An Innovative Enterostomal Therapy Nurse Model of Community Wound Care Delivery

Transcription

1 WJ350203_ qxp 3/3/08 4:01 PM Page 169 J Wound Ostomy Continence Nurs. 2008;35(2): Published by Lippincott Williams & Wilkins WOUND CARE An Innovative Enterostomal Therapy Nurse Model of Community Wound Care Delivery A Retrospective Cost-Effectiveness Analysis Connie Harris Ronald Shannon PURPOSE: A Canadian specialty nursing association identified the necessity to examine the role and impact of enterostomal (ET) nursing in Canada. We completed a retrospective analysis of the cost-effectiveness and benefits of ET nurse driven resources for the treatment of acute and chronic wounds in the community. DESIGN: This was a multicenter retrospective pragmatic chart audit of 3 models of nursing care utilizing 4 community nursing agencies and 1 specialty company owned and operated by ET nurses. An analysis was completed using quantitative methods to evaluate healing outcomes, nursing costs, and cost-effectiveness. MAIN OUTCOME MEASURES: Kaplan-Meier estimates were calculated to determine the average time to 100% healing of acute and chronic wounds and total nursing visit costs for treatment in a community setting. Average direct nursing costs related to management of each wound were determined by number of nursing visits and related reimbursement for each visit. A Monte Carlo simulation method was used to help account for costs and benefits in determination of cost-effectiveness between caring groups and the uncertainty from variation between patients and wounds. RESULTS: Three hundred sixty chronic wounds and 54 acute surgical wound charts were audited. Involvement of a registered nurse (RN) with ET or advanced wound ostomy skills (AWOS) in community-level chronic and acute wound care was associated with lower overall costs mainly due to reduced time to 100% closure of the wound and reduced number of nursing visits. The differences in health benefits and total costs of nursing care between the ET/AWOS and a hybrid group that includes interventions developed by an ET nurse and followed by general visiting nurses that could include both RNs and registered practical nurses is an expected reduction in healing times of 45 days and an expected cost difference of $ per chronic wound treated. When outcomes were broken into ET/AWOS involvement categories for treatment of chronic wounds, there was a significantly faster time to 100% closure at a lower mean cost as the ET/AWOS involvement increased in the case. For acute wound treatment, the differences in health benefits and total costs between the ET/AWOS and a hybrid nursing care model were an expected reduction in healing times of 95 days and an expected cost difference of $ per acute wound treated. Again, there was a significant difference in healing times and reduced mean cost as the ET/AWOS became more involved in the treatment. The financial benefit to the Ontario Ministry of Health and Long-Term Care is estimated to increase as the involvement of nurses with ET/AWOS specialty training increases. CONCLUSIONS: The greater the involvement both directly and indirectly of an ET/AWOS nurse in the management of wounds, the greater the savings and the shorter the healing times. Background The Canadian healthcare system is currently facing pressure to control healthcare costs while maintaining a high quality of care. There is an increased demand for home care, which now includes both chronic and acute disease models. Nursing visits and the cost of wound care dressings absorb a significant percentage of community healthcare budgets, with wound care delivery accounting for more than 50% of nursing visits. 1 Current research and Connie Harris, MSc (c), RN, ET, Senior Clinical Specialist, Ostomy and Wounds CarePartners ET NOW Division, Kitchener, Ontario, Canada. Ronald Shannon, MPH, Freelance Healthcare Economist, Clifton Park, New York. Corresponding Author: Connie Harris, MSc (c), RN, ET, Ostomy and Wounds CarePartners ET NOW Division, Unit B Frobisher Dr, Waterloo, Ontario, Canada N2V 2C9 (connie.harris@ carepartners.ca). Copyright 2008 by the Wound, Ostomy and Continence Nurses Society J WOCN March/April

