Turkish Nurses' Opinions of the Braden and Waterlow Pressure Ulcer Risk Assessment Scales: A Descriptive Pilot Study

Risk evaluation scales are used as part of prevention strategies for pressure ulcers (PUs). Two of the more used scales, the Braden and Waterlow Pressure Ulcer Risk Assessment Scales (B-PURAS and W-PURAS, respectively) are used in Turkey and worldwide, and their validity in terms of sensitivity, specificity, and predictive validity have been examined in various studies. To determine nurses’ opinions of B-PURAS and W-PURAS in terms of administration time, practicality, clarity, and perceived ability to encompass PU risks and accurately predict PUs, a descriptive study was conducted from October 15, 2011 to November 20, 2011 at a university hospital among nurses who volunteered to participate.Demographic information collected and assessed included age, highest degree of education completed, and practice area. Participants were trained to use both scales by researchers during a 1-hour session and asked to use them for 2 weeks in their daily practice. The nurses then completed a paper-and-pencil, 12-item questionnaire measuring agreement with general questions about PURAS with options to provide comments. In addition, the questionnaire contained 3 open-ended questions on scale preference, 1 question to rate perceived scale accuracy in predicting PUs, and the opportunity to recommend changes to the scales. Data were analyzed using SPSS 20.0. Frequencies, percentages, and Spearman’s rank correlations were calculated. Eighty-three (83) registered nurses (mean age 27.46 ± 3.73, mean nursing experience 6.53 ± 3.25 [range 0.5–16] years) participated; 18.1% of the nurses had prior experience using such scales, and none of the clinics in the study facility had used PU risk scales previously. Most participants (71, 85.5%) had a bachelor’s degree, 62 (74.7%) worked in a facility that admitted patients at high risk for developing PU, but 66 (almost 80%) saw fewer than 4 ulcers per week. Each nurse performed an average of 22 assessments during the study for a total of 1,826 assessments. Participants generally found both scales practical and appropriate; almost 75% thought the B-PURAS was acceptable for use in all clinics, compared to 51% for the W-PURAS; only 20 participants questioned the scales’ accuracy in predicting PUs. While 43% preferred the B-PURAS for precision, clarity, and practicability, 25% preferred the W-PURAS because it was more comprehensive. Also, the older the nurse, the more likely the preference for the B-PURAS (P <0.019). Some nurses (13%) suggested adding different risk factors such as serum albumin. Overall, 61.4% stated they would prefer to use the B-PURAS over the W-PURAS. Nurses’ recommendations should be considered for practice and new scale development and testing, ideally in different patient populations. 

Pressure ulcers (PUs) are frequently encountered in individuals with acute and chronic diseases.^1-5 According to a bottom-up costing approach study by Bennett et al⁶ and a literature review by Ayello and Braden,⁷ the care and management of PUs are difficult, they are costly to treat, and they require extensive recovery time, prolonging the duration of hospitalization. In terms of quality, cost, and nursing care, preventing PUs is easier and cheaper than treating them.^5-7

PU prevalence rates vary markedly among different countries. In the United States, the National Patient Care Safety Monitoring Study,⁸ which covered more than 51,000 patients, found 4.5% of Medicare beneficiaries developed a PU during their hospital stay and 5.8% had a PU on admission. In Canada, PU prevalence was estimated at 25% in acute care settings.⁹ In the UK, a point-prevalence study for estimating the prevalence of wounds in general (including surgical wounds, diabetic leg and foot ulcers, cancer, and PUs) found the prevalence of PUs alone was 17%.¹⁰ In descriptive PU prevalence studies^11-15 in Turkey (N = 84), the prevalence of surgery-related PUs was 54.8%, and in intensive care units (ICUs) the rate was between 14.3% and 20.56%; point-prevalence studies revealed a prevalence between 8.1% and 8.3%. Although the literature overall (including descriptive and prospective studies and literature reviews^7,16-26) suggests PUs are largely preventable, they remain a considerable problem. In the current authors’ opinion, the lack of a preventive approach and the much greater focus on the pathophysiology of PUs may be considered factors contributing to the overall failure to find a desirable solution to this problem.

