Chronic wounds are wounds that fail to heal within the expected timeframe, respond poorly to treatment, and frequently recur [1]. They represent a major clinical challenge because prolonged healing may lead to functional impairment, reduced quality of life, and increased healthcare burden. In the United States alone, chronic wounds affect more than six million individuals and are associated with annual healthcare costs exceeding USD 9.7 billion [2].

Accurate wound assessment is essential for monitoring the progression of healing, guiding treatment decisions, and evaluating treatment response. However, wound healing is a complex biological process that involves overlapping phases of inflammation, proliferation, and remodeling [3]. Because wound characteristics may change dynamically during healing, standardized wound assessment tools are needed to facilitate consistent evaluation and communication among healthcare providers. Nevertheless, many existing scoring systems remain relatively complex and are not routinely applied in daily clinical practice [3].

Two scoring systems commonly used in chronic wound assessment are the Falanga wound bed score and the Perdanakusuma II score. The Falanga score, introduced in the United States in 2006, evaluates multiple wound bed parameters related to wound bed preparation and healing potential and has been widely recognized in wound assessment practice [4]. However, its application may be relatively complex because it requires assessment of multiple clinical parameters and familiarity with structured scoring interpretation. In contrast, the Perdanakusuma II score was developed as a simplified wound assessment tool based on wound color, inflammation, and exudate to facilitate rapid and practical clinical evaluation [5]. This scoring system has demonstrated high reproducibility across various healthcare practitioners, with reported Cronbach alpha and intraclass correlation coefficient (ICC2) values of 0.953 and 0.952, respectively [5].

Although several other wound assessment systems have been proposed, including the Pressure Ulcer Scale for Healing (PUSH), Bates-Jensen Wound Assessment Tool (BWAT), and Tissue-Infection/inflammation-Moisture imbalance-Edge of wound (TIME) acronym, each has important limitations in routine clinical application. PUSH is primarily intended for pressure ulcers, BWAT may be relatively time-consuming because of its extensive parameters, and TIME functions mainly as a clinical framework rather than a quantitative scoring system [6-8]. In resource-limited and high-volume clinical settings, particularly in developing countries, wound assessment tools that are simpler and easier to interpret may offer advantages for routine clinical use and surgical training. However, comparative evidence regarding the practical performance of simplified and conventional wound scoring systems in structured surgical education settings remains limited. This study aimed to compare the classification accuracy and agreement with reference standards of the Perdanakusuma II and Falanga scoring systems for chronic wound assessment within a structured training setting among surgical residents.

A total of 41 surgical residents were included, with a mean age of 29.44±2.92 years (median, 29; range, 26–37) (Table 1). Most participants were male (63.4%). Residents were drawn from multiple surgical specialties, with the largest proportions from oral and maxillofacial surgery and general surgery (each 17.1%), followed by cardiothoracic and vascular surgery (14.6%) and plastic, reconstructive and aesthetic surgery (12.2%). Across both pretest and posttest assessments, the Perdanakusuma II score consistently demonstrated higher classification accuracy than the Falanga score. Following structured training, accuracy improved in both systems, with the Perdanakusuma II score maintaining superior performance. The mean accuracy of the Falanga score increased from 41.71%±12.48% pretest to 54.02%±10.62% posttest, while the Perdanakusuma II score improved from 56.59%±13.62% to 67.44%±13.56% (Table 2).

Table 3 presents the distribution of wound category classifications using the Falanga scoring system before and after training. Pretest assessments were predominantly skewed toward the severe category, indicating a tendency toward overestimation of wound severity. The highest pretest accuracy was observed for image #2 (36/41, 87.8%), followed by image #13 (35/41, 85.4%), whereas image #7 showed no correct responses (0/41, 0%). Following training, classification performance improved, with a redistribution toward more appropriate categories. The highest posttest accuracy was observed for image #6 (39/41, 95.1%), while the lowest was for image #16 (3/41, 7.3%). Overall, these findings indicate improved accuracy and better discrimination of wound severity after training (Table 3).

