Benchmarking Chronic Wound Healing Outcomes

Background. Benchmarking chronic wound outcomes (comparing outcomes achieved in practice to those reported elsewhere) begins with documenting and summarizing individual and cohort outcomes of one’s practice, continues with searching for comparable results, and is completed by improving care if benchmark outcomes exceed one’s own. Objective. Literature reviewed illustrated this cycle of document-summarize-search-compare outcomes of a recognized healing measure—percent completely healed within 12 weeks of care. Methods. MEDLINE and Association for the Advancement of Wound Care (AAWC) Venous and Pressure Ulcer Guidelines and Evidence Table searches identified example healing benchmarks from randomized controlled trials (RCTs) with 100 subjects/group with a diabetic foot ulcer (DU), venous leg ulcer (VU), or pressure ulcer (PU). Graphed benchmarks were compared to outcomes reported for same-etiology cohorts to illustrate institutional and patient-level benchmarking. Results. Cohorts in some settings reported results similar to the best RCT results for DU and VU. More adequately powered PU RCTs are needed to provide full- and partial-thickness PU benchmarks. RCTs have strengths and limitations as benchmarking resources. Risk-adjusted analyses would improve benchmarking. Conclusion. Documenting and comparing outcomes to published results can support current interventions, highlight opportunities for improvement, or ensure that interventions applied are working to meet individual patient and wound care goals.

Clinicians use wound care outcomes every day to A) document and report results of care providing feedback to professionals that can improve results¹; B) monitor progress toward achieving clinical and patient-oriented goals, keeping patients on the path to healing, and C) choose interventions with best available evidence of efficacy in achieving desired goals of patient and wound care (Figure 1). The third step is called benchmarking, which means using published wound outcomes reported by others as a framework within which to assess their own outcomes and potential need for improved care. Benchmarking begins with documenting and summarizing individual and cohort outcomes of one’s own practice, continues with searching for results on similar patients, and is completed by improving care if the benchmark outcomes exceed one’s own.

Objective
A recognized chronic wound healing outcome²—percent of wounds completely healed after 12 or more weeks of care—was used to illustrate benchmarking in clinical practice. Example published healing benchmarks for venous ulcers (VU), pressure ulcers (PU), or diabetic foot ulcers (DU) were summarized to provide a framework within which to evaluate individual or institutional outcomes of care to assess needs for improvement.

The MEDLINE reference database and the Evidence Tables for the AAWC Venous³and Pressure⁴Ulcer Guidelines were searched for examples of randomized controlled trials (RCTs) reporting percent of DU, PU, or VU completely healed during at least 12 weeks of care as an example of a commonly reported healing outcome. MEDLINE search terms included: “heal” and “topical” and “random” and [“chronic wound” or venous ulcer or pressure ulcer or diabetic ulcer”]. The AAWC Venous and Pressure Ulcer Evidence Tables were searched using the “Find” function in Microsoft® Word software, and using “heal” as the search term.   

RCTs were included if they compared at least two topical interventions with at least 100 relevant subjects per group and reported percent healed after 12 or more weeks of VU, PU, and DU care with a standardized protocol of care. Treatment group size of at least 100 was selected because a prospective study of chronic wounds generally requires at least 100 subjects per treatment arm to be adequately powered to reveal statistically significant effects.⁵   

Healing outcomes for each intervention were summarized and presented graphically as benchmarks⁶ for VU, PU, and DU healing. These benchmarks were described as a framework within which to evaluate outcomes from an individual clinician’s or an institution’s clinical practice.   

One cohort example for each etiology (DU, PU, or VU) was presented in the corresponding graph to illustrate its comparison to related RCT benchmark framework of reported healing outcomes.   

A clinical case scenario described use of “percent wound area reduction from baseline” as a measured, documented healing outcome, illustrating its use as an early warning that a VU was not responding to care. The scenario described use of the content-validated, evidence-based AAWC Venous Ulcer Guideline 2010 update,³ and its related Evidence Table, to inform clinical decisions in order to optimize VU practice at the individual patient level and improve progress toward meeting that patient’s VU care goals.

Eight RCTs and one meta-analysis qualified as benchmark resources for DU, PU, and VU. Five RCTs^7–11 represent healing benchmarks presented in Figure 2 for DU. One meta-analysis⁵ summarized RCT 12-week percent completely healed results qualifying as PU benchmarks in Figure 3. Three RCTs^12–14 and the same meta-analysis⁵ applied to VU provided the benchmarks for VU in Figure 4.   