The first preventive intervention involves assessment of the risk of developing a PU.^7,16-22 According to a literature review by Ayello and Braden,⁷ if a risk assessment is performed systematically, PU occurrence can decrease by 60% commensurate with the risk level, and the cost of care also decreases markedly. According to various descriptive studies and literature reviews,^3,6,20 identifying the risk of PU development decreases the immediate care costs and allows for higher quality and more effective patient care.

Many nurse investigators have developed various PU risk assessment scales (PURASs) to determine patient risk of developing PUs.^21-24 The first PURAS was developed by Doreen Norton in 1962 for the selection of elderly participants for research on PUs.^25,26 Numerous PURASs were developed in the 1980s; currently, at least 40 different PURASs are available.¹⁷ Hidalgo et al’s²⁶ systematic review reported the most commonly known scales are the Braden (1987), Norton (1962), Gosnell (1973), and Waterlow Pressure Ulcer Risk Assessment Scales (1985); among these, the Norton and Braden scales are most commonly used throughout the world. However, although guidelines typically recommend the use of PURASs, these scales are not without critics. For instance, the scales do not include all risk factors associated with PUs, they sometimes are used as alternatives rather than to supplement clinical evaluations, and they have generally poor validity and reliability.^25,26 As such, hospital guidelines nowadays recommend the use of clinical assessment and a PURAS together.^27-30

Nurses can employ one of these scales while providing care or can develop one of their own. Developing a new scale and testing its reliability and validity requires expertise on the issue or the support of experts. Furthermore, the importance of the clinical decision-making of nurses themselves in risk evaluation cannot be overlooked, especially the clinical decision-making ability of experienced nurses who are experts in their fields. Thus, it may be easier to use scales already developed for nurses in clinical practice.¹⁹

Clinical and home care nurses who are certified in wound, ostomy, and continence nursing are legally responsible for the prevention and care of PUs in Turkey.^31,32 Although the importance of using a PURAS and preventive care is frequently stressed in the literature, PURASs are not commonly used in acute care units or long-term care facilities in Turkey; furthermore, only limited care aimed at preventing PUs is generally provided because assessments are not frequently performed.^2,13,19 In some countries, making a risk evaluation within the first 6 or 24 hours following admission may not be required. In Turkey, only the Braden, Norton, and Waterlow PURASs, which have undergone reliability and validity analyses in Turkey, are accepted for use.^33-35

To the authors’ knowledge, around the world no studies have been conducted to examine nurses’ opinions regarding the use of these scales. The clinical observations and expertise of nurses are important for developing scales for PU risk evaluation, because nurse opinions are critical in determining the risk factors to be included in a scale. Nurse input is also necessary for determining the characteristics of the scale, such as the clarity and practicability of items that need to be included.^{7,17,19,20,25} Therefore, instead of merely asking nurses to accept the developed scales as wholly complete and accurate, researchers might obtain nurse opinions and allow them to contribute to scale development, which in turn may improve the buy-in and benefit of these scales to nursing care. In the authors’ opinion, PURASs based on nurses’ opinions tend to cover a more extensive range of relevant risk factors and are more applicable and responsive to nurses’ needs in planning patient care; furthermore, nurses’ opinions affect the use of evidence-based PU risk assessment.

The purpose of this descriptive pilot study was to determine the opinions and preferences of nurses regarding the Braden and Waterlow PURASs (B-PURAS and W-PURAS, respectively).

The study was performed at a 1,100-bed university hospital. Written permission to conduct the study was obtained from the hospital directorate, which is also the hospital’s institutional review board. Study data were collected between October 15 and November 20, 2011 from participating nurses working in ICUs at the university hospital (including the anesthesia, neurosurgery, internal medicine, general surgery, chest diseases, cardiology, cardiovascular surgery, and neurology ICUs) as well as in the physical therapy, rehabilitation and orthopedics clinics. Nurses who worked in these areas  determined PU risk using their own professional knowledge and experience. Nurses were informed about the study and were invited to volunteer. Nurses who consented to participate in the study were given individual training by one of the researchers, who provided a 1-hour training session that addressed PUs, PU risk assessment, and use of the Braden and Waterlow PURASs. Following training, the nurses administered these 2 scales at the same time each day on the same patients for 2 weeks. Both scales were completed for each patient. After these 2 weeks, the questionnaires were completed by the participants and collected by the researchers.