2 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 best practice has served as a catalyst for the development of advanced wound dressings over the past 25 years. The promotion of best practices such as moist wound healing may be used correctly, incorrectly, or not at all. Lack of knowledge of best practice and inconsistencies in application may result in unnecessary daily community nursing visits, prolonged healing times, frequent physician appointments, and hospital admissions due to complications. Healthcare organizations need to understand and position the best qualified healthcare professional who can provide leadership to maximize the direct and indirect management of patients with wounds. Today s challenges are to make the best possible use of limited human and economic resources while ensuring the delivery of excellent patient care. Research positively influences the choice of the best treatment for a wound. Using meta-analysis, Heater and colleagues 2 studied the contribution of research-based practice to patient outcomes, and reported that patients who receive research-based nursing care make sizable gains in changing their behavior to improve their health (behavioral knowledge), actual improvements in body functions (physiological), and emotional (psychosocial) outcomes compared with those receiving care that is not research based. The Canadian Association for Enterostomal Therapy (CAET) was recognized as a distinct nursing specialty by the Canadian Nurse s Association in March The association s Enterostomal Therapy Nursing Education Program (ETNEP) is a postgraduate program for baccalaureate prepared registered nurses (RNs) that provides specialized and expanding knowledge and clinical expertise in managing skin, wound, ostomy, and continence care. Primary responsibilities of an enterostomal (ET) nurse are to assess, direct, deliver, coordinate, and evaluate individualized patient care by measuring the responses and outcomes to evidence-based management. Purpose As a concurrent activity with the process to develop specialty certification, the CAET identified the need to examine the role and impact of ET nursing in Canada. This stemmed from a need to market the ET specialty and to definitively demonstrate its value. However, the association realized that it did not have outcomes data to support the role. In 2005, the CAET partnered with a unique ET-owned and ET-operated community nursing agency and a healthcare economist to conduct a retrospective pragmatic analysis of the cost-effectiveness and benefits of ET nurse driven resources for the treatment of acute and chronic wounds in the community. Methods An analysis of treatment outcomes of patients with acute and chronic wounds in the community by ET and registered nurses with advanced wound and ostomy skills (AWOS) was completed. A retrospective review of patient charts was completed and quantitative analyses were used to evaluate healing outcomes, nursing costs, and costeffectiveness. Outcomes used to measure effectiveness of wound management were (1) the time required for wound healing, (2) time required for the patient to be discharged (healing discharge times), and (3) condition at discharge. Patient charts were reviewed from 4 community nursing agencies and an ET-owned and ET-operated community nursing agency specializing in the delivery of evidencebased care of wounds. Participating Sites Five nursing agencies in 2 local community care access centers (CCACs) participated in the study. All of these nursing agencies provide community care under contract to the CCAC, which is funded under the Province of Ontario Ministry of Health and Long-Term Care. The 3 general community nursing agencies who participated in the Waterloo Region portion of the study did not have ET nurses or AWOS on staff, but worked collaboratively with the specialty ET nursing agency. Referrals for consultation to the ET/AWOS agency utilized the following criteria, which were developed collaboratively with representatives of all nursing agencies and the CCAC (Figure 1). 3 The general nursing agency in the Oxford-Region CCAC used the acronym FUN to identify indications for an ET/AWOS consultation. When applies to the clinical setting, F indicates the need to consult the ET/AWOS nurse when the frequency of dressing changes is not less than 3 times per week within 4 weeks. U indicates the need to consult the ET/AWOS nurse when the etiology of the wound is unknown, and N indicates the need to consult the ET/AWOS nurse if the size of the wound has not decreased by 20% to 30% within 3 to 4 weeks of initiating treatment. These criteria were established by a joint CCAC collaborative initiative that several CCAC and service providers participated in across the province and included a wound education component for the general visiting nurses. 4 The visits by the specialty ET nursing agency were 50% ET and 50% AWOS. RN ETs did all of the admission visits and would carry more complex clients than the AWOS nurses, but both groups are able to initiate wound care below the dermis, including sharp nonviable debridement. The AWOS works in collaboration with the ET nurse as needed, whereas the ET has a more autonomous practice. Both groups inform the physician of care plans as a valued team member. The model for care delivery for each of the 4 general agency sites could include all of the following: 1. The hybrid model where the specialty nursing agency coordinated visits with RN and registered practical nurse (RPN) visiting nurses from other community nursing agencies. Both RNs and RPNs study from the same body of nursing knowledge. RNs study

3 WJ350203_ qxp 3/3/08 4:01 PM Page 171 J WOCN Volume 35/Number 2 Harris & Shannon 171 FIGURE 1. Triage for ET consultation Waterloo Region CCAC, for a longer period of time, allowing for a greater knowledge base in clinical practice, decision making, critical thinking, leadership, research, and resource management. RPNs study for a shorter period of time, resulting in a more focused body of knowledge in the same areas. 5 The RPNs scope of practice is broadly similar to that of a licensed practical or licensed vocational nurse in the United States. 2. The ET/AWOS only model where patients were seen exclusively by the specialty agency. 3. The RN/RPN only model with no ET/AWOS involvement. The model of care for the RN ET/AWOS agency could include the first and second models as described above, with the RN ET/AWOS either being the first to assess and treat or being asked in after the general agency had admitted the patient. Sample Discharged charts representing all of the major wound etiologies seen in the participating sites between January 1, 2006, and April 15, 2006, were reviewed. Additional discharged charts representing both the hybrid group and the ET/AWOS group were obtained from 2003 to Wound

