Reliability and Validity of the Chinese Version of DESIGN-R, an Assessment Instrument for Pressure Ulcers
Lack of a valid and reliable assessment instrument may have contributed to a lack of wound healing research in China. The DESIGN-R tool scores the severity of pressure ulcers (PUs) and monitors healing using the observable state assessment; it was developed by the Scientific Education Committee of the Japanese Society of Pressure Ulcers (JSPU) and tested for reliability and validity.
A Chinese version of DESIGN-R was developed based on Brislin’s model of translation and tested for validity and reliability. Using a purposive sampling method, 44 practicing registered nurses (RNs) and 11 physicians (MDs) were recruited from 52 departments of Nanfang Hospital, Guangdong, China. Based on their experience, they were classified as general medical staff (gMS) or experienced medical staff (eMS). All used the Chinese version of DESIGN-R to assess eight photographs of PUs and descriptors. In addition, eight eMS also used the Bates-Jensen Wound Assessment Tool (BWAT) to assess the same wounds. Inter-rater reliability was high (total ICC score = 0.960). ICC inflammation/infection scores were 0.530 and 0.759 for gMS and eMS, respectively; granulation ICC scores were 0.532 and 0.794 in gMS and eMS, respectively. The correlation coefficients between the BWAT and DESIGN-R tool were >0.80 for all eight raters. The results suggest the Chinese version of DESIGN-R is valid and reliable and may be a useful scoring tool for RNs and MDs to monitor PU status in daily clinical practice. Additional research is warranted, and clinical instruments for inflammation/infection and granulation assessment must be developed for gMS.
Potential Conflicts of Interest: Part of this work was presented at the 17th Annual Scientific Meeting of Japan Academy of Gerontological Nursing, Kanazawa, Japan, July 14, 2012.
Pressure ulcers (PUs) are a serious health problem; they are unpleasant, upsetting, challenging to treat, and increase mortality and patient length of stay.1,2 Therefore, healthcare professionals use a range of techniques to prevent PUs.3 Despite preventive efforts, PU prevalence has been estimated to be 0.4% to 28% in the United States,1 9% to 18.1% in Europe,2,3 and 4.18% and 0.44% to 45.5% in Japan4 and Korea,5 respectively, depending on the facility and type of patient. Although no nationwide statistics on the incidence of PUs in China are available, given the size of China’s population, the number of patients with PUs can be predicted to be very large.
For effective and efficient PU management, it is essential to measure and describe wound healing progress precisely. The basic foundation of an assessment instrument is to document various aspects of PU appearance and change in wound status over time. A PU assessment instrument should provide objective information to facilitate appropriate wound assessment and management by the wound care clinician and the interdisciplinary team. A variety of PU-related measurement tools for healing are available and include the Pressure Ulcer Scale for Healing (PUSH Tool)6 developed by the National Pressure Ulcer Advisory Panel (NPUAP), the Bates-Jensen Wound Assessment Tool (BWAT),7,8 DESIGN-R9-11 developed by the Scientific Education Committee of the Japanese Society of Pressure Ulcers (JSPU), and others.
An ideal assessment instrument needs to be inexpensive, time-saving, and simple enough to be used regularly in clinical settings by different healthcare professionals. Moreover, the instrument must be simple to learn, easy to use, sensitive to change in the wound healing process, and safe for patients. The BWAT12 contains 13 items that assess wound size, depth, edges, undermining, necrotic tissue type, necrotic tissue amount, granulation and epithelialization tissue, exudate type and amount, surrounding skin color, edema, and induration, providing a detailed clinical status. However, widespread adoption of this instrument has been limited by the large number of items that, according to a descriptive study and a critical review,9 require 3.8 to 15 minutes to complete and by the user’s need for professional training in wound care.9 Thus, it has been found to be more appropriate as a research than as a clinical tool.13 At the same time, the PUSH Tool is easy to use; it has been weighted based on principal component analysis using clinical data in a retrospective study.14 However, because fewer ulcer characteristics are assessed and Stage I and Stage II PUs are excluded, two retrospective studies14,15 involving the assessment of the validity of the PUSH Tool to monitor healing of PUs have shown the scoring points of the PUSH Tool demonstrate limited variation (only 58% to 74% of the variation across 10 weeks) in all data. The PUSH tool is neither a research tool nor a tool to guide treatment decisions; rather, it serves as a bedside tool to evaluate wound progress.6,14
DESIGN-R (see Figure 1) was developed as a PU healing assessment tool by the Scientific Education Committee of the Japanese Society of Pressure Ulcers (JSPU) and contains six items: wound depth, exudate, size, infection, granulation, and necrosis (DESIGN). P is added to the acronym when a pocket (undermining) is present. It is widely used in Japan and other Asian countries.11 Each section of DESIGN-R is scored according to observable state. The score of the tool ranges from 0 (healed) to 66 (greatest severity). The reliability and validity of DESIGN-R have been reported9 where the values of the items have been weighted to reflect PU healing rates by providing the grade indicating the most severe status with higher points. Multivariable Cox’s regression models were used to assess the weighting of each item for PU healing in this research. The DESIGN-R tool can classify PU severity, quantify healing patterns as an aid in determining the best intervention for each item, and monitor detailed changes in wound status. With this information, the DESIGN-R tool provides the potential to guide treatment selection.9 Also, a descriptive study9 evaluating validity and reliability of DESIGN has shown that it has a shorter completion time (average, only 2.6 minutes versus 3.4 minutes using the PSST), important to busy clinicians who use it regularly in clinical settings.
