The Braden Scale for Predicting the Outcome and Prognosis of Pressure Injuries in Older Inpatients: A Multicenter, Retrospective Cohort Study

Introduction. The Braden scale is used to assess the risk of patients with pressure injuries (PIs), but there are limitations to the prediction of PI healing. There is a lack of tools for evaluating PI healing and outcome in clinical practice. Objective. The purpose of this study was to examine the ability of the Braden scale to predict the outcome and prognosis of PIs in older patients. Materials and Methods. Outcome indicator was the wound healing rate of patients with PIs at discharge. The receiver operating characteristic (ROC) and Hosmer-Lemeshow goodness-of-fit test were used to evaluate the discrimination and calibration. Results. Completed data were available for 309 patients, 181 of whom (58.6%) were male. The Braden scale had poor discrimination to predict the outcome and prognosis of PIs with an area under the curve (AUC) of 0.63 (95% CI, 0.56–0.70; P = .01). Subgroup analyses showed the Braden scale had low diagnostic value for patients aged over 90 years (AUC^ROC = 0.56; 95% CI, 0.17–0.96; P = .738), patients with respiratory diseases (AUC^ROC = 0.51; 95% CI, 0.37–0.65; P = .908), and digestive system diseases (AUC^ROC = 0.59; 95% CI, 0.42–0.75; P = .342). The level of calibration ability by Hosmer-Lemeshow goodness-of-fit test was acceptable, defined as P >.200 (χ² = 6.59; P = .473). In patients aged more than 90 years (χ² = 4.88; P = .431) and female patients (χ² = 7.03; P = .425), the Braden scale was also fitting. It was not suitable for patients with respiratory diseases (χ² = 11.35; P = .078). Conclusions. The Braden scale had low discrimination for predicting the outcome and prognosis of PIs in older inpatients. The development of a new tool is needed to predict healing in patients with preexisting PIs.

Introduction

Pressure injuries (PIs) are chronic, refractory wounds with a long healing cycle and require difficult treatment courses that result in a low rate of healing. The prevalence of PIs in the older population is high. Pressure injuries are one of the most significant problems in medical institutions, both domestically and abroad.¹ With the accelerated growth of the aging population, the numbers of people with PIs are likely to increase.² One study showed that 70% of PIs occurred in people over 70 years of age,³ and a cohort study of 323 nursing home residents with advanced dementia found the prevalence of PIs to be as high as 38.7%.⁴ A multicenter cross-sectional observation study in the United Kingdom found the prevalence of community PIs was 7.7%, with 84% of those affected being patients 65 years and older.⁵

Current treatments of PIs mainly rely on traditional debridement, flap reconstruction, and negative pressure drainage.^6,7 However, PIs appear to be highly resistant to treatment, for which the outcomes are slow or even ineffective. A retrospective study of 78 patients with PIs by Karahan et al⁸ showed the rate of healing was only 35.6% at discharge or death. The refractory nature of PIs has led to the formation of a relatively large population of patients with PIs,⁹ and as such the approach to healing PIs needs to be improved. Despite advances in treating PIs, an actual cure for them has proven to be elusive. A study by Payne et al¹⁰ showed patients with stage 3 and 4 PIs would require 110 weeks to achieve healing.

Once PIs develop, the clinical treatments are not only difficult but also costly, bringing a heavy financial burden to patients and society. According to statistics published in 2011, annual medical expenses in the United States related to PI care amounted to $362 million to $3.3 billion.¹¹ Pressure injuries can also reduce quality of life.¹² Related literature has reported that the factors affecting the healing of PIs are numerous and complex.^13,14 These factors include adequate nutrition, circulatory disorders, moisture, artificial ventilation, and tissue perfusion changes. In addition, an increasing number of studies have indicated that albumin levels and blood count are also associated with the healing of PIs.^15,16

