Mobile Wound Management System Application: A Three-year Retrospective Study of its Effect on Quality of Coding Pressure Injury at Three Acute Care Hospitals

Pressure injuries (PIs) continue to burden the health care industry, affecting approximately 2.5 million patients and accounting for approximately 60 000 deaths annually in the United States.¹ Pressure injuries also have a negative financial effect on the US health care system and are associated with an additional annual cost of $43 000 per related hospital stay and a $25 billion overall total cost nationally.¹ Although there are existing PI staging criteria to aid in patient care, these criteria have shortcomings.

The Braden Scale is an assessment tool used to predict low, medium, and high risk for PI to aid in identifying early prevention efforts that can be implemented to prevent PI; it is not used for staging wounds or wound status.² The Bates-Jensen Wound Assessment Tool (BWAT), the Pressure Ulcer Scale for Healing (PUSH), and the Healing Progression Rate tool (HPR [a modified version of the PUSH tool]) are used to assess wound status.^3,4 Each tool includes questions and criteria for assigning a score. To begin, a baseline score is determined, after which the particular tool is repeated periodically to document wound status over time (same, better, or worse), depending on how the score fluctuates. None of the aforementioned tools assist with wound staging to accurately comply with the National Pressure Injury Advisory Panel (NPIAP) criteria, however.

The primary purpose of this study is to assess the consistency in coding the reported severity of PIs present on arrival (POA) and the severity of hospital-acquired PIs (HAPIs) before and after the implementation of a redesigned wound assessment tool. The original tool was developed as a paper cutout that was to be placed adjacent to/bordering a wound to aid in PI assessment. With this in mind, a wound management system (WMS) was designed to electronically integrate the tool within a mobile application. The WMS combines both electronic photography and documentation in a mobile application that was used alongside a developed expert assessment team focused on a consistent formula for assessing and coding PIs.

A previous version of the aforementioned physical tool, the NE1 Wound Assessment Tool (WAT) (Medline Industries, LP), has been shown to be highly reliable, validated, and easily implemented by registered nurses (RNs).^3,5,6(The study device was invented by author N.E.) To facilitate easier use of wound photography and management of PIs, a smartphone application that integrates the WAT was developed. The NE1 Wound Management System mobile application (WMS app; Medline Industries, LP) was designed to simplify wound classification and better assist health care workers in the assessment of skin alterations, wounds, and PIs. The updated WMS was studied in years 2 and 3 in this 3-year retrospective review. During year 1, the original WAT was used.

The Centers for Medicare & Medicaid Services (CMS) incentivizes the practice of quality care. With changes made to the reimbursement policy in the past decade for PIs, the CMS provides an additional payment based on the Medicare Severity-Diagnosis Related Group (MS-DRG) code for the care of more severe PIs (stage 3 and 4) that are POA.^7,8 This reimbursement MS-DRG code is not available if the PI is a HAPI or a hospital-acquired condition (HAC).⁸ Therefore, it has become increasingly important to capture an adequately staged and coded PI that is POA. The WAT has been shown to improve the accuracy of PI staging with a resultant increase in reimbursement with use of MS-DRG coding.⁷ However, the financial implications of using the WMS app in the acute care setting have yet to be adequately studied in the clinical literature.

A 3-year retrospective review was conducted to determine whether the use of the WMS app affected trends in coding quality and MS-DRG–related reimbursement potential for PIs across 3 acute care facilities (200-, 400-, and 700-bed). Notable patterns related to the number of PIs POA were analyzed to determine changes in coding and quality before and after implementation of the WMS app at these facilities.

Platform description

The WMS app is an integrated platform that provides an accurate, valid, and reliable wound classification system. The foundation of the system design is based on the contribution of critical expertise and knowledge of certified wound care specialists. The WMS app is built on and integrates the WAT, with an electronic iteration of the original tool displayed on the main page of the app interface. A set of tissue-type pictures and written descriptions representative of PIs for each NPIAP stage are displayed at the top of the tool. These pictures aid in matching and referencing wound tissue, type, and color, in addition to the stage. The WMS has 4 green calibrated markings that allow the software to electronically measure wound surface area (length × width, cm²) accurately and consistently.

Upon patient presentation, specific identifiers such as name, date of birth, age, sex, location, admission date, and time are automatically integrated into the WMS app by scanning the patient’s wristband. The smartphone is then used to photograph the wound, and the software automatically measures and records the size of the wound along with a timestamp. This photographic documentation can be used for later reference and can be incorporated into the patient’s electronic medical record (EMR).

The health care worker can complete the Quick Start form on the mobile WMS app. This form was created by advanced wound care experts and is based on an algorithm consisting of the wound history, anatomy, tissue type, and touch/view details (HATT) for use in wound classification and in determining the severity of tissue type damage. History or cause of the wound encompasses the primary cause of the wound (PI or other) and any contributing factors. The anatomic site or body part is the location of the wound. Details on tissue type are used to match the wound tissue types with the picture scale to enable assignment of a value. The touch/view integrates the wound details (ie, intact, nonintact, textures, visible or palpable wound base, blanch test, temperature). Examples of the HATT wound classification and severity level of tissue type damage are provided in Figure 1.

