Reducing hospital acquired pressure injury in a learning health center: Making the case for quality

Abstract Introduction The purpose of this descriptive study is to examine a learning health system (LHS) continuous improvement and learning approach as a case for increased quality, standardized processes, redesigned workflows, and better resource utilization. Hospital acquired pressure injuries (HAPI) commonly occur in the hospitalized patient and are costly and preventable. This study examines the effect of a LHS approach to reducing HAPI within a large academic medical center. Methods Our learning health center implemented a 6‐year series of iterative improvements that included both process and technology changes, with robust data and analytical reforms. In this descriptive, observational study, we retrospectively examined longitudinal data from April 1, 2018 to March 31, 2022, examining the variables of total number of all‐stage HAPI counts and average length of stay (ALOS). We also analyzed patient characteristics observed/expected mortality ratios, as well as total patient days, and the case‐mix index to determine whether these factors varied over the study period. We used the Agency for Healthcare Research and Quality cost estimates to identify the estimated financial benefit of HAPI reductions on an annualized basis. Results HAPI per 1000 patient days for FY 20 (October 1‐September 30) and FY 21, decreased from 2.30 to 1.30 and annualized event AHRQ cost estimates for HAPI decreased by $4 786 980 from FY 20 to FY 21. A strong, statistically significant, negative and seemingly counterintuitive correlation was found (r = −.524, P = .003) between HAPI and ALOS. Conclusions The LHS efforts directed toward HAPI reduction led to sustained improvements during the study period. These results demonstrate the benefits of a holistic approach to quality improvement offered by the LHS model. The LHS model goes beyond a problem‐based approach to process improvement. Rather than targeting a specific problem to solve, the LHS system creates structures that yield process improvement benefits over a continued time period.