Currently, the Braden scale is one of the most commonly used PI risk assessment scales.^8,17,18 Medical staff can quickly predict the likelihood of PIs, identify high-risk patients with PIs, and perform early intervention to reduce the incidence of PIs.¹⁹ The positive role of the Braden scale in preventing the occurrence of PIs is widely recognized. The Braden scale includes sensory perception, nutrition, activity, mobility, moisture, and shear/friction. It produces a total risk score ranging from 6 to 23.²⁰ The Braden scale is primarily used in the risk assessment of PIs but has limitations with regard to predicting healing of PIs that have already occurred. Results of a study showed that only the friction or shear force in the entries of the Braden scale was related to the prognosis of PI healing.²¹ The findings suggested that the Braden scale was not a useful tool for guiding care planning of patients who already have stage 1 PIs.²¹ Clinical treatment and care of patients with PIs are mostly based on personal experiences. There is no scientific basis for using the Braden scale to predict the prognosis for patients with PIs. In addition, a review found the activity and nutrient subscales of the Braden scale are not suitable for predicting PI development.²² Subjective factors are more likely to affect the results of the Braden scale.²³ In addition, to the authors’ knowledge, no research has evaluated the value of the Braden scale in predicting the outcome and prognosis of PIs.

There are some alternative PI evaluation tools that are used in clinical practice, such as the Pressure Ulcer Scale for Healing (PUSH); the Bates-Jensen Wound Assessment Tool; and the Depth, Exudate, size, Inflammation/Infection, Granulation, Necrotic Tissue (DESIGN) tool.²⁴ These models are applicable to describing PIs in the development process, but these evaluations do not lend themselves to the task of predicting wound healing.²⁵ Existing tools for risk assessment and monitoring of healing are inadequate to meet an identified deficit in care planning.²⁶

Whether the Braden scale can judge the prognosis of PIs in older patients remains to be studied. Moreover, there is a lack of tools in clinical practice to assess the outcome and prognosis of PIs. The aim of this study was to examine the utility of the Braden scale in predicting the outcome and prognosis of PIs as quantified by receiver operating characteristic (ROC) and calibration curve.

Methods

Study design

A multicenter, retrospective observational study of inpatients with PIs in three grade 3a hospitals was conducted from January 2017 to December 2017. The 3 hospitals were Affiliated Hospital of Nantong University, Nantong First People’s Hospital, and Nantong Third People’s Hospital. The study protocol was approved by the Institutional Review Board of each hospital.

Study population

For this study, the following inclusion criteria were specified: (1) patients 65 years of age or older²⁷; (2) patients with existing stage 2, 3, or 4 deep tissue injury and unstageable wounds that met the guidelines of the definition and staging of PIs by the National Pressure Ulcer Advisory Panel in 2016²⁸; and (3) hospitalized patients whose risk of PI had been assessed using the Braden scale.

Patients were excluded if they (1) had a category 1 PI and intact skin; (2) had a PI for which the specific category could not be determined²⁹; (3) died before a PI developed; or (4) if their record was controversial or incomplete.

Estimation of sample size

A small pre-experiment was conducted. The rate of PI healing at discharge was approximately 30%. The estimated sample content was 6 × 10 ÷ 30%, which were 200 cases. Considering some samples would be rejected due to missing data during collection and entry, it was necessary to appropriately increase the reserve of 10%. The final sample size was no fewer than 220 cases.

Data collection

Pressure injury registration data were collected from the nursing adverse event management system of the electronic medical records of the 3 hospitals according to the conditions of each hospital. The collected data included 2 parts: (1) demographic characteristics, which included hospitalization number, name, age, sex, and disease category; and (2) information related to PIs, including Braden scores and PI category. The results of the Braden scores and the category of PI were taken into the initial evaluation results of the admission. A case where multiple PIs occurred in the same part of the body during the analysis was counted as 1 case. For patients with more than 1 PI, the highest category was included in the analysis. The degree of PI healing at the time of discharge was the primary focus of the current study. The 3 hospitals used the same standard of care interventions: wound management measures, the use of mattresses and protective dressings, the frequency of turning over the patient and slapping the back, and nutritional support.

Outcomes

Complete wound healing rate at discharge was considered the main outcome, and this information was obtained from the nursing record sheet or from the PI reporting form in the case data. The standard of healing was complete epithelialization of the wound with no exudate.³⁰ The PIs of all parts of 1 patient reached the healing standard and were confirmed as a case of healing.

