The Incidence of Pressure Ulcers in the Emergency Department: A Meta-analysis

Objective. The aim of this meta-analysis was to estimate the incidence of pressure ulcers (PUs) in the emergency department (ED). Methods. The authors searched PubMed, Web of Science, and CINAHL databases for all time leading up to February 2016. The incidence of PUs in the ED were collected, and the pooled incidence with 95% confidence interval (CI) was calculated by meta-analysis. Results. Six studies with 8 cohorts were included in this study. The incidence of PUs in the ED ranged from 0.38% to 19.1%. The pooled incidence was 6.310% (95% CI 4.658%–7.962%; Z = 7.49, P = 0.000) by the random-effects model (I² = 98.2%, P = 0.000). Funnel plot, Begg’s test (z = 0.37, P = 0.711) and Egger’s test (t = 2.73, P = 0.034) suggested there was significant publication bias. Sensitivity analysis that included prospective studies showed the pooled incidence was 11.992% (95% CI 5.569%–18.415%; Z = 3.66, P = 0.000) by the random-effects model (I² = 93.9%, P = 0.000). Conclusion. Even for a short stay in the ED, developing a PU is a common complication. The figures of this meta-analysis may be useful in the estimation of the PU burden in the ED.

Although much progress has been made in the prevention and the treatment of pressure ulcers (PUs),¹ they are still a common problem for patients. Pressure ulcers are more likely to occur in patients who are critically ill; patients who have had a prolonged surgery, a stroke, or a spinal injury; and frail elderly patients and individuals using wheelchairs in long-term care settings. In intensive care settings over the past 10 years, PU prevalence ranged from 4% to 49%.² Also in the past decade, patients with surgery-related PUs had an incidence of 15%, with 95% confidence intervals (CI) 14%–16%.³ For patients who had a stroke over the past 10 years, the PU prevalence ranged from 17% to 28%.⁴ In long-term care settings, again over the past decade, the prevalence of PUs ranged from 3.5% to 16.9%.^5-7 

In the authors’ clinical practice, they found that PUs were also common in the emergency department (ED), and the incidence had increased in recent years. Pressure ulcers in the ED were also reported by other researchers.^8-10 Santamaria and colleagues⁸ reported the incidence of PUs in critically ill patients and those with trauma in the ED ranged from 3.1% to 13.1%. Dugaret and coauhthors⁹ reported 12.8% of patients (77/602) in the ED had prevalent PUs at admission and 19.1% (115/602) at discharge, while Denby and Rowlands¹⁰ reported the PU incidence was 0.38% in another ED. Although PUs have always been assessed in critically ill patients, prolonged surgery patients, patients who have had a stroke, and elderly patients, there is an increasing focus on PUs in the ED in recent years due to the increased PU incidence. However, the authors still did not find any meta-analysis to investigate the incidence of PUs in the ED worldwide. 

In this study, the authors aim to systematically review the incidence of PUs in the ED with the goal of the results providing a benchmark to evaluate the incidence of PUs in this setting. 

Data sources and search strategy. The authors searched PubMed, Web of Science, and CINAHL databases for relevant studies. The search terms included “pressure ulcer,” “pressure sore,” “bedsore,” and “emergency department.” The strategy was used in PubMed listed as follows: (“pressure ulcer”[MeSH Terms] OR “pressure ulcer*”[Title/abstract] OR “Bedsore*”[Title/abstract] OR “Pressure Sore*”[Title/abstract]) AND (“Emergency Department” [Title/abstract] OR “emergency service” [MeSH Terms]). In Web of Science advanced search, the authors used the strategy of “TS = pressure ulcer* AND TS = Emergency Department.” In CINAHL, “pressure ulcer” and “emergency department” were searched as keywords. The search time span was set from all time leading up to February 26, 2016 in all databases. All searches were limited to English by database-supplied limits, and the search was also broadened by reviewing relevant references.

Study selection criteria. The inclusion criteria incl-uded: 1) studies that investigated the incidence of PUs in the ED, 2) studies that provided a number of patients who acquired PUs in the ED and provided number of total included patients, and 3) studies that provided the PU incidence with 95% CI. The study selection criteria can  be described as: 1 AND (2 OR 3).  Animal studies, case reports, and reviews were excluded. 

Study data extraction and quality appraisal. The    authors used a uniform data extraction tool to collect the following data from included studies: first author, published year, country, research design, patient’s age, gender, number of events, sample size, and PU incidence with 95% CI. 

Quality appraisal was conducted according to research design. The Jadad scale was used for randomized, controlled clinical trials¹¹ and the Newcastle-Ottawa Scale for observational studies.¹² Two reviewers independently extracted data and appraised quality. Disagreements were resolved by consensus and discussion with a third reviewer.