4 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 etiologies included surgical wounds, diabetic foot ulcers, venous stasis ulcers, pressure ulcers, and other diabetic wounds. The goal was to have each of the 3 groups (hybrid, ET/AWOS, and RN/RPN) have a minimum of 20 charts from each wound type surgical, diabetic foot, venous, pressure, other diabetic wounds, and other chronic wounds, with a goal of 140 for each group and 420 total charts. Inclusion and Exclusion Criteria Inclusion criteria included patients treated for a wound with a documented etiology (surgical wound, diabetic foot ulcer, venous stasis ulcer, pressure ulcer, or other diabetic wound) and wound measurement documented over at least 2 visits. Exclusion criteria included private referral, long-term care or acute care patients, less than 2 serial wound measurements, less than 2 visits, first visit made in an acute care facility, or other wound etiologies outside of the inclusion criteria. Ethics Approval and Privacy Considerations Ethics committee review and approval was obtained prior to data collection. Informed consent was not required because this was considered a quality review process. However, 1 nursing agency requested patient consent to be obtained and consents were obtained at that site. Patient identifiers were not revealed in the final report, and hard copy and computer data were secured at all times. All researchers and student assistants entered into an agreement regarding patient privacy and confidentiality, conforming to the regulations as outlined by Federal and Provincial Privacy Legislation in Canada. Objectives We sought to determine if there were significant differences in time to complete wound healing, condition of the wound at the end of care, and wound etiology and time to end of care based on the 3 care delivery models (ET/AWOS only, no ET/AWOS care, and hybrid) defined previously. Specific study objectives were (1) to demonstrate outcomes of ET/AWOS nurse as compared with general visiting nurses for managing wounds in the community; (2) to determine the cost savings and health benefits when an ET/AWOS nurse manages a wound in the community setting; and (3) to provide a budget impact analysis of using ET/AWOS nurse services. Data Collection Patient charts were selected by etiology for the specialty agency charts. The general nursing agency charts were collected as the patient was discharged, and later sorted based on etiology by nursing student auditors who were trained by the researchers. Data extracted from the charts included (1) patient demographic information (age, gender, etc), (2) primary medical diagnosis, (3) date of first home visit, (4) discharge date from home health, (5) wound etiology, size, and depth, (6) duration of wound (expressed in weeks) prior to onset of nursing treatment, (6) wound assessment completed at each visit (wound measurements, tissue description, odor, antibiotic use, Bates Jensen Wound Assessment Tool score if available, patient teaching done, pressure redistribution devices), (7) number of nursing visits by care agency and nursing type, (8) length of time on service, and (9) condition of the wound on discharge. After collecting data from 237 charts, it was discovered that there were elements missing in some records essential to making an adequate comparison between the care delivery models. Analysis of these charts revealed an inadequate sample size to accommodate the power needed to determine significant differences. Therefore, a second phase of data collection was undertaken to compile adequate sampling with adequate information needed to comprise a robust sample. Unfortunately, the sample size of the no ET (RN/RPN only) remained very small and charts frequently lacked serial wound measurements. As a result, this model could not be adequately represented within this study. Because the general nursing agencies did not store discharged charts onsite and did not have a database by which to search for charts by etiology, they were limited to collecting charts as patients were discharged. Five out of 5 agencies participated in the initial phase of data collection, but only 2 agencies participated in the second phase due to time and staffing restrictions. Wound Outcomes For the purposes of this study, health benefits were defined as the difference in healing trajectories and mean time to 100% wound closure. Healing trajectories have been indicated by the US Food and Drug Administration and the USbased Wound Healing Society as the most stringent criteria to determine the efficacy of a new wound healing agent as achieving 100% wound closure. 6 Wound healing follows an exponential course, with the rate of change of wound area progressively decreasing as the residual wound area approaches total closure. Kaplan-Meier survival methodology was used in this study to create dynamic healing trajectories and compare mean time to 100% closure for statistical significance. 7 The method has been reported for wound healing studies and is useful when there are a significant number of patients who do not reach the end point under investigation (eg, complete healing). 8,9 Therefore, wound healing was quantified based on (1) time required to complete wound healing, (2) time required to discharge, and (3) condition at discharge. Conditions at discharge included (1) wound healed (closed), (2) independent in self-care and healing, (3) patient discharged but wound not healing, (4) wound-related hospital readmission, (5) non woundrelated hospital readmission, (6) patient expired, (7) patient lost to service, and (8) wound healing and patient discharged to other nursing agency.

5 WJ350203_ qxp 3/3/08 4:01 PM Page 173 J WOCN Volume 35/Number 2 Harris & Shannon 173 Economic Analysis A cost-effectiveness analysis was completed to determine the incremental differences between nursing visit costs and time to 100% closure of the wound. Current average fees for service were provided by the CCAC of Waterloo Region, Ontario, for the 2005 nursing rates. They were $53.00 (CN) for an RN visit and $43.00 (CN) for an RPN visit. The ET specialty nursing agency provided actual billing fees; they were $61.09 (CN) for an ET visit and $48.75 (CN) for an AWOS visit. This fee structure was used to determine the nursing costs of wound treatment in the cost-effectiveness models. These were used as a standard for economic evaluation for the whole project. The cost of medical resources including pressure redistribution devices, dressings, and other adjunctive services was not quantified in this analysis. Decision Analysis Probabilistic decision analysis, a method to help clinicians develop policies and make decisions on how best to treat individual patients, was used as the conceptual model for economic data analysis. This analysis was selected to demonstrate the relative best or most valued outcome of each of the care provider groups. Decision analysis software TreeAge Professional-Healthcare Module assisted in creating the decision trees and running a Monte Carlo simulation. 11 Monte Carlo disease simulation estimates the effect of variability among patients in both underlying disease progression patterns and individual responsiveness to treatments. 11 The output information is presented in the form of distributions, which was used to estimate mean costs and mean time to healing. These results can be used to compute cost-effectiveness ratios between the 3 provider models (ET/AWOS only, no ET/AWOS care, and hybrid). Distributions were fitted to healing time and cost results for each caring group and simulated using iterations. A 95% confidence interval and graphical representation of the results demonstrated the incremental costs and healing benefits. Budget Impact Analysis Budget impact analysis (BIA) is an important part of a comprehensive economic assessment of a healthcare technology or program. 12 The purpose is to estimate the financial consequences of adoption and diffusion of a healthcare intervention within a healthcare system given inevitable resource constraints. In this study, BIA was used to predict how an increase or a decrease in ET/AWOS involvement in a wound case impacted the trajectory of spending on nurse visits. Statistical Analysis All calculations and graphing were performed using SPSS 8.0 for Windows. 13 Kaplan-Meier survival analysis was used to measure healing trajectories. Trajectories were constructed for ulcers that totally healed (100% closure) and those that did not ( 100% closure) over the period of care offered by each nursing model. The percentage of patients achieving total healing versus length of treatment time per care provider model was plotted. For each wound type, a log-rank statistic was used to determine mean time to healing and demonstrate statistical differences between healing trajectories for each caring model. Categorical data were compared using the 2 test and continuous data were compared using the t test or Mann-Whitney U test. Results Subjects Four hundred ninety-six charts were audited in the 2 phases of the study. If the general visiting nursing agency s charts was missing data for charts in the hybrid model, the audit was augmented by access to that client s chart in the ET nursing agency files (each agency uses its own proprietary documentation system). Further risk adjustment was completed on the data for optimal comparison of caring groups. A total of 360 chronic wounds were evaluated in this study; 154 were treated using the ET/AWOS model and 206 treated by the hybrid model. An additional 54 acute surgical wounds included 8 treated by the ET/AWOS model and 46 treated by the hybrid model. The sample size for the ET/AWOS acute wound group was too small to make a direct comparison, so analysis was made by factoring the percentage involvement of the ET/AWOS in the treatment. Missing data and risk adjustment in the surgical wound type left a total of 43 wounds for comparative intervention analysis. Patients in the ET/AWOS model were comparable to their counterparts in the hybrid model with respect to age and gender, wound size and depth at initial visit, and use of pressure redistribution interventions (Table 1). However, the 2 groups differed on the wound etiology and location. Wounds in the hybrid group were more likely to be diabetic (55.6% vs 39.6%, P.029) and located on the toe (32.4% vs 20.1%, P.010). Chronic Wounds Pooled analysis of chronic wounds indicated a faster healing response when the ET/AWOS managed the wound exclusively. Kaplan-Meier survival analysis shows a significantly faster and consistent healing trajectory (100% closure) for the ET/AWOS treated wounds when compared with the hybrid group managed wounds (Figure 2). The mean time to 100% closure was approximately 99 and 143 days for the RN ET/AWOS and hybrid groups, respectively (Table 2, P.0006). Analysis of the wounds by etiology yielded similar results with the exception of venous ulcers (Table 3). In the hybrid group, the ET/AWOS involvement was calculated dividing the number of ET/AWOS visits into the