Table 4 summarizes the distribution of wound category classifications using the Perdanakusuma II score before and after training. During the pretest, the highest accuracy was observed for image #4, with all participants correctly classifying the wound (41/41, 100%), followed by image #2 (40/41, 97.6%). In contrast, the lowest accuracy was found in images #7 and #8, each with only 2 correct responses (2/41, 4.9%). Following training, the highest accuracy was observed for images #6, #10, and #13, each achieving 39/41 correct classifications (95.1%). The lowest posttest accuracy was observed for image #8 (6/41, 14.6%). Overall, these findings indicate an improvement in classification accuracy after training (Table 4).

For comparison, using the Falanga score, 14 images demonstrated an accuracy of <60% in the pretest, which decreased to 10 images in the posttest. Notably, no images achieved an accuracy of >90% during the pretest, whereas three images reached this threshold following training (Table 3). In contrast, with the Perdanakusuma II score, 10 images had an accuracy of <60% in the pretest, which decreased to eight images in the posttest. Furthermore, the number of images achieving >90% accuracy increased from four in the pretest to six in the posttest (Table 4). These findings indicate a more favorable distribution of accuracy using the Perdanakusuma II score compared with the Falanga score.

Table 5 compares the classification accuracy between the Falanga and Perdanakusuma II scoring systems based on posttest results. Overall, the Perdanakusuma II score demonstrated higher accuracy across several images. Statistically significant differences were observed in images #1 (P=0.024), #3 (P=0.001), #9 (P=0.002), #14 (P<0.001), #16 (P=0.012), and #20 (P=0.002), all favoring the Perdanakusuma II score. For the remaining images, no statistically significant differences were found between the two scoring systems. These findings suggest that the Perdanakusuma II score may provide superior classification accuracy in selected wound scenarios compared with the Falanga score.

Agreement analysis demonstrated consistently higher ICC values for the Perdanakusuma II score compared with the Falanga score, both before and after training (Table 6). Before training, the Perdanakusuma II score demonstrated good agreement with the reference standard (ICC2=0.73; 95% confidence interval [CI], 0.60–0.85), whereas the Falanga score showed lower agreement (ICC2=0.39; 95% CI, 0.27–0.59). Following training, agreement improved in both systems, with ICC2 increasing to 0.62 (95% CI, 0.48–0.78) for the Falanga score and 0.76 (95% CI, 0.65–0.87) for the Perdanakusuma II score.

MAE analysis further demonstrated lower scoring deviation from the reference standard in the Perdanakusuma II score compared with the Falanga score (Table 7). Before training, the MAE was 3.04 for the Falanga score and 1.74 for the Perdanakusuma II score. Following training, the MAE decreased in both systems, reaching 2.10 and 1.42, respectively, indicating improved scoring accuracy and consistently lower deviation in the Perdanakusuma II system.

Fig. 4 demonstrates a significant increase in scores from pretest to posttest for both the Falanga and Perdanakusuma II scoring systems (Wilcoxon signed-rank test, P<0.001 for both). In addition, the Perdanakusuma II scores were significantly higher than the Falanga scores in both the pretest and posttest assessments (Wilcoxon signed-rank test, P<0.001 for both comparisons), indicating consistently superior performance of the Perdanakusuma II system (Fig. 4).

The findings of this study demonstrate that the Perdanakusuma II scoring system showed consistently higher classification accuracy and stronger agreement with the expert reference standard, as well as lower scoring deviation compared with the Falanga score within a structured training setting among surgical residents. Although both scoring systems demonstrated improvement following structured training, the Perdanakusuma II score maintained superior performance across both pretest and posttest assessments. These findings suggest that simpler and more visually interpretable scoring frameworks may facilitate more consistent application during early-stage surgical training, particularly when cognitive processing demands are reduced during clinical decision-making [11].