After these key characteristics are input, the image is filed in the system, thus creating a worklist for the Advanced Wound Care Team (AWCT) Department to manage within the patient’s EMR. A depiction of the WMS app with fictitious patient details is shown in Figure 2.

Staff training

Prior to the introduction of the WMS app, the health care staff at all 3 facilities received training on how to use the system. Rollout of the WMS app consisted of training courses on the integrated platform launched in September 2018 on the health care facilities secure smartphones. The goal was to train experts (so-called wound champions) and develop highly reliable organization-wide fundamentals. The program training was designed to successfully integrate teamwork principles into the health care system from admission to discharge, reduce adverse patient events (ie, PIs) related to communication breakdown and/or lack of teamwork, optimize patient outcomes by improving communication (electronic integration of wound image assessments by trained experts), and develop other teamwork skills among health care professionals.

The training courses began with every member of the wound care team watching 5 videos on the basics of the WMS app and the use of the HATT algorithm documentation. This training for the wound care team is intended for RNs, certified nursing assistants, physical therapists (PTs), PT assistants, occupational therapists (OTs), certified OT assistants, speech therapists, respiratory therapists, and registered dietitians.

The AWCT consists of providers (ie, doctor of medicine, doctor of osteopathic medicine, doctor of podiatric medicine, and hospitalists) and PTs or RNs who are certified to provide wound care or working toward wound certification. The AWCT members watched the same 5 training/competency video assignments as the wound care team. A hands-on skills day and/or a 10-minute training session was then administered so each clinician could practice taking wound photographs using the WMS app. Finally, the AWCT evaluated 30 wound test cases to assess their ability to correctly classify PIs and other wound types using the WMS app.

The AWCT Department members are assigned role levels based on competency. A level 1 supervisor is wound certified and has all access privileges to the program, review audits, and administrative mail stops. A level 2 staff member is wound certified and, after passing 50 reviewed wound photographic reports signed off by the supervisor, is allowed to perform second reviews of photographic assessments for themselves as well as level 3 staff members. A level 3 staff member is non-wound certified but is on a path to be certified; these staff members are allowed to perform wound photographic assessments, but a second review by a level 1 or 2 team member is required.

In the final portion of the training, providers (ie, doctor of medicine, doctor of osteopathic medicine, doctor of podiatric medicine, and hospitalist) were given a handout or attended a short meeting in which their responsibilities for reviewing and electronically signing photographic assessment reports were explained. Providers could agree or disagree with the wound findings. Once the provider signs the photographic assessment report, it is electronically filed in the patient’s EMR and is ready to be coded. The PI documentation and workflow in the WMS app is shown in Figure 3.

Data collection

After receiving institutional review board approval, 3 years of data were collected for all adult patients admitted to 3 acute care hospital facilities (200-, 400-, and 700-bed). Data from year 1 (October 1, 2017, through September 30, 2018) were collected before the WMS app was implemented. Data from year 2 (October 1, 2018, through September 30, 2019) and year 3 (October 1, 2019, through September 30, 2020) were collected after implementation of the WMS app at the 3 facilities.

Patient data collected included discharge date, diagnosis code, diagnosis description, diagnosis POA, MS-DRG, MS-DRG description, DRG, DRG description, and insurance type. Patients were stratified based on whether or not a PI was coded as POA. Pressure injuries that subsequently occurred during the inpatient hospital stay were classified as HAPIs. Stage 1, stage 2, or deep tissue PI remained categorized as a HAPI. However, any PI that subsequently progressed to a more severe type (ie, stage 3, stage 4, or unstageable HAPI) met the added criteria for categorization as an HAC. Wounds that were initially coded as not POA that later progressed to PI but that were clinically undeterminable (coding = W) were omitted from the data analysis.

Statistical methods

The objective of this study was to compare the total number of PIs coded as POA or HAPI before and after implementation of a WMS app. The first period consists of year 1 (October 1, 2017, through September 30, 2018), prior to implementation of the WMS app. The second period consists of year 2 (October 1, 2018, through September 30, 2019) and year 3 (October 1, 2019, through September 30, 2020), after implementation of the WMS app.

To determine if there were drastic changes in the patient population over time, the average daily census (ADC) by month for each year was compared for each facility individually using an analysis of variance adjusted for unequal variance, with post hoc Tukey-Kramer pairwise comparisons.

For each facility, a chi-square test with a 95% significance level was used to determine whether the number of PIs POA among all PIs differed significantly between the period before implementation of the WMS (preimplementation) and the period after implementation of the WMS (postimplementation). In addition, a chi-square test with a 95% level of significance was used to compare the number of reported PIs that were POA among all PIs reported for each pair of years (ie, years 1 and 2, years 1 and 3, years 2 and 3) at each individual facility.