Benchmarking Healing Outcomes at the Institutional Level
These outcomes can serve as benchmarks against which to assess the results achieved within one’s practice. For example, using Figure 2 as a benchmark for DU outcomes, a multidisciplinary team applying optimal wound care and consistent off-loading of DU reported more than 70% of both insulin-dependent and noninsulin-dependent patients (area within outline in Figure 2) with a Wagner Grade 1, 2, or 3 DU healed during 12 weeks, based on Kaplan-Meier survival curves.¹⁵ Results of these practices matched or exceeded the benchmarks for percent healed at 12 weeks in Figure 2,^7–11 providing a sound rationale to continue practices applying optimal wound care and consistent offloading by a multidisciplinary team.   

No single RCT with at least 100 subjects with a PU per study arm was found, so the only benchmark for PU in Figure 3 was a meta-analysis of smaller controlled studies. Documented healing outcomes of 57% healed in 12 weeks for a cohort of 507 patients with partial- or full-thickness thickness (Stage II–Stage III) pressure ulcers were similar to the 61% reported for Stage II-III pressure ulcers in the meta-analysis for groups managed optimally with a hydrocolloid dressing.¹⁶ Currently, this appears to be the best benchmark to aim for regarding PU. The cohort of patients with full-thickness (Stage III or Stage IV) PU experienced lower 12-week percents healed (36% of 373 full-thickness PU healed in 12 weeks) than the 61% of 134 partial-thickness Stage II PU reported as healed in 12 weeks from the same settings. Documenting full- and partial-thickness PU healing outcomes separately revealed a need to change future PU RCT designs, separating full- and partial-thickness PU outcome analyses to improve PU benchmarking. There appears to be no qualifying RCT benchmark for full-thickness PU percent healed after 12 weeks of care.   

Illustrating use of Figure 4 as a benchmarking tool for VU management, a specialized VU service engaging in evidence-based practice¹⁷ had increased VU 12-week healing from 12% in a community-managed cohort to 53% in a subsequent cohort diagnosed and managed in the clinic. The specialized clinic 12-week healing outcomes match the best reported in Figure 4 for VU, highlighting the exemplary value of its multidisciplinary leg ulcer service.   

Figure 4 also illustrates the value of risk-adjusted benchmarking for recipients of care. The large honey RCT included predominantly VU £ 6-month duration, £ 5 cm² area, with low risk of delayed healing.¹⁸ A similar RCT on 54 subjects per group with larger, longer duration, sloughy VU dressed with secondary foam dressings and 4-layer compression, reported 44% healed at 12 weeks in subjects receiving primary Manuka honey for 4 weeks and 33% healed using primary hydrogel (P £ 0.037). Study summaries accessible in the AAWC VU Evidence Table or cited articles can help identify studies relevant to one’s clients or patients with their own particular level of risk for delayed healing. Table 1 lists some common risk factors for delayed healing of DU, PU, and VU.^16,19–21 Using these, clinical staff can adjust expectations and resource management for the level of delayed healing risk experienced within their practice setting and recognize key areas on which to focus care for a specific patient to avert healing delays.

Benchmarking Outcomes for an Individual Patient.
As an example of benchmarking outcomes for individual patient care, Mrs. G, has a low-exudate 3 cm² venous ulcer (VU) just above her left medial maleolus. Though she is controlling her lower leg edema well with the multilayer graduated high compression stockings you prescribed, Mrs. G complains of pain and ulcer bleeding on removing the saline gauze wound dressing from her ulcer for her daily dressing changes. While documenting wound dimensions on your institution’s medical records, you realize that her wound reduced in area only 5% during the last 4-weeks. This alerts you that Mrs. G is not likely to heal in 12²² or even 24²³ weeks if you continue the same protocol of care for her.   

You both agree to change to a different dressing regimen. Which of the many dressings with claims of “moist wound healing” should you prescribe? You consult the content-validated, evidence-based Association for the Advancement of Wound Care (AAWC) Venous Ulcer Guideline at aawconline.org/Professional Resources/Guidelines summarizing recommendations from all VU guidelines listed on the National Guideline Clearinghouse at www.guidelines.gov. Searching the AAWC Venous Ulcer Guideline for “dressing” you find several recommendations relevant to manage Mrs. G’s pain and healing challenges (Table 2). You choose a dressing that reduces venous ulcer pain and dressing frequency, while improving healing rates. These effects are listed in bold print in the guideline if supported by A-level evidence (that is two or more RCT on venous ulcers).   