Quality control

The authors collected patient data by obtaining support and assistance from relevant hospitals departments. The assisting data collectors were all graduate students in nursing. General information in the PI nursing report registration and patient medical records were strictly checked, and attending health care professionals were consulted as needed to review handwritten elements of the care records. Patients with inconsistent general information and unknown causes were excluded.

The patients were selected and analyzed according to the inclusion and exclusion criteria. All the case data included in the analysis were photographed and recorded by the researcher and another assistant researcher. After the data were entered, 5% of the original data were randomly selected for quality inspection, such as error checking and trapping. After data collection was completed, the authors reviewed the electronic medical record system and the data in the file to avoid errors.

The information was obtained from the nursing record sheet or from the PI report registration form in the case data, which complemented each other, to obtain more comprehensive information; contradictory cases were eliminated to obtain more accurate information.

Statistical analyses

Statistical analysis was performed using SPSS version 22.0 (IBM Corporation) and STATA version 11 (StataCorp LLC) software. The results were presented as mean ± standard deviation (normal distribution or approximate normal distribution data). Count data were expressed in terms of frequency and percentage.

First, the authors investigated the discrimination of the Braden scale for prediction of PI healing by ROC. The independent variable (test item) was Braden scores, and the dependent variable was the condition of PI healing at discharge (ie, yes or no). An area under the curve (AUC^ROC) of 0.5 to 0.7 indicated low predictive value, an AUC^ROC of 0.7 to 0.9 indicated medium predictive value, and an AUC^ROC of 0.90 to 1.00 represented a higher predictive value; an AUC^ROC of 0.5 indicated that the test had no predictive ability.³¹

Second, the calibration degree of the Braden scale for predicting PI healing was tested by using a calibration curve. The predicted probabilities of each PI were sorted from small to large and divided into 8 groups. The actual healing values for each group and the predicted healing values for the model were listed separately. The actual healing rate (actual healing value divided by total number of cases) and predicted healing rate (predicted healing value divided by total number of cases) of each group were calculated. The calibration curve was drawn according to the actual healing rate and the predicted healing rate of each group. For patients with more than 1 PI, all PIs reached the healing standard defined as healing.

Subgroup analyses by patients’ age, sex, and disease categories were also performed. The significance level alpha was .05, with P < .05 considered a statistically significant goodness-of-fit test (P > .200 for good fit).

Results

Patient characteristics

A multicenter retrospective collection of report forms and medical records of inpatients with PIs in 3 hospitals was conducted from January 2017 to December 2017 (Figure 1). A total of 309 inpatients with PIs met the inclusion criteria. Participant age ranged from 65 to 99 years (mean, 78.7 ± 7.4). The average length of stay was 19.1 ± 16.5 days. Most patients had stage 2 PIs (73.8%), followed by stage 3 PIs (20.1%); stage 4 PIs were least common (6.1%). Of the 309 patients, 181 were male and 128 were female. Among the 309 patients, 28.5% had a respiratory disease diagnosis; other common diagnoses were nervous system disorders (17.2%), digestive system disorders (16.2%), and cardiovascular disease (10%). Ninety-one patients had PIs categorized as healed at discharge; the healing rate was 29.4%. Braden scores ranged from 7 to 23 (mean, 12.2 ± 2.8).

Discrimination of Braden scale for predicting the outcome and prognosis of PIs

This study examined the discrimination of PIs by using the Braden score. The Braden scale has a lower performance value in the overall sample with an AUC^ROC of 0.63 (95% CI, 0.56–0.70; P = .01; Figure 2A).

The ability to predict the outcome and prognosis of PIs in different age subgroups was explored. The Braden scale showed greater utility in patients aged 65 years to 74 years (AUC^ROC, 0.63; 95% CI, 0.51–0.74; P = .036) than in those aged 75 years to 89 years (AUC^ROC, 0.61; 95% CI, 0.52–0.71; P = 0.048). The Braden scale may not be suitable for predicting PI healing in patients older than 90 years (AUC^ROC, 0.56; 95% CI, 0.17–0.96; P = .201).