In China, two prospective studies16,17 regarding the application of PUSH in wound management were conducted; sample size was limited (N=12 and 48, respectively) due to the lack of a precise assessment tool to follow the healing process. The inter-rater reliability and concurrent criterion validity of PUSH have not been tested yet in the Chinese population. Moreover, according to a literature review,18 in China most PU healing process assessments are performed without an assessment tool. For some serious cases, PU status is recorded using photographs. No literature has reported the popularity of scoring tools such as BWAT, PUSH, and DESIGN-R in China. To help improve the quality of wound management in China, researchers developed a Chinese version of DESIGN-R and introduced it into China.
Reliability and validity should be tested when a new assessment tool is introduced. The authors’ clinical experience suggested that PU assessment was a complex process requiring substantial visual and physical assessment skill, combined with clinical judgment and experience.10 These factors may influence the inter-rater reliability of a PU assessment tool. For instance, a prospective outcomes management study19 reported the ability of the average, nonexpert clinician to correctly stage PUs is poor (23% to 58% of answers were correct). In the authors’ previous study, inter-rater reliability of the DESIGN varied among nurses with different experience in wound care.9 It has remained unclear whether inter-rater reliability would be affected by different professional levels (eg, physicians versus nurses). Furthermore, the introduction of a worthwhile instrument into China would involve training on correct use. It is also necessary to provide a different but well-established instrument as a comparative reference to assess the validity of the new tool. Therefore, the purpose of the study was two-fold: 1) to determine whether differences in the reliability of the Chinese version of DESIGN-R would be found among doctors and nurses with different wound care experience; and 2) to determine whether description of PU status would differ between the Chinese version of DESIGN-R and the BWAT. Accordingly, in this study, the validity and the reliability of the Chinese version of DESIGN-R among different Chinese professional medical staffs were examined.
Methods and Procedures
Translation. The authors translated the English version of DESIGN-R10,11 into Chinese based on Brislin’s model for translation.20 Three translations from English to Chinese were performed by three different and independent translators. The first Chinese version of DESIGN-R was synthesized from the three translations and then translated back to English by three other translators who were blind to the published English version. All translators whose native language was Chinese had a good command of English and had adequate knowledge of wound care. The second Chinese version was amended by comparing the published English version and the back-translations. Remaining discrepancies were mediated by a bilingual (Japanese and Chinese) wound care specialist who reviewed the Chinese version and the original Japanese version.
Participants. To test the reliability and validity of the Chinese version of DESIGN-R, an observational study was conducted in February 2012 in Nanfang Hospital, a 2,000-bed, comprehensive hospital in Guangdong, China. One hundred (100) registered nurses (RNs) and medical doctors (MDs) were asked to participate in the testing by purposive sampling. Based on the numbers of RNs in each department, one or two RNs were selected from 52 different departments for a total of 87, from which 44 RNs were recruited in final analysis. Six RNs were members of the PU management team who had at least 2 years’ experience in wound care and had received professional training in wound care; this group was expected to test the ability of experienced wound care RNs in using DESIGN-R. The remaining RNs were expected to test the ability of general RNs to use DESIGN-R. Ten residents from different departments who had treated patients with PUs were selected as general MDs, and two plastic surgeons and one physician who had received professional training were selected as experienced wound care MDs. According to their clinical experience and professional training in wound care management, all participants were separated into experienced medical staff (eMS) and general medical staff (gMS).