6 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 TABLE 1. Patient and Chronic Wound Characteristics, By Caring Group RN ET /AWOS Specialty Nurse Involvement RN ET/AWOS Collaborating With Characteristic Specialty Service Alone General Nursing Agency P Age, mean (SD), y (15.76) (13.99).426 Women, n (%) 60 (39) 97 (46.9).134 Men, n (%) 94 (61) 110 (53.1) Wound type, n (%) Diabetic foot ulcer 61 (39.6) 115 (55.6).029 Pressure ulcer 30 (19.5) 31 (15) Venous stasis ulcer 33 (21.4) 32 (15.5) Other diabetic ulcer 30 (19.5) 29 (14.0) Wound location, n (%) Foot 55 (35.7) 68 (32.9).010 Leg 40 (26.0) 34 (16.4) Mid-region (sacrum, ischial tuberosity, coccyx, 23 (14.9) 22 (10.6) trochanter) Toe 31 (20.1) 67 (32.4) Other 5 (3.2) 16 (7.7) Wound size (area length width) 4 cm (74.2) 143 (69.8) to 16 cm 2 34 (22.5) 52 (25.4) 16.1 to 36 cm 2 2 (1.3) 6 (2.9) 36.1 to 80 cm 2 2 (1.3) 2 (1.0) 80 cm 2 1 (0.7) 2 (1.0) Wound depth Nonblanchable erythema on intact skin 12 (7.9) 10 (5.0).575 Partial-thickness skin loss involving epidermis 58 (38.2) 68 (33.8) and/or dermis Full-thickness skin loss involving damage or 61 (40.1) 86 (42.8) necrosis of subcutaneous tissue; may extend down to but not through underlying fascia; and/or mixed partial- and full-thickness and/or tissue layers obscured by granulation tissue Obscured by necrosis 16 (10.5) 28 (13.9) Full-thickness skin loss with extensive destruction, 5 (3.3) 9 (4.5) tissue necrosis or damage to muscle, bone, or supporting structures Pressure relief, % Orthotic use for diabetic foot ulcers Pressure redistribution for pressure ulcers Compression bandaging for venous ulcers Abbreviations: AWOS, advanced ostomy and wound skills; ET, enterostomal; RN, registered nurse; SD, standard deviations. total number of visits for each patient/wound case. The percentages were grouped into categories (high involvement: ET/AWOS involved in 75% to 100% of visits; moderate involvement: ET/AWOS involved 50% to 74% of visits; mild involvement: ET/AWOS involved 25% to 49% of visits, and minimal involvement: ET/AWOS involved in 24%). Healing trajectories were seen to improve, as there was more involvement by the ET/AWOS (Figure 3). A 1-way analysis of variance showed a statistically significant difference in mean time to complete wound closure with higher levels of ET/AWOS involvement (Table 4, P.0002). Figure 4 compares mean nursing costs in the pooled analysis of chronic wounds. This was calculated by multiplying the number of nurse visits by CCAC reimbursement provided for each nursing qualification and summing them to a total for each wound case. Where there were multiple wounds on a patient, the total cost was divided by the number of wounds treated on the patient. The

7 WJ350203_ qxp 3/3/08 4:01 PM Page 175 J WOCN Volume 35/Number 2 Harris & Shannon 175 FIGURE 2. Kaplan-Meier time to 100% closure (chronic wounds). TABLE 2. Mean Time to 100% Closure of Pooled Chronic Wounds, by Caring Group Significance (Log-Rank Mean (SE) Time to 95% Confidence Statistic for Equality Intervention Sample Size 100% Closure, d Interval of Distributions) RN ET/AWOS (9.17) (80.60, ).0006 Hybrid (13.20) (117.51, ) TABLE 3. Subgroup Results, Mean Time to 100% Closure, By Wound Type and Caring Group Sample Mean (SE) Time to 95% Confidence Significance Intervention Wound Type Size 100% Closure, d Interval Heal Size Depth RN ET/AWOS Diabetic foot (13.80) (77.69, ) Hybrid (17.02) (116.40, ) RN ET/AWOS Pressure ulcers (12.80) (52.71, ) Hybrid (37.05) (113.67, ) RN ET/AWOS Venous stasis ulcers (26.43) (89.95, ) Hybrid (19.49) (65.53, ) RN ET/AWOS Other diabetic wound types (7.94) (32.42, ) Hybrid (19.47) (67.02, ) Abbreviations: AWOS, advanced ostomy and wound skills; ET, enterostomal; RN, registered nurse; SE, standard error.

