Skin integrity is identified as a measure of nursing care quality, yet pressure ulcers (PU) are a major health problem. Increasingly, patients admitted to acute care hospitals are older and possess multiple risk factors for PU development. The primary goal of this quality improvement project was to improve PU prevention and management. A secondary goal was to improve nursing documentation of PU prevention and intervention strategies. A preintervention, intervention, postintervention design was employed; the intervention included formal and informal education. Data were collected in 2 phases; Phase I occurred in a critical care unit and Phase II occurred on a medical-surgical unit. Demographics, PU incidence, risk factors, and comorbidity data were collected, along with nursing specific data. The results of this quality improvement project suggest that educational programs improve nursing care and documentation, and they raise additional questions regarding the concept of "avoidable versus unavoidable" PU in high-risk populations.
Background
Pressure ulcers (PU) represent a major health challenge, particularly for older adults in acute care settings.1 Given the aging of the population, PU are likely to become a problem of increasing proportion in the near future.2 While reported PU incidence rates vary widely, a recent national study of 116 acute care facilities in 34 states (n = 17,560) reported that 7% of patients developed PU.3 The majority (73%) occur in patients aged 65 years and older. PU prevalence in intensive care unit (ICU) patients has been reported at 28.7%4 as compared to 21.6% in acute care patients.5 PU are associated with significant economic burden, and costs continue to rise.6 After adjustment for admission characteristics associated with increased costs, the mean adjusted hospital costs for patients with PU has been reported at $14,2606; expenses for the 1 million patients with PU in the United States annually are estimated at 6.5 billion dollars a year.7 Affected patients experience increased length of stay,8 pain, increased morbidity and mortality,9,10 and diminished quality of life.11 Research by Brown12 is consistent with other research that identifies a link between nosocomial PU and mortality. Therefore prevention, early detection, and optimal treatment of PU are essential activities that traditionally lie exclusively within the nursing domain.
The Institute for Healthcare Improvement (IHI) has identified PU prevention as one of their primary goals in their "Save 5 million lives" campaign. The IHI believes that significant reductions in PU can be realized if hospitals implement evidence-based best practices.13 The Centers for Medicare and Medicaid Services (CMS) has identified PU prevention as a goal for 2007. PU are identified as potential care management events, and under the proposed plan, Stage 3 or 4 PU acquired after admission to a healthcare facility will result in reduced or eliminated payment for hospital-acquired pressure ulcers (HAPU).14 The American Nurses Association (ANA)15 identifies skin integrity as a measure of nursing care quality. Recently, the National Quality Forum16 endorsed a set of 15 nursing-sensitive performance measures, including PU incidence as a means of measuring nursing care quality. However, considerable debate is evident in the literature as to whether all PU are preventable, and wound care specialists also question the healing potential of specific PU.2,12
PU Causes and Risk Factors
PU are areas of localized tissue destruction caused by compression of soft tissue between a bony prominence and an external surface for a prolonged period of time.17 The most important contributing factors in the pathogenesis are unrelieved pressure and shear. Development of PU is directly linked to tolerance of tissues for pressure as well as the intensity and duration of that pressure. Tissue tolerance is determined by both intrinsic factors (those that influence the integrity of the skin) and extrinsic factors, which reflect the degree to which skin is exposed to moisture, friction, and shear.18 PU can develop within 24 hours but can take as long as 5 days to present,17 and tend to be more extensive than they appear initially.2 The current staging system, developed by the National Pressure Ulcer Advisory Panel,19 is the most commonly used method, but considerable debate related to it is evident.2,20 The National Pressure Ulcer Advisory Panel has recently updated the pressure ulcer staging system, and those recommendations are now being incorporated into CMS and other appropriate documents.21 The revised system identifies suspected deep tissue injury, adds further descriptions for each of the 4 stages, and also includes criteria of unstageable PU.