The first author provided a 30-minute explanation of DESIGN-R tool use to the participants. Participants were asked about their professional certifications as well as years of experience in clinical settings and in wound care, the average number of wounds he/she took care of every month in the past 1 year, whether he/she received professional training in wound care, and other information using a demographic form.
Reliability and validity testing. Photos frequently have been used in dermatological assessment tool research.11,21 For this study, eight photographs with brief case descriptions were used. To assess amount of wound exudate, additional photographs of the dressings and information on dressing materials and changing times was provided. A rectilinear scale next to the wound was included in the photograph and, where undermining had occurred, a dotted line was added to indicate extent of undermining.
Inter-rater reliability. In the inter-rater reliability test, participants assessed the PUs on their own. All participants were asked to complete one DESIGN-R answer sheet for each case scenario. Participants also were asked to consider the degree of difficulty in using DESIGN-R to assess wound healing using the terms difficult, moderate degree of difficulty, and easy. The answer sheets were collected by the principal researcher after the participants had finished the assessment and were identified by serial number. Data from participants with incomplete answers in the DESIGN-R answer sheets were excluded. Two RNs and one plastic surgeon in the experienced medical staff (eMS) participated in the test-retest reliability test 1 month later using the same eight photographs.
Validity. A criterion validity test was conducted. The BWAT (English version) was used as a standard against which to measure the validity of the Chinese version of DESIGN-R, because the reliability and validity of the BWAT have been reported by a previous study.8 Because experienced wound specialists are required for PU evaluation using the BWAT,22 eight bilingual raters (Chinese and English) among the eMS, including the six trained RNs and the two plastic surgeons, who were the same participants from the reliability raters, were recruited for the validity test. They were provided a 40-minute explanation about how to use these assessment tools, and all participants evaluated the same eight photographs using the BWAT. Data from participants with incomplete answers in the BWAT answer sheets were excluded.
Statistical analyses. Data were entered into Microsoft Excel for Windows 2010 and rechecked by two research nurses. All statistical analyses were performed using SPSS 20.0 for Windows (IBM). The Mann-Whitney test was used to compare demographic characteristics between the two groups of raters (eMS and gMS, MDs and RNs). The intra-class correlation coefficient (ICC) was calculated for the inter-rater reliability test and the test-retest reliability test in the different groups of medical staff members. Spearman’s rank related coefficient was calculated in the criterion validity test of the Chinese version of DESIGN-R against the BWAT. Percentage was used to describe “ease of use.” The level of significance was set at P <0.05 for all the analyses.
Ethical considerations. Written permission was given by the developers of the tool (JSPU) in November 2011 to use DESIGN-R in this academic study. Study approval was obtained from Nanfang Hospital. The first author gave an instructional session to the participants regarding ethical consideration of this research, and participants were informed they could withdraw from the research at any time by a letter of informed consent. Returning the completed answer sheets indicated consent to participate.
Participants. Fifty-five (55) medical staff members recruited for this research participated: 44 RNs and 11 MDs (see Table 1). Among the 44 RNs, six were experienced wound care RNs with professional training in wound care; the other 38 RNs were general nurses without professional training in wound care. The 11 MDs included two plastic surgeons, one physician, and eight residents. Based on their clinical experience in wound care management, for the purpose of this study, the two plastic surgeons, one physician (from the endocrine department), and six trained RNs were considered as eMS, and the other participants were regarded as gMS. One of the six trained RNs contributed to the translation process from English to Chinese as described above. None of the raters had any experience in using wound healing assessment tools.
With respect to years of work experience, differences were significant between the gMS (median = 4.0, interquartile range = 1.0–10.0) and eMS (median = 17.0, interquartile range =12.5–20.0) (P = 0.001) and between the RNs (median = 7.5, interquartile range = 3.0–16.5) and MDs (median = 1.0, interquartile range = 1.0–10.0) (P = 0.015). Regarding years of wound care experience, the eMS (median = 8.0, interquartile range = 2.5–10.0) had significantly more than the gMS group (median = 2.0, interquartile range = 1.0–6.5) (P = 0.019), but the difference was not significant between the RNs and MDs. No significant difference was found between the gMS and eMS and between the RNs and MDs in the number of wounds they cared for per month in the past 1 year
. Inter-rater reliability. The inter-rater reliability of the Chinese version of DESIGN-R was analyzed among RNs and MDs and among eMS and gMS, respectively. The results showed a high ICC of 0.960 (95% confidence interval [CI], 0.912–0.990) for the total score. For each of the items, the ICC ranged from 0.570–1.0 for all the raters (see Table 3). Because of the small number of participants (trained RNs, plastic surgeons, and physician), data were not analyzed according to their specialties. The eMS ICCs concerning inflammation/infection (0.759 in eMS versus 0.530 in gMS) and granulation (0.794 in eMS versus 0.532 in gMS) were higher than those of the gMS, but no differences were found in these ICCs between the MDs and the RNs overall. At the same time, the eMS ICC for depth was lower than that of the gMS, with no significant difference between the MDs and the RNs. Test-retest reliability results showed an acceptable correlation, with ICCs of more than 0.8 in most of the items (see Table 4).
