Skip to main content

A Comprehensive Program to Prevent Pressure Ulcers in Long-Term Care: Exploring Costs and Outcomes

Empirical Studies

A Comprehensive Program to Prevent Pressure Ulcers in Long-Term Care: Exploring Costs and Outcomes

Index: Ostomy Wound Manage. 2002;48(4):52-62.

   Pressure ulcers contPressure ulcers continue to be prevalent and costly for long-term care facilities. A recent document published by the National Pressure Ulcer Advisory Panel revealed a pressure ulcer incidence rate of 2.2% to 23.9% in long-term care.1

Although the cost of pressure ulcer prevention remains elusive, costs associated with their treatment have been conservatively estimated to range from $500 to $50,000 per ulcer,2 with more severe wounds being significantly more expensive to manage than less severe ulcers.3 These costs do not account for the pain and suffering commonly associated with these ulcers. Presently, approximately 1.5 to 3 million adults suffer with pressure ulcers.4 Given the high incidence rates, the need to address pressure ulcer prevention has become paramount. Most recently, the U.S. Centers for Medicare and Medicaid Services (formerly the Health Care Financing Administration) included pressure ulcers as one of three sentinel events for long-term care5; therefore, the formation of a pressure ulcer or subsequent deterioration of a pressure ulcer can lead to significant monetary penalties (maximum $10,000/day) in long-term care.

   Pressure ulcers have become so common in long-term care that federal regulations now articulate pressure ulcer standards or guidelines of care and prevention. In May 1992, the Agency for Health Care Research and Quality (formerly the Agency for Health Care Policy and Research) released Clinical Practice Guidelines6 for the prevention of pressure ulcers. These guidelines provide the healthcare community with current practice parameters based on expert opinion and synthesis of scientific evidence. The Joint Commission for Accreditation of Health Care Organizations (JCAHO) recommends use of the AHRQ clinical practice guidelines. Moreover, the Centers for Medicare and Medicaid Services are using the guidelines to create policy and reimbursement criteria and to direct the federal and state survey process of long-term care facilities.

   The AHRQ guidelines for pressure ulcer prevention are meant to be living documents -- that is, providers should implement these guidelines in a cost-effective manner that offers intelligent wound care based on available evidence. How best to implement the standards in long-term care continues to be a challenge, even for discerning administrators and health professionals attempting to maximize resource utilization and balance quality pressure ulcer care. Thus, the purpose of this study was to examine the effectiveness of comprehensive protocols of care (SOLUTIONS®, ConvaTec, a Bristol-Myers Squibb Company, Princeton, NJ) focused on risk factors identified by the Braden Scale to prevent pressure ulcers in two long-term care facilities.

Methods

   Study design. This quasi-experimental study was conducted in two phases. In Phase I, retrospective medical record abstraction methods were used to ascertain usual (control) pressure ulcer prevention care before the implementation of the comprehensive prevention protocol of care. The retrospective medical record review also was used to determine the cumulative incidence of pressure ulcers for both long-term care facilities. In Phase II, a consistent patient risk assessment (Braden Scale) and the comprehensive protocols of care were introduced in both long-term care facilities (experiment). This study was approved by the Yale Institutional Research Review Board before data collection.

   Participants and setting. After obtaining appropriate Internal Review Board approval, a convenience sample was used to procure participants in Phase I (January 1999 to July 1999) and Phase II (August 1999 to December 1999) of the study. The only inclusion criterion for the study was that residents were identified at risk for pressure ulcers by having at least one pressure ulcer risk factor (identified by the literature) or were identified at risk by a validated risk assessment tool (Norton Scale [Phase I]7and Braden Scale8 [Phase II]). Approval by the resident's primary care physician and written informed consent also were obtained. Because no pressure ulcer prediction tool captures all risk factors, the authors included additional risk factors noted in the literature.9

   The study was conducted in two long-term care facilities with total bed capacities of 150 (Facility A) and 110 (Facility B), respectively. These two long-term care facilities were selected because they were owned by the same corporation, had a high incidence of pressure ulcers, and did not use comprehensive protocols of care. All units (eg, rehabilitation, subacute care, and skilled care) were involved. Because both facilities had received pressure ulcer citations from the Connecticut Department of Health, the administrative, medical, and nursing staffs were motivated to improve their pressure ulcer prevention programs.

   Data collection. In Phase I, a medical record abstraction was conducted by a trained research nurse to identify all nursing home residents at risk for pressure ulcers and to determine the usual pressure ulcer preventive care provided at both long-term care facilities before the intervention.

   The medical record abstraction included all medical records from January 1, 1999 to June 30, 1999. Medical records contained information obtained from the Minimum Data Set (MDS) and ICD-9 code for pressure ulcers, including resident demographics, medical diagnoses, and clinical characteristics. All sections of the medical records (eg, nursing assessment forms, medical and nursing notes, and laboratory results) were used to ascertain pressure ulcer data. Demographic data included: age, gender, race, and source of facility admission. Primary and secondary medical diagnoses also were collected. The clinical characteristics data obtained included: mobility status, urinary and fecal incontinence, motor deficits, dry skin, history of fractures or pressure ulcers, and use of pressure ulcer risk assessment tools. Specific preventive interventions (eg, turning every 2 hours and use of support surfaces) and data related to absence or presence of a pressure ulcer (for establishing incidence rate) also were obtained.

   To calculate the percentage of at risk population, the medical records of all residents residing in the two long-term care facilities at the time were included. A participant was determined to be at risk by the research nurse if at least one risk factor for pressure ulcer development as identified in the literature was noted (eg, bedbound, chairbound, urinary/fecal incontinence, albumin < 3.5mg/dL). Participants also were identified as at risk by the Norton scale used by both long-term care facilities.

