INTRODUCTION
An estimated 2.5 million hospitalized patients have experienced pressure ulcers (PrUs) in the United States.1 Expenses to care for patients with PrUs are estimated to be in the billions ($10.2 billion), and these patients are more likely to need long-term care after discharge from the hospital.2 Government healthcare programs cover 9 of 10 patients admitted with PrUs.3 Hospital days for patients with PrUs are 2 to 3 times as long as patients without PrUs.4 In October 2008, the Centers for Medicare & Medicaid Services (CMS) launched payment restrictions for categories of physical conditions that were deemed high volume/high cost, resulted in a higher payment when added as a secondary diagnosis during hospitalization, and could have been prevented if evidence-based guidelines had been followed.5 This policy change means hospitals lose reimbursement for services related to caring for persons with Stages III and IV PrUs when the PrU was considered to be acquired during the hospital stay. Even if the PrU was present on admission (POA) but not documented, it is considered as a hospital-acquired condition (HAC) and not reimbursable. The loss of repayment for HACs, such as Stages III and IV PrUs, increases the importance of having an accurate process to identify and document PrUs that are POA.
The POA condition is identified and documented in the physician's admission history and physical. On patient discharge, the medical records and billing coders review the physician's discharge summary. If the PrU is present but not documented in the admission history and is found on the discharge summary, it is classified as an HAC by CMS and is coded accordingly.
The study hospital had started the process to achieve Magnet designation. As the Magnet's "Forces of Magnetism"6 were reviewed, several nursing-specific activities, such as quality of care, quality improvement, and advancing the culture of safety, were prioritized for initial Magnet action. Data from CMS and the National Database of Nursing Quality Indicators (NDNQI) reports provided evidence to support nursing practice changes. A previous internal CMS report of the hospital's PrU POA/HAC data identified several hospital-acquired ulcers that were not identified as POA and properly noted in the physicians' admission history. The result of missing documentation of POA PrUs means the facility did not receive reimbursement for the care provided related to PrUs. Next, the 2012 NDNQI comparison report for facilities similar to the study facility demonstrates a national benchmark for hospital-acquired pressure ulcers (HAPUs) (Stages I-IV) at 2.75%, and this hospital has a rate of 4.95%.
The purpose of this study was to determine if an electronic medical record (EMR) improvement would enhance documentation and communication of POA PrUs. Currently, the condition of the patients' skin is assessed by the nurse while conducting the nursing admission physical assessment, and those findings should be communicated to the physician. This communication may be verbal or written, but there is no designated method to document that the information had been shared or acknowledged. In order to document the communication of this information to the admitting physician, the wound assessment screen in the EMR was revised to include a new section for indicating whether a PrU was POA; if the "yes" box was checked, the EMR prompted the nurse to notify the physician. Once the notification of the admitting physician was complete (by phone or in person), the nurse updated a "provider notification screen" to document the communication occurred. Nurses and physicians were notified of this new process through e-mails, postings, and unit meeting announcements. A similar electronic process was already in place for provider notification of laboratory values and changes in patient condition.
A review of both the Cumulative Index to Nursing & Allied Health Literature and MEDLINE databases revealed the majority of literature focused on risk factor analysis, prevention programs, intervention comparison between POA and HAPUs, and a likelihood to have prevention interventions consistently in place to prevent further skin breakdown among high-risk patients. The amount of research currently found in the literature specific to POA documentation of PrUs is sparse.7-12
The paper format or physician sign-off seems to be the most frequently suggested method to communicate findings to the physician.13-15 Those interventions may be successful but do not provide electronic traceable evidence. In the EMR format, a traceable intervention can be researched by unit, time period, nurse, and intervention type. As more documentation is converted from paper to electronic, the retrieval of information can be accomplished by requesting a specific report from the EMR. The proposed intervention in this project provided an electronically traceable intervention from the EMR.
Reading this article will help increase clinicians' awareness of the importance of documenting PrUs that are POA.