The Falanga scoring system, which has been widely validated and demonstrates predictive value for wound healing outcomes, offers a comprehensive evaluation through eight parameters, with each incremental score increase associated with a 22.8% higher probability of wound healing [4]. This level of detail enhances its clinical robustness and prognostic value; however, it simultaneously introduces a higher cognitive burden and requires greater clinical experience to apply consistently. In the context of surgical trainees, particularly those in early stages of training, this complexity may limit immediate usability, as also reflected in the persistent variability of post-training accuracy across certain wound cases. These findings are consistent with simulation-based training literature, where complex clinical tools require repeated exposure and structured learning to achieve mastery [12].

In contrast, the Perdanakusuma II scoring system adopts a simplified framework based on three core parameters: wound color, inflammation, and exudate. These parameters are fundamental, visually intuitive, and routinely taught from early medical education, allowing faster pattern recognition and more consistent interpretation, consistent with experimental evidence highlighting the importance of early-phase wound healing dynamics in determining overall repair outcomes [13]. This likely explains the higher and more stable performance observed, as well as its superiority across both baseline and post-training assessments.

This pattern was further supported by agreement analysis, in which the Perdanakusuma II score demonstrated consistently higher ICC values than the Falanga score across both pretest and posttest assessments. Although structured training improved agreement in both systems, the Perdanakusuma II score maintained superior consistency, suggesting lower interpretative complexity and greater reproducibility among surgical residents. This finding was further supported by MAE analysis, which demonstrated consistently lower scoring deviation from the reference standard in the Perdanakusuma II score compared with the Falanga score across both pre- and posttest assessments.

The results align with previous validation studies demonstrating that the Perdanakusuma II score achieves more balanced diagnostic performance and stable area under the curve values over time compared to the Falanga score [5]. From an educational standpoint, this supports cognitive load theory, where simplified information processing enhances learning efficiency, retention, and application in clinical settings [14].

Beyond structural simplicity, the difference between the two scoring systems also lies in their conceptual orientation. The Perdanakusuma II score more directly reflects wound healing progression through clinically intuitive and visually identifiable parameters, whereas the Falanga score places greater emphasis on wound bed characterization and readiness for closure without explicitly defining a healed state. This distinction may influence how trainees interpret clinical findings and make decisions, particularly during the early phases of training, when clear and visually salient cues are important for developing accurate mental representations and supporting clinical judgment through observational learning [15,16]. In addition, simpler and more visually interpretable scoring frameworks may be more readily integrated into routine clinical workflows, whereas more complex systems often require structured training and repeated exposure to ensure consistent interpretation [17]. In this context, the Perdanakusuma II score appears more adaptable to routine practice, whereas the Falanga score is more dependent on structured documentation and training to achieve reliable interpretation across users.

Although structured training improved agreement in both systems, the Perdanakusuma II score demonstrated higher baseline agreement and maintained superior consistency following training, suggesting lower cognitive complexity and greater interpretability. These findings are noteworthy because variability in chronic wound interpretation has also been reported even among experienced wound care experts. Greco et al. [18] demonstrated only fair-to-slight agreement among international wound specialists when applying commonly used wound description terminology and Falanga appearance classifications, highlighting the inherent subjectivity of chronic wound assessment. Interestingly, although previous validation studies of the Perdanakusuma II score reported excellent internal consistency and inter-rater agreement, with Cronbach alpha and ICC values exceeding 0.95 [5], the agreement values observed in the present study were comparatively lower. This difference may reflect the inclusion of PGY-1 surgical residents from multiple surgical specialties who were still in the early phase of training and had heterogeneous prior exposure to wound assessment. Nevertheless, despite involving relatively inexperienced participants, the Perdanakusuma II score still demonstrated higher agreement and lower scoring deviation than the Falanga system following structured training.

These differences have important clinical implications. Accurate wound assessment is essential for guiding appropriate management, as misclassification may lead to inappropriate treatment, delayed healing, and increased healthcare costs [19]. Early and accurate identification of wound progression is also a key prognostic factor, with studies showing that a 40% reduction in wound area within 4 weeks predicts favorable healing outcomes [20]. In this context, a scoring system that is not only valid but also consistently applicable across users becomes highly valuable. The higher accuracy and consistency of the Perdanakusuma II score suggest its potential to reduce diagnostic variability and improve clinical decision-making.