   Before data collection, three charts were randomly selected to measure interrater reliability between the research nurse and the principal investigator. Reliability coefficients were high for all data elements, with Kappa values ranging from 0.8 to 1.0, indicating excellent reliability. The staff nurses were consulted only when a discrepancy in the medical record was noted. Each medical record was reviewed up to 5 months retrospectively from the date of the review to obtain the usual pressure ulcer preventive care provided by the two long-term care facilities. During Phase II of the study, resident demographics (eg, age and gender), medical diagnoses (primary and secondary), comorbidities, Braden Scale for patient pressure-ulcer risk, and the Pressure Sore Status Tool10 to monitor pressure ulcer status were collected. In this phase, independent medical record data on preventive care were collected automatically as part of the comprehensive protocols of care employed.

   Pressure ulcer protocols of care (Phase II).
  1. The program
  The wound and skin care program was developed and validated11 by wound care experts on behalf of the sponsoring company and provided the foundation for the comprehensive pressure ulcer protocols of care. This comprehensive wound program presents protocols of care to help prevent and manage chronic wounds (eg, pressure ulcers, diabetic ulcers, arterial ulcers, and venous stasis ulcers). The program encourages the systematic use of patient risk, skin/wound assessment, and nutritional assessment through validated algorithms that offer care plans based on each type of assessment. The program does not recommend specific brand products related to dressings, pressure-relieving devices, or nutritional supplements. Rather, broad classifications are recommended (ie, hydrocolloid), enabling the long-term care facility staff to customize the care plan to their specific products that correlate with the recommended classification.

  This study only employed the protocols and care plans section related to pressure ulcer prevention. This section is comprised of recommendations from the AHRQ guidelines on the prevention of pressure ulcers and wound expert opinions, as well as literature related to the prevention of pressure ulcers. More specifically, the program for pressure ulcer prevention is comprised of the Braden Scale for the Prediction of Pressure Sores8 and its corresponding care plan algorithms (based on the level of pressure ulcer risk). A major strength of this program is that the pressure ulcer risk assessment and associated algorithms are validated.2, 11-14 The nutritional supplementation and pressure-relieving devices were standardized for both long-term care facilities. Finally, the program also includes a continuing education module for nurses and physicians on the treatment and prevention of chronic ulcers. Note: Before implementing the program, neither facility consistently used assessment tools or any pressure ulcer prevention protocols.

  2. Education
  Two months before implementing the comprehensive prevention program, the research team completed a series of educational sessions for the nursing staff. Employees on all three shifts at both long-term care facilities attended mandatory sessions. Registered nurses (RNs) were educated on how to use the Braden Scale for the Prediction of Pressure Sores and how to generate the prevention care plans. According to published research, the Braden Scale is most predictive when used by registered nurses.15 Before implementing the Braden scale, both facilities were using the Norton scale haphazardly. Registered nurses and licensed practical nurses (LPNs) were educated on a variety of skin care products and the correct use of these products. Educational sessions also were conducted for the certified nursing assistants (CNAs). These sessions included content on basic pressure ulcer prevention and skin care. Demonstrations on how to give a bath with return demonstrations also were provided to ensure proper use of skin care products. One educational session was held for the medical staff to review the study protocol at both long-term care facilities.

  3. Skin consultations
  Coupled with the comprehensive protocols of care, a certified Wound, Ostomy and Continence Nurse was available to the two long-term care facilities on a limited basis (2 hours/week or 8.67 hours/month) for consultation as part of the sponsoring company's normal practice for institutions starting the program. To help empower the nursing staff, the CWOCN consultant made recommendations only.

   Phase Two procedures. Each resident in the two long-term care facilities was evaluated for pressure ulcer risk level by a trained RN using the Braden scale. Braden scale scores were collected bimonthly for high-risk (Braden Scale ≤ 18) and monthly for low-risk residents (Braden Scale ≥ 19). A care plan was generated for all residents with three or more pressure ulcer risk factors and for those identified at risk for pressure ulcer development using the Braden scale.

   The potential areas of prevention that could be addressed by the care plan, depending on the Braden score (level of risk), included: frequency of skin/pressure ulcer assessment, mobility, pressure relief, nutrition, moisture management, friction and shear relief, and general skin care. Data regarding preventive interventions were recorded on the care plan (by the facility RN) and validated by direct observation (by the research nurse). Each study participant was examined weekly by the facility nurse for the presence or absence of pressure ulcers and date of occurrence.

   Each facility selected its own essential skin care prevention system to implement on a daily basis, when soiling occurred, and/or as needed. The essential skin care prevention system included a skin cleanser, moisturizer, and moisture repellent. This skin care prevention system has been recommended as essential for providing basic skin care.6, 16-18 Facility A selected products from ConvaTec: Aloe Vesta® 2-n-1 Perineal/Skin Cleanser (cleanser), Aloe Vesta® 2-n-1 Skin Conditioner (moisturizer), and Aloe Vesta® 2-n-1 Protective Ointment (barrier) for their essential skin care program. Facility B used the same moisturizer and barrier cream but substituted Aloe Vesta® 2-n-1 Perineal/Skin Cleanser with Aloe Vesta® 3-n-1 Cleansing Foam (cleanser), because it could also be used as a shampoo.