An essential step in preventing PU is identification of persons who are at risk.22 Studying 75,168 older people, Margolis and colleagues23 identified numerous medical conditions associated with PU development, and urged clinicians to be aware of the higher risk of even ambulatory medical patients. Intrinsic factors include, but are not limited to: advanced age, altered sensory perception, altered mental state, impaired circulation, anemia, comorbidities, diabetes mellitus, longer length of stay in a facility, malnutrition, immobility, male gender, black race, surgical intervention, weight,2,4,23-32 and emergency room stays.33 Critical care patients are at particularly high risk; they often experience multisystem failure and generally have limited mobility.34 PU in critical care units may represent the single most underrated medical problem in critically ill patients.32 While use of specialized beds may be beneficial,35 use of pressure-redistribution surfaces alone may not be adequate to prevent skin breakdown in this at risk group.34 Specific ICU risk factors include emergent admission, repeated surgical interventions, altered level of consciousness, days without nutrition, inability to turn or mobilize, length of stay greater than 7 days, reduced sensory perception, and sepsis.36-39 Extrinsic risk factors include friction, shear, and moisture.18 When multiple risk factors coexist, the risk for PU development increases accordingly. While most PU may be preventable, the inherent complexity of patients typically admitted to hospitals renders the initiation of prevention strategies increasingly challenging and, in some cases, perhaps impossible. Thirty to 50% of hospitalized patients have evidence of malnutrition,40 which is a major risk for mortality and morbidity,41 including PU development and delayed wound healing. No single diagnostic measure is available to assess malnutrition, which occurs in both overweight and underweight people,41 making detection difficult.
While more research is clearly needed to fully elucidate the risk factors for PU formation among various patient populations, we currently have evidence-based tools for predicting PU risk. Current guidelines for PU prevention include PU risk assessment upon admission to a healthcare facility and at least every 48 hours thereafter (or when there is a change in patient condition).17 For example, the Braden scale,18,42,43 which consists of 6 subscales, has been validated for the prediction of PU risk.44-48 "Cut" scores (scores at which action is required) and frequency of assessments have been recommended. Interrater reliability has been reported at r = .99 (Pearson Product moment correlations) for RNs (P < .001)42 completing the scale, and it has the advantage of being easy to use and understand. Predictive validity has been examined in various samples and settings. Bergstrom and coworkers44 noted that admission assessments were highly predictive, but assessments completed at 48 to 72 hours postadmission were even more predictive of risk.
It is critically important that available, reliable, and valid risk assessment tools are used, and used correctly, by nurses to identify patients at risk. The nurse must then promptly initiate a comprehensive program for PU prevention based on current guidelines. Prevention and treatment guidelines are readily available from such organizations as IHI13 and WOCN.16 One approach to reducing PU has been to institute educational programs targeted at nursing staff. A recent survey49 analyzed responses of 692 nurses from 48 states who were readers of Advances in Skin and Wound Care and Nursing in 2004 about preventing and treating PU. Review of selected items revealed that 44% of respondents did not know the purpose of the Braden scale, 30% were unsure about staging, and 70% stated that their basic education was not sufficient in the area of chronic wounds.
Multiple studies have examined the impact of nurse education on clinical outcomes.50-52 Hopkins and colleagues50 implemented a multicomponent intervention, including best practices and research-based protocols, facilitated by varied educational activities with nursing staff. Improvements in PU prevalence rates and severity were reported. Sinclair and associates51 reported on the effect of an evidence-based education program for PU prevention. A quasi-experimental design was used to measure nurses' knowledge before and after a standardized educational workshop. The program included risk assessment, prevention, and wound staging, and was effective in increasing nurses' knowledge scores. Wolverton and coinvestigators39 implemented a performance improvement project that included weekly skin care rounds, specific staff education, and reinforcement of skin care policies and standards. Incomplete nursing documentation hindered data collection, and was identified as a focus area. Though the literature is limited, it appears that nurse-focused, comprehensive educational interventions have the potential to improve clinical outcomes. The primary goal of this quality improvement (QI) project was to improve PU prevention and management. A secondary goal was to improve nursing documentation of skin prevention and intervention strategies.