Standardization of wound assessment further enhances communication and coordination among healthcare professionals. In multidisciplinary wound care settings, the use of a consistent and easily interpretable scoring system facilitates shared understanding, improves documentation quality, and supports evidence-based management [21]. Effective communication frameworks are essential in both clinical and public health contexts, influencing not only patient care but also broader healthcare outcomes and system efficiency [22].

From an educational perspective, these findings reinforce the importance of aligning teaching tools with learner readiness. Surgical residents from diverse subspecialties, typically during the early stages of professional training, must master substantial clinical knowledge within time-limited educational environments, making learning efficiency essential [23]. Spaced repetition and other cognitively optimized strategies have been shown to improve long-term retention and support efficient knowledge acquisition, although their adoption in surgical training remains inconsistent due to lack of standardization and integration into curricula [24]. In this context, simpler and more intuitive tools such as the Perdanakusuma II score may better support early learning by reducing cognitive load and facilitating consistent application, whereas more complex systems such as Falanga may require greater repetition and structured training before achieving reliable use.

Taken together, the comparison between the two scoring systems highlights that performance of clinical assessment tools is not determined solely by comprehensiveness or validation status, but also by usability, cognitive demand, and contextual fit within training and clinical environments. As summarized in Table 8, the Perdanakusuma II score offers advantages in simplicity, interpretability, and educational applicability, whereas the Falanga score provides greater detail and established prognostic validity. A complementary approach that integrates both systems according to clinical and educational needs may therefore represent the most optimal strategy for improving wound assessment practice.

This study has several strengths. First, the study incorporated both classification accuracy and agreement analysis using ICC and MAE, allowing more comprehensive evaluation of scoring performance relative to an expert reference standard. Second, the use of standardized wound images evaluated through expert consensus improved consistency of the assessment dataset. Third, the study involved surgical residents from multiple surgical specialties within a tertiary academic center, reflecting the multidisciplinary nature of wound management in routine surgical training. In addition, the timed assessment format using 1-minute image evaluation intervals may better reflect real-world clinical conditions, where wound assessment often needs to be performed efficiently in high-volume practice settings [25]. Despite this time-constrained setting, both scoring systems, particularly the Perdanakusuma II score, demonstrated acceptable classification performance and agreement with the reference standard.

Nevertheless, several limitations should be acknowledged. The study used a single-center quasi-experimental design with a relatively limited sample size, which may affect generalizability. In addition, wound assessment was performed using static clinical images rather than direct bedside examination, which may not fully represent real-world clinical evaluation. The study also evaluated short-term performance following structured training and did not assess long-term retention of scoring competency. Furthermore, although agreement analysis was performed, the present study was not designed as a formal psychometric validation study.

Future studies should evaluate the performance of the Perdanakusuma II score in broader clinical settings involving nurses, general practitioners, medical students, and non-specialist healthcare providers. Prospective multicenter studies with direct bedside wound assessment and longitudinal follow-up would also be valuable to evaluate long-term educational retention and clinical applicability. Given the increasing burden of chronic wounds, simpler and more interpretable assessment tools may contribute to improved standardization of wound evaluation, particularly in resource-limited healthcare settings and multidisciplinary clinical environments.

In conclusion, both the Perdanakusuma II and Falanga scoring systems demonstrated improved classification performance following structured training. However, the Perdanakusuma II score showed consistently higher agreement, accuracy, and ease of application, making it particularly suitable for early-stage surgical training and routine clinical use. Structured training significantly improved participants’ cognitive competence in wound evaluation, highlighting the importance of systematic educational approaches. The integration of the Perdanakusuma II score into clinical training and practice may enhance healthcare provider competency and support the standardization and quality of chronic wound management in Indonesia.