   Nutrition was standardized using the Mead Johnson Nutritionals (A Bristol-Myers Squibb Company, Evansville, Ill.) portfolio. The facilities implemented a policy to give all residents extra protein during medication pass (unless contraindicated by the physician). Thus, all participants received Boost® during medication rounds. If extra supplementation (per dietary consult) was needed, Boost Plus® for oral and Deliver 2.0 for tube feeding was provided. Support surfaces from Kinetic Concepts Inc.® (San Antonio, Tex.) and Hill-Rom® (Batesville, Ind.) were used for all study participants. Participants who had a Braden scale score of ≤ 12 received an alternating air mattress and those scoring ≥ 13 received a foam overlay. Although both facilities were using support surfaces, no formalized plan to stratify support surfaces correlated to risk assessment score was used before Phase II.

   The research nurse monitored compliance with the Braden scale and generation of the prevention care plans on a monthly basis. The research nurse also randomly monitored three participants each week to determine the accuracy of the care plan to preventive interventions carried out by the nursing staff. The Director or Assistant Director of Nursing monitored the care plans on a monthly basis.

   Data analysis. Statistical analyses were performed using the SPSS System (SPSS, Inc., Chicago, Illinois). Univariate analyses were performed to describe the retrospective and prospective samples. Pressure ulcer incidence was calculated using the cumulative incidence method. Specifically, the number of subjects with new ulcers was divided by the total number of at risk residents in the sample over the 5-month period for the retrospective and prospective data points.

   Chi-square analyses were computed to examine the association between preventive interventions and level of pressure ulcer risk. Kaplan-Meier survival analysis was used to determine differences in pressure ulcer incidence over the 5-month intervention period. The log-rank test was used to determine significant differences in the stratification of the survival distribution. Kappa statistics were used to measure interrater reliability.

   Cost-effectiveness. Cost associated with pressure ulcer prevention was computed using an activity-based costing model.19 The activity-based costing module was conducted 2 months into the prevention program to decrease the Hawthorne effect --that is, the effects due to special attention. A stratified random sampling obtained a representative sample of 20 participants at high risk for pressure ulcers. A die was cast to identify 10 high-risk residents in Facility A and Facility B to complete the activity-based costing module. The activity-based costing model captured all costs attributable to prevention of the pressure ulcer.19 Thus, costs associated with all labor (eg, registered nurses) and supplies (eg, skin care products) for 1 month per resident were incorporated into the economic model.

   The cost of each specific type of support surface used on beds and wheelchairs, as well as supplies used for preventing the pressure ulcer, was procured from the purchasing department at each long-term care facility. The cost of labor for repositioning was based on the frequency of turning, nursing staff time to perform the task, the number of personnel required, and the average hourly salary of nursing staff in both facilities (RN = $22.52, LPN = $20.26, CNA = $13.20). The frequency of repositioning was determined by reviewing each subject's care plan for a 24-hour period. The nursing time required for repositioning was established at 2 minutes (via time and motion) and estimated every 2 hours based on the care plans. The cost for the nurse to perform a Braden scale assessment and generate the prevention care plan was established at 5 minutes (via time and motion), which was multiplied by the mean salary for the licensed staff at both facilities. The average costs of prevention for each subject for a month were calculated for six different categories of preventive interventions: 1) risk assessment labor, 2) skin care labor and supplies, 3) mattress support surface, 4) chair support surfaces, 5) nutrition, and 6) CNA repositioning labor.

   To determine the monthly cost of pressure ulcer preventive care for pressure-ulcer-free, high-risk residents to the facilities, a two-step process was used. First, the average cost to prevent a pressure ulcer in one high-risk resident who developed an ulcer was used. This cost was divided by the total number of high-risk residents who did not develop a pressure ulcer; this figure yielded the monthly prevention cost per resident.

Results

   Phase I. A total of 302 resident charts were reviewed during this phase of the study. Of these, 203 residents were identified as being at risk for pressure ulcers. Of the 203 at-risk residents, 65 (32%) arrived at the long-term care facilities with pressure ulcers. The mean age of residents was 82 years (SD = 12.38). The majority of residents was female (72%), white (97%), and transferred to the long-term care facilities from hospitals (78%). The most common medical diagnoses for participants at risk for pressure ulcers included connective tissue disease (68%) and coronary artery disease (66%). Clinical characteristics revealed that dry skin (90%) and urinary incontinence (76%) were the most prevalent pressure ulcer risk factors. Documentation of pressure ulcer risk for residents found to be at risk during the chart review was noted for 136 (67%) residents.

   A variety of interventions was used in the 136 residents identified to be at risk for pressure ulcers (see Table 1). For those residents at high risk (as identified by the Norton scale or having at least three risk factors) for pressure ulcers (n = 127), the majority had a skin assessment (n = 124, 97.6%), received nutritional consult on admission (n = 83, 65.4%), was placed on a support surface (n = 82, 64.6%), and was documented to be repositioned every 2 hours (n = 78, 61.4%). However, few charts documented whether the support surface was 4 inches or greater (n = 37 , 29.1%), or used an alternating air mattress (n = 27, 21.3%).

   The association between preventive interventions and level of pressure ulcer risk revealed no statistically significant association between pressure ulcer risk and nutritional consult (P= 0.564), placement on an overlay (P = 0.525), placement on an overlay ≥ 4 inches thick (P = 0.757), repositioning every 2 hours (P = 0.168), or skin assessment (P = 0.998). Thus, the findings suggest that preventive interventions were not selected based on risk assessment scores or pressure ulcer risk factors. Finally, the cumulative incidence rate during the Phase I medical record review covering January to June of 1999 was 43.3%.