Questionnaires.
Demographic data. The questionnaire on demographic characteristics comprised 12 items on nurses’ personal characteristics (eg, age, highest degree completed), their status in PURAS training, and their previous experience with using PURASs. The nurses completed the paper-and-pencil instrument independently (see Table 1). 

Risk scale comparison. Researchers (experts interested in PUs and scale development) developed this questionnaire based on a review of the literature; it was not tested for validity or reliability. The survey instrument included 12 statements that allowed participants to compare the 2 scales/assessment on time to completion, practicality, clarity, coverage of PU risks, and perceived ability to predict PUs. Participants indicated their level of agreement with survey items (“I agree,” “I partly agree,” and “I do not agree”) and were asked to explain their answers when they selected “I partly agree” or “I do not agree” (see Table 2). Participants also indicated their level of agreement with the statement, “Risk assessment scales that have established validity and reliability should be used to prevent PUs.” The questionnaire also included 1 item that asked nurses how accurately they believed the scales predicted PUs; this item was rated using a visual analog scale ranging from 0 (worse at predicting) to 10 (better at predicting). Questions included, “Which PURAS do nurses prefer by area of practice?” “What are reasons for preferring one scale over the other?” “What are risk factors nurses believe should be included PURAS?” 

The questionnaire initially was sent to 5 experts on PUs and scale development who evaluated the items in terms of comprehensibility, scientific content and criteria, ability to quantify the questions, and rating scale method. Necessary revisions were made according to the experts’ opinions. 

Data analysis. The data obtained in the study were analyzed using Statistical Packages for Social Sciences (SPSS) for Windows, Version 20.0 (SPSS Inc, Chicago, II, IL, USA). The frequency and percentages of nurses’ responses and Spearman’s rho were calculated to analyze the correlation between scale preference, work experience, and age. The open-ended questions were individually assessed and similar statements were grouped; data on the frequency and percentages were calculated  by SPSS.

Out of a total of 122 nurses in the participating departments, 83 (68%, mean age 27.46 ± 3.73, mean nursing experience 6.53 ± 3.25 [range 0.5–16] years) took part in the study. Nonparticipation was due to workload (27) or being on leave during the study period (12). None of these nurses used any kind of PURAS in their daily work, but 18.1% stated they had prior experience using such scales. Participants’ demographic characteristics are summarized in Table 1. Most nurses (71, 85.5%) had a bachelor’s degree in nursing, 62 (74.7%) had worked previously at clinics where patients with a high risk of PUs were admitted, and 40 (48.2%) provided care to an average of 1 to 2 patients with PUs weekly. Most had not received additional PU prevention/treatment (67, 80.7%) or training on risk assessment scales (81, 97.6%) after graduation. Each nurse performed an average of 22 assessments during the study for a total of 1,826 assessments. Most participants (80.7%) agreed with the statement, “Risk assessment scales that have established validity and reliability should be used to prevent PUs.”

B-PURAS. Most participants (66, 79.5%) thought the B-PURAS was easy to understand, 62 (74.7%) thought the B-PURAS was practical in terms of time consumption, and 56 (67.4%) thought it was practical to use. Only 8 (9.6%) of the participants thought the B-PURAS did not accurately identify patients at risk of PUs. The majority of participants (59, 71%) did not have difficulty choosing from among answer options while using the B-PURAS, and (19) 23% thought the B-PURAS was not suitable for all clinics (see Table 2).

W-PURAS. In assessing the W-PURAS, 53 (63.8%) thought it encompasses all risks associated with PUs, 22 (26.5%) had difficulty choosing from among the answer options, 36 (43.3%) thought it was important for identifying patients at risk but 12 (14.4%) thought it does not accurately identify which patients are at risk of PUs, and 42 (51.2%) found it suitable for use in all clinics (see Table 2).