8 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 FIGURE 3. Kaplan-Meier time to 100% closure (RN ET/AWOS specialty nurse involvement, chronic wounds). primary reason for the provision of nursing visits was related to a chronic wound as identified in the patient chart. The mean cost of nursing visits in the hybrid group was significantly higher than the ET/AWOS (Figure 4, P.000). Figure 5 provides a cost analysis between the provider models based on wound etiology. The same trend was seen in each wound type. Interestingly, there was a higher cost of nurse visits in the hybrid group for the treatment of venous ulcers, although the time to healing, as previously mentioned, was faster. Figure 6 displays the differences in nursing costs as the ET/AWOS becomes more involved in the case. Economic Analyses A cost-effectiveness analysis was performed to determine the cost to achieve complete closure among subjects with chronic wounds. Costs were calculated using 2005 contract levels. Probabilistic decision analysis and Monte Carlo simulation were used to estimate the expected costs and time to complete closure of chronic wounds treated by the ET/AWOS alone or hybrid care models. The nursing costs and healing times assumed a normal distribution. The results indicate that the ET/AWOS model can expect to provide a patient with approximately 45 more ulcer-free days at an expected cost savings of $ (CN) per wound compared to the hybrid care model (Table 5). Analysis also revealed that the greater the ET/AWOS involvement in wound management, the greater the response to healing and the lower the cost (Table 5). With minimal ET/AWOS involvement, the cost was approximately $ (CN) per chronic wound as compared a cost of $ (CN) per wound with high ET/AWOS involvement. Expected time to complete closure was 97 days compared to 158 days. These results indicate that the ET/AWOS only care model was more cost-effective in healing wounds to complete closure when compared to the hybrid care model. TABLE 4. Mean Time to 100% Closure of Chronic Wounds, by Caring Group and RN ET/AWOS Involvement Mean (SE) Time to 95% Confidence Involvement Sample Size 100% Closure, d Interval Significance High (75% 100%) (8.87) (80.64, ).0002 Moderate (50% 74%) (8.79) (59.03, 93.50) Mild (25% 49%) (39.16) (84.21, ) Minimal (0% 24%) (15.16) (124.31, ) Abbreviation: SE, standard error.

9 WJ350203_ qxp 3/3/08 4:01 PM Page 177 J WOCN Volume 35/Number 2 Harris & Shannon 177 Figure 7 shows the results of analysis of the mean difference in nursing visit cost and healing time. The majority of point estimates fall in the northeast and southeast quadrants of the cost-effectiveness plane, suggesting that the hybrid care model is associated with a longer time to achieve complete closure of chronic wounds. A majority of the point estimates in those quadrants lie in the northeast plane where the nursing visit cost is higher for the hybrid model than the ET/AWOS only model. FIGURE 4. Mean cost of nursing visits per chronic wound, by caring group. FIGURE 5. Mean cost of nursing visits for each chronic wound, by caring group and wound type. Acute Wounds Acute wounds treated by visiting nurses tended to be surgical wounds and were associated with few, if any, complications. However, in this study, only acute wounds that met the indicators for an ET consultation (stalled healing due to intrinsic or extrinsic factors) were included. 14 For a statistical comparison, acute wounds were categorized into ET/AWOS involvement 50% and those with 50% ET/AWOS involvement. Patients in ET/AWOS care model did not differ from those managed by the hybrid care model with respect to age, gender, wound size and depth at initial visit, or location (Table 6). Kaplan-Meier survival analysis revealed a faster and more consistent healing trajectory when the ET/AWOS FIGURE 6. Mean cost of nursing visits per chronic wound, by RN ET/AWOS involvement in treatment. TABLE 5. Cost-Effectiveness Results (Chronic Wounds), Monte Carlo Simulation Mean (SD) Incremental Incremental Effectiveness Time Effectiveness Time Cost- Caring Group Mean (SD) Cost Cost to 100% Closure, d to 100% Closure, d Effectiveness Ratio RN ET/AWOS $ ($ ) (112.75) $12.00 Hybrid $ ($ ) $ (187.74) $49.00 High involvement $ ($ ) (113.52) $14.00 Moderate involvement $ ($ ) $ (31.69) $20.00 Mild involvement $ ($ ) $ (226.20) $49.00 Minimal involvement $ ($ ) $ (171.35) 0.66 $67.00 Abbreviation: SD, standard deviations.