   Phase II. Following implementation of the comprehensive prevention protocols of care, the cumulative 5-month incidence rate decreased from 43.3% to 28.5% (for both facilities combined). Specifically, with 121 high-risk residents in August 1999 and 125 in December 1999, the pressure ulcer incidence in Facility A significantly decreased from 13.2% (August 1999) to 1.7% (December 1999, P = 0.02) -- an 87% incidence reduction. Facility B, with 83 high-risk residents (August 1999) to 85 (December 1999) high-risk residents, also experienced a significant (P = 0.02) reduction in pressure ulcer incidence (from 15% in August 1999 to 3.5% in December 1999). No significant demographic differences were found between Phase I and Phase II. Of interest, Facility A was able to decrease its pressure ulcer incidence rates despite significant senior management changes (removal of two administrators and one Director of Nursing) in September and December (see Figure 1). In total, an estimated 614 ulcer-free residents were gained in Facility A in the 5-month intervention period (see Table 2). For Facility B, 412 ulcer-free residents were gained during the same 5-month period (see Table 2).

   The activity-based model (see Table 3) revealed an average cost-per-month to prevent pressure ulcers in one high-risk resident of $796.73 (including a one-time charge for mattress and chair overlay). This model revealed that the most expensive component of pressure ulcer prevention was the labor cost of $277.15 per month (licensed staff and CNAs). Thus, the 5-month cost to prevent pressure ulcers in one high-risk resident was $2,875.65 ($519.73 x 5 months, plus $252.00 for mattress overlay and $25.00 for chair cushion). This is a total of $2,875.65, divided by 5 months, or $575.14 pressure ulcer prevention costs per high-risk resident per month to reduce the incidence of new pressure ulcers from 13.2% or 15% per month to 1.7% or 3.5% per month. For those high-risk residents who remained ulcer-free, the cost of monthly prevention was calculated as the total number of pressure ulcer-free residents divided by cost of residents who developed pressure ulcers. This cost varied from $5.10 (August 1999) to $4.35 (December 1999) for Facility A and $7.53 (August 1999) to $6.53 (December 1999) for Facility B.

Discussion

   Using a consistent protocol of care including this comprehensive prevention program significantly reduced the incidence of pressure ulcers in two long-term care facilities. Overall, the facilities had decreases in their pressure ulcer incidence rates of 87% and 76%, respectively. Facility A's pressure ulcer incidence rates decreased from 13.2% in August 1999 to 1.7% in December 1999. Facility B's pressure ulcer incidence rate decreased from 15% in August 1999 to 3.5% in December 1999. Of note: Facility B had an outbreak of influenza in December 1999. The lack of further decreases in pressure ulcer incidence rates in Facility B may be attributed to an influenza outbreak at this facility in December 1999.

   This study supports several studies in which pressure ulcer prevention protocols were implemented and significant decreases in the incidence of pressure ulcers occurred.20-22 The use of a validated pressure ulcer prediction tool coupled with effective prevention algorithms, evidence-supported products, and staff education was essential in decreasing the pressure ulcer incidence rates in both long-term care facilities.

   The use of monitoring by the research nurse and nursing administrators may have increased the adherence by nursing staff to the protocols of care. The nursing administrators were highly motivated to decrease their pressure ulcer incidence rates; thus, some organizational cultural changes occurred. Evidence supports the hypothesis that changes such as those reported here may be transient, depending on interest by administration.20 However, despite organizational turmoil in one of the facilities (Facility A), the protocol of care implemented remained in place and pressure ulcer incidence continued to decrease. This may be related to the ease of use of the program protocols, the streamlining of products used, and/or the effectiveness of the staff education program. Further investigation into facility specific characteristics may increase understanding of the results.

   Although the literature contains several reports of effectively implemented comprehensive pressure ulcer prevention programs, studies investigating the cost of providing pressure ulcer preventive care are rare. Only one study that reports the cost of implementing an intensive pressure ulcer prevention program in a 77-bed long-term care facility was found.20 These researchers found a pre-intervention incidence rate of 23% and a post-intervention incidence rate of 5%. More interestingly, researchers in this study determined the mean cost for prevention/resident to be $91.56 + $135 for 6 months. This is far less than the authors' monthly costs estimates of $575.14 The discrepancy in total costs may be attributed to the different time period of data collection (1994 compared to 1999), region of the country (Midwest compared to Northeast), use of all residents in calculating costs compared to only high-risk residents in the present study, and variables entered into the cost models.20 However, pressure ulcer prevention can be quite costly; thus, targeted prevention should be initiated based on level of risk.

Limitations

   The study had several limitations. The staff time used in estimating labor costs may have been different in the unmeasured residents. The authors' activity-based costing module was completed well into the study period; thus, the nursing staff was familiar with all the protocols. The cost might have been higher if the authors had implemented the activity-based costing module in the beginning of Phase II when the nursing staff were getting familiarized with the protocols and products. Also, activity-based costing by the research team and long-term care staff was only completed during the day shift; conceivably, capturing staff time on the evening and/or night shifts may have yielded different results. This potential effect was minimized by using random selection in the activity-based costing model as well as standardizing the treatments on all three shifts. Finally, the Hawthorne effect may have been a factor accounting for the high level of adherence to the protocols.

Conclusion

   Pressure ulcers continue to be a common health problem in long-term care. Given the impending capitated system and the increasingly litigious nature in long-term care, preventing pressure ulcers is essential.22 The "true" cost of preventing pressure ulcers remains elusive. Additional studies are needed to investigate whether prevention is more cost-effective than treatment.
The authors demonstrated that the use of validated, comprehensive prevention protocols of care, including the program described here, significantly reduced the incidence of pressure ulcers in two long-term care facilities. In addition, they found that standardizing pressure ulcer preventive care with targeted interventions based on risk level was much more cost-effective over the 5-month intervention period.

