Catherine VanGilder, BS, MT, CCRA; Gordon D. MacFarlane, PhD; and Stephanie Meyer

Transcription

1 FEATURE Results of Nine International Pressure Ulcer Prevalence Surveys: 1989 to 2005 Catherine VanGilder, BS, MT, CCRA; Gordon D. MacFarlane, PhD; and Stephanie Meyer Pressure ulcers continue to be a significant problem for patients and healthcare facilities. Since 1989, results from the International Pressure Ulcer Prevalence surveys observational, cross-sectional cohort studies conducted by Hill-Rom, Batesville, Ind, have been used to document aggregate prevalence rates and provide acute care, long-term acute care, and longterm care facilities with internal and external benchmarks of process improvement. During each of the nine surveys conducted between 1989 and 2005, clinical teams in participating facilities predominantly in the US (some facilities in Canada, Saudi Arabia, and Australia participated after 2003) assessed admitted patients on assigned study dates. For this study, trends using all records (n = 447,930; average, 49,770 per year) were reviewed. The majority of facilities in each survey year were in the US (99% overall). Overall and nosocomial pressure ulcer prevalence rates ranged from 9.2% and 5.6% in 1989 to 15.5% and 10% in 2003 and 2004, respectively. The highest prevalence was documented in long-term acute care (27.3% overall, 8.5% nosocomial). Most commonly, ulcers were located at the sacrum (28%), heels (23.6%), and buttocks (17.2%). Ulcers were more commonly assessed as Stage I and Stage II (>70%). However, in patients with dark skin tone (2004 and 2005 data, n = 162,296), 13% of identified ulcers were Stage I compared to 32% in patients with medium and 38% in patients with light skin tone. Using the most complete data sets (2003, 2004, and 2005), more severe pressure ulcer prevalence (Stage III+) was not found to be age-related. Approximately 48% of all patients who had pressure ulcers and 48% of patients with nosocomial pressure ulcers were assessed at mild or no risk (Braden scale score >14). Prevalence within the Braden Score risk categories aligned with risk for developing pressure ulcers. Despite increased attention to the pressure ulcer problem, prevalence rates from the last five survey years are relatively unchanged. KEYWORDS: pressure ulcers, pressure ulcer prevalence, nosocomial pressure ulcer, risk assessment, skin tone Ostomy Wound Management 2008;54(2):40 54 Ms. VanGilder, Dr. MacFarlane, and Ms. Meyer are Manager, Clinical Research; Director of Clinical and Scientific Research; and Clinical Information Analyst, respectively, Hill-Rom, Batesville, Ind. Please address correspondence to: Catherine VanGilder, BS, MT, CCRA, 8150 Jadewood Drive, Wilmington, NC 28411; vangilder@hill-rom.com. The authors disclose they are employees of Hill-Rom, Batesville, Ind. Hill-Rom provided funding for the study. 40 OstomyWound Management

2 Pressure ulcers are a significant burden to the healthcare environment in the US, Europe, and other countries. 1-6 Over the last 5 to 6 years, healthcare facilities have increased their focus on prevention and several accrediting agencies and professional societies in the US view pressure ulcer prevalence as an overall quality indicator for facilities. Related prevention endeavors include the Joint Commission on Accreditation of Healthcare Organization s (JCAHO) 2007 Long Term Care National Safety Goal #14, the National Pressure Ulcer Advisory Panel s Board of Directors directives, and the American Nursing Association s Safety and Quality initiatives. 7 9 However, data are limited as to whether this increased focus has reduced the number of nosocomial pressure ulcers (NPU) in healthcare facilities. In order to assist healthcare facilities benchmark their pressure ulcer prevalence against like institutions, as well as internally for quality improvement initiatives, Hill-Rom, Inc., Batesville, Ind, facilitated the International Pressure Ulcer Prevalence survey among acute (AC), long-term acute care (LTAC), and long-term care (LTC) facilities. Acute care facilities in this report comprise inpatient hospital facilities that predominantly treat patients for relatively short amounts of time for crisis or episodic illness. Longterm, acute care facilities include hospitals focused on patients who require intense, specialized treatment for a longer time (usually, 20 to 30 days). Long-term care facilities offer continuing maintenance and inpatient health services to chronically ill, disabled, or developmentally disabled patients. Survey data were collected from 1989 to The study included all patients admitted to or residing in the reporting facilities (the at risk population). Data were collected during a specific 24-hour pre-selected period within a 2- to 3-day window determined by the sponsor. The purpose of this report is to provide an indepth analysis of the International Pressure Ulcer Prevalence Survey data collected from 1989 to Prevalence of Pressure Ulcers The National Pressure Ulcer Advisory Panel (NPUAP) 8 has defined prevalence as a cross-sectional count of the number of cases at a specific point in time, or the number of people with pressure ulcers who exist in a patient population at a given point in time. Prevalence is calculated using the following formula 10 and reported as a percentage: [Number of patients with pressure ulcers] x 100 Number of patients surveyed Overall prevalence includes both pre-existing and NPUs. Pressure ulcers are considered to be nosocomial if the ulcer was not documented on facility admission, regardless of whether it developed during the current admission; prevalence is calculated as specified, with only the number of NPU patients included in the numerator. Therefore, skin assessment on facility admission is critical for accurate quality assessment and NPU reporting. Measuring the prevalence of pressure ulcers over time allows facilities to use data benchmarks to compare process improvements within their own facility and against facilities of similar size and patient acuity. Measuring unit-specific (eg, intensive care) prevalence facilitates identification of areas of concern within the facility and implementation of targeted improvement programs. The increasing use of Ostomy Wound Management 2008;54(2):40 54 KEY POINTS Because each method to document the extent of the pressure ulcer problem has its own advantages, disadvantages, and limitations, aggregate data from a variety of sources is needed. While prevalence data cannot be used to ascertain incidence or trends, the sample size and time frame of nine international pressure ulcer prevalence studies are unique and confirm that pressure ulcer prevalence rates are relatively unchanged. The data also suggest that the use of risk assessment instruments has increased but that current scales may have to be supplemented with clinical judgment. Finally, the assessment of Stage I ulcers remains challenging, especially in patients with dark skin tones. February 2008 Vol. 54 Issue 2 41