In the sex subgroups, there was no significant difference between male (AUC^ROC, 0.65; 95% CI, 0.56–0.74; P = .002) and female (AUC^ROC, 0.60; 95% CI, 0.49–0.71; P = .070) patients. The discrimination value of the Braden scale for predicting the outcome and prognosis of PIs was not particularly obvious.

In different disease subgroups, the Braden scale was not valid in individuals who had respiratory diseases (AUC^ROC, 0.51; 95% CI, 0.37–0.65; P = .908) and digestive system diseases (AUC^ROC, 0.59; 95% CI, 0.42–0.75; P = .342).

The subgroups predictive validity of the Braden scale is listed in Table 1 and eFigure 1. A summary forest map for all subgroup analyses of ROC for Braden scale in predicting PI healing is shown in eFigure 2.

Calibration of Braden scale for predicting the outcome and prognosis of PIs

There was no statistically significant difference (χ² = 6.59; P = .473) between the predicted values and the actual observations. Specifically, the Braden scale had good calibration when used in patients older than 90 years of age (χ² = 4.88; P = .431).

The Braden scale showed greater utility in male patients (χ² = 1.99; P = .921) than in female patients (χ² = 7.03; P = .425). Subgroup results showed the Braden scale was not suitable for older patients with respiratory diseases (χ² = 11.35; P = .078).

Calibration curves were drawn based on the actual healing rate and predicted healing rate for each subgroup, as shown in eFigure 2. According to the Hosmer-Lemeshow goodness-of-fit test, results suggested that the calibration of the Braden scale was acceptable. However, some subgroups did not fit (Table 2 and Figure 3).

Discussion

The present results showed the area under the ROC curve based on the Braden scale was 0.63 (95% CI, 0.56–0.70; P = .001). The goodness-of-fit test showed that χ² equaled 6.59 (P = .473). The Braden scale had a low ability to predict outcomes in older patients who already had PIs. The study by Alderden et al³² supports these conclusions,demonstrating that the Braden scale was not a useful tool for guiding care planning for patients who already have a stage 1 PI.

Patients with PIs can be given a good prognosis after being given individualized treatment and care^33,34; however, a comprehensive evaluation system for predicting the outcome of PIs is currently lacking in clinical practice. In addition, the US population older than 65 years of age is expected to double in the next 25 years; therefore, the number of people with PIs may increase exponentially.³⁵ It is necessary to establish and test a comprehensive and objective evaluation system to predict the outcome of PIs.

Not only is the Braden scale not recommended for use in older patients who already have PIs, but its value in predicting the development risk of PIs is also suspect. He et al³⁵ found that use of the Braden scale in surgical patients who developed PIs after unrelated surgery had low overall diagnostic accuracy and diagnostic ability. The area of ROC was 0.6921, and the pooled values for sensitivity and specificity were 0.42 and 0.84, respectively.^36,37 One study suggested clinical judgment may also be useful for predicting risk, and experienced nurses should be able to identify patients who are high-risk without the Braden scale. The implementation of the assessment scale has been shown to reduce the incidence of PIs, but it is not clear whether this is because the scale itself is valuable, or if the education and training accompanying the scale has improved the clinical judgment of those health care professionals using it.³⁸ The appropriateness of the Braden scale in long-term care was questionable due to its low specificity.³⁹ Another study showed that the lowest Braden score has no significant correlation to the development of PIs.¹³

In the current study, a subgroup analysis indicated that the Braden scale was not suitable for people older than 90 years of age (ROC, 0.56; P = .738) or for patients with respiratory (ROC, 0.51; P = .908) and digestive diseases (ROC, 0.59; P = .342). One explanation may be that patients over the age of 90 and those with respiratory or digestive diseases may have poor physical function and their condition may not be conducive to the development of PIs. The development of a new tool is needed to predict healing in the 90-year-old or older adult with preexisting PIs. In contrast to the results of this study, a retrospective study by Park et al⁴⁰ found that the Braden score in the healing group was lower than that in the nonhealing group, and the difference was statistically significant. The reason may be that the study population was limited to patients with stage 2 PIs,³⁹ whereas the current study included patients with stage 2 through 4 PIs; this may have resulted in the same level of the Braden score for the included patients and partially eliminated the effect of scoring on PI development. Subgroup analysis in the study by Park et al⁴⁰ also showed little difference in predicting the outcome and prognosis of PIs using the Braden scale between male and female patients. However, one study has shown that the utility of the Braden scale did not differ between male and female patients when used to predict the risk of developing PIs.⁴¹