Statistical analysis. For each of the selected studies, the incidence with 95% CI was calculated. The overall pooled incidence with 95% CI was meta-analyzed by a fixed-effects or random-effects model. If heterogeneity was found between included studies, the authors chose the fixed-effects model; otherwise, the random-effects model was used. The heterogeneity was analyzed by I² statistical analysis. Significant heterogeneity was indicated when I² > 50% with a P value < 0.05. Publication bias analysis was conducted by funnel plot. An asymmetric funnel plot indicates publication bias, while a symmetrical funnel plot indicates no publication bias. Begg’s test and Egger’s test were also used to ascertain publication bias, with P < 0.05 indicating publication bias. Sensitivity analysis was conducted but only included prospective studies. All analyses were performed using Stata 12.0 (StataCorp LP, College Station, TX).

Eligible studies. The authors initially retrieved 30 potentially relevant articles from PubMed, 59 from Web of Science, 62 from CINAHL, and 5 from relevant references reviews.  A total of 25 repeated articles were excluded. Of these 131 articles, 125 articles did not meet the eligibility criteria of the current study and were excluded. Thus, 6 articles were included for analysis (Figure 1). Study sample sizes ranged from 152 to 32 664. Among 6 studies, 2 came from Australia, 2 from the United States, and the other 2 studies were from France and Canada, respectively.  There were 4 prospective studies and 1 cross-sectional study, and the quality appraisal indicated a high quality of the included studies. Characteristics of the included studies are presented in Table 1.

Pooled incidence of PUs in the ED. Six studies^8-10,13-15 with 8 cohorts provided enough information for meta-analysis. The PU incidence of the included studies ranged from 0.38% to 19.1%. Significant heterogeneity was found in the included studies (I² = 98.2%, P = 0.000). The pooled incidence of PUs in the ED was 6.310% (95% CI 4.658%–7.962%; Z = 7.49, P = 0.000) by random-effects model. Figure 2 shows the forest plot. 

Publication bias analysis. The funnel plot was asymmetric, indicating a possible publication bias (Figure 3).  Egger’s test (t = 2.73, P = 0.034) showed a significant publication bias, but Begg’s test (z = 0.37, P = 0.711) indicated no publication bias.

Sensitivity analysis. Sensitivity analysis with only the included prospective studies showed the pooled incidence of PUs in the ED was 11.992% (95% CI 5.569%–18.415%; Z = 3.66, P = 0.000) by random-effects model (Figure 4).

In the current meta-analysis, the authors found the pooled incidence of PUs in the ED was 6.310% (95% CI 4.658%–7.962%). Sensitivity analysis also confirmed the pooled incidence was 11.992% (95% CI 5.569%–18.415%). The results showed PUs have been a common complication in the ED over the past decade, and the authors discovered some possible reasons. First, the demographics of patients visiting the ED have changed over this timeframe. Patients older than 60 years account for almost 24% of all ED visits worldwide.¹⁶ Second, patients admitted in the ED usually present with a serious medical illness, multiple comorbidities, and poor functional status with limited mobility and activity.¹⁷ Moreover, the length of time spent in an ED has increased in the past decade. These studies from the past 10 years showed 99.2% of patients in the United States had an ED length of stay greater than 2 hours prior to hospital admission, suggesting patients in the ED are vulnerable to PUs.¹⁰

This meta-analysis pooled the incidence of PUs in the ED, and the PU stages in accordance with the 2014 National Pressure Ulcer Advisory Panel Guidelines included Stages I–IV.  However, the authors cannot pool the incidence of PUs in every stage. Dugaret et al⁹ reported Stage I PUs were the most common to occur during the patient’s ED stay. The incidence of PUs at ED discharge were 80.9% (Stage I), 9.6% (Stage II), 7.0% (Stage III), and 2.6% (Stage IV), respectively.⁹ Denby and Rowlands¹⁰ also found Stage I PUs to be the most common stages, with 61.6% Stage I PUs and 38.4% Stage II PUs.  Although the authors cannot pool the incidence of PUs in every stage, the results indicated that the authors’ pooled incidence was mostly from Stage I and Stage II PUs. 

Pressure ulcers are one of the common complications in the ED, which makes prevention important, especially in patients who are elderly, have a serious illness, are immobile, or are nutritionally compromised.¹⁸ Cost-effectiveness analysis also supported early PU prevention in the ED. Early prevention was effective, with 0.0015 quality-adjusted life days gained, and less costly, with a mean cost savings of $32 per patient.¹⁴ 

This meta-analysis has certain limitations. First, it presents substantial heterogeneity between included studies, which is a common limitation for pooling incidence. Second, the authors only included studies published in English. A significant publication bias, the most common reporting bias,¹⁹ existed in this meta-analysis. Third, only 6 studies were included in this meta-analysis, making for a small sample size. These limitations will lead to bias and need to be considered when evaluating the conclusions.

In summary, the findings of this study suggest that PUs are a common complication in the ED, even for a short stay. Appropriate monitoring and treatment for PUs in the ED must be performed. Furthermore, the figures of this meta-analysis may be useful in the estimation of the burden of PUs in the ED.

From Nantong University, Nantong, Jiangsu, China

Address correspondence to:
Hong-Lin Chen, MD
Nantong University
Qixiu Road 19#
Nantong, Jiangsu 
China
pphss@126.com 

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