3 TABLE 2 PRESSURE ULCER PREVALENCE : TOTAL STUDY POPULATION Year Participating facilities Total patients surveyed 34,987 33,063 31,530 39,874 42,817 33,907 61,427 84,487 85,838 International patients surveyed 123 1,183 3,176 Patients with pressure ulcers 3,219 3,703 3,501 4,027 6,358 4,977 9,550 13,136 13,027 Patients with nosocomial NA NA NA NA 3,040 2,903 4,155 6,533 6,283 pressure ulcers Overall prevalence 9.2% 11.2% 11.1% 10.1% 14.8% % 15.5% 15.2% Prevalence minus Stage I 5.6% 6.9% 6.9% 6.6% 9.4% 9.2% 10.2% 10.3% 10.0% Nosocomial pressure ulcer NA NA NA NA 7.1% 8.6% 6.8% % prevalence Nosocomial pressure ulcers NA NA NA NA NA 4.3% 3.5% 4.1% 3.9% prevalence minus Stage I pressure ulcer prevalence data to assess and compare the quality of facility care has prompted the development and refinement of risk-adjusted prevalence scores Large variations may exist between different risk adjustment procedures, as well as variations in how those procedures are applied and analyzed. Correlations between risk-adjusted prevalence, clinical outcomes, and quality of care as reflected by facility deficiency citations are just beginning to be assessed. 13 However, until some standardization occurs in the procedures for risk adjustment, comparisons between assessments should be viewed with caution. History of the Survey TABLE 1 VARIABLES INCLUDED IN SURVEY ASSESSMENTS Year Assessment includes Department level assessment X X X X X X Nosocomial differentiation X X X X Nutritional risk X X Pressure ulcer risk X X Unable to stage X X In 1989, the first International Pressure Ulcer Prevalence Study (the Study) in this series was conducted in response to a challenge from the NPUAP; 148 US healthcare facilities participated. Although the study was to be a one-time event, many facilities requested a follow-up study, which was conducted in 1991 and included 168 treating facilities. Additional variables were added over time (see Table 1) and participation continues to grow. More than 85,000 patients in 651 facilities were surveyed in 2005 (see Table 2). Amlung et al 14 published a brief analysis of the survey data. The information contained herein updates the previous publication and provides a broader analysis of this ongoing study. Methods Setting. Facilities elected to participate in the Study by signing up on the Hill-Rom website ( The survey is available to all healthcare facilities regardless of Hill-Rom customer status. Interest in the survey is solicited through advertising at wound healing and nursing society meetings, as well as directly through sales teams. The facility registers to participate as an AC, LTAC, or LTC institution; 42 OstomyWound Management

4 TABLE 3 PRESSURE ULCER PREVALENCE BY CARE SETTING 2003, 2004, AND 2005 (US DATA ONLY) Year Parameter AC 1 LTAC 2 LTC 3 AC 1 LTAC 2 LTC 3 AC 1 LTAC 2 LTC 3 Participating facilities Patients surveyed 55,885 1,604 3,477 76,291 2,005 4,830 74,401 1,983 6,242 Patients with pressure ulcer 8, , , Patients with nosocomial 3, , , pressure ulcer Overall prevalence 15.4% 23.9% 13.2% 15.3% 27.3% 13.6% 14.6% 27.3% 14.4% Prevalence minus Stage I 10.0% 18.5% 9.2% 10.0% 22.4% 10.6% 9.4% 23.3% 11.4% Nosocomial pressure ulcer 6.9% 6.2% 4.2% % 6.0% 7.3% 7.0% 5.6% prevalence Nosocomial pressure ulcer 3.6% 4.3% 2.2% 4.1% 5.9% 4.1% 3.8% 5.1% 3. prevalence minus Stage I 1 = Acute care; 2 = Long-term acute care; 3 = Long-term care data were analyzed according to that specification. The facility s desig- PRESSURE ULCER PREVALENCE BY STAGE : TABLE 4 nated coordinator TOTAL STUDY POPULATION (N = 447,930)* received study materials, Year which included data Stage forms, educational materials, and general instruc- II 39% 3 36% 39% 39% % 3 I 39% 38% 38% 35% 3 38% 39% 33% 34% tions. Participating facilities were instructed to perform III IV 14% 8% 11% 9% 10% 8% 6% 8% 8% the survey during a Eschar NA 8% 9% 9% 10% 11% 9% 6% 6% Unable NA NA NA NA NA NA NA 9% 10% pre-determined 24-hour * Percent of patients with most severe ulcer reported per patients period within a pre-selected 2- to 3-day window. For example, the 2005 survey was performed between March 1 and March 3, The goal of the survey was to perform skin assessments on all admitted patients during the 24-hour period; however, 100% patient inclusion was not mandated for participation. A few patients in most participating facilities commonly are either in long operative procedures or otherwise unavailable for skin assessment and are not included in this survey. Procedures. Demographics assessed included age and skin tone (dark, medium or light, a subjective assessment left to the survey team s perception as variations of skin tone exist within ethnic groups). Stage and quantity of ulcers and type of pressure ulcer risk assessment performed (when applicable), the risk assessment value, and time between admission into the healthcare facility and assessment of risk were recorded and reported to the study sponsor. All data were self-reported by each participating facility. If incomplete records were generated, the fields present were incorporated into the data set. An indepth description of study data collection methods has previously been published. 14 Determining prevalence. Overall prevalence in this study represents the percentage of patients identified as having a pressure ulcer relative to the number of admitted patients assessed. Overall prevalence excluding Stage I also was calculated because it has been argued that significant errors and subsequent data problems associated with assessing Stage I pressure February 2008 Vol. 54 Issue 2 43