Currently, there are no specific tools for predicting the prognosis of PIs in clinical practice. The PUSH evaluation tool is only available for wound condition descriptions. One study has shown that PUSH assessment results were not significantly correlated with traditional PI nursing observations.²⁶ A new scale is urgently needed. The authors’ research demonstrated that the Braden scale also could not reliably predict outcome and prognosis. A new predictive healing scale, combined with such tools as PUSH, will be better used for the assessment of PI healing.

Limitations

Due to the limitations of a retrospective design, the data collected in this study may not be as reliable as prospective data. This study only collected data in 3 hospitals, which may not fully represent the actual situation of PIs in all hospitalized patients in China. The extrapolation of the conclusion needs to be further verified by multiregional and multicenter expansion of sample size.

Conclusions

The Braden scale has been shown to have less utility in predicting the outcome and prognosis of PIs in older patients. For older patients who have already developed PIs, the Braden scale is not recommended. Development of a new assessment tool to predict the outcome and prognosis of PIs is important. A new predictive prognosis scale, combined with such tools as PUSH, will be better used for the assessment of PI outcomes

Acknowledgments

Authors: Yi-Ping Song, MM¹; Man-Li Zha, MM²; Hong-Wu Shen, BM²; Yang Li, BM³; Lin Du, MM¹; Ji-Yu Cai, MM¹; Yi Qin, MM²; and Hong-Lin Chen, MD¹

Affiliations: ¹Nantong University, Nantong, Jiangsu, China; ²Affiliated Hospital of Nantong University, Nantong, Jiangsu, China; and ³School of Public Health, Nantong University

Correspondence: Hong-Lin Chen, MD, Dean, Nantong University, School of Nursing, QiXiu Road 19#, Nantong, Jiangsu 226001 China; honglinyjs@126.com

Disclosure: The authors disclose no financial or other conflicts of interest.

References

1. Dinorah GA, Thomas DR. Assessment and management of chronic pressure ulcers in the elderly. Med Clin N Am. 2006;90(5):925–944. doi:10.1016/j.mcna.2006.05.018

2. Watson R. Europe’s Aging Population. BMJ. 1996;312(7044):1442–1442.

3. Bansal C, Scott R, Stewart D, Cockerell CJ. Decubitus ulcers: a review of the literature. Int J Dermatol. 2010;44(10):805–810. doi:10.1111/j.1365-4632.2005.02636.x

4. Mitchell SL, Teno JM, Kiely DK, et al. The clinical course of advanced dementia. New Engl J Med. 2009;361(16):1529–1538. doi:10.1056/NEJMoa0902234

5. Stevenson R, Collinson M, Henderson V, et al. The prevalence of pressure ulcers in community settings: an observational study. Int J Nurs Stud. 2013;50(11):1550–1557. doi:10.1016/j.ijnurstu.2013.04.001

6. Cushing CA, Phillips LG. Evidence-based medicine: pressure sores. Plast Reconstr Surg. 2013;132(6):1720–1732.

7. Zuo XL, Meng FJ. A care bundle for pressure ulcer treatment in intensive care units. Int J Nurs Sci. 2015;2(4):340–347. doi:10.1016/j.ijnss.2015.10.008

8. Karahan A, AAbbasoğlu A, Işık SA, et al. Factors affecting wound healing in individuals with pressure ulcers: a retrospective study. Ostomy Wound Manage. 2018;64(2):32–39.

9. Scott JR, Gibran NS, Engrav HL, Mack CD, Rivara FP. Incidence and characteristics of hospitalized patients with pressure ulcers: State of Washington, 1987 to 2000. Plast Reconstr Surg. 2006;117(2):630–634. doi:10.1097/01.prs.0000197210.94131.39

10. Payne W, Bhalla R, Hill D, Pierpont Y, Robson M. Wound healing trajectories to determine pressure ulcer treatment efficacy. Eplasty. 2011;11:e1.