More than half of the participants (51, 61.4 %) stated they would prefer to use the B-PURAS in their clinics. However, 7 participants (77.7%) working in the cardiovascular surgery ICU preferred the W-PURAS. All of the nurses working at the physiotherapy rehabilitation clinic, 87.5% of those working at the coronary ICU, and 83.3% of those working at the general surgery ICU preferred the B-PURAS (see Table 3). 

Open-ended questions. Open-ended items regarding reasons for preferring one scale over the other were classified under the following categories: precision, clarity, and practicability.

 Among all participants, 43% preferred the B-PURAS for its precision, clarity, and practicability, while 25% reported their reason for preferring the W-PURAS was its elaborateness and comprehensiveness (see Table 3). The mean score on the visual analog scale regarding the B-PURAS’s perceived ability to predict PUs was 7.4 ± 1.7, while for W-PURAS it was 6.8 ± 1.8 (see Table 3). Twelve (12) nurses recommended that in addition to the risk factors found in the 2 scales, factors such as serum albumin value, arterial blood pressure, infection, connection with respirator, amputation, and dehydration should be added.

A significant correlation was found between participant preference for the B-PURAS and age (P <0.05) (see Table 4). Although a significant correlation between years of service and scale preference was not found (P >0.05), a significant negative correlation was found between age and B-PURAS preference (P <0.05); namely, with increasing age, nurses expressed a stronger preference for the B-PURAS (see Table 4).

Although they had limited experience using risk assessment scales, the majority of the participants thought PURASs should be employed in the prevention of PUs. The importance of PURASs and preventive care practices has been frequently emphasized in the literature. The Agency for Health Care Policy Research²⁷ issued the first guidelines on PU prevention, which stated PURASs with established validity and reliability should be used for the prevention of PUs. Guidelines^27-30 also recommended the use of PURASs to identify and manage the risk factors of PUs. According to a systematic review by Walsh and Demsey,³⁶ offering care commensurate with risk level by using a PURAS decreases the risk of PUs and improves the cost-effectiveness of care. Furthermore, recording and performing regular evaluations of a patients’ care requirements facilitates quantification of the patient’s improvement.^26,37 Notably, a correlational study³⁸ of critically ill patients indicated nurses’ subjective global risk estimates and total scores on risk assessment scales shared a 60% variance, suggesting nurses’ subjective assessments of risk exposure differ markedly from their risk estimates derived using standardized risk assessment tools. One explanation for this may be nurses draw on factors other than those captured by the standardized tools when making clinical decisions or they weigh certain patient conditions differently.³⁸ It should be noted that recommending the use of scales does not necessarily mean that nurses’ identification and classification of patients’ risk according to their own knowledge and experiences is unimportant.

One third of participants thought the B-PURAS covered PU risks better than the W-PURAS. A possible explanation for this result is the B-PURAS is based upon risk factors for patients who are cared for at home,³⁹ although with proven validity and reliability, the scale now is used widely in acute and long-term care settings. Braden and Bergstrom³⁹ regarded the intensity of the pressure and the tolerance of the tissue to pressure as the main etiological factors, and thereby developed the B-PURAS with 6 subscales. B-PURAS has been found to have the highest validity and reliability of all PURASs.⁴⁰ Specifically, the B-PURAS has satisfactory reliability and validity and appears to be effective for assessing risk of PUs.^33,34,41-45 According to Hidalgo et al’s²⁶ review, 22 studies have been performed to assess the validity and reliability of the B-PURAS. In these studies, the coefficient of consistency between observers was found to be between 0.83 and 0.99; the sensitivity was between 38.9% and 100%; the specificity was between 26% and 100%; the positive predictive value was 4%–100%; and the negative predictive value was 50%–100%.

In this study, most participants (79.5%) stated B-PURAS was easy to understand, while only 38.5% of them thought the same about the W-PURAS. For a PURAS to be easily comprehended, the statements should be clear and the scale itself should yield consistent results. The results of numerous studies^33,34,40-44 on B-PURAS with varying sample sizes have indicated the scale can consistently assess the risk of PUs successfully. In contrast, fewer validity and reliability studies have been performed on the W-PURAS; it has commonly been criticized for having a low level of reliability.³⁶ Based on the results of their methodological review, Papanikolaou et al⁴⁴ stated varying interpretations can evolve from the W-PURAS results because its subscales are not clear. In the current study, nurses’ opinions were compatible with these criticisms of the scale.