5 TABLE 5 ANATOMIC LOCATIONS OF REPORTED PRESSURE ULCERS ENTIRE 2005 STUDY SAMPLE (N = 85,838) All ulcers (%) All ulcers (minus Stage I) (%) Nosocomial pressure ulcers (%) Nosocomial pressure ulcers (minus Stage I) (%) Location Total AC 1 LTAC 2 LTC 3 Total AC 1 LTAC 2 LTC 3 Total AC 1 LTAC 2 LTC 3 Total AC 1 LTAC 2 LTC 3 Hand Scapula Occiput Scrotum Knee/peri knee Trochanter Ischium Malleolus Elbow Back Ear Leg Foot Heel Buttocks Sacrum ulcers can occur. 15 Nosocomial pressure ulcers are defined as ulcers found during the survey which were not documented in the patients initial skin assessment upon admission to the facility. If patients had more than one pressure ulcer, the total number of both nosocomial and community-acquired pressure ulcers was recorded; however, only the highest staged ulcer was used for analysis in the aggregate data set. The NPUAP staging system 16 was used for staging definitions. Additional data. Variables were added over time, including department type (1995); a differentiation between nosocomial and pre-existing pressure ulcers (2001); nutritional status, pressure ulcer risk, and the unable to stage category for pressure assessment (2004) (see Table 1). The Braden Risk Assessment tool 17 was most commonly used (more than 80% of the time over the last 2 years) for facilities that report risk scores. Therefore, to analyze risk scores and pressure ulcer prevalence, only data from patients who were assessed using the Braden tool were included. Data analysis. Data were extracted from the original data sets, sorted by desired fields, and analyzed using simple descriptive statistics. Results 1 = Acute care; 2 = Long-term acute care; 3 = Long-term care Overall results. In 1989, 148 facilities with 34,987 patients participated in the Study; by 2005, participation had grown to include 651 facilities with 85,838 patients (see Table 2). The majority (623) of 44 OstomyWound Management

6 TABLE 6 ANATOMIC LOCATIONS WITH HIGHEST NOSOCOMIAL PRESSURE ULCER PREVALENCE RATES (EXCLUDING STAGE I ULCERS) Location (n = 61,427) (n = 84,487) (n = 85,838) Foot Heel Buttocks Sacrum participants in 2005 are US facilities, largely in AC (533 facilities, 74,401 patients), followed by LTC (52 facilities, 6,242 patients), and LTAC (38 facilities, 1,983 patients) (see Table 3). Participants from the international community comprise 123 Canadian patients in 2003 (0.2% of the total that year). When the survey was offered outside of North America in 2004, 1,183 were patients from Canada, Australia, and Saudi Arabi combined (1.4% of total). In 2005, participants included 3,176 patients (3.4% of total) from Canada, Saudi Arabia, and the United Arab Emirates. The international component of this study is very small and represents only 1% of the total sample size. The participating Canadian facilities (18 AC, seven LTAC, and one LTC, totaling 2,977 patients) are included in a separate discussion. Prevalence. Between 1999 and 2005, pressure ulcer prevalence has remained at approximately 15% of the total sample in reporting healthcare facilities. Similarly over the same time period, NPUs have remained relatively unchanged at approximately 7.5% (see Table 2). Before 1999, pressure ulcer prevalence ranged from 9.2% to 11.1%. If Stage I pressure ulcers are excluded, overall prevalence is still relatively unchanged (approximately 9.8%), while NPU prevalence, excluding Stage I, has been approximately 4% over this time period. Figure 1. Surveyed and pressure ulcer population distribution by age (2005 data, n = 84,686). February 2008 Vol. 54 Issue 2 45

7 Figure 2. Age group distributions by ulcer prevalence, depth and gender (2005 data, n = 84,686).* *Stage III+ includes the following assessments: Stage III, Stage IV, and eschar. Number of pressure ulcers. The average number of pressure ulcers per patient from 2003 to 2005 ranged from 1.7 to 1.8 for overall prevalence and 1.5 to 1.6 for NPUs. If Stage I pressure ulcers are excluded, the average number of ulcers per patient ranged from 1.6 to 1.7, with an average of 1.4 NPUs reported per patient. Stage. By most severe stage, the most commonly reported ulcers were Stage I and Stage II, accounting for slightly more than two thirds of all ulcers (see Table 4). From 1999 to 2005, Stage III and Stage IV pressure ulcer distribution has remained at approximately 8% and, respectively. The refinement of staging classifications to include eschar and unable to stage options has not impacted the distribution of Stage III and Stage IV ulcers. Anatomical location. Using all 2005 survey data, the most commonly reported anatomic location of all pressure ulcers was the sacrum (28.3%), followed by the heel (23.6%) and buttocks (17.2%). Nosocomial pressure ulcers follow this same general trend (26.6%, 25.9%, 17.1%, sacrum, heel buttocks, respectively) (see Table 5). The trends in NPUs excluding Stage I at the foot, buttocks, and sacrum have remained essentially constant for the 2003 to 2005 surveys (see Table 6) in contrast to a slight drop in the percentage of ulcers at the heel over this time period. Age. The 2005 survey included 39,889 men and 44,797 women. The average age of all patients surveyed was 65.3 years. The majority of patients (78%) with identified pressure ulcers are age 61 years and above (see Figure 1); 64% of all patients surveyed were >61 years old (see Table 7). The percentage of any given age group with a pressure ulcer increased as age increases, which is especially apparent in the >80 years category relative to the number of surveyed patients (see Figure 1). However, the percentage of subjects within an age group with a Stage III, Stage IV, or eschar/unable to stage was high in the younger age groups, peaked within the 31 to 40 years age group, and declined as age increased (see Figure 2). Men had 46 OstomyWound Management

