadership [33], which may have contributed to its limited impact. Perhaps the main pitfall of the study is that it was aimed at a general population of nurse practitioners and not at a selected public motivated by wounds. Most post‐graduate training courses are free to choose, and there is a motivational bias towards training nurses who take the step of enrolling in a training course dedicated to wound care. This was not applicable in our study. According to Bandura [34], an individual's sense of self‐efficacy determines the goals set and the effort exerted. However, the intervention did not significantly impact APNS' self‐reported satisfaction, suggesting that perceived efficacy may not align with actual performance.

4.3. Comparison Between Bandage Types

The intervention's impact was contingent upon the type of bandage utilised. Several reasons could explain it. At baseline, the majority of APNSs in the system A control group demonstrated conformity with installation protocols. In contrast to the system B, the system A, which includes two bands and instructions in a single package, facilitates adherence to recommended practices. The observed higher B1 pressures in the system A groups can be attributed to the inherent differences between the two systems; specifically, the greater rigidity of the multilayer bandage due to the successive overlapping of layers [8, 13, 35]. In France, the system A is predominantly employed for the management of open‐stage venous leg ulcers. Consequently, more experienced APNSs in the system A group likely had greater practical familiarity with this kit [10].

4.4. Sustainability of the Training Intervention

Although the training intervention had no immediate significant effect overall, slight improvements were observed at 1 month, particularly in the system B group. However, these gains were not sustained in all groups after 3 months, suggesting that a single brief training session may not be sufficient to ensure long‐term retention of compression bandaging skills [36].

4.5. Application for Clinical Practice

Application time was longer for system A due to the multi‐layer system. The increase in application time after the procedure could reflect APNSs' commitment to properly apply and the effect of repetition. They mobilise the theoretical elements of the intervention in action and are attentive to their gesture. As the application time decreases after 1 and 3 months, we can assume that APNSs have integrated the criteria of conformity. In current practice, even though multi‐layer bandages are recommended as the first‐line treatment for VLUs [9], the time required to apply them could be an obstacle to their use by home‐care nurses.

4.6. Strengths and Limitations

This study has several strengths. It was based on an original 2 × 2 factorial randomised trial design. In the same study, this design made it possible to assess the effect of the brief training intervention on the quality of compression installation and to compare the effect of the intervention according to the type of bandage [20]. We were able to evaluate directly the quality of the application technique and the achievement of sufficient pressure, which is a guarantee of healing in VLUs and not just the increase in theoretical knowledge, which may not be reflected in practice.

However, we cannot exclude some biases. The first one concerns the rigidity of the mannequin. The pressure values measured may have been overestimated compared with those for human subjects, whose tissues have different elasticity and compressibility. In practice, if the pressure was too high, the patient would alert the nurse, who would reapply the bandage [37]. To strengthen the external validity of the study, the expert (CR) applied each compression bandage to the mannequin three times in order to verify that the pressure values at points A and B1 matched the expected levels. The resulting pressures, which ranged from 37 to 60 mmHg, were consistent with international and HAS recommendations. However, the variability observed may still limit the external validity of the study and the transferability of these results to real clinical conditions. Secondly, we cannot exclude contamination bias, common in randomised trials in education [38]. APNSs could have had the opportunity to exchange information with one another during the study on bandage application techniques. This bias may have led to a lack of statistical power to detect a significant effect of the intervention [38]. Thirdly, the halo effect may have influenced participants between the two evaluations, as some may have attributed expertise to peers with compression‐specific training or experience, potentially relying on their practices and creating a power dynamic [39, 40]. This could explain the differing impacts on installation conformity between the system B and system A groups. Fourthly, the Hawthorne effect may have altered results, as participants' behaviour might have changed due to awareness of being observed and a desire to perform well, independent of the experimental manipulation [41, 42]. However, the extent and direction of this effect are difficult to quantify and may have affected all trial arms. Fifth, the use of the last observation carried forward (LOCF) method to handle missing data may introduce bias by overestimating the intervention effect, particularly when data are not missing at random [23]. However, only one value was missing in the intervention group at the 3‐month assessment, and no significant effect was observed. Therefore, the potential impact of this method on the study results is likely minimal. Sixth, the students' prior experience and familiarity with one of compression systems may have contributed to the lack of significance of the results. Due to the small sample size per factorial arm, further studies are needed to generalise the results.

5. Conclusion

APNSs with 3 years' experience nurses had a sub‐optimal level of skill in applying venous compression. This study in a population of health professionals not selected by their motivation nevertheless seems to be a reflection of competence in terms of applying compression. Under the conditions of this brief intervention, behaviourist‐based training did not appear to produce a statistically significant improvement in the CCB score. These findings highlight that single, short sessions are insufficient for developing complex psychomotor skills. Effective continuing education should be longitudinal on at least 6 months, including repeated practice, structured feedback on performance, and clearly defined learning objectives. Further research, ideally via randomised controlled trials, is needed to assess alternative training interventions.

Funding

This work was supported by the Université Grenoble Alpes.

Ethics Statement

On the 4th of September 2023, The Grenoble Alpes Research Ethics Committee (CERGA) delivered a favourable opinion on this study, reference «CERGA‐Avis‐2023‐22».

Conflicts of Interest

Blaise Sophie reports a relationship with Urgo laboratories SL that includes: consulting or advisory, speaking and lecture fees, and travel reimbursement, a relationship with ConvaTec France that includes: speaking and lecture fees and a relationship with Sigvaris SAS that includes: consulting or advisory.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.