METHODS
The research question posed to study participants was: Did a formalized electronic process designed to identify and communicate assessment findings of POA PrUs from clinical nurses to the admitting physicians (1) reduce instances of missed identification of POA PrUs, (2) improve communication processes between clinical nurses and physicians, (3) improve physicians' documentation of POA PrUs, and (4) improve reimbursement to the hospital for care of the patient with a PrU?
This before-and-after study design was conducted at a single site. The "before" time frames were November 2010 to February 2011 and November 2011 to February 2012. The postintervention period is November 2012 to February 2013. The study took place in a 333-bed, acute-care hospital in the Midwest. Nursing units involved in the data collection included critical care, medical-surgical, orthopedics, and oncology.
The hospital's quality improvement department's PrU/HAC report was used as the initial source of data. It gathers data based on CMS inclusion criteria, including all qualifying patients (aged >=18 years) admitted to this hospital (inpatients) for 5 or more days, with the exception of patients admitted to the obstetrics unit, patients with a principal diagnosis of a PrU, patients with paralysis, or those admitted directly from a skilled care facility.5
The second source of data for this project was the medical records for the patients on the units listed on the quality improvement department's report. Specific sections of the EMR were reviewed for physician and nursing documentation entries relevant to POA PrU identification and communication of POA assessment findings. The tool featured in the Agency for Healthcare and Research Quality's (AHRQ's) "Preventing Pressure Ulcers in Hospitals: A Toolkit for Improving Quality of Care"16 was used as a guide to abstracting information from the medical record. This tool was designed to measure key processes of care. Among the information abstracted is POA detail and nurse/physician notification. The tool is not specific to either CMS or NDNQI criteria but collects staging and notification information for all PrUs.
The NDNQI report was used as an additional source of data to complement the primary data source and guide the review of the medical record. The NDNQI patient inclusion/exclusion criteria included all patient age groups and no minimum length of stay. All stages of PrUs are reported.17 Excluded are wounds that are not pressure related, patients who are actively dying, and PrU prevention is no longer a treatment goal, well-baby, levels I and II neonatal units, obstetrics, psychiatric (includes geriatric psychiatrics), pediatric rehabilitation, and emergency department. It was vital to include the NDNQI definitions and quality outcomes from the Magnet journey standpoint as the final results and conclusions are reported to the hospital to help keep the focus on all PrUs and quality nursing care, not just the CMS reimbursement aspects.
The study population was nursing staff because they document the assessment findings into the EMR. The indirect study population was the physician staff given that they receive the contact from the nurse. The EMR prompt to contact the physician of POA/PrU provided verification of the communication process. This project was approved by the institutional review boards from the university and the hospital.
Data analyses of the preintervention and postintervention POA and HAC rates were conducted using a test of the difference between 2 independent proportions. A software program called Decision Analyst that runs STATS 2.0 was used for the calculations
The medical record electronic prompt was launched on November 1, 2012. The hospital's PrU POA/HAC reports from each month in the preintervention time frames (November 2010 to February 2011 and November 2011 to February 2012) and the postintervention time frame (November 2012 to February 2013) were reviewed. Next, in all of the time frames, the medical records were reviewed using the tool from the AHRQ's toolkit, the Montana Rural Healthcare Performance Improvement Network tool.16 Special attention was paid to the descriptions of PrUs in the nursing and physician documentation for consistency, completeness, and use of terminology.
RESULTS
The breakdown of the age and sex of the population admitted to the hospital with PrUs from the preintervention and postintervention data collection periods is shown in Table 1. The average age of the patient was between 74 and 77 years old, and the majority of the patients were female, except in 2012-2013. In all years (2010, 2011, and 2012), Medicare was the primary insurance payer (79%-88%), with private insurance for 7% to 15% and Medicaid covering the remainder. Table 2 shows the numbers of PrU POA and HAC for the preintervention and postintervention time frames.