8 TABLE 7 AGE AND GENDER DISTRIBUTION 2005 (N = 84,646) Ages: to to to to to to 80 >80 Total men 1,258 1,259 1,954 4,201 6,745 7,718 9,160 7,594 Total men with pressure ulcers ,080 1,532 1,577 Total men Stage III, Stage IV, eschar/unable to be staged Men with a pressure ulcer (%) % 7.6% 9.8% 11.6% 14.0% % Men with a Stage III, Stage IV, % % 4.5% 4.5% 4.5% 4.6% eschar/unable to stage Severe pressure ulcers within 35.6% 48.6% 48.3% % 31.9% 26.8% 22.3% each age category (5) Total women 1,243 1,499 2,242 4,170 5,692 7,377 10,351 12,223 Total women with pressure ulcers ,601 2,305 Total women Stage III, Stage IV, eschar/unable to stage Women with a pressure ulcer (%) 2.5% 2.8% 4.2% 6.3% 9.4% % 18.9% Women with a Stage III, Stage IV, % 2.9% 3.5% 3.6% 3.8% eschar/unable to stage Severe pressure ulcers within 29.0% 23.8% 41.1% 31.1% % 23.3% 20.4% each age category * All study data with completed gender variable are used. Distribution results were similar in 2003 and 2004 Figure 3. Prevalence and pressure ulcer severity by age and gender (2005 data, N = 84,686). February 2008 Vol. 54 Issue 2 47

9 TABLE 8 PRESSURE ULCER PREVALENCE BY SKIN TONE: 2004 AND 2005 STUDY RESULTS* 2004 and 2005 Total # Total # PU Stage I Stage II Stage III Stage IV Eschar Unable to Combined Patients Patients Stage Light 110,966 17,257 6,514 6,355 1, ,472 Medium 25,840 3,740 1,183 1, Dark 25,490 3, , Total 162,296 24,620 8,167 9,325 1,710 1,676 1,458 2,284 Percentage of Total Patients Light % 0.96% 0.86% 0.81% 1.33% Medium 4.58% 5.70% 0.98% 0.98% 0.94% 1.30% Dark 1.84% % 1.84% 1.26% 1.86% Percentage of Pressure Ulcer Patients Light 37.75% 36.83% % 5.19% 8.53% Medium 31.63% 39.39% 6.76% 6.74% % Dark % 10.82% 12.92% 8.86% 13.11% * US study data only. One ulcer (highest stage) per patient only Figure 4. Pressure ulcer prevalence and severity by skin tone (2004 and 2005 data, N = 162,296). more and greater severity pressure ulcers than women in all age categories (see Figure 3). The percentage difference between men and women for Stage III, Stage IV, or eschar/unable to stage ulcers was greatest (25%) in participants 21 to 30 years but essentially equivalent (2%) in those >80 years (see Figure 2). Skin tone. Reviewing US data from the time the variable unable to stage was added (2004 and 48 OstomyWound Management

10 TABLE 9 BRADEN SCALE RISK SCORE AND PRESSURE ULCER (PU) PREVALENCE Braden Scale Risk Data Combined (Number of patients assessed Total with Braden Risk assessment tool) Overall all 39,346 58,489 64, ,095 Acute care 35,169 53,261 57, ,774 Long-term care 2,675 3,556 5,281 11,512 Long-term acute care 857 1,535 1,626 4,018 Braden Scale Risk Assessments and PU Prevalence ( data combined) All study patients Total PU NPU Total Total Overall PU Nosocomial PU Patients Assessed Prevalence Prevalence Very high risk (Braden 6 9) 2,112 (8%) 774 (6%) 4,337 (3%) 49% 18% High risk (Braden 10 12) 6,086 (24%) 2,613 (22%) 13,789 (8%) 44% 19% Moderate risk (Braden 13 14) 5,255 (20%) 2,399 (20%) 15,651 (10%) 34% 15% Mild risk (Braden 15 18) 8,684 (34%) 4,295 (36%) 50,764 (31%) 1 8% No risk (Braden 19 23) 3,544 (14%) 2,003 (1) 77,554 (48%) 5% 3% TOTALS 25,681 12, ,095 Acute care Very high risk (Braden 6 9) 1,925 (8%) 737 () 3,851 (3%) 50% 19% High risk (Braden 10 12) 5,345 (23%) 2,373 (21%) 11,656 (8%) 46% 20% Moderate risk (Braden 13 14) 4,567 (20%) 2,155 (19%) 13,036 (9%) 35% 1 Mild risk (Braden 15 18) 7,710 (34%) 3,956 (36%) 44,799 (31%) 1 9% No risk (Braden 19 23) 3,253 (14%) 1,884 (1) 72,432 (50%) 4% 3% TOTALS 22,800 11, ,774 Long-term care Very high risk (Braden 6 9) 89 (6%) 22 (3%) 307 (3%) 29% High risk (Braden 10 12) 402 (25%) 160 (25%) 1,409 (12%) 29% 11% Moderate risk (Braden 13 14) 395 (24%) 163 (25%) 1,901 (16%) 21% 9% Mild risk (Braden 15 18) 559 (34%) 226 (35%) 4,246 (3) 13% 5% No risk (Braden 19 23) 184 (11%) 78 (12%) 3,649 (32%) 5% 2% TOTALS 1, Long-term acute care Very high risk (Braden 6 9) 93 (8%) 15 (5%) 165 (4%) 56% 9% High risk (Braden 10 12) 321 (29%) 76 (25%) 684 (1) 4 11% Moderate risk (Braden 13 14) 270 (24%) 76 (25%) 654 (16%) 41% 12% Mild risk (Braden 15 18) 356 (32%) 100 (33%) 1,433 (36%) 25% No risk (Braden 19 23) 84 (8%) 33 (11%) 1,082 (2) 8% 3% TOTALS 1, ,018 * Braden Risk scores were available for 64,260 of 85,838 patients in 2005; 58,489 of 84,487 in 2004; 39,346 of 61,427 in ), the number of Stage I pressure ulcers identified in patients with dark skin tone (average 13%) was lower than medium skin tone patients (32%) and light skin tone patients (38%) (see Table 8). The overall distribution of identified ulcers for dark skin tone patients is slightly shifted to higher stages as compared to light or medium skin tone patients (see Figure 4). Risk assessment. Of all 2005 survey participants (international data also were evaluated), 89% had a pressure ulcer risk assessment of some kind performed during their facility stay; of those, 49% were February 2008 Vol. 54 Issue 2 49