Validity. For the second part of the study, the criterion validity of the Chinese version of DESIGN-R was tested against the BWAT. The correlation coefficients of the BWAT versus the DESIGN-R of all eight experienced raters showed a strong correlation, with all values >0.80, ranging from 0.807 to 0.939 (see Table 5).
Test use. Among all the raters in the reliability test, 15 (27.3%) thought the DESIGN-R was easy to use in the clinical setting, 32 (58.2%) rated it as having a moderate degree of difficulty, and eight (14.5%) believed it was difficult.
Healthcare staff contend PU treatment will benefit from an easily and accurately measured parameter that is positively correlated with a wound healing outcome.7 DESIGN-R, a revised version of DESIGN, is a clinical PU healing assessment tool shown to have sufficient validity and reliability for use in clinical practice. It has adequate content validity, including seven items that cover the central concepts of PU healing, and it takes less time to administer than other assessment instruments used.9-11 The subscores have values weighted by statistical analysis from 2,598 patients to describe PU healing and to compare wound-healing progress for different PUs in different patients.10 Results of two multicenter cohort studies involving 3,196 patients showed the total score can predict PU healing up to 90 days in advance.11 This means that DESIGN-R provides an accurate, simple, and quick way for evaluating PU healing in clinical settings. Although it has been tested in a variety of clinical settings by different professional clinicians, it remains unclear whether the inter-rater reliability of DESIGN-R can be influenced by different raters, especially among different disciplines, such as MDs and RNs.
In this study, DESIGN-R was translated into Chinese and the reliability and validity of the Chinese version were evaluated. DESIGN-R was selected for introduction into China because it is inexpensive, saves time, and is sufficiently practical to be used regularly in different clinical settings. In addition, this study was the first attempt to test the reliability of DESIGN-R between RNs and MDs with different experience in wound care management. The researchers also anticipated the results of this study would provide some guidance as to how to effectively implement DESIGN-R in China.
Overall, the Chinese version of DESIGN-R showed acceptable reliability and validity. The ICCs of inter-rater reliability and test-retest reliability of DESIGN-R total score were always >0.9 whether the raters were MDs, RNs, eMS, or gMS. This finding strongly supports the idea that, even though none of the Chinese participants had any experience using this scoring tool to assess wounds, they could use the tool appropriately after 30 minutes of instruction. Moreover, only eight (14.5%) participants reported experiencing difficulty in applying DESIGN-R in wound assessment. This implies the Chinese version is easy to learn and to use for Chinese clinicians.
The results of the criterion-related validity test in the eight experienced participants were satisfactory, showing correlation coefficients of 0.8 between DESIGN-R and the BWAT tool. The BWAT has been adopted among advanced wound specialists for a variety of chronic wounds. Clinometric testing has been used to establish its heterogeneity, clinical sensitivity, face validity, and rating for subjective components.8 Therefore, the strong correlation between the two scoring tools suggests the Chinese version of DESIGN-R is sufficiently accurate to evaluate PU status to approximately the same degree as the BWAT. Similar results were found in the authors’ previous study.9
The results of the inter-rater reliability of the seven items of the Chinese version of DESIGN-R showed scores for depth, inflammation/infection, and granulation were affected by the participants’ experience in wound care, but not by profession (whether the participants were MDs or RNs). The ICC values for inflammation/infection and granulation were lower in the gMS (0.530 and 0.532, respectively) than in the eMS (0.759 and 0.794, respectively). The present result is partly consistent with the authors’ previous evaluation of the older version of DESIGN,9 where the ICCs for granulation tissue and necrotic tissue were 0.40 and 0.54, respectively, in real clinical setting assessments. In the present study, several factors may have affected the assessment of inflammation/infection and granulation. One possibility is that the participants in the gMS without professional wound care training have inadequate knowledge about granulation and inflammation/infection. The other possibility may be that photographs, not actual wounds, were used. The participants in the gMS with limited wound care experience may have had difficulty evaluating inflammation/infection and granulation due to limited information about temperature, smell, and flexibility or induration of the wound bed, which cannot be represented by photographs. In addition, the assessment would be affected by the quality of the photographs, such as resolution and color saturation. The ICC of depth was slightly lower in the eMS (0.763) than in the gMS (0.901), probably due the small sample of eMS. From these results, the authors conclude that additional clinical instruments must be developed for granulation and inflammation/infection assessment, especially for the gMS, to improve the inter-rater reliability in clinical application. Furthermore, it is particularly noteworthy that these differences were not observed between the RNs and the MDs. Therefore, DESIGN-R can be considered as a “common language” in the multidisciplinary team approach to PU and can be used by MDs and RNs and possibly by other professionals such as physiotherapists, pharmacists, nutritional therapists, and so on.