Compared with the preintervention time frames from November 2010 to February 2011, a reduction in the instances of missed POA PrUs had been achieved (Table 3). The postintervention results compared with the preintervention time frames show a statistically significant change (2010/2011) compared with 2012/2013: P < .01, z = 2.507; 2011/2012 compared with 2012/2013: P < .01, z = 2.632). The number of POA-reported PrUs had increased from the preintervention time frame to the postintervention time frame. Table 4 shows the significance of the reduction in HAC, preintervention compared with postintervention. The HAC rates were 0% for all months in the postintervention period of November 2012 to February 2013. For this time period, there were a total of 57 patients admitted with PrUs (POA), and no patients developed HAC PrUs by discharge. Statistically, the reduction in HAC was significant in 2010-2011 compared with 2011-2012 and with 2012-2013. However, this change cannot be explained by the implementation of the electronic prompt because the data demonstrated the most significant reduction occurred prior to the intervention.
In addition, the electronic prompt to document the nurses' contact with the physician specific to the discovery of a POA PrU was not frequently used. The data indicated it was used only 10 times (17%) during the postintervention time frame. The best-case scenario would have been for the electronic prompt to be used consistently (at the rate of 90% or better).
Examination of the medical records demonstrated that communication improved between the nurse and the physician. This was evidenced by the physicians' admission histories and physical examinations, which included skin assessment documentation acknowledging any POA PrUs. A comparison of the physician notes to the nurse notes (comparison made for any descriptions on the condition of the skin and/or the presence of any type of PrU) showed that 77% of the time similar descriptions were found in both the physician and nurse admission notes in the 2012-2013 postintervention time frame.
In 2011-2012, a comparison of the physician notes to the nurse notes (compared for any documentation on the condition of the skin and/or the presence of any type of ulcer) shows 50% of the time there were similar descriptions in both the physician and nurse admission notes. In 2010-2011, a comparison of the physician notes to the nurse notes (compared for any documentation on the condition of the skin and/or the presence of any type of ulcer) shows that 24% of the time there were similar descriptions in both the physician and nurse admission notes.
Over the preintervention and postintervention time frames, the quality of the physicians' documentation has improved (use of terminology). In the 2010-2011 time frames, the physician notes indicated "reddened areas of skin" or "problems with heels." By 2012-2013, the physicians' notes indicated "red sacral area POA." The use of the POA term shows an evolution in physicians' awareness of the skin status at admission and, more specific, the use of terminology indicating aspects of the patients' condition.
The effects on reimbursement can be demonstrated in the 4-month postintervention time period. Of the 57 documented POA PrUs, there were a total of 9 that were Stage III or IV. Had any of these not been properly identified on admission, according to the CMS guidelines, the cost to the hospital would be $43,180 per PrU2 for a potential loss in reimbursement of $388,620.
DISCUSSION
The increase in the reported POA-identified PrUs showed an improvement in communication between the physician and nurse based on how documentation contained more consistent descriptions in the postintervention time frame compared with the preintervention time frame. Although the use of the EMR nurse prompt to contact the physician with admission assessment findings of a PrU was not impressive, there was an improvement in documentation overall demonstrated by the increase in the consistency, completeness, and specific terminology used in comments found in both physician and nurse documentation.
Even with the significant amount of documentation in both the physician and nurse notes, the quality of the documentation was not equal. The nurse documentation was more descriptive (based on EMR-structured nursing admission physical assessment fields with terminology choices) compared with the physician documentation. At the time of this study, the physicians were dictating their documentation using a basic history and physical outline. With only the basic dictation outline, the physician documentation lacked choices in wound description terminology, staging choices, and differentiation between skin changes related to pressure versus other skin integrity issues (such as those related to venous insufficiency or diabetes).
A broader look at preintervention and postintervention time frames revealed that the process to accurately capture POA PrUs had been evolving and improving over time. This study captured a portion of the change, and the results have demonstrated the EMR prompt intervention was not sufficient to have created the improvement. While focusing on the growth of the process to capture POA information, the study uncovered 6 key complementary components that occurred during the same time frame: EMR upgrades, shared governance, unit-based nursing councils, Magnet journey, NDNQI participation, and growth of unit-based skin care champions.