11 conducted within the first 12 hours following admission. An additional 25% reported assessments between 12 and 24 hours following admission. Of all the patients surveyed (using all risk scales), 41.2% were determined to be at risk for pressure ulcer development utilizing an identified risk scale, while 56.3% were deemed not at risk (2.5% were not documented). In 2005, most reporting facilities (86.5%) utilized the Braden risk assessment tool. Scores were available for 64,260 of 85,838 patients in 2005, 58,489 of 84,487 in 2004, and 39,346 of 61,427 patients in Using combined 2003, 2004, and 2005 data, Braden Risk scores followed prevalence trends with 49% of patients at very high risk (Braden 6 9), 44% at high risk (Braden 10 12), 34% at moderate risk (Braden 13 14), 1 at mild risk (Braden 15 18), and 5% at no risk (Braden 19 23) (see Table 9). Nosocomial pressure ulcer prevalence did not follow this trend: 18% had very high risk, 19% had high risk, 15% had moderate risk, 8% had mild risk, and 3% had no risk (see Table 9). However, of specific interest is that 48% of all patients who had a pressure ulcer identified and 52% of all patients where NPUs were identified were considered at mild or no risk at the time of the survey. Acute care. A total of 74,401 AC patients (533 facilities) were assessed in the US in the 2005 survey; 10,857 (14.6%) had a pressure ulcer. Average age of participants was 64 years (range 63.7 to 64.2 years, 2003 to 2005); 4 (range 46.2% to 47.6%) were men, 53% (range 52.4% to 53.8%) were women. Nosocomial pressure ulcers were present in 5,395 (7.3%) patients. If Stage I ulcers are excluded, the overall prevalence is 9.4%; 3.8% were NPU patients. Since 2001, intensive care units (ICUs) in the participating facilities have consistently reported the highest NPU prevalence, ranging from to 15%. In 2005, ICUs (n = 5,938 patients) reported overall prevalence ranging from 22.4% in surgical ICUs to 25.9% in medical ICUs (MICUs), with NPU prevalence between 12.8% and 15.3%, respectively. Among all participating AC facilities, 19.1% reported a <2% NPU prevalence rate in In 2005, the most common ulcers identified in AC were Stage II (3); 34% were Stage I, were Stage III, were Stage IV, and 10% were unable to be staged. Pressure ulcers at the sacrum were most common (28.9%), followed by the heel (23.) buttocks (17.8%), and foot (5.3%). Nosocomial pressure ulcer prevalence, excluding Stage I ulcers, was 32.0% at the sacrum, 23.2% at the buttocks, and 16.3% at the heel (see Table 5). Using 2005 data, 49% of all pressure ulcer risk assessments were performed in AC within 12 hours of admission; another 25% were performed between 13 and 24 hours, 6% were performed the second day after admission, and 30% were performed after 48 hours. Of all AC admissions, 41% were found to be at risk, 56% were not, and 2.5% were undocumented. In 2005, Braden risk scores were available for 57,344 out of 73,401 patients (78%); in 2004, 53,261 out of 76,291 patients (70%); and in 2003, 35,169 out of 55,885 patients (63%). For the three survey years combined, 8% of patients with an identified pressure ulcers had been assessed as being at very high risk, 23% at high risk, 20% at moderate risk, 34% at mild risk, and 14% at no risk. Similar trends were found for NPUs. Prevalence was higher in each more severe risk category. Long-term acute care. In 2005, 38 US facilities with 1,983 patients participated in the prevalence survey. The mean age was 69.7 years, which was relatively unchanged over the preceding 3 years. Among the patients, 58% were men and 42% were women, similar to 2004 data (5 and 43%, men and women, respectively) but dissimilar to 2003 information where 75% and 25% were men and women, respectively. Risk assessment data (primarily Braden scale) were available from 86.3% of facilities and 63% of LTAC patients assessed were deemed at risk for pressure ulcer development. The Norton scale was used in 10.5% of the surveyed facilities; 3.2% of facilities reported using other risk assessment tools. Data for Braden risk assessment were available for 1,626 of 1,983 patients (82%) in 2005, 1,535 of 2,005 patients (7) in 2004, and 857 of 1,604 patients (53%) in Only 34.6% of all patients had a risk assessments performed within 12 hours of admission; another 14. were performed between 12 and 24 hours of admission. An additional 22.1% of the risk assessments were performed within the first week of admission and 28.5% were performed after the first week of admission. Of all patients, 63% were found to be at risk, 36% were not, and 1.6% were undocumented. 50 OstomyWound Management