Extrapolating options. The instrument also may be useful in telemedicine and other distance consultation formats. Electronic technology, such as Visitrak (Smith & Nephew Wound Management Inc, Largo, FL), VEV MD system (Courtesy of Vista of Medical Ltd, Manitoba, Canada),23 and Canvas24 system, are used to assess wound status. Some of those methods employ a stereophotogrammetry technique to facilitate comparison of images obtained at different times. The accuracy of the software23,24 provides an objective technique for wound evaluation, measurement, and tracking. Even though time and cost restraints have restricted the use of these electronic methods in daily practice, electronic devices may eventually provide more precise wound information and a visual record of wound status. DESIGN-R, BWAT, and these more advanced electronic options can be used easily at bedside; each has its advantages and disadvantages.
The external validity of this research may be limited because participants were not randomly selected and the number of participants in the eMS group, especially MDs, was small. They may not represent the larger community of eMS, gMS, MDs, and RNs. Another limitation is that photographs rather than real PUs were used. It is difficult and complex to use real PUs in a testing situation because of changes in clinical status, patient burden, and priority of care. Photographs have been used in many PU healing process assessment studies, suggesting that using photographs for this purpose may be reliable. However, it is still necessary to evaluate the reliability and validity of the Chinese version of DESIGN-R in real PU healing processes.
In addition, one of the members who contributed to the translation process of DESIGN-R also was included among the six trained RNs, and thus participation-related bias could not be ruled out. None of the participants had experience using the wound healing assessment tools in this study. Future studies should be conducted by Chinese medical staffs who are expert in using DESIGN-R and other assessment instruments.
Accurate assessment is a key component of wound management.8 Clinical research is still needed to improve assessment accuracy, especially of inflammation/infection and granulation assessment among gMS, in order to apply DESIGN-R in China to improve clinical practice.
The results of this study demonstrated the Chinese version of DESIGN-R is a practical assessment tool with acceptable accuracy, comparable to the BWAT, among MDs and RNs for describing PU status in clinical settings. When the user of DESIGN-R is an inexperienced wound care clinician, additional clinical instruments for inflammation/infection and granulation assessment must be employed. Additional education, training, and research are warranted to improve the ability of this tool to guide practice.
This study was partially supported by the Japan China Sasakawa Medical Fellowship. The authors thank Ms. Zhou Hongzhen and Mr. Liu Jie for their help in organizing the tests at Nanfang Hospital. The authors also are grateful to all the nurses and doctors who participated in the study.
Ms. Zhong is a Registered Nurse, Department of Nursing, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, China; and a Researcher, Department of Gerontological Nursing/Wound Care Management, Division of Health Sciences and Nursing, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. Dr. Nagase is a Medical Doctor, Kunitachi Aoyagi-en, Tokyo, Japan; and a Visiting Researcher, Department of Gerontological Nursing/Wound Care Management, Division of Health Sciences and Nursing, Graduate School of Medicine, The University of Tokyo. Dr. Huang is a Project Researcher; Ms. Yamamoto is a doctoral course student; Mr. Kanazawa is a master course student; and Dr. Sanada is a Professor, Department of Gerontological Nursing/Wound Care Management, Division of Health Sciences and Nursing, Graduate School of Medicine, The University of Tokyo. Dr. Kaitani is an Associate Professor, Department of Adult Nursing, School of Nursing, Sapporo City University, Sapporo, Japan. Dr. Iizaka is a Public Health Nurse, Kawasaki, Japan. Please address correspondence to: Professor Hiromi Sanada, PhD, RN, WOCN, Department of Gerontological Nursing/Wound Care Management, Division of Health Sciences and Nursing, University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan; email: email@example.com.