Setting and Background
This QI project occurred at The Miriam Hospital (TMH), a 247-bed, not-for-profit, acute care hospital in Providence, Rhode Island. TMH is 1 of 4 healthcare affiliates of the Lifespan system, created in 1994, and is a Brown Medical School teaching facility. The facility provides a broad range of primary, secondary, and tertiary clinical services, with nationally recognized research and teaching programs. TMH was the ninth hospital to obtain Magnet status, and earned a second recertification in March 2006.
Several factors influenced the development of this project. Reducing PU had been identified as a quality improvement goal by Lifespan. A study was on-going related to examining the occurrence, presentation, and timing of PU development in patients undergoing cardiac surgery. That study, reported in detail elsewhere,52 included an extensive educational intervention with nursing staff and was successful in improving clinical outcomes. Also, the Certified Wound and Ostomy Nurse (CWON) and Clinical Nurse Specialist (CNS) were in the process of implementing a comprehensive preventive plan that included moving to daily Braden scale assessment, revising all relevant policies and procedures to reflect an evidence base, and instituting documentation forms for on-going use. The decision was made to extend the educational intervention reported in the prior study, in particular to include both formal and informal training and on-going unit-based support.
Sample
The sample consisted of patients admitted to 2 units, a 16-bed critical care unit and a 30-bed general medical-surgical unit. The critical care unit is a medical-surgical unit with a high proportion of cardiac patients but other diagnoses as well, and is 1 of 4 critical care units. Convenience sampling was used; all patients admitted to these units during the study period were included.
Procedures
The QI project was divided into 2 phases. Phase I took place in critical care unit (Unit I) and Phase II occurred in a 30-bed medical-surgical unit (Unit II). Both phases occurred over a 4-month time period, from November 2004 to February 2005 (Phase I), and April 2005 to July 2005 (Phase II). Within each phase, preintervention data collection occurred over a 4- to 6-week period, followed by a formal and informal educational intervention. It took about 4 to 6 weeks for the majority of RN staff to attend the formal component of the intervention. After the formal program was completed, postintervention data were collected over a 4- to 6-week time period.
All data were collected by a nurse hired specifically for this project. This RN received 1 month of intensive training, predominantly by the CWON. Selection of study variables was based on review of the literature and clinical experience. Variables of interest included demographic data and factors associated with PU risk such as age, altered mental/sensory status, emergent admission, inadequate nutritional status, length of stay, multiple comorbidities such as diabetes mellitus, peripheral vascular disease, sepsis, surgical intervention, transfer from another facility, and ability to ambulate or turn. Nursing specific data included Braden scores at admission and on-going nursing interventions designed to preserve skin integrity such as back care, repositioning, use of special pressure redistribution surfaces, presence of a PU prevention/treatment plan, and use of an available wound documentation form. It is important to note that we were relying on nursing documentation to identify these nursing strategies, which was often challenging. Related to variables such as back care, we were looking for documentation of any kind of skin-related back care such as inspection or back rub. In the ICU Phase (I), the serum albumin and prealbumin levels were determined. Clinically significant malnutrition was diagnosed if the serum albumin was less than 3.5 mg/dL. Prealbumin has a very short half-life and is particularly useful in monitoring acute changes in nutritional status of the critically ill; normal levels range from 15 to 32 mg/dL. While reviewing the record, the research nurse recorded the Braden score documented by the nurse. She then completed a second Braden score as soon as possible, usually within 24 to 48 hours. Informal comparison of these scores revealed that the research nurse tended to score patients as at higher risk than the score assigned by staff nurses.
Medical record review was completed on actual or suspected PU related to unit and date of origin, staging and progression during the hospital stay, nutritional status, skin care assessment, planning, documentation, any in-hospital transfers, and discharge disposition. While the records were reviewed in significant detail, no attempt was made to complete a formal root cause analysis and to determine whether PU were clinically avoidable or unavoidable.