12 Overall, LTAC pressure ulcer prevalence in 2005 (27.3%), 2004 (27.3%), and 2003 (23.9%) is consistently the highest recorded in any care segment surveyed. However, the NPU prevalence was 7.0% in 2005, 8.5% in 2004, 6.2% in 2003 similar to AC at 7.3%, 7., and 6.9%, respectively, and more than LTC at 5.6%, 6.0%, and 4.2%, respectively. Excluding Stage I, overall prevalence was 23.3% in 2005, 22.4% in 2004, and 18.5% in 2003, indicating that almost one out of every four to five patients in LTAC has a Stage II or higher pressure ulcer. In 28.9% of all participating LTACs, a NPU prevalence of <2% was reported in Pressure ulcers at the sacrum were most common (28.3%) in LTACs, followed by heel (22.5%), buttocks (16.5%), and foot (8.6%). Nosocomial pressure ulcer prevalence, excluding Stage I ulcers, was 24.1% at the buttocks, 19.3% at the sacrum, and 17.9% at the heel (see Table 5). Long-term acute care patients who had pressure ulcers were most commonly identified as having mild or no risk at the time of the survey (61%), which was also true for NPU patients (55%). Although overall prevalence was higher at higher risk categories, NPU prevalence was highest in the moderate risk category (see Table 9). Long-term care. Of the 6,242 patients in 52 LTC facilities surveyed in 2005, 897 (14.4%) had pressure ulcers. Nosocomial pressure ulcer prevalence was 5.6%; 3. of those ulcers were greater than Stage I. Braden risk scores were reported by 91% of participating facilities and 63% of residents surveyed were identified as at risk for pressure ulcers. Braden scores were available for 5,281 out of 6,242 patients (85%) in 2005, 3,556 out of 4,830 patients (74%) in 2004, and 2,675 out of 3,477 patients (7) in Patients who had pressure ulcers most commonly were deemed at least moderate risk (54%); this trend held for NPU patients (53%). Prevalence was highest overall as risk score increased; however, this trend did not hold for NPU prevalence. The average age of the surveyed residents over the last 3 years was 75 years; the number of men and women was approximately the same. The anatomic distribution of pressure ulcers overall included the sacrum (23%), heel (22.9%), and buttocks (16.2%), similar to other facilities. However, NPU location percentages varied from those in AC and LTAC settings. The proportion of NPUs, excluding Stage I, at the sacrum was approximately half that seen in other care settings (17.6% versus 32%) and the proportion of ulcers at the foot was approximately three times higher than that in other care settings (20.2% versus 7.9% in LTAC and 4.4% for AC) (see Table 5). The majority of pressure ulcers identified as NPUs in LTC were located on the lower extremities (51.9%). Nosocomial pressure ulcer prevalence rates were relatively unchanged during each survey year; 13.2% in 2003, 13.6% in 2004, and 14.4% in Skilled nursing units had a lower NPU rate (4.4%) than nonskilled units (7.). Of the 52 LTC facilities reporting, 12 (23%) documented <1% NPU prevalence; another five facilities (10%) had a NPU prevalence of <2%, indicating considerable variability in NPU prevalence in LTC facilities. Canadian data. In Canada, 18 AC, one LTC, and seven LTAC facilities were assessed during the 2005 survey period, comprising a total of 2,977 patients. Overall prevalence across all care segments was 23., with NPU prevalence at 12.8%. If Stage I ulcers are excluded, prevalence was 13.1% and NPU prevalence was 6.3%. Discussion Many countries have conducted pressure ulcer prevalence studies, indicating the global problem of pressure ulcers. The Canadian healthcare data in this study show an overall prevalence of 23. and a NPU prevalence of 12.8%. These figures are lower than the overall prevalence of 26% published in a review of published and unpublished data by Woodbury and Houghton. 18 In their study, 4,831 AC patients, 3,390 non-acute patients, and 4,200 mixed healthcare setting patients were assessed. Pressure ulcer prevalence was 25.1% in AC facilities, 29.9%, in non-acute care facilities, and 15.1%. in community/home care. Mixed healthcare settings (which may include both acute and postacute care settings) reported a pressure ulcer prevalence of 22.1%. The authors state, The data suggest that pressure ulcers are a significant concern in all healthcare settings across Canada. Additionally, a February 2008 Vol. 54 Issue 2 51

13 recent report 5 of Quebec home care agencies indicates a 1.4% overall prevalence of wounds; 34.8% were defined as pressure ulcers. Although an effort was made in the 2005 Study to include home care agencies, an insufficient number of agencies participated to complete a representative sample. Bours et al 19 reported a pressure ulcer prevalence of 23.1% among 16,344 patients across multiple healthcare settings in The Netherlands using the European Pressure Ulcer Advisory Panel (EPUAP) staging classification system. In this study, prevalence was 13.2% in university hospitals, 23.3% in general hospitals, 32.4% in nursing homes, 34.8% in institutions for the physically handicapped, 21.3%, in home care, and 15.6% in residential home care. The author also reports that 568 out of the 3,782 patients (15%) who had pressure ulcers were assessed as not at risk or a Braden Score >20. This is similar to the present study s 13.8% of patients assessed as not being at risk (Braden >19) who had a pressure ulcer. Bours et al 20 also reported a 28. prevalence in 850 ICU patients, which is similar to the present study s ICU overall prevalence range of 22% to 25%. In the Study, Braden Risk assessment scores followed overall pressure ulcer prevalence trends, especially in AC facilities. However, 13.8% of all patients who had pressure ulcers and 16.6% of NPU patients had been assessed as not being at risk, while 33.8% of all ulcers, and 35.5% of NPUs were found in patients identified as being at mild risk. Whether the present Braden Risk Score accurately reflects overall patient condition throughout the course of hospitalization may need to be addressed and other potentially relevant variables studied. An additional consideration might be whether pressure ulcer preventive care on the unit should be based on current risk rather than risk across the overall course of hospitalization. Pressure ulcer prevention programs should be initiated if deemed clinically relevant regardless of Braden Risk Score. A report of German hospitals and nursing homes by Lahmann et al 4 noted that 21.1% of all patients identified as at risk for pressure ulcer development (defined as Braden <20) had a pressure ulcer. For a relative comparison, when considering at-risk patients defined as Braden <19 in the present overall study group, 22,137 out of 84,541 patients (26.2%) have a pressure ulcer. Lahmann also reports that excluding Grade I pressure ulcers (EPUAP scoring system), the prevalence rate in this study was 10.2% with higher rates in AC (24.6%, mean age 63.6 years) than in LTC facilities (13.9%, mean age 81.9 years). The distribution of pressure ulcers by age group in the Study (see Figures 2 and 3) suggests that overall ulcer prevalence in the facilities surveyed is not age-dependent for patients <80 years old. However, severe ulcers (Stage III and above) appear to be more common by percentage in the younger age groups. While the actual number of patients in the older age groups is considerably higher, the risk of developing a pressure ulcer would appear to be greater in the younger age groups. As has been discussed in NPUAP staging discussion forums, Stage I pressure ulcers are more difficult to diagnose in patients with dark skin than in those with lighter skin. This may be a factor that contributes highly to data errors in assessing Stage I pressure ulcers. In 1989, the NPUAP 8 set a national goal to reduce the incidence of pressure ulcers by 50% by the year But according to a 2003 Agency for Health Care Research and Quality (AHRQ) report, 21 pressure ulcers are increasingly common among hospitalized patients in the US the number of admissions in which pressure ulcers were noted increased 63% from 1993 to The difference between the Study prevalence results during this time period was approximately 4.4 % (see Table 2). The methods used are different from those used in the AHRQ report and both prevalence studies and record reviews have unique limitations. However, the difference between the Study results and the AHRQ numbers is large, suggesting that the latter may in part be explained by an increase in pressure ulcer documentation rather than a substantial increase in the number of patients. Clearly, the goal to reduce the incidence of pressure ulcers remains elusive and may require more substantive changes than originally anticipated. One of the Healthy People 2010 objectives 22 is to reduce the proportion of nursing home residents with current diagnosis of pressure ulcers. Similarly, one of JCAHO 7 goals is to prevent health care-associated pressure ulcers (decubitus ulcers). 52 OstomyWound Management