11. Van D, Rustagi K, Gray T, Halford M, Ziemkiewicz E, Shreve J. The $17.1 billion problem: the annual cost of measurable medical errors. Health Aff (Millwood). 2011;30(4):596–603. doi:10.1377/hlthaff.2011.0084

12. Gorecki C, Brown J, Nelson E, et al; European Quality of Life Pressure Ulcer Project Group. Impact of pressure ulcers on quality of life in older patients: a systematic review. J Am Geriatr Soc. 2009;57(7):1175–1183. doi:10.1111/j.1532-5415.2009.02307.x

13. Pender L, Frazier S. The relationship between dermal pressure ulcers, oxygenation and perfusion in mechanically ventilated patients. Intensive Crit Care Nurs. 2005;21(1):29–38. doi:10.1016/j.iccn.2004.07.005

14. Jones K, Fennie K. Factors influencing pressure ulcer healing in adults over 50: an exploratory study. J Am Med Dir Assoc. 2007;8(6):378–387. doi:10.1016/j.jamda.2007.02.011

15. Iizaka S, Sanada H, Matsui Y, et al. Serum albumin level is a limited nutritional marker for predicting wound healing in patients with pressure ulcer: two multicenter prospective cohort studies. Clin Nutr. 2011;30(6):738–745. doi:10.1016/j.clnu.2011.07.003

16. Sugino H, Hashimoto I, Tanaka Y, Ishida S, Abe Y, Nakanishi H. Relation between the serum albumin level and nutrition supply in patients with pressure ulcers: retrospective study in an acute care setting. J Med Invest. 2014;61(1–2):15–21. doi:10.2152/jmi.61.15

17. Park KH. A retrospective study using the Pressure Ulcer Scale for Healing (PUSH) tool to examine factors affecting stage II pressure ulcer healing in a Korean acute care hospital. Ostomy Wound Manage. 2014;60(9):40–51.

18. Lane C, Selleck C, Chen Y, Tang Y. The impact of smoking and smoking cessation on wound healing in spinal cord-injured patients with pressure injuries: a retrospective comparison cohort study. J Wound Ostomy Continence Nurs. 2016;43(5):483–487. doi:10.1097/WON.0000000000000260

19. Pancorbo-Hidalgo P, Garcia-Fernandez F, Lopez-Medina I, Alvarez-Nieto C. Risk assessment scales for pressure ulcer prevention: a systematic review. J Adv Nurs. 2006;54(1):94–110. doi:10.1111/j.1365-2648.2006.03794.x

20. Bergstrom N, Braden B, Kemp M, Champagne M, Ruby E. Predicting pressure ulcer risk: a multisite study of the predictive validity of the Braden Scale. Nurs Res. 1998;47(5):261–269. doi:10.1097/00006199-199809000-00005

21. Alderden J, Zhao YL, Zhang Y, et al. Outcomes associated with stage 1 pressure injuries: a retrospective cohort study. Am J Crit Care. 2018;27(6):471–476. doi:10.4037/ajcc2018293

22. Cox J. Predictive power of the Braden scale for pressure sore risk in adult critical care patients: a comprehensive review. J Wound Ostomy Continence Nurs. 2012;39(6):613–621. doi:10.1097/WON.0b013e31826a4d83

23. Alderden J, Zhao Y, Zhang Y, Wang J, Huai BS. Braden scale (ALB) for assessing pressure ulcer risk in hospital patients: a validity and reliability study. Appl Nurs Res. 2017;33:169–174. doi:10.1016/j.apnr.2016.12.001

24. Harris C, Bates-Jensen B, Parslow N, Raizman R, Singh M, Ketchen R. Bates-Jensen wound assessment tool: pictorial guide validation project. J Wound Ostomy Continence Nurs. 2010;37(3):253–259. doi:10.1097/WON.0b013e3181d73aab

25. Zhong X, Nagase T, Huang L, et al. Reliability and validity of the Chinese version of DESIGN-R, an assessment instrument for pressure ulcers. Ostomy Wound Manage. 2013;59(2):36–43.