   Pressure ulcers continue to be prevalent and costly for long-term care facilities. A recent document published by the National Pressure Ulcer Advisory Panel revealed a pressure ulcer incidence rate of 2.2% to 23.9% in long-term care.1 Although the cost of pressure ulcer prevention remains elusive, costs associated with their treatment have been conservatively estimated to range from $500 to $50,000 per ulcer,2 with more severe wounds being significantly more expensive to manage than less severe ulcers.3 These costs do not account for the pain and suffering commonly associated with these ulcers. Presently, approximately 1.5 to 3 million adults suffer with pressure ulcers.4 Given the high incidence rates, the need to address pressure ulcer prevention has become paramount. Most recently, the U.S. Centers for Medicare and Medicaid Services (formerly the Health Care Financing Administration) included pressure ulcers as one of three sentinel events for long-term care5; therefore, the formation of a pressure ulcer or subsequent deterioration of a pressure ulcer can lead to significant monetary penalties (maximum $10,000/day) in long-term care.

   Pressure ulcers have become so common in long-term care that federal regulations now articulate pressure ulcer standards or guidelines of care and prevention. In May 1992, the Agency for Health Care Research and Quality (formerly the Agency for Health Care Policy and Research) released Clinical Practice Guidelines6 for the prevention of pressure ulcers. These guidelines provide the healthcare community with current practice parameters based on expert opinion and synthesis of scientific evidence. The Joint Commission for Accreditation of Health Care Organizations (JCAHO) recommends use of the AHRQ clinical practice guidelines. Moreover, the Centers for Medicare and Medicaid Services are using the guidelines to create policy and reimbursement criteria and to direct the federal and state survey process of long-term care facilities.

   The AHRQ guidelines for pressure ulcer prevention are meant to be living documents -- that is, providers should implement these guidelines in a cost-effective manner that offers intelligent wound care based on available evidence. How best to implement the standards in long-term care continues to be a challenge, even for discerning administrators and health professionals attempting to maximize resource utilization and balance quality pressure ulcer care. Thus, the purpose of this study was to examine the effectiveness of comprehensive protocols of care (SOLUTIONS®, ConvaTec, a Bristol-Myers Squibb Company, Princeton, NJ) focused on risk factors identified by the Braden Scale to prevent pressure ulcers in two long-term care facilities.

Methods

   Study design. This quasi-experimental study was conducted in two phases. In Phase I, retrospective medical record abstraction methods were used to ascertain usual (control) pressure ulcer prevention care before the implementation of the comprehensive prevention protocol of care. The retrospective medical record review also was used to determine the cumulative incidence of pressure ulcers for both long-term care facilities. In Phase II, a consistent patient risk assessment (Braden Scale) and the comprehensive protocols of care were introduced in both long-term care facilities (experiment). This study was approved by the Yale Institutional Research Review Board before data collection.

   Participants and setting. After obtaining appropriate Internal Review Board approval, a convenience sample was used to procure participants in Phase I (January 1999 to July 1999) and Phase II (August 1999 to December 1999) of the study. The only inclusion criterion for the study was that residents were identified at risk for pressure ulcers by having at least one pressure ulcer risk factor (identified by the literature) or were identified at risk by a validated risk assessment tool (Norton Scale [Phase I]7and Braden Scale8 [Phase II]). Approval by the resident's primary care physician and written informed consent also were obtained. Because no pressure ulcer prediction tool captures all risk factors, the authors included additional risk factors noted in the literature.9

   The study was conducted in two long-term care facilities with total bed capacities of 150 (Facility A) and 110 (Facility B), respectively. These two long-term care facilities were selected because they were owned by the same corporation, had a high incidence of pressure ulcers, and did not use comprehensive protocols of care. All units (eg, rehabilitation, subacute care, and skilled care) were involved. Because both facilities had received pressure ulcer citations from the Connecticut Department of Health, the administrative, medical, and nursing staffs were motivated to improve their pressure ulcer prevention programs.

   Data collection. In Phase I, a medical record abstraction was conducted by a trained research nurse to identify all nursing home residents at risk for pressure ulcers and to determine the usual pressure ulcer preventive care provided at both long-term care facilities before the intervention.

   The medical record abstraction included all medical records from January 1, 1999 to June 30, 1999. Medical records contained information obtained from the Minimum Data Set (MDS) and ICD-9 code for pressure ulcers, including resident demographics, medical diagnoses, and clinical characteristics. All sections of the medical records (eg, nursing assessment forms, medical and nursing notes, and laboratory results) were used to ascertain pressure ulcer data. Demographic data included: age, gender, race, and source of facility admission. Primary and secondary medical diagnoses also were collected. The clinical characteristics data obtained included: mobility status, urinary and fecal incontinence, motor deficits, dry skin, history of fractures or pressure ulcers, and use of pressure ulcer risk assessment tools. Specific preventive interventions (eg, turning every 2 hours and use of support surfaces) and data related to absence or presence of a pressure ulcer (for establishing incidence rate) also were obtained.

   To calculate the percentage of at risk population, the medical records of all residents residing in the two long-term care facilities at the time were included. A participant was determined to be at risk by the research nurse if at least one risk factor for pressure ulcer development as identified in the literature was noted (eg, bedbound, chairbound, urinary/fecal incontinence, albumin < 3.5mg/dL). Participants also were identified as at risk by the Norton scale used by both long-term care facilities.