14 According to the AHRQ, 21 pressure ulcers are associated with an average length of stay of between 10.2 and 14.1 days, compared to a 4.6-day average length of stay for patients without pressure ulcers. Costs to treat pressure ulcers have been estimated at $500 to $40,000 depending on the severity of the wound. 23 The acquisition of a pressure ulcer during a hospitalization or a nursing home stay has been identified as an area of necessary quality improvement. Although no quantifiable change has occurred in the overall prevalence of pressure ulcers since initiating the Study, 19.1% of AC, 28.9% of LTAC, and 32. of LTC facilities report an NPU rate of <2% compared to an average of 7.5%, suggesting considerable variations in outcomes between participating facilities. In part because of these differences, risk-adjustment strategies have been devised to allow more direct comparisons between facilities. However, no consensus exists for this adjustment or how it should be applied. It is unknown whether the difference described above would remain if the survey results were risk-adjusted for each facility. Pressure ulcer prevalence has been used as a benchmark for the success of process improvement programs. 3,24-26 Many authors report significant reductions in pressure ulcer rates following implementation of a wound care process improvement (PI) effort. Hiser 3 describes a facility-wide NPU prevalence of 9.2% before PI and 6.6% 3 years later with specific improvements in NPU prevalence in the MICU (from 29.6% to 0% for the four quarters after PI). Granick 24 reports decreases in pressure ulcer prevalence from 22.6% in 1993 to 8. by 1996 after the implementation of a multidisciplinary wound program that provided early intervention for patients either at risk for wounds or presenting with early lesions. Hopkins 25 notes an initial overall pressure ulcer prevalence of 18% in 1996 and, after implementation of a process improvement program, the prevalence rate decreased to 10% in 1997 and to 9% in Similarly, O Brien et al, 26 describing the results of three prevalence studies conducted in one AC facility, noted a decrease in NPUs. In 1993, 18% of 313 patients had an ulcer and 14% had NPUs, in 1995 the overall prevalence was 10% (7.5% NPUs), and in 1997 the overall prevalence was 10% with a 5.1% rate of NPUs. According to the authors, It seems reasonable to conclude that the above-mentioned programs (education, rounds, required routine skin assessments, and support surface criteria development) and changes have contributed to the considerable decrease of nosocomial ulcers, particularly Stage II, Stage III, and Stage IV wounds in our facility. 26 The first step in many of these studies was to perform a prevalence study for benchmark purposes and identification of root causes of the problem. Once problem areas are identified and root cause analyses performed, the implementation of preventative measures for at-risk patients can aid the facility in an overall reduction in NPU pressure ulcers. Despite the overall focus on pressure ulcer prevention, an overall decrease in pressure ulcer prevalence has not been observed in this or other studies. Potential explanations may include overall higher patient acuity levels or less-than-optimal RN staffing levels both of which may suppress overall efforts to prevent NPUs. Much work is needed to move toward a healthcare system that provides protection from pressure ulcer formation during hospitalization or residence. Many institutions are making great efforts to improve overall quality of care. Prevalence studies are helpful to document present status and provide internal and external benchmarks to measure targeted process improvements. Limitations This observational, cross-sectional cohort study has several limitations. The data are self-reported by facilities and the results were not validated by the study sponsor. Moreover, the variables collected have changed over time, limiting ability to compare some results from one year to the next and because the data were not risk-adjusted, at-risk correlations cannot be made. Finally, the aggregate study data were not collected to evaluate process improvement strategies; only individual facilities are able to perform this function as they receive more inclusive data pertaining to their particular institution. Conclusion Results of this study underscore that pressure ulcers remain a problem in all age groups and all healthcare February 2008 Vol. 54 Issue 2 53