10 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 FIGURE 7. Incremental cost-effectiveness plane of using the hybrid model for the healing of chronic wounds. was more involved in treatment (Figure 8). The mean time to 100% closure was approximately 86 days when the ET/AWOS participated more than 50% of the time as compared to 165 days when they participated in less than 50% of visits (Table 7, P.0704). Statistically, the healing trajectories were equivalent, possibly owing to the small sample size and wide variability in healing times. Economic Analyses Figure 9 compares mean nursing costs in the ET/AWOS intervention analysis for acute wounds. Costs were calculated by multiplying the number of nurse visits by CCAC reimbursement provided for each nursing qualification and summing them to a total for each wound case. Where there were multiple wounds on a patient, we divided the total cost by the number of wounds treated on the patient. The mean cost of nursing visits when the ET/AWOS were involved in more than 50% of visits was $ (CN) per wound treated as compared to $ (CN) when the ET/AWOS was involved in less than 50% of visits. Probabilistic decision analysis and Monte Carlo simulation were used to estimate the expected costs and time to complete closure of acute wounds referred to the ET/AWOS for consultation. Results indicate that a greater than 50% involvement by the ET/AWOS can be expected to provide a patient with approximately 77 more woundfree days at an expected cost savings of $ (CN) per wound when compared to a model of ET/AWOS involvement in less than 50% of visits (Table 8). Figure 10 displays the mean difference in nursing visit cost and healing time for acute wounds referred to the ET/AWOS model. The majority of point estimates fall in the southwest quadrant of the cost-effectiveness plane, suggesting that the ET/AWOS involvement in more than 50% of visits is associated with a shorter time wound closure and a lower cost than involvement by the ET/AWOS in less than half of visits. Budget Impact Analysis Statistics were acquired from the CCAC of Waterloo region regarding the number of wound patients serviced from TABLE 6. Patient and Acute Wound Characteristics, by RN ET/AWOS Involvement RN ET /AWOS Specialty RN ET /AWOS Specialty Service Involvement Service Involvement Characteristic (0% 50% Involvement) (51% 100% Involvement) P Age, mean (SD), y (19.85) (30.80).740 Women, n (%) 14 (48) 8 (57).586 Men, n (%) 15 (52) 6 (43) Wound size (area length width ), n (%) 4 cm 2 12 (41.4) 4 (28.6) to 16 cm 2 13 (44.8) 6 (42.9) 16.1 to 36 cm 2 4 (13.8) 4 (28.6) Wound depth, n (%) Nonblanchable erythema on intact skin 0 (0.0) 0 (0.0).473 Partial-thickness skin loss involving epidermis and/or dermis 2 (7.2) 0 (0.0) Full-thickness skin loss involving damage or necrosis of 23 (88.5) 11 (84.6) subcutaneous tissue; may extend down to but not through underlying fascia; and/or mixed partial- and full-thickness and/or tissue layers obscured by granulation tissue Obscured by necrosis 1 (3.8) 2 (15.4) Full-thickness skin loss with extensive destruction, tissue 0 (0.0) 0 (0.0) necrosis or damage to muscle, bone, or supporting structures Abbreviations: AWOS, advanced ostomy and wound skills; ET, enterostomal; RN, registered nurse; SD, standard deviations.

11 WJ350203_ qxp 3/3/08 4:01 PM Page 179 J WOCN Volume 35/Number 2 Harris & Shannon 179 FIGURE 8. Kaplan-Meier time to 100% closure (RN ET/AWOS specialty nurse involvement, acute wounds). May 1, 2006, to March 31, Using these statistics, a set of prevalence statistics were interpolated representing the number of wounds encountered in the Province of Ontario over a 1-year period. Using a rate of 2.0 per 1000 patients and assuming a chronic wound lasts 4 months, we estimated about 45, 000 venous stasis ulcer patients in 1 year. Diabetic foot ulcer prevalence is equivalent according to the rate per 1000 patients of venous stasis ulcers. Pressure ulcer prevalence estimates were increased 40% over venous stasis estimates (63,000), and other diabetic wounds were estimated as slightly higher than diabetic foot ulcers (50,000). Surgical wounds were estimated to occur more frequently than chronic wounds. The number of acute wound patients were increased by 400%, reflecting the relative prevalence from regional to provincial estimates (200,000). Approximately 70% of the chronic wounds would qualify for ET/AWOS consultation based on the indicators for ET consultation from the CCAC Waterloo. We then estimated the total budget impact to the Ministry of Health in the Province of Ontario by ET/AWOS involvement in acute and chronic wound management (Table 9). The predicted savings to the Ministry of Health is approximately $1.3 billion (CN) per year for chronic wounds and $575 million (CN) per year for acute wounds when the involvement of the ET/AWOS nurse is more than 50%. TABLE 7. Mean Time to 100% Closure of Acute Wounds, by RN ET/AWOS Involvement Significance Mean (SE) (Log-Rank Statistic Time to 100% 95% Confidence for Equality of Intervention Sample Size Closure, d Interval Distributions) RN ET/AWOS (0% 50% (28.45) (109.06, ).0704 involvement) RN ET/AWOS (51% 100% (17.66) (51.43, ) involvement) Abbreviations: AWOS, advanced ostomy and wound skills; ET, enterostomal; RN, registered nurse.