   Before data collection, three charts were randomly selected to measure interrater reliability between the research nurse and the principal investigator. Reliability coefficients were high for all data elements, with Kappa values ranging from 0.8 to 1.0, indicating excellent reliability. The staff nurses were consulted only when a discrepancy in the medical record was noted. Each medical record was reviewed up to 5 months retrospectively from the date of the review to obtain the usual pressure ulcer preventive care provided by the two long-term care facilities. During Phase II of the study, resident demographics (eg, age and gender), medical diagnoses (primary and secondary), comorbidities, Braden Scale for patient pressure-ulcer risk, and the Pressure Sore Status Tool10 to monitor pressure ulcer status were collected. In this phase, independent medical record data on preventive care were collected automatically as part of the comprehensive protocols of care employed.

   Pressure ulcer protocols of care (Phase II).
  1. The program
The wound and skin care program was developed and validated11 by wound care experts on behalf of the sponsoring company and provided the foundation for the comprehensive pressure ulcer protocols of care. This comprehensive wound program presents protocols of care to help prevent and manage chronic wounds (eg, pressure ulcers, diabetic ulcers, arterial ulcers, and venous stasis ulcers). The program encourages the systematic use of patient risk, skin/wound assessment, and nutritional assessment through validated algorithms that offer care plans based on each type of assessment. The program does not recommend specific brand products related to dressings, pressure-relieving devices, or nutritional supplements. Rather, broad classifications are recommended (ie, hydrocolloid), enabling the long-term care facility staff to customize the care plan to their specific products that correlate with the recommended classification.

   This study only employed the protocols and care plans section related to pressure ulcer prevention. This section is comprised of recommendations from the AHRQ guidelines on the prevention of pressure ulcers and wound expert opinions, as well as literature related to the prevention of pressure ulcers. More specifically, the program for pressure ulcer prevention is comprised of the Braden Scale for the Prediction of Pressure Sores8 and its corresponding care plan algorithms (based on the level of pressure ulcer risk). A major strength of this program is that the pressure ulcer risk assessment and associated algorithms are validated.2, 11-14 The nutritional supplementation and pressure-relieving devices were standardized for both long-term care facilities. Finally, the program also includes a continuing education module for nurses and physicians on the treatment and prevention of chronic ulcers. Note: Before implementing the program, neither facility consistently used assessment tools or any pressure ulcer prevention protocols.

  2. Education
   Two months before implementing the comprehensive prevention program, the research team completed a series of educational sessions for the nursing staff. Employees on all three shifts at both long-term care facilities attended mandatory sessions. Registered nurses (RNs) were educated on how to use the Braden Scale for the Prediction of Pressure Sores and how to generate the prevention care plans. According to published research, the Braden Scale is most predictive when used by registered nurses.15 Before implementing the Braden scale, both facilities were using the Norton scale haphazardly. Registered nurses and licensed practical nurses (LPNs) were educated on a variety of skin care products and the correct use of these products. Educational sessions also were conducted for the certified nursing assistants (CNAs). These sessions included content on basic pressure ulcer prevention and skin care. Demonstrations on how to give a bath with return demonstrations also were provided to ensure proper use of skin care products. One educational session was held for the medical staff to review the study protocol at both long-term care facilities.

  3. Skin consultations
   Coupled with the comprehensive protocols of care, a certified Wound, Ostomy and Continence Nurse was available to the two long-term care facilities on a limited basis (2 hours/week or 8.67 hours/month) for consultation as part of the sponsoring company's normal practice for institutions starting the program. To help empower the nursing staff, the CWOCN consultant made recommendations only.

   Phase Two procedures. Each resident in the two long-term care facilities was evaluated for pressure ulcer risk level by a trained RN using the Braden scale. Braden scale scores were collected bimonthly for high-risk (Braden Scale ≤ 18) and monthly for low-risk residents (Braden Scale ≥ 19). A care plan was generated for all residents with three or more pressure ulcer risk factors and for those identified at risk for pressure ulcer development using the Braden scale.

   The potential areas of prevention that could be addressed by the care plan, depending on the Braden score (level of risk), included: frequency of skin/pressure ulcer assessment, mobility, pressure relief, nutrition, moisture management, friction and shear relief, and general skin care. Data regarding preventive interventions were recorded on the care plan (by the facility RN) and validated by direct observation (by the research nurse). Each study participant was examined weekly by the facility nurse for the presence or absence of pressure ulcers and date of occurrence.
Each facility selected its own essential skin care prevention system to implement on a daily basis, when soiling occurred, and/or as needed. The essential skin care prevention system included a skin cleanser, moisturizer, and moisture repellent. This skin care prevention system has been recommended as essential for providing basic skin care.6, 16-18 Facility A selected products from ConvaTec: Aloe Vesta® 2-n-1 Perineal/Skin Cleanser (cleanser), Aloe Vesta® 2-n-1 Skin Conditioner (moisturizer), and Aloe Vesta® 2-n-1 Protective Ointment (barrier) for their essential skin care program. Facility B used the same moisturizer and barrier cream but substituted Aloe Vesta® 2-n-1 Perineal/Skin Cleanser with Aloe Vesta® 3-n-1 Cleansing Foam (cleanser), because it could also be used as a shampoo.

   Nutrition was standardized using the Mead Johnson Nutritionals (A Bristol-Myers Squibb Company, Evansville, Ill.) portfolio. The facilities implemented a policy to give all residents extra protein during medication pass (unless contraindicated by the physician). Thus, all participants received Boost® during medication rounds. If extra supplementation (per dietary consult) was needed, Boost Plus® for oral and Deliver™ 2.0 for tube feeding was provided. Support surfaces from Kinetic Concepts Inc.® (San Antonio, Tex.) and Hill-Rom® (Batesville, Ind.) were used for all study participants. Participants who had a Braden scale score of ≤ 12 received an alternating air mattress and those scoring ≥ 13 received a foam overlay. Although both facilities were using support surfaces, no formalized plan to stratify support surfaces correlated to risk assessment score was used before Phase II.