15 settings in the US and participating Canadian facilities despite increased emphasis on pressure ulcer prevention and increased use of pressure ulcer risk assessment instruments. While the majority of patients with a pressure ulcer also were assessed as being at increased risk for developing pressure ulcers, results of this and other studies suggest that patients with a high Braden Score (low risk) may still develop pressure ulcers. Assessing early stage pressure ulcers continues to be a clinical challenge, especially in patients with dark skin tone. Of the three patient care settings studied, LTAC consistently had the highest pressure ulcer prevalence rate. Overall, most ulcers identified were in the sacral area but in LTC facilities the majority of NPUs (51.9%) were located on the lower extremities. Determining prevalence rates provides benchmarks to facilitate assessment of prevention programs; in turn, such data may assist healthcare facilities and agencies improve care. Although the aggregate study data cannot be used to calculate correlations, the prevalence rates in participating facilities have remained relatively unchanged, suggesting either that despite best efforts some pressure ulcers may be inevitable or that additional strategies are needed. Continued research and documentation of efforts and outcomes are needed to successfully address the pressure ulcer problem. - OWM References 1. Chaves LM, Grypdonck MHF, Defloor T. Pressure ulcer prevention in homecare: Do Dutch homecare agencies have an evidence-based pressure ulcer protocol? JWOCN. 2006;33(3): Gunningberg L. EPUAP pressure ulcer prevalence survey in Sweden: a two-year follow-up of quality indicators. JWOCN. 2006;33(3): Hiser B, Rochette J, Philbin S, Lowerhouse N, TerBurgh C, Pietsch C. Implementing a pressure ulcer prevention program and enhancing the role of the CWOCN: impact on outcomes. Ostomy Wound Manage. 2006;52(2): Lahmann NA, Halfens RJG, Dassen T. Pressure ulcers in German nursing homes and acute care hospitals: prevalence, frequency, and ulcer characteristics. Ostomy Wound Manage. 2006;52(2): Rodrigues I, Megie MF. Prevalence of chronic wounds in Quebec home care: an exploratory study. Ostomy Wound Manage. 2006;52(5): Schwien T, Gilbert J, Lang C. Pressure ulcer prevalence and the role of negative pressure wound therapy in home health quality outcomes. Ostomy Wound Manage. 2005;51(9): Joint Commission on Accreditation of Healthcare Organizations Long Term Care National Patient Safety Goals, Goal #14. Available at: /. Accessed January 25, National Pressure Ulcer Advisory Panel Board of Directors: Cuddigan J, Berlowitz DR, Ayello EA (eds). Pressure ulcers in America: prevalence, incidence, and implications for the future. An executive summary of the National Pressure Ulcer Advisory Panel Monograph. Adv Skin Wound Care. 2001;14(4): Nursing facts. Nursing-sensitive quality indicators for acute care settings and ANA s Safety and Quality Initiative. Available at: Accessed January 25, Frantz RA. Measuring prevalence and incidence of pressure ulcers. Adv Skin Wound Care. 1997;10(1): Bours GJ, Halfens RJ, Berger MP, Huijer Abu-Saad H, Grol RT. Development of a model for case-mix adjustment of pressure ulcer prevalence rates. Med Care. 2003; 41(1): Berlowitz DR, Brandeis GH, Anderson JJ, et al. Evaluation of a riskadjustment model for pressure ulcer development using the Minimum Data Set. J Am Geriatr Soc. 2001;49(7): Dellefield ME. Organizational correlates of the risk-adjusted pressure ulcer prevalence and subsequent survey deficiency citation in California nursing homes. Res Nurs Health. 2006;29(4): Amlung SR, Miller WL, Bosley LM. The 1999 national pressure ulcer prevalence survey: a benchmarking approach. Adv Skin Wound Care. 2001;14(6): Halfens RJ, Bours GJ, VanAst W. Relevance of the diagnosis Stage 1 pressure ulcer : an empirical study of the clinical course of Stage I ulcers in acute care and long-term care hospital populations. J Clin Nurs. 2001;10(6): National Pressure Ulcer Advisory Panel. Pressure Ulcer Stages Revised by NPUAP. Available at: Accessed August 3, Braden BJ. Risk assessment in pressure ulcer prevention. In: Krasner DL, Rodeheaver GT, Sibbald RG. Chronic Wound Care: A Clinical Source Book for Healthcare Professionals. Wayne, Pa: HMP Communications;2001: Woodbury MG, Houghton PE. Prevalence of pressure ulcers in Canadian healthcare settings. Ostomy Wound Manage. 2004;50(10): Bours GJJW, Halfens RJG, Abu-Saad HH, Grol RTM. Prevalence, Prevention, and Treatment of Pressure Ulcers: Descriptive Study in 89 Institutions in the Netherlands. Res Nurs Health. 2002;25(2): Bours GJJW, Laat ED, Halfens RJG, Lubbers M. Prevalence, risk factors and prevention of pressure ulcers in Dutch intensive care units. Intensive Care Med. 2001;27(10): Russo CA, Elixhauser A. Hospitalizations related to pressure sores. Statistical Brief #3, AHRQ Healthcare cost and utilization project, April Available at: Accessed January 25, Department of Health and Human Services. Objective In: Healthy People Washington DC: US Department of Health and Human Services; Available at: Accessed January 3, Lyder CH. Pressure ulcer prevention and management. Ann Rev Nurs Res. 2002;20: Granick MS, McGowan E, Long CD. Outcome assessment of an inhospital cross-functional wound care team. Plast Reconstr Surg. 1998;101(5): Hopkins B, Hanlon M, Yauk S, Sykes S, Rose T, Cleary A. Reducing nosocomial pressure ulcers in an acute care facility. J Nurs Care Qual. 2000;14(3): O Brien SP, Wind S, van Rijswijk L, Kerstein MD. Sequential biannual prevalence studies of pressure ulcers at Allegheny-Hahnemann University Hospital. Ostomy Wound Manage. 1998;44(3 suppl A): 78S 89S. 54 OstomyWound Management