You select a dressing from among those with A-Level evidence, confident that these dressings will give Mrs. G better results than the saline gauze dressings previously used. To assure that this is the right procedure for Mrs. G and her VU, you find that many subjects like Mrs. G were included in the RCTs listed as supporting evidence summarized in the AAWC Venous Ulcer Guideline Evidence Table (also at the Professional Resources/Guidelines section of aawconline.org).   

You discuss instructions for use of the dressing with Mrs. G, share the evidence that it has helped to heal others with similar VU while reducing their pain and dressing change frequency, and give her a copy of the VU patient brochure “Dress and Compress for Success” downloaded from the same website. She gratefully uses the new dressing beneath her multilayer compression stockings, which, as the brochure reminds her, aid venous return and reduce lower leg swelling.   

At her next visit to your office, Mrs. G gratefully reports outcomes similar to that of the hundreds of patients like her in the RCTs supporting her new protocol of care: less pain on dressing removal, which was needed only once or twice per week, and 50% area reduction in the 4 weeks since her last office visit.   

Mrs. G’s case illustrates the useful roles that measured wound outcomes play (Figure 1) in documenting and monitoring wound progress, alerting wound care professionals to problems and use of available resources to inform patient and professional wound care decisions with the best available evidence.

Outcomes were summarized for the example healing outcome, percent healed after 12 weeks of care, as a framework for benchmarking example clinical results for DU, PU, and VU in Figure 2, Figure 3 and Figure 4, respectively. RCTs with 100 subjects per group were used for benchmarking in order to track clearly defined intervention effects. Readers who find this process useful are invited to add similar RCTs to Figure 2, Figure 3, and Figure 4 and use them as continuously improving benchmarking tools accessible at aawconline.org.   

Healing is only one measure of chronic wound outcomes. One can follow this “document, search, summarize, and compare” benchmarking procedure for any outcome of interest, including those described by Gottrup et al.²⁴ Hopefully, future benchmarking publications will summarize the state of the art for measuring other important outcomes.   

A limitation of this work is that it relied only on RCTs and meta-analyses to identify chronic wound healing benchmark interventions. As Carter et al²⁵ notes, RCTs often include only a portion of relevant patients, excluding “outliers” who would increase variability and obscure significant effects of the interventions studied. For example, the cohort study in Figure 2 included Wagner Grade 1, 2, or 3 DU, while some of the RCTs included only Wagner Grade 1 and 2 DU. Off-loading techniques differ among DU studies, causing further variability. Subjects off-loaded with a total contact cast healed 90% of DU in 12 weeks compared to only 65% for subjects receiving similar care off-loaded using a removable walker boot.²⁶ Such variability in inclusion criteria and basic care techniques across studies is why the results are described only as frameworks for benchmarking one’s results, not a cross-study comparison of treatment efficacy. As wound registries expand and report risk-adjusted healing outcomes, related benchmarks will be clarified.   

RCTs were used as a source of benchmarks for healing outcomes because they standardize essential aspects of the intervention(s) studied, such as compression for venous ulcers or offloading for pressure or neuropathic diabetic ulcers. This allows within-study efficacy comparisons, but between-study differences in procedures and subject, and/or wound inclusion criteria, make cross-study outcome comparisons difficult.   

Another important source of benchmarking data lies in growing wound registries, such as that described in this issue of WOUNDS by Dr. Fife and colleagues²⁷ or by the Agency for Healthcare Research and Quality (AHRQ).²⁸ The strength of using registry results for benchmarking is in its compelling call for a change in practice if one’s current healing outcomes for similar chronic wounds falls short of those displayed for similar wounds managed otherwise. A limitation of registries may be variability in procedures of care and adherence to protocol, so those using registry results for benchmarking purposes may be less aware of the exact protocols followed to achieve a registry result. As registries expand, analyses of variables in care that affect outcomes will help clarify ideal care.

Measured, documented wound outcomes can be powerful tools for improving practice and progress toward patient and wound goals. Benchmarking one’s outcomes compared to robust results reported in the literature can either identify opportunities for improving practices or justify adhering to current practice. Only if one’s current healing outcomes for a relevant wound fall short of corresponding benchmarks is there reason to consider changing practice.

Dr. Bolton is an Adjunct Associate Professor, Department of Surgery, University of Medicine and Dentistry of New Jersey.

Address correspondence to: Laura L. Bolton, PhD, FAPWCA 15 Franklyn Place Metuchen, NJ 08840 llbolton@gmail.com