12 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 FIGURE 9. Mean cost of nursing visits per acute wound, by RN ET/AWOS involvement in treatment. Discussion Our findings support 3 major conclusions: (1) the ET/AWOS only care model provides more rapid wound healing times than a hybrid model that combines ET/AWOS with RN/RPN; (2) the ET/AWOS only model is more cost-effective than the hybrid model; and (3) within the hybrid model, cost and wound healing outcomes were improved when ET/AWOS saw the patient on 50% of visits or more frequently. These findings were achieved, even though the wounds treated by the ET/AWOS care model were larger in size than wounds treated by the RN/RPN (no ET/AWOS) model, supporting the assumption that specialized knowledge and clinical expertise are beneficial when managing larger and more complex wounds. In 1994, Arnold and Weir 15 compared wound healing outcomes achieved by non-et nurses and ET nurses in the same setting. They demonstrated that ET/WOC nurses delivering evidence-based wound care improved the healing rate by 100% (36.3% non-et vs 78.5% ET nurses) within the same time frame. This study included 344 wounds cared for by ET/AWOC nurses and 464 wounds cared for by staff nurses. However, this study did not include a hybrid model of ET/AWOC and staff nurses working collaboratively. Nevertheless, we believe that the hybrid model we describe is relevant because it most closely reflects reality in community nursing settings, where the ET nurse serves as the wound care expert, assessing and applying evidence-based interventions. White 16 described the activity of avoiding delayed healing as one of optimizing healing. This begins with a detailed and regular assessment, using a validated and reliable wound assessment tool, and planning and delivering treatment that is evidence based. Standards of care for wound healing include principles of wound bed preparation, maintenance of a moist wound healing environment, use of compression for venous ulcers, or pressure redistribution for diabetic foot ulcers or venous ulcers. The ET specialty agency involved in this study used measures described by DiCenso and Cullum 17 to provide clinical expertise and knowledge of current clinical evidence to deliver care that led to enhanced wound healing rates in a cost-effective manner. Venous healing rates were higher in the hybrid care model when compared to the ET/AWOS only model. There are 2 possible explanations for this outcome. One patient with many years history of recurrent venous ulcerations in the ET/AWOS group was nonadherent with high-compression therapy and had an extended length of stay. This patient should have been transferred to the general nursing agency for care and was not. While this type of case would be excluded in a prospective trial, it represents the real-world cases captured in a retrospective analysis. All patients with venous leg ulcers, whether pure venous or mixed arterial/venous etiology, are assessed for compression therapy by the ET nurse who either each carries a Doppler to perform Ankle Brachial Indexes or suggests more definitive studies for clients who have the risk of calcified vessels or who have signs of arterial disease. Compression safe for the client s vasculature is initiated as part of the plan of care, which means that all clients whether in the hybrid or the RN ET/AWOS group would be receiving appropriate therapy. It may also be that tighter inclusion/exclusion criteria for an audit of venous etiology ulcers, stipulating that clients be adherent to the plan of care, or that ulcers be of less than a certain age would produce different results in the audit. Strengths This is the first ever cost-effectiveness study of ET/AWOS nursing practice. There are no preexisting comparable TABLE 8. Cost-Effectiveness Results (Acute Wounds), Monte Carlo Simulation Incremental Effectiveness Cost- Incremental Mean (SD) Effectiveness Time to 100% Effectiveness Caring Group Mean (SD) Cost Cost Time to 100% Closure, d Closure, d Ratio RN ET/AWOS $ ($ ) (73.61) $26.00 (51% 100%) RN ET/AWOS $ ($ ) $ (153.50) $73.00 (0% 50%) Abbreviations: AWOS, advanced ostomy and wound skills; ET, enterostomal; RN, registered nurse; SD, standard deviations.

13 WJ350203_ qxp 3/3/08 4:01 PM Page 181 J WOCN Volume 35/Number 2 Harris & Shannon 181 care delivery based on their observations. The more frequently the ET/AWOS nurse visited the patient, the more positive were the healing outcomes and the cost savings in nursing time. It also demonstrates successful collaboration among different representations of healthcare delivery. The CAET is a professional nursing association representing and fostering the development of its ET nursing members by utilizing research and theory; reflective practice and experience in the ETNEP; offering continuing educational support; and the recognition as a nursing specialty within the Canadian Nurses Association for its members. These qualities are illustrated in Harrison s Evidence for Practice Framework. 18,19 Limitations FIGURE 10. Incremental cost-effectiveness plane of RN ET/AWOS specialty nurse involvement for the healing of acute wounds. data, and as such this study will be subject to much scrutiny. It can serve as a model for further study and analysis not only within this specialty but also for other nursing specialties. In addition, this model can provide a template for further study for other models of ET/AWOS practice. The results supported what the specialty agency believed to be true about the uniqueness of the models of It was not possible to meet the aim of a minimum of 20 charts per care provider group for each type of wound, with a total of 140 patients for each care provider group (all wounds). This was directly influenced by the fact that the visiting nursing agencies did not store their discharged charts on site and did not have a computerized database to list charts by wound etiology. The RN/RPN only (no ET/AWOS) model did not yield enough cases for analysis, partly because many nurses did not record serial wound measurements. The small cross-section of practice within 2 locales in Ontario may not be representative of all practice sites. The unique ET/AWOS nurse delivery model lacks a comparator TABLE 9. Predicted Financial Impact on the Ontario Ministry of Health and Long-Term Care Budgets for RN ET/AWOS Involvement in Community Wound Management RN ET/AWOS Involvement (Chronic Wounds) High Moderate Mild Minimal Expected cost per $ $ $ $ wound treatment Chronic ulcer prevalence (1 y) a Estimated number of patients qualifying for referral (70%) Total yearly cost $ $ $ $ (predicted) RN ET/AWOS Involvement (Acute Wounds) 50% Involvement 50% Involvement Expected cost per wound treatment $ $ Acute wound prevalence (1 y) a Estimated number of patients qualifying for referral (30%) Total yearly cost (predicted) $ $ a Assume 1 wound per patient. Abbreviations: AWOS, advanced ostomy and wound skills; ET, enterostomal; RN, registered nurse.