No true agreement was reached as to whether either scale was more practical. In this study, nurses thought the scales took an approximately equal amount of time to complete. Practicality and time for completion are interrelated — PURASs that can be quickly completed may be considered more practical. These results may be attributed to the fact the B-PURAS is easier to understand because it has fewer subscales than the W-PURAS, where “neurological deficits” and “medication” subscales require more nurse interpretation and the “skin type/visual risk areas” and “tissue malnutrition” subscales contain items for which multiple answers can be given. The majority of the participants stated they had difficulty choosing answer options when using the W-PURAS; some subscales require nurses’ interpretations and more than one option must be scored for a given item, which may have been influenced nurses’ answers.

Arguably, the most important findings of the present study are that nurses rated both scales as 7 (out of 10) on perceived predictive ability for PUs and that the items of these scales are not best prediction (10) for determining the risk of PUs in the patients they care for at the clinics. Nevertheless, more than half of the nurses believed PURASs that have established validity and reliability through empirical study should be used in the prevention of PUs. Nurses also recommended (with similar frequency for all items) that risk factors such as serum albumin level, blood pressure, amputation, infection status, and connection with ventilator should be added to these scales. Based on these findings, further studies aiming to develop new PURASs that include other risk factors in accordance with nurses’ recommendations should be performed to determine if additional independent risk factors should be considered.

More than half of the participants preferred the B-PURAS, with the remainder preferring the W-PURAS. Nurses who favored the B-PURAS ascribed their preference to its greater clarity and practicality, while those who favored the W-PURAS cited its comprehensiveness. The nurses working at the cardiovascular surgery ICU preferred the W-PURAS. In a literature review by Feuchtinger et al⁴⁵ on the risk of PUs in patients undergoing cardiovascular surgery, factors such as age; time spent on the operating table; having diabetes mellitus, multiple organ failure, anemia, or neurological conditions; and smoking status — all of which are subscales of the W-PURAS — were deemed significant. Hence, these nurses’ preferences may be associated with the fact there are more risk factors for PUs in cardiovascular surgery patients. Additionally, the W-PURAS layout includes a scoring system to determine risk level on its front side, while on the back side, items on the prevention of PUs, nursing practices, and classifications of PUs are presented. This arrangement provides guidance on nursing care, types of preventative aids associated with the 3 levels of risk status, wound assessment, and dressings.

The nurses whose opinions were solicited were limited to those working in ICUs and orthopedics and physiotherapy-rehabilitation clinics of a single university hospital. As such, the views of nurses working at other clinics or in state and private hospitals were not included, which makes generalization of the results more difficult.

Another limitation is the small sample size and number of assessments. However, despite this, the general results were similar to those reported in national statistics.

Future research to determine whether nurses are able to score these scales accurately and then act according to patients’ results would be beneficial.

The findings of this descriptive pilot study indicated the majority of nurses working at clinics where the risk of PUs was high believed use of a PURAS is necessary, and as they became  familiar with the 2 PURAS utilized during the study, most of them preferred using the B-PURAS over the W-PURAS. The opinions of a larger sample of nurses who have used PURASs in clinics should be obtained and the findings of this study should be verified in a larger study population. 

This research was supported by Scientific Research Projects of Gazi University, Ankara, Turkey. (Project No.47/2011-0).

Ms. Avsar is a research assistant, Yıldırım Beyazıt University Faculty of Health Sciences, Department of Nursing, Ankara, Turkey. Dr. Karadag is a Professor, Koç University School of Nursing, Istanbul, Turkey.

Please address correspondence to: Pinar Avsar, RN, MSN, Faculty of Health Science, Yıldırım Beyazıt University, Bilkent Yolu 3.Km Bilkent/Çankaya, Ankara 06010 Turkey; email: p.avsar.ank@gmail.com.