| INTRODUCTION
The Learning Health System (LHS) is characterized as an organization that integrates internal and external data and evidence to change processes and practices, leading to higher quality, and safer patient care. 1 LHSs have additional characteristics that include leadership commitment to continuous improvement and a culture of learning, a systematic approach to gathering data and evidence, and use of technology to improve decision support, where data are used to refine processes and create a feedback loop for continuous improvement. 2,3 Another important characteristic of LHSs is their patient-centeredness, or the inclusion of patients and families as members of the care team. 4 The results of such an approach are designed to achieve higher quality, safe, and more efficient organizations that attract and retain patients and talented health care workers. 2 Within an LHS, leaders respond to problems or focus areas systematically in three phases: (1)  In a LHS, health care quality is dependent on a myriad of factors that can be categorized according to Donabedian's model of structure, process, and outcomes. 6 The organizational structure within which care delivery occurs is composed of the physical environment, people, policies, and culture, and physical resources, such as equipment and supplies, all contributing to the quality of care. Processes or workflows are the engine for care delivery, the mechanics of quality, while the outcomes are the clinical, financial, and operational effects of quality.
Since the seminal publication "Crossing the Quality Chasm," health care leaders were encouraged to commit to the pursuit of high quality and excellent patient outcomes. 7 Although the health care industry's value proposition is often calculated equally by quality and cost, when there is a lapse in quality, costs increase. 8 Likewise, when there is an excessive focus on cost-cutting, quality generally suffers. Despite the positive implications for quality care delivery, there is a cost associated with high quality in health systems. Resources and human effort must be expended to attain excellence in quality care. [9][10][11] From a patient perspective, quality health care is most prominently defined through clinical outcomes and the absence of adverse events. While costs are important, the patient's return to a higher or better state of health determines a safe, quality experience. There are specific and nationally recognized outcome measures within the health care industry for patient safety, including but not limited to rates associated with the development of hospital acquired pressure injuries (HAPI), patient falls, and infections associated with care.
Patient safety and quality are inter-related, and while they remain a focus for health care providers and leaders, there are indications that in the policy arena patient safety in the United States may be waning and that decline is beginning to be felt in health care settings. 12,13 In developing this study, we posed a central, critical question.
Does the LHS approach create a strong case for the value of quality and safety in a health care organization? To address this question, we specifically evaluated if one acute care facility's HAPI improvements were sustained between October 1, 2019 to March 31, 2022, during the pandemic (Centers for Disease Control and Prevention, 2021). 14 The purpose of this descriptive study is to examine one organization's continuous improvement and learning approach as a case for increased quality, standardized processes, redesigned workflows, and better resource utilization. We will describe the relationships that exist between structural and organizational characteristics that have an impact on outcomes specific to one adverse event, HAPI, within one organization. in the United States health care industry, patient expenditures for pressure injuries range from $9.1 to $11.6 billion dollars annually. 16 Effective evidence informed clinical care by providers may mitigate the incidence of pressure injuries, thus funding agencies consider HAPI a never-event hospital acquired condition (HAC). [17][18][19] Our LHS problem of interest and focus area was improving the safety and quality of our patient care by reducing the incidence of HAPI.
Our organization has committed significant resources to HAPI reduction. We began our continuous improvement and organizational learning journey in 2016 with a small improvement research pilot to understand our internal status of the HAPI challenge, specifically staff competencies and HAPI stage attribution. 20,21 Using the 3-phased LHS approach described earlier (Figure 1), we collected, analyzed, and interpreted data to discover that pressure injury staging was inaccurate, in some cases by as much as 50%, and our pressure injury reporting structures and systems were flawed. From this effort we had two significant learnings: first, developing staff competencies in practice is important, but we must utilize the expertise of those with the most competence in making decisions, such as the Wound, Ostomy, Continence (WOC) nurse for staging pressure injuries. Second, improvement cannot occur without actionable and accurate data collected from the point of care. These learnings directly focused our team on a series of iterative improvements, including the development of an analytical platform and resource reforms.
We used data from the initial pilot to represent, manage, and apply knowledge to implement a series of interventions that dramatically changed practices in our hospital. In 2017, the organization resourced a 16-member WOC Team (WOCT) with 12 WOC nurses, a dedicated nurse practitioner, a physician medical director, and two wound care technicians. With this fully staffed WOCT, our WOC nurse-to-bed ratio decreased from 1:300 to 1:100, resulting in more comprehensive patient care and focused staff support for proactive mitigation of skinrelated injuries. In addition, the WOCT supported all-stage attribution of pressure injuries, placing the assessment of HAPI at the level of the expert, with a resultant higher degree of accuracy in pressure injury staging data. We gained significant learnings from this iterative improvement effort. First, resources are necessary for improvement, and while adding human resources may increase cost initially, the long term benefits will largely pay for the investment. Second, we found that the investment in professional development and front-line staff competency for clinical care is a worthy endeavor that is developed over time with expert mentoring and continuous coaching, but also creates a mechanism for additional learning and improvement.
The LHS model naturally aligns with continuous improvement.
We used the phased approach iteratively to initiate continuous HAPI improvement that has been sustained for over 6 years ( Table 1). The data inaccuracies identified during the initial pilot led to the development of a HAPI analytical system 21 that transformed pressure injury reporting in our medical center, leading to actionable data available at the front line of care delivery. Our team integrated a business intelligence tool into our processes and thus converted real-time, expert attributed HAPI staging from the point of care into transformational improvement data. From these data, we learned that we had specific opportunities to improve care delivery for higher risk and vulnerable patients. For example, tracheostomy patients were found to be at higher risk for HAPI associated with trach plates, a process that was improved using standardized protocols with specific treatment plans.
Our clinical leaders began making real-time adjustments for management and prevention of pressure injuries at the earliest stages of skin injury after determining heel and sacral pressure injury were significant risks for our patients, particularly those patients in intensive care units. Integrating preventative processes and measures, such as rounding, heel floating, observational scanning for devices, and equipment, with conjunctive use of silicone adhesive foam dressings mitigated risks. Our iterative learnings prompted our team to institute a significant feedback process using root cause analysis techniques to collect data on all of our HAPI occurrences, thus providing an in-depth analysis of patterns in HAPI outcomes that could be mitigated and processes that were amenable to improvement.
During the COVID-19 pandemic, our sustained improvement efforts focused on the impact of the coronavirus on HAPI. We used our HAPI data to adjust patient care delivery processes 24 to reduce pressure injuries on COVID patients in our medical intensive care units and to evaluate opportunities for health disparities in our patient populations during the pandemic. 25 We used our learnings from the root cause analysis process to quickly identify that patient therapeutics, such as proning were increasing the risk for upper body pressure injuries, particularly on the face and head. Clinical teams were able to mitigate this risk with silicone adhesive foam dressings that we had learned reduced HAPI in other anatomical locations. Our significant efforts to reduce HAPI, learn, and innovate have led us to examine the current impact of the learning health system model for this specific problem of interest.