The number of PIs categorized as either stage 3 or stage 4 POA among all reported PIs was counted. The Fisher exact test with a 95% level of significance was used to compare the number of reported stage 3 or stage 4 PIs that were POA among all PIs reported between the period prior to implementation of the WMS and the period after implementation at each individual facility (200-bed, 400-bed, and 700-bed). In addition, the Fisher exact test with a 95% level of significance was used to compare each pair of years.

Any stage 3, stage 4, or unstageable PI that was not POA was categorized as an HAC. The Fisher exact test with a 95% level of significance was used to compare the number of HACs among all PIs reported between the preimplementation period and the postimplementation period at each individual facility. In addition, the Fisher exact test with a 95% level of significance was used to compare each pair of years.

All statistical analyses were conducted using SAS software (version 9.4 for Microsoft Windows; SAS Institute Inc).

In the 200-bed facility, the ADC for each of the 3 years was significantly different from each of the other years studied (P <.05 for all pairwise combinations). In the 400-bed facility, only years 2 and 3 were not significantly different from each other (P =.1158), whereas years 1 and 2 and years 1 and 3 were statistically significantly different from each other (P <.0001 for both comparisons). In the 700-bed facility, the average monthly ADC was not statistically significantly different across the 3 years studied.

Pressure injuries coded as POA

There was a significant difference in the number of PIs determined to be POA between years 1 and 2 at the 200-bed facility (P =.0221) and at the 400-bed facility (P =.0138) as well as between years 2 and 3 at the 400-bed facility (P <.0001). However, there was no significant difference in the number of PIs determined to be POA between any of the years at the 700-bed facility (Table 1). The results should be considered in context, because statistical significance (or lack thereof) does not necessarily equate to practical clinical significance when identifying PIs that are POA to the hospital.

Stage 3 or 4 PIs coded as POA

When comparing the number of stage 3 and stage 4 PIs among all PIs reported at the 200-bed facility, evidence was sufficient to suggest that there was a difference between preimplementation and postimplementation (P =.0456), between years 1 and 3 (P =.0068), and between years 2 and 3 (P =.0382). At the 400-bed facility, evidence was sufficient to suggest that there was a difference in the number of stage 3 or stage 4 PIs among all PIs reported between preimplementation and postimplementation (P =.0072), between years 1 and 2 (P =.0284), and between years 1 and 3 (P =.0076). At the 700-bed facility, there was not enough evidence to suggest a difference in the number of stage 3 or stage 4 PIs among all PIs reported either between preimplementation and postimplementation or between any pairs of years (Table 2).

Pressure injuries coded as HAC

At the 200-bed facility, there was not enough evidence to suggest a difference in the number of HACs among all reported PIs between preimplementation and postimplementation or between years in any pairing. At the 400-bed facility, there was enough evidence to suggest a difference in the number of HACs among all reported PIs between years 1 and 2 (P =.0427) and between years 2 and 3 (P =.00009). At the 700-bed facility, evidence was sufficient to suggest that there was a difference in the number of HACs among all reported PIs between preimplementation and postimplementation (P =.00004), between years 1 and 2 (P =.0013), and between years 1 and 3 (P =.0038) (Table 3).

Research has shown that the WAT is a simple tool that can improve the skin assessment ability of RNs with little training.³ Nurses are considered the frontline for assessing PIs that are POA; thus, they were a key focus of the training regimen that was rolled out with the WMS app at the 3 facilities studied. The WMS app integrates assessment responsibilities equally across all health care personnel, helping to facilitate proper skin assessment from all entities into the EMR systems more easily and accurately, from the moment the patient arrives at the hospital throughout their inpatient hospital stay.

Wound assessment by nurses or other qualified frontline staff is quickly achieved through a simple time-stamped wound photograph capture using the WMS app. The process begins with documentation of wound history and anatomy (HA of the HATT method). The photograph and information are electronically passed along to the AWCT worklist to add further expert assessment information to document the more severe tissue type and touch/view details (TT of the HATT method) for correct wound classification and care recommendations. This workflow that allows for proper wound classification and staging of all types of wounds is a key characteristic of the WMS app. The use of evidence-based clinical documentation (ie, HATT method) eliminates the imperfect practice of guessing PI stage and other wound classification levels of injury. Certified wound care experts undergo evidence-based training and have access to evidence-based documentation software, whereby they can quickly and accurately assess and adequately stage suspected PIs for accurate coding. Evidence has shown that the creation of quality improvement teams that combine the expertise of health care workers has considerable potential to improve wound management documentation.⁹ The WMS app was created with this concept in mind, allowing for the development of a dynamic team with shared responsibilities for wound classification, thus providing for more accurate coding.

A common misconception is that HAPIs reflect negatively on a facility.¹⁰ This misconception indirectly results in an overall decrease in the reporting and coding of HAPIs, even though HAPIs are an overall nursing quality indicator.¹⁰ Underreporting of HAPIs is detrimental to patient care and, as a result, PIs that were not captured at the preliminary assessment may progress to an HAC, a coded condition that negatively affects the facility if not captured on arrival.⁸

Reporting of hospital quality metrics is required throughout the US health care system by accrediting bodies and regulatory agencies such as the National Quality