A decision to improve the EMR structure for nursing documentation had already occurred and continues to unfold as part of the bigger picture to upgrade the EMR functionality. The chief nursing officer began a shared governance program, unit-based nursing councils, and the Magnet journey in the 3 years (2009-2012) prior to the intervention portion of this study. Participation in NDNQI quality indicators was initiated in 2008. The growth of unit-based skin care champions had been nurtured since mid-2010. Quarterly meetings and educational offerings provided these unit clinical nurse skin champions with additional resources to help their peers develop and evaluate individualized patient care plans for persons at risk of or who had actual skin breakdown. Combined, these influences and process changes affect the capture of POA PrUs.
LIMITATIONS
There were no controls for patient acuity. Also, at the start of the study period, a new group of hospitalists started at the study hospital. This may have been a confounding factor because the skill and knowledge of the new group of hospitalists may have differed from the group present during the preintervention period. This study represented a specific point in time, and these results were a snapshot of this one point that had a unique mix of patients and care scenarios.
Other limitations included differentiating between PrU Stages I and II and maceration-associated skin damage (MASD) or deep tissue injury (DTI).18 Education on the definitions and clinical presentation of MASD and DTI had not been highlighted, which may cloud the correct identification of the cause of the skin ulcer. The National Pressure Ulcer Advisory Panel (NPUAP) had updated the staging criteria for Stages I and II to specifically differentiate Stage 1 PrUs from DTI (bruising color) and MASD tissue presentations from a Stage II PrU.19 At the time of this writing, CMS is focused on Stages III and IV HAC, and reporting of DTIs is not required. However, as the tissue damage of DTIs begin to show, they are more likely to present as a Stage III or IV PrU.20
RECOMMENDATIONS
The terminology used to describe skin changes related to PrUs and the details of wound descriptors need to be reviewed and reinforced with all of the clinical nursing staff. Although documentation from the physician and nurse may have been describing the same wound or skin condition, it was often very confusing on what specifically was being represented. Standardized terms useful in PrU and skin-issue-related documentation can be found in guidelines published by the NPUAP/European Pressure Ulcer Advisory Panel.
Next, education of the clinical nursing staff about the Braden scale for patient risk assessment with focus on the subscales: the definitions of each subscale, patient evaluation, and interventions specific for individual patient care scenarios, would be helpful to provide a baseline for accurate risk assessments specific to the individual patient. One of the themes in the nurse documentation review was the recognition of patient incontinence as documented in the shift-to-shift notations, but the risk assessment subscale score was graded for no moisture issues. Another theme noted was confusion of the cause of redness from pressure and redness from incontinence issues and how to differentiate. Incontinence-associated skin issues often require additional and ongoing education to clarify the difference between pressure-related issues and moisture-related skin issues. Use of the unit skin care champion and additional expert resources can create interventions customized for each type of patient skin care issue identified.
Physicians would benefit from an introduction to terminology specific to PrUs and associated skin conditions. In addition, building an admitting physicians' skin assessment history and physical EMR structure that imports the nurses' initial physical assessment data would provide consistency in terminology, location, and skin descriptions automatically.
From individual to unit-based activities, executive to physician, each staff member involved in the care of patients can provide support and consistency in helping with identification and prevention of PrUs. No one individual item created the accuracy; it is the succession of each step that built this process. These components became part of a process that continued to achieve system-wide accuracy in capturing POA PrUs and preventing the development of HACs such as the PrU.
PRACTICE PEARLS
* Prompt identification of skin conditions and follow through on condition communication between clinical nurses and admitting physicians enhance patient safety and hospital reimbursement.
* Do not assume the admitting physician is aware of the patients' skin condition.
* Educate providers (nursing and medicine) on terminology specific to pressure ulcer and associated skin conditions.
* MASD/incontinence-associated dermatitis-specific education helps clarify the difference between pressure-related skin issues and moisture-related skin issues.
* Unit-based skin care champions can function to enhance the process of proper identification of skin conditions.
REFERENCES