14 WJ350203_ qxp 3/3/08 4:01 PM Page Harris & Shannon J WOCN March/April 2008 elsewhere in country, but provided ease of access to a large number of charts of clients who had received this model of care. To generalize to other settings may be limited by virtue of constraints within regional healthcare systems. Implications for Clinical Practice The results of this study support a paradigm shift (Figure 11) to redirect some of the costs of existing general nursing to an increased budget for ET/AWOS nurses, who are qualified nursing wound specialists, performing comprehensive baseline and monitored assessments, comprehensive and interprofessional development of treatment plan, consultative in complex wounds, application and dissemination of wound research, and development. They demonstrate both clinical practice and economical advantages on significant patient care outcomes. The greater the knowledge and skills of the nurse, the better the patient outcomes (cost outcomes, debridement, documentation of teaching, and use of advanced wound care dressings). This allows for cost savings and other resources to be redirected into other valuable healthcare delivery issues, that is, more staffing, staffing education, product, and research. Indirect involvement of the ET nurse may influence practice of others through mentoring, as seen with the ET/AWOS model of the specialty agency. This agency is led by ET nurses who provide both direct and indirect management of wounds for all of their patients. Fifty percent of visits are made by RN ETs and 50% by RN AWOS who are educated and skill prepared by ET nurses with ongoing supervision and consultation by the ETs. At no point does the RN with AWOS treat a patient from admission to discharge independently. Indirect impact by the ET nurse can also include education, program development and management, policies and procedures, care paths, assessment, documentation and teaching tools, product selection, participation in wound research, and development best practice guidelines. FIGURE 11. Paradigm shift with respect to ET practice. Our results suggest that limiting the number of ET visits per patient may result in less positive patient outcomes for patients with acute and chronic wounds. In contrast, an increased frequency of visits and consistent presence of an ET/AWOS nurse will better support the accountabilities of nursing agencies when managing patients with wounds. Conclusions ET/AWOS nurses positively impact healing trajectories and cost outcomes of chronic and acute wounds as direct care providers and when practicing in a hybrid model that combines ET/AWOS care with care delivered by RN/RPN. These specialty practice nurses deliver skilled care to patients with complex wounds, and the results of this study provide further evidence that ET/AWOS nurses are essential leaders on the wound care team. ACKNOWLEDGMENTS The authors acknowledge and thank the Canadian Association for Enterostomal Therapy for their collaboration and funding of this study. They also acknowledge Lauren Alexander and Cathy O Brien, McMaster University Conestogo College BScN program; Lisa Parks, Julie Straus, and Jodie Perkins, ET NOW; Kim Voelker and Pam Hurrell, CCACWW; Tally Hill ParaMed, Sue Wideman, and Deb Dalton, CarePartners; Deb Kauk, Lani Mavin, and Kelly Baechler, Comcare in Waterloo Region; Anita Coles, Oxford CCAC; Cathy Walker, CarePartners Oxford County; Dr Steven Abdool, Homewood Health Centre Research Ethics Board; and Dr R. Gary Sibbald and Dr Douglas Queen, Provincial Outcomes Project. The authors are appreciative of the valued editorial assistance from Kathryn Kozell, MScN, RN, ACNP, ET, President of the CAET. References 1. Knight L. Personal discussion re: visit volume and characteristics for Care Partners visiting nursing agencies in the province of Ontario Heater BS, Becker AM, Olson RK. Nursing interventions and patient outcomes: a meta-analysis of studies. Nurs Res. 1988; 37(5): Waterloo Region Community Care Access Centre. CCAC090 Enterostomal Therapy (ET) Consultation Request form Mar 3/ Orridge C, Purbhoo D. Wound Care: A Guiding Framework: A Joint CCAC Initiative in Collaboration With Their Service Partners. Toronto, Ontario: Wound Review Project, Toronto CCAC; Expectations of Registered Nurses and Registered Practical Nurses. The College of Nurses of Ontario Web site. cno.org/international_en/intro/expectations.htm. Accessed December 18, Robson MC, Maggi SP, Smith PD, et al. Ease of wound closure as an endpoint of treatment efficacy. Wound Repair Regen. 1999;7(2):90 96.

15 WJ350203_ qxp 3/3/08 4:01 PM Page 183 J WOCN Volume 35/Number 2 Harris & Shannon Kaplan E, Meier P. Nonparametric estimation from incomplete observation. J Am Stat Assoc. 1958;53: Polansky M, van Rijswijk L. Utilizing survival analysis techniques in chronic wound healing studies. Wounds. 1994; 6: Peto R, Pike C, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient; part II: analysis and examples. Br J Cancer. 1977;35: TreeAge Software, Inc Richter A, Mauskopf J. Monte Carlo simulation in health care models. Value Health. 1998;1(1): Mauskopf JA, Sullivan SD, Annemans L, et al. Principles of good practice for budget impact analysis: report of the ISPOR task force on good research practices budget impact analysis. Value Health. 2007;10(5): SPSS, Inc Sussman C, Bates-Jensen B. Assessment of the skin and wound. In: Wound Care a Collaborative Practice Manual. 3rd ed. Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins; 2007: Arnold N, Weir D. Retrospective analysis of healing in wounds cared for by ET nurses versus staff nurses in a home setting. J Wound Ostomy Contin Nurses. 1994;21(4): White R. Delayed wound healing: who, what, when and why? Nurs Stand Suppl. 2006: Dicenso A, Cullum N, Ciliska D. Implementing evidence-based nursing: Some misconceptions. Evidence-Based Nursing. 1998, 1 April: Harrison MB. Queen s Nursing 301 Course, Professional Practice Framework. 19. Harrison MB. Paper presented at: Evidence Based Nursing: Theory Science and Practice Workshop; December 2003; Queen s University, Kingston, Ontario, Canada. Call for Authors: Wound Care Review articles, case studies, case series, and original research reports focusing on the potential role of unprocessed honey in wound healing Review articles or original research reports focusing on the antibacterial properties of silver Continuous Quality Improvement projects, research reports, or institutional case studies focusing on innovative approaches to reduce facility-acquired pressure ulcers Case studies, case series, review articles, and original research reports focusing on topical therapies for pressure ulcers, vascular ulcers, or neuropathic (diabetic foot) ulcers Original research reports focusing on the histologic and clinical effects of negative pressure wound therapy