Intervention
A formal continuing education unit (CEU) component was developed by the CWON. The CEU included review of the objectives of the program, followed by discussion of the background and rationale for the project, including the clinical issues leading to its development. Data collection procedures were identified, and key literature related to PU risks was reviewed. A video, "The Braden Scale for Assessing Pressure Ulcer Risk53" that provides detailed information about use of the Braden Scale18,42,43 was viewed and discussed. After viewing the video, relevant research was briefly reviewed, followed by presentation of a case study. Participants were asked to use the Braden Scale to rate the patient presented in the case study. This case study also provided an opportunity to highlight and discuss risk factors not specifically evaluated when using the Braden Scale, such as comorbidities, age, length of stay, occurrence of multiple surgeries during hospital course, and previous history of neuropathy or foot ulcers. Nurses were specifically encouraged to use their clinical experience and intuition while employing the Braden Scale scoring system and to consider hidden risk factors while independently scoring the patient.
Nurses were encouraged to carefully review patients' history and make some assumptions about the possibility of risk factors that might not be readily apparent or reported. For example, based on the literature and clinical experience, patients with a long history of diabetes mellitus have a significant incidence of neuropathy but are often unaware of diminished sensation. Assuming that such a patient is at potentially higher risk would trigger interventions to prevent heel PU. Another example would be assuming that a patient transferred from another facility had been exposed to some degree of immobility prior to the current admission and thus would benefit from immediate implementation of prevention strategies even if no signs of breakdown were apparent. Nurses were asked to use the Braden scale to score the presented case, and participants' scores were compared. Extensive discussions about rating differences occurred. Nurses were then asked to score the case again, and the range of Braden scores narrowed in all cases. Further individualized instruction about optimal use of the Braden scale with actual patients occurred in the next stage of the intervention.
The formal CEU also included interventions for prevention of PU. Specific strategies included heel elevation; removing compression hose early in the postoperative period; use of chair cushions and padding; repositioning in chairs and limiting position to 2 hours; encouraging out of bed for all meals or limited sitting times at 90 degree elevations; importance of early nutrition; talking about and asking about skin at handoffs.
The final component of the formal education program focused on documentation of PU assessment and related care. Controversies regarding staging were reviewed, and emphasis was placed on the importance of documenting a PU as unstageable if the base of the wound is not visible. Participants were shown slides of several wounds along with the definitions of stages; then the group was asked to discuss staging on numerous slides. Staging over areas with very thin skin such as ankles as well as wounds that are obviously deep, but with small openings, along with methods to assist in the staging of these wounds were discussed. Techniques for documenting and measuring existing wounds were reviewed. Concepts of wound length and width as related to head-to-toe orientation or hip-to-hip orientation were discussed. A uniform way to measure depth and undermining was illustrated, with accompanying illustrative examples. Difficulties associated with describing wounds with complex surface configurations or variable depths were discussed in detail. The importance of accurate documentation and compliance with the hospitals' policies was also emphasized. A Wound Documentation Form was issued to each participant, and facility policies related to wound documentation were reviewed. Staff were encouraged to include the skin (the largest organ) in their narrative nursing documentation, intershift reports, transfer reports as patients move from unit to unit, and in completion of discharge forms. Participants were asked to offer interventions that might be used with their patient population to prevent pressure ulcers, and a turning log and adjustments to the clinical documentation records were suggested.
Concurrent with the formal educational intervention, and on-going during the postintervention period, the project nurse initiated and maintained a "shoulder-to-shoulder" presence on the units. A team-building approach was used, and emphasis was placed on developing trusting relationships in a supportive environment. The project nurse served as an informal educator and role model, facilitating routine skin assessment as well as preventive and treatment-oriented skin interventions. She routinely rotated to all 3 shifts, working directly with the staff nurse providing care to assess skin and then develop a plan of care for prevention and treatment. Emphasis was placed on preventive measures and subsequent documentation, including conducting and using Braden scores to tailor care; frequent turning and early mobilization; importance of adequate nutrition; importance of early, consistent intervention with Stage I ulcers; management of PU by stage; encouragement to consult the CWON; using pressure reducing devices in beds and chairs. An outgrowth of this work was development of standardized, skin-specific care plans. Digital photography, routinely used by the CWON for wound documentation, was used to monitor progress of treatment modalities. The project nurse instituted a skin care products cart on each nursing unit, and product representatives were invited to provide in-services for the nursing staff. Frequently used products requiring a physician order were added to the automated medication and supply management system to facilitate access. The project nurse encouraged the use of low air loss mattress overlays in combination with specially designed underpads when appropriate. Educational materials were obtained and stored on the nursing unit. The project nurse strove to create credibility and trust when interacting with staff nurses. Because staff nurses were supported in a nonthreatening manner, they sought out the project nurse freely and willingly.
Reliability checks related to Braden scores were conducted at regular intervals. Specifically, the research nurse routinely completed Braden scores with nursing staff when possible, and provided individualized instruction and feedback as indicated. The research nurse noted that, over time, scoring of staff nurses more closely corresponded to that of the research nurse.
Data Analysis
Descriptive statistics appropriate to the level of QI project variables were measured. Differences between preintervention and postintervention groups were analyzed using ANOVA.
Results
Phase I (Critical Care Unit)
Data were collected on 62 patients during the preintervention and postintervention periods. Thirty-two men and 30 women comprised the preintervention group, and 35 men and 27 women formed the postintervention group. About half in each group had been transferred from another unit or another facility. Comparisons of mean scores of preintervention and postintervention groups on selected variables are summarized in Table 1. Significant differences were found in the length of stay. Mean weights, while admittedly a crude measure, suggested that neither group was underweight. Patients in both groups had low albumin levels. They averaged 5 comorbidities and 5 PU risk factors. Mean admission Braden scores did not differ between the groups. More than twice as many people in the postintervention group died (22 vs 10; P = NS). PU incidence by stage and selected factors associated with PU risk are illustrated in Table 2. Nine patients developed 12 PU in the preintervention group, and 8 patients developed 13 PU in the postintervention group; these differences were not statistically significant. The mean number of PU risk factors for both groups was 5 and approximately 75% of patients in both groups required mechanical ventilation.
Analyses were then performed to compare patients that did not develop PU to those that did (Table 3). The mean length of stay (LOS) for patients in the preintervention group who developed HAPU was significantly greater (P = .03) than those who did not develop PU; the same was found in the postintervention group (P = .007). The mean hospital length of stay for those who developed a PU was longer than for the group overall (overall mean LOS preintervention = 9.9; overall mean LOS postintervention 12.0; P = NS). Based on timing and presentation, we suspected that one HAPU found in the preintervention group and 2 nosocomial PU in the postintervention group may have originated during surgery. Significant differences included age and length of stay. Patients with HAPU were significantly younger, but they had significantly longer length of stay than those who did not develop PU. Though not statistically significant, patients in the preintervention group who developed HAPU tended to weigh more and have lower Braden scores. Patients in the postintervention group that developed HAPU had a significantly longer length of stay and more comorbidities than those who did not develop a PU.
Data specific to various aspects of nursing care are illustrated in Table 4. We used interrater reliability scores to determine whether the formal and informal educational intervention improved nurses' ability to calculate a Braden score that matched the project nurse's score. As can be seen, the staff nurses' scores more closely approximated the experts' score postintervention. Pressure ulcer prevention/treatment care plans existed in 33% of patients in the preintervention group as compared to 77% in the postintervention group. Use of overlays, documentation related to bony prominences, and use of our Wound Documentation Form for patients with PU increased to 100% postintervention.
In the preintervention group, albumin levels were drawn for 8 out of 9 patients with HAPU. Values ranged from 1.1 to 2.4 grams/dL (mean: 1.65 grams/dL). The number of albumins drawn over the patients' hospital course ranged from 0 to 6, with a mean of 2.1 measurements. Prealbumin levels were drawn on 3 of the 9 patients. Nutritional interventions were initiated on average within 1.6 days of admission. Postintervention, albumin levels were drawn for 5 of 8 patients who developed HAPU. Levels ranged from 1.5 to 2.9 grams/dL, with a mean of 1.94 grams/dL. Prealbumins were obtained for 4 of the 8 patients, and ranged from 5.3 to 22.9 grams/dL, with a mean of 12.7 grams/dL. The average time to initiation of nutrition was 2.14 days in the postintervention group.
Phase II: Medical-Surgical Unit
Data were collected on a total of 71 patients during the preintervention and postintervention periods. There were 44 men and 27 women in the preintervention group, and 26 men and 45 women in the postintervention group. Table 5 compares characteristics of patients in the preintervention and postintervention groups. Significant differences were detected between length of stay, weight, and emergency department hold time. About one-third of patients in the preintervention group and about half of the postintervention group had diabetes mellitus. Almost three-fourths of patients in both groups were aged 65 years and older and had peripheral vascular disease. More than half had impaired mobility, nearly one-third suffered from impaired mental status, and half were incontinent of urine or stool. Approximately 25% were transferred from other facilities and nearly 75% had been held in the emergency department.
Table 6 summarizes HAPU by stage. Fewer PU occurred in the postintervention group, despite a higher number of associated factors and substantially longer length of stay. Nevertheless, differences were not statistically significant. One patient in each group developed a PU within 2 to 3 days after cardiac catheterization. Statistically significant differences were found in the postintervention group for documentation of repositioning (P < .001), out of bed activity (P = .006), heel elevation (P = .03), back care (P = .002), and bony prominence assessment (P < .001).
Table 7 compares characteristics of Phase II subjects who developed HAPU to subjects who remained free of PU. Patients in the preintervention group who developed nosocomial PU were significantly older and they had significantly lower Braden scores. They tended to weigh less than those that did not develop a PU, but this difference was not statistically significant. Postintervention group patients who developed PU had a significantly longer length of stay and lower admission Braden scores.
Discussion
More than half of patients in the critical care (Phase I) group were transferred from other facilities. This is noteworthy, since prior admission to a health care facility and its associated length of stay has been shown to increase the risk of PU development.54 Subjects in this study had multiple factors associated with an increased risk for PU including diabetes mellitus, peripheral vascular disease, and low serum albumin levels. More than half of the sample were mechanically ventilated and approximately 75% of those who developed a HAPU were mechanically ventilated. This finding corroborates that of other researchers55 who also found that mechanically ventilated patients were at higher risk of developing PU.
We also found that "clashing standards of care" potentially influenced the likelihood of PU development. During the intervention, nurses were encouraged to keep head of bed (HOB) at, or below, 30[degrees] or at the lowest degree possible, consistent with WOCN Clinical Practice Guidelines for the prevention of PU.17 In contrast, evidence-based recommendations for the prevention of ventilator-associated pneumonia include semirecumbent positioning (elevation of HOB to 45[degrees]).56 The dilemma is clear; how to resolve it at the bedside is not. Further research is needed in this area. In the interim, nurses need to be aware of practice recommendations relevant to a patient's care but they must also tailor the plan of care to the unique characteristics and risk factors of the individual patient. An interdisciplinary approach to clinical problem solving would also be beneficial.
Patients who experienced HAPU had longer LOS and higher mortality rates. We observed that approximately one-third of preintervention patients and two-thirds of postintervention group patients with HAPU expired. Though limited, some research has supported that PU are significantly associated with mortality in ICU57 as well as in community residing58 older adults. In a few cases, PU developed in patients with multisystem failure who were clearly "end-stage" despite their ICU status. The ability to prevent PU in these cases is questionable.59 Langemo and Brown60 described end-stage skin failure as a type of skin death that frequently occurs in complexly ill people despite good care. Though nutrition interventions were initiated in a reasonable time period, albumin levels were suggestive of prolonged malnutrition, an important PU risk factor.