   The research nurse monitored compliance with the Braden scale and generation of the prevention care plans on a monthly basis. The research nurse also randomly monitored three participants each week to determine the accuracy of the care plan to preventive interventions carried out by the nursing staff. The Director or Assistant Director of Nursing monitored the care plans on a monthly basis.

   Data analysis. Statistical analyses were performed using the SPSS System (SPSS, Inc., Chicago, Illinois). Univariate analyses were performed to describe the retrospective and prospective samples. Pressure ulcer incidence was calculated using the cumulative incidence method. Specifically, the number of subjects with new ulcers was divided by the total number of at risk residents in the sample over the 5-month period for the retrospective and prospective data points.

   Chi-square analyses were computed to examine the association between preventive interventions and level of pressure ulcer risk. Kaplan-Meier survival analysis was used to determine differences in pressure ulcer incidence over the 5-month intervention period. The log-rank test was used to determine significant differences in the stratification of the survival distribution. Kappa statistics were used to measure interrater reliability.

   Cost-effectiveness. Cost associated with pressure ulcer prevention was computed using an activity-based costing model.19 The activity-based costing module was conducted 2 months into the prevention program to decrease the Hawthorne effect --that is, the effects due to special attention. A stratified random sampling obtained a representative sample of 20 participants at high risk for pressure ulcers. A die was cast to identify 10 high-risk residents in Facility A and Facility B to complete the activity-based costing module. The activity-based costing model captured all costs attributable to prevention of the pressure ulcer.19 Thus, costs associated with all labor (eg, registered nurses) and supplies (eg, skin care products) for 1 month per resident were incorporated into the economic model.

   The cost of each specific type of support surface used on beds and wheelchairs, as well as supplies used for preventing the pressure ulcer, was procured from the purchasing department at each long-term care facility. The cost of labor for repositioning was based on the frequency of turning, nursing staff time to perform the task, the number of personnel required, and the average hourly salary of nursing staff in both facilities (RN = $22.52, LPN = $20.26, CNA = $13.20). The frequency of repositioning was determined by reviewing each subject's care plan for a 24-hour period. The nursing time required for repositioning was established at 2 minutes (via time and motion) and estimated every 2 hours based on the care plans. The cost for the nurse to perform a Braden scale assessment and generate the prevention care plan was established at 5 minutes (via time and motion), which was multiplied by the mean salary for the licensed staff at both facilities. The average costs of prevention for each subject for a month were calculated for six different categories of preventive interventions: 1) risk assessment labor, 2) skin care labor and supplies, 3) mattress support surface, 4) chair support surfaces, 5) nutrition, and 6) CNA repositioning labor.

   To determine the monthly cost of pressure ulcer preventive care for pressure-ulcer-free, high-risk residents to the facilities, a two-step process was used. First, the average cost to prevent a pressure ulcer in one high-risk resident who developed an ulcer was used. This cost was divided by the total number of high-risk residents who did not develop a pressure ulcer; this figure yielded the monthly prevention cost per resident.

Results

   Phase I. A total of 302 resident charts were reviewed during this phase of the study. Of these, 203 residents were identified as being at risk for pressure ulcers. Of the 203 at-risk residents, 65 (32%) arrived at the long-term care facilities with pressure ulcers. The mean age of residents was 82 years (SD = 12.38). The majority of residents was female (72%), white (97%), and transferred to the long-term care facilities from hospitals (78%). The most common medical diagnoses for participants at risk for pressure ulcers included connective tissue disease (68%) and coronary artery disease (66%). Clinical characteristics revealed that dry skin (90%) and urinary incontinence (76%) were the most prevalent pressure ulcer risk factors. Documentation of pressure ulcer risk for residents found to be at risk during the chart review was noted for 136 (67%) residents.

   A variety of interventions was used in the 136 residents identified to be at risk for pressure ulcers (see Table 1). For those residents at high risk (as identified by the Norton scale or having at least three risk factors) for pressure ulcers (n = 127), the majority had a skin assessment (n = 124, 97.6%), received nutritional consult on admission (n = 83, 65.4%), was placed on a support surface (n = 82, 64.6%), and was documented to be repositioned every 2 hours (n = 78, 61.4%). However, few charts documented whether the support surface was 4 inches or greater (n = 37 , 29.1%), or used an alternating air mattress (n = 27, 21.3%).

   The association between preventive interventions and level of pressure ulcer risk revealed no statistically significant association between pressure ulcer risk and nutritional consult (P= 0.564), placement on an overlay (P = 0.525), placement on an overlay ≥ 4 inches thick (P = 0.757), repositioning every 2 hours (P = 0.168), or skin assessment (P = 0.998). Thus, the findings suggest that preventive interventions were not selected based on risk assessment scores or pressure ulcer risk factors. Finally, the cumulative incidence rate during the Phase I medical record review covering January to June of 1999 was 43.3%.

