Min Wei1, Ling Wu2, Yan Chen3, Qiaomei Fu4, Wenyue Chen1, Dongliang Yang5. 1. Department of Orthopedics, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, Nanjing, Jiangsu, China. 2. Wound Care Center, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, Nanjing, Jiangsu, China. 3. Nursing Department Office, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, Nanjing, Jiangsu, China. 4. Surgery Branch, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, Nanjing, Jiangsu, China. 5. Cangzhou Medical College, Cangzhou, Hebei, China.
Abstract
BACKGROUND: Pressure ulcers (PUs) are encountered in all types of care settings. The incidence of PUs in the intensive care unit (ICU) is higher for a variety of reasons. The Braden Scale is a widely used tool to assess the risk of PU, but its predictive power is controversial in ICU settings. In this systematic review and meta-analysis, we aimed to evaluate the predictive accuracy of the Braden Scale for measurement of risk of PU in adult ICU patients. METHODS: A comprehensive literature search in English databases (PubMed, Cochrane Library, OVID, and Web of Science), Chinese databases (SinoMed, CNKI, and Wanfang), and gray literature was performed. Studies assessing the predictive value of the Braden Scale for risk of PU in adult ICU patients were searched. Quality of the studies was scored with Quality Assessment of Diagnostic Accuracy Studies-2. Country, study design, setting, blinding, and characteristics of included studies were extracted. RESULTS: Eleven full-text articles containing 10 044 patients, comprising 1058 patients with PUs were included. The pooled sensitivity and specificity of the Braden Scale for predicting PU risk in ICU adults were 0.89 (95% CI, 0.87-0.91; I2 = 94.9%, P = .0000) and 0.28 (95% CI, 0.27-0.29; I2 = 99.2%, P = .0000), respectively. The pooled DOR was 6.29 (95% CI: 4.09-9.68). The overall weighted AUC was 0.7812 ± 0.0331 (95% CI: 0.7163-0.8461) and the Q* value was 0.7196 ± 0.0283 (95% CI: 0.6641-0.7751). Significant heterogeneity was noted among the included studies. Meta-regression analysis showed that there was no heterogeneity in blinding (P = .074), study design (P = .679), or cut-off value (P = .821). CONCLUSIONS: This meta-analysis demonstrated that the Braden Scale had a moderate predictive validity with good sensitivity and low specificity in adult critically ill patients. Further development and modification of this tool or generation of a new tool with higher predictive power is warranted for use in ICU populations. RELEVANCE TO CLINICAL PRACTICE: The first step in prevention of PU is risk assessment. In this meta-analysis, we aimed to evaluate the predictive power of the Braden Scale for assessing risk of PU in ICU adult patients, which could potentially guide clinical practice.
BACKGROUND: Pressure ulcers (PUs) are encountered in all types of care settings. The incidence of PUs in the intensive care unit (ICU) is higher for a variety of reasons. The Braden Scale is a widely used tool to assess the risk of PU, but its predictive power is controversial in ICU settings. In this systematic review and meta-analysis, we aimed to evaluate the predictive accuracy of the Braden Scale for measurement of risk of PU in adult ICU patients. METHODS: A comprehensive literature search in English databases (PubMed, Cochrane Library, OVID, and Web of Science), Chinese databases (SinoMed, CNKI, and Wanfang), and gray literature was performed. Studies assessing the predictive value of the Braden Scale for risk of PU in adult ICU patients were searched. Quality of the studies was scored with Quality Assessment of Diagnostic Accuracy Studies-2. Country, study design, setting, blinding, and characteristics of included studies were extracted. RESULTS: Eleven full-text articles containing 10 044 patients, comprising 1058 patients with PUs were included. The pooled sensitivity and specificity of the Braden Scale for predicting PU risk in ICU adults were 0.89 (95% CI, 0.87-0.91; I2 = 94.9%, P = .0000) and 0.28 (95% CI, 0.27-0.29; I2 = 99.2%, P = .0000), respectively. The pooled DOR was 6.29 (95% CI: 4.09-9.68). The overall weighted AUC was 0.7812 ± 0.0331 (95% CI: 0.7163-0.8461) and the Q* value was 0.7196 ± 0.0283 (95% CI: 0.6641-0.7751). Significant heterogeneity was noted among the included studies. Meta-regression analysis showed that there was no heterogeneity in blinding (P = .074), study design (P = .679), or cut-off value (P = .821). CONCLUSIONS: This meta-analysis demonstrated that the Braden Scale had a moderate predictive validity with good sensitivity and low specificity in adult critically ill patients. Further development and modification of this tool or generation of a new tool with higher predictive power is warranted for use in ICU populations. RELEVANCE TO CLINICAL PRACTICE: The first step in prevention of PU is risk assessment. In this meta-analysis, we aimed to evaluate the predictive power of the Braden Scale for assessing risk of PU in ICU adult patients, which could potentially guide clinical practice.
Authors: Jenny Alderden; Susan M Kennerly; Andrew Wilson; Jonathan Dimas; Casey McFarland; David Y Yap; Lucy Zhao; Tracey L Yap Journal: Comput Inform Nurs Date: 2022-10-01 Impact factor: 2.146