| Setting
The setting for this project is an urban academic medical center

| Institutional Review Board
Exempt Institutional Review Board approval was obtained from the organization's IRB for conducting this study.

| Study design
In this descriptive, observational study, we retrospectively examined

| Data collection and analysis
An Excel 27 spreadsheet was used by the team to collect organizationally reported counts of all stage HAPI and ALOS. HAPI data are collected from our electronic health record and stored in our HAPI analytic database, a business intelligence platform. Data for ALOS were obtained from the organization's financial decision support system, another business intelligence tool. The data were collected and reported monthly, using hospital level aggregates. We also analyzed patient characteristics, observed to expected (O/E) mortality ratios, as well as total patient days, and the case-mix index to determine whether these factors varied over the study period.
We used the AHRQ cost estimates of $14 506 per event for HAPI. 28 For this analysis, we used the fiscal year, October-September (2020-2021), and created an average number of events for each fiscal year, which was multiplied by the AHRQ estimates for each event. Total HAPI costs per fiscal year were then annualized for 12 months.
To analyze these data, the study team used the functionality within the Excel spreadsheet to create a summary, descriptive statistics, and graphs. 29 We created statistical process control charts using an Excel add-in, QI Macros. 30 We also used SPSS 31 to analyze the correlation between HAPI counts and ALOS. A correlation analysis was used to analyze the relationship between HAPI and ALOS ( Table 2). A statistically significant, strong negative correlation, r = À.524, P = .003, was found.

| The value equation
The value equation in health care improves when an organization invests in improving quality, while either sustaining or reducing costs. 33  Interprofessional collaboration is necessary for achieving sustainable improvements. While the human resource expertise of the WOC Nurses created the workflow and changed processes, adjusted the WOC staffto-bed ratios, and built the interprofessional team to lead the charge; the proficiency of the database experts and improvement scientists who knew how to create sustainable change in a LHC was necessary. However, ultimately, to create a learning system, providers and stakeholders, specifically nurses, are essential to integrating standardized workflow and evidence-based processes for HAPI reduction into practice.

| Implications for practice
These results demonstrate the benefits of a holistic approach to quality improvement offered by the LHS model. The LHS model goes beyond a problem-based approach to process improvement. Rather than targeting a specific problem to solve, the LHS system creates structures that yield sustained process improvement benefits over a prolonged time period. The early investment and development of these structures demonstrated their added benefit of positioning the organization for continued improvement during a period of national decline in overall safety and quality during the pandemic. 32 These improvements have also resulted in cost savings during a period of financial strain noted in health systems nationwide. 34 These findings highlight the benefit of a commitment to a learning infrastructure in bolstering organizational resilience and sustained performance when crisis hits. Leaders must advocate for and invest in structures that support continued organizational learning rather than emphasizing individual strategies for solving specific problems.

| CONCLUSIONS
Becoming a learning health system is an iterative process, developing over time with strong leadership, effective use of data, and a culture and workforce committed to continuous learning and improvement. Developing a LHS culture and stance is imperative in our current health care landscape to effectively achieve safe, highquality care and outcomes. Despite the commitment that is required to evolve into a LHS, leaders effectively completing the journey may ultimately achieve value, defined as the highest quality of care delivery provided at the lowest cost, and an optimal patient experience.
Our experience with HAPI reduction is evidence of an LHS transformational outcome.