26. George-Saintilus E, Tommasulo B, Cal CE, et al. Pressure ulcer PUSH score and traditional nursing assessment in nursing home residents: do they correlate? J Am Med Dir Assoc. 2009;10(2):141–144. doi:10.1016/j.jamda.2008.10.014

27. Organization-WHO TWH. The United Nations World Health Organization proposes a new age segment. Published 2013. Accessed December 1, 2020. https://baike.baidu.com/redirect/0bd4RCYxMvY1H9fOzYam3CqNWUSJt2VZlno1yNVNvxIfXdmvDZTLTjFDihbMS7XpIl4BgZYmelU-cE3EfFg_bl-mqjS2KdbuXo7cLX5RVRAPoV5lsF0

28. NPUAP. Prevention and Treatment of Pressure Ulcers: Clinical Practice Guideline. Published 2016. Accessed November 20, 2020. http://www.internationalguideline.com/downloads

29. Ayello E, Cordero G, Sibbald R. Survey results from Canada and some Latin America countries: 2016 National Pressure Ulcer Advisory Panel changes in terminology and definitions. Adv Skin Wound Care. 2017;30(2):71–76. doi:10.1097/01.ASW.0000511443.83861.d1

30. Jiang QX, Li XH, Zhou X, et al. Efficacy evaluation of triethanolamine cream wet wound wound and feasibility analysis of healing score. J Nurs. 2012;19(13):53–57.

31. Swets JA. Measuring the accuracy of diagnostic systems. Science. 1988;240(4857):1285–1293. doi:10.1126/science.3287615

32. Roerecke M, Tobe SW, Kaczorowski J, et al. Sex-specific associations between alcohol consumption and incidence of hypertension: a systematic review and meta-analysis of cohort studies. J Am Heart Assoc. 2018;7(13):e008202. doi:10.1161/JAHA.117.008202

33. Bredesen I, Bjøro K, Gunningberg L, Hofoss D. The prevalence, prevention and multilevel variance of pressure ulcers in Norwegian hospitals: a cross-sectional study. Int J Nurs Stud. 2015;52(1):149–156. doi:10.1016/j.ijnurstu.2014.07.005

34. Black J, Edsberg L, Baharestani M, et al. Pressure ulcers: avoidable or unavoidable? Results of the National Pressure Ulcer Advisory Panel Consensus Conference. Ostomy Wound Manage. 2011;57(2):24–37.

35. National Institute on Aging. Dramatic Changes in U.S. Aging Highlighted in New Census, NIH Report. March 9, 2009. Accessed December 1, 2020. https://www.nih.gov/news-events/news-releases/dramatic-changes-us-aging-highlighted-new-census-nih-report

36. He W, Liu P, Chen H. The Braden Scale cannot be used alone for assessing pressure ulcer risk in surgical patients: a meta-analysis. Ostomy Wound Manage. 2012;58(2):34–40.

37. Liao Y, Gao G, Mo L. Predictive accuracy of the Braden Q Scale in risk assessment for paediatric pressure ulcer: a meta-analysis. Int J Nurs Sci. 2018;4(4):419–426. doi:10.1016/j.ijnss.2018.08.003

38. Anthony D, Parboteeah S, Saleh M, Papanikolaou P. Norton, Waterlow and Braden scores: a review of the literature and a comparison between the scores and clinical judgement. J Clin Nurs. 2008;17(5):646–653. doi:10.1111/j.1365-2702.2007.02029.x

39. Wilchesky M, Lungu O. Predictive and concurrent validity of the Braden scale in long-term care: a meta-analysis. Wound Repair Regen. 2015;23(1):44–56. doi:10.1111/wrr.12261

40. Park KH. A retrospective study using the pressure ulcer scale for healing (PUSH) tool to examine factors affecting stage II pressure ulcer healing in a Korean acute care hospital. Ostomy Wound Manage. 2014;60(9):40–51.

41. Fernandes L, Caliri M. Using the Braden and Glasgow scales to predict pressure ulcer risk in patients hospitalized at intensive care units. Rev Lat Am Enfermagem. 2008;16(6):973–978. doi:10.1590/s0104-11692008000600006