Prealbumin levels, thought by some to be more valuable as a method for measuring nutritional status,61 were found to significantly predict nutritional status whereas albumin was not. We believe that routine prealbumin screening would be a cost-effective means to identify hospitalized patients at high risk of developing PU secondary to malnutrition.
There was not a statistically significant difference in PU incidence in either group following the educational intervention, though there was a trend toward improvement. It is speculated that the small sample size of the HAPU groups limited our ability to identify statistically significant differences. Statistically significant differences in nursing care, including skin-specific care plans, use of overlays, skin-care documentation, and use of the Wound Documentation Form, increased after the intervention. These findings, as well as our anecdotal experience, support the hypothesis that many HAPU are indeed unavoidable and require a team-based interdisciplinary approach.
PU prevention and management is extremely complex and often not exclusively under nursing control. For example, in our sample, OR and cardiac catheterization experiences, many of which were preceded by hospitalization at another facility and associated with prolonged immobility, contributed to PU risk. In addition, while recommendations for dietary interventions were generally present, they were often not adopted by the medical team despite recommendations by nursing and dieticians, a finding also reported by Black.62 An interesting observation from the critical care phase in particular was that skin assessment (other than presence of incisions/tubes) or specific inclusion of PU in the physical examination or in documentation within the medical problem list was lacking. This tendency was also supported by Black62 who reported that physician documentation of PU risk or presence is sparse or absent.
Phase Two (Medical-Surgical)
Patients who developed HAPU in both preintervention and postintervention tended to be transferred from other facilities, incontinent of urine or stool, immobile, and have impaired mental status. Though the PU incidence did not change significantly following the intervention, improvement in documentation of nursing prevention strategies, including out of bed activity, repositioning, elevation of heels, back care, and bony prominence assessment, all directly attributable to nursing care, did occur. We believe that at least some of these PU occurred despite appropriate nursing intervention and were thus unavoidable. These findings further support Black's62 assertions that unavoidable pressure ulcers, those that develop despite accurate risk assessment and preventive care, do indeed occur.
During the preintervention period in both the critical care and medical-surgical groups, several cases were identified where patients were admitted to the hospital with a PU; however, that occurrence was not documented by nursing within the first 24 hours of admission, and was thus considered to be a HAPU. This problem was identified and did not occur in either postintervention group. Accurate and consistent documentation of PU preventive strategies is critical, since effective in 2008 with the pending CMS initiative concerning payment for hospital errors, HAPU will not be reimbursed unless they are considered to be unavoidable14 (Box).
Conclusions
The primary goal of this project was to improve PU prevention and management; a secondary goal was to improve nursing documentation of skin prevention and intervention strategies. We believe that these goals were accomplished. While the actual number of PU preintervention and postintervention did not change dramatically, several critical factors must be considered. We believe that nurse-sensitive PU indicators were positively impacted; improvements in documentation of nursing assessments, preventative strategies, and interventions were demonstrated postintervention in both phases. Modifications to nursing documentation forms were implemented to facilitate more accurate, consistent charting.
KEY POINTS
[check mark] Development of pressure ulcers is complex and multifactorial.
[check mark] A comprehensive educational intervention with application and follow-up at the bedside can positively impact nurse-sensitive pressure ulcer indicators.
[check mark] Despite accurate risk assessment and preventive care, some pressure ulcers are unavoidable and challenge traditional thinking and practices.
[check mark] A comprehensive, interdisciplinary approach to prevention and management is needed.
ACKNOWLEDGEMENTS
The authors would like to acknowledge funding support provided by the Lifespan Risk Management Department, Providence, Rhode Island. The authors also wish to thank all of the staff nurses at The Miriam Hospital who participated in this project; without their support, the project could not have been completed and would not have been a success. Support provided by the Department of Nursing is also acknowledged.
References