   Phase II. Following implementation of the comprehensive prevention protocols of care, the cumulative 5-month incidence rate decreased from 43.3% to 28.5% (for both facilities combined). Specifically, with 121 high-risk residents in August 1999 and 125 in December 1999, the pressure ulcer incidence in Facility A significantly decreased from 13.2% (August 1999) to 1.7% (December 1999, P = 0.02) -- an 87% incidence reduction. Facility B, with 83 high-risk residents (August 1999) to 85 (December 1999) high-risk residents, also experienced a significant (P = 0.02) reduction in pressure ulcer incidence (from 15% in August 1999 to 3.5% in December 1999). No significant demographic differences were found between Phase I and Phase II. Of interest, Facility A was able to decrease its pressure ulcer incidence rates despite significant senior management changes (removal of two administrators and one Director of Nursing) in September and December (see Figure 1). In total, an estimated 614 ulcer-free residents were gained in Facility A in the 5-month intervention period (see Table 2). For Facility B, 412 ulcer-free residents were gained during the same 5-month period (see Table 2).

   The activity-based model (see Table 3) revealed an average cost-per-month to prevent pressure ulcers in one high-risk resident of $796.73 (including a one-time charge for mattress and chair overlay). This model revealed that the most expensive component of pressure ulcer prevention was the labor cost of $277.15 per month (licensed staff and CNAs). Thus, the 5-month cost to prevent pressure ulcers in one high-risk resident was $2,875.65 ($519.73 x 5 months, plus $252.00 for mattress overlay and $25.00 for chair cushion). This is a total of $2,875.65, divided by 5 months, or $575.14 pressure ulcer prevention costs per high-risk resident per month to reduce the incidence of new pressure ulcers from 13.2% or 15% per month to 1.7% or 3.5% per month. For those high-risk residents who remained ulcer-free, the cost of monthly prevention was calculated as the total number of pressure ulcer-free residents divided by cost of residents who developed pressure ulcers. This cost varied from $5.10 (August 1999) to $4.35 (December 1999) for Facility A and $7.53 (August 1999) to $6.53 (December 1999) for Facility B.

Discussion

   Using a consistent protocol of care including this comprehensive prevention program significantly reduced the incidence of pressure ulcers in two long-term care facilities. Overall, the facilities had decreases in their pressure ulcer incidence rates of 87% and 76%, respectively. Facility A's pressure ulcer incidence rates decreased from 13.2% in August 1999 to 1.7% in December 1999. Facility B's pressure ulcer incidence rate decreased from 15% in August 1999 to 3.5% in December 1999. Of note: Facility B had an outbreak of influenza in December 1999. The lack of further decreases in pressure ulcer incidence rates in Facility B may be attributed to an influenza outbreak at this facility in December 1999.

   This study supports several studies in which pressure ulcer prevention protocols were implemented and significant decreases in the incidence of pressure ulcers occurred.20-22 The use of a validated pressure ulcer prediction tool coupled with effective prevention algorithms, evidence-supported products, and staff education was essential in decreasing the pressure ulcer incidence rates in both long-term care facilities.

   The use of monitoring by the research nurse and nursing administrators may have increased the adherence by nursing staff to the protocols of care. The nursing administrators were highly motivated to decrease their pressure ulcer incidence rates; thus, some organizational cultural changes occurred. Evidence supports the hypothesis that changes such as those reported here may be transient, depending on interest by administration.20 However, despite organizational turmoil in one of the facilities (Facility A), the protocol of care implemented remained in place and pressure ulcer incidence continued to decrease. This may be related to the ease of use of the program protocols, the streamlining of products used, and/or the effectiveness of the staff education program. Further investigation into facility specific characteristics may increase understanding of the results.

   Although the literature contains several reports of effectively implemented comprehensive pressure ulcer prevention programs, studies investigating the cost of providing pressure ulcer preventive care are rare. Only one study that reports the cost of implementing an intensive pressure ulcer prevention program in a 77-bed long-term care facility was found.20 These researchers found a pre-intervention incidence rate of 23% and a post-intervention incidence rate of 5%. More interestingly, researchers in this study determined the mean cost for prevention/resident to be $91.56 + $135 for 6 months. This is far less than the authors' monthly costs estimates of $575.14 The discrepancy in total costs may be attributed to the different time period of data collection (1994 compared to 1999), region of the country (Midwest compared to Northeast), use of all residents in calculating costs compared to only high-risk residents in the present study, and variables entered into the cost models.20 However, pressure ulcer prevention can be quite costly; thus, targeted prevention should be initiated based on level of risk.

Limitations

   The study had several limitations. The staff time used in estimating labor costs may have been different in the unmeasured residents. The authors' activity-based costing module was completed well into the study period; thus, the nursing staff was familiar with all the protocols. The cost might have been higher if the authors had implemented the activity-based costing module in the beginning of Phase II when the nursing staff were getting familiarized with the protocols and products. Also, activity-based costing by the research team and long-term care staff was only completed during the day shift; conceivably, capturing staff time on the evening and/or night shifts may have yielded different results. This potential effect was minimized by using random selection in the activity-based costing model as well as standardizing the treatments on all three shifts. Finally, the Hawthorne effect may have been a factor accounting for the high level of adherence to the protocols.

Conclusion

   Pressure ulcers continue to be a common health problem in long-term care. Given the impending capitated system and the increasingly litigious nature in long-term care, preventing pressure ulcers is essential.22 The "true" cost of preventing pressure ulcers remains elusive. Additional studies are needed to investigate whether prevention is more cost-effective than treatment.
The authors demonstrated that the use of validated, comprehensive prevention protocols of care, including the program described here, significantly reduced the incidence of pressure ulcers in two long-term care facilities. In addition, they found that standardizing pressure ulcer preventive care with targeted interventions based on risk level was much more cost-effective over the 5-month intervention period.

   This study was funded by ConvaTec, a Bristol-Myers Squibb Company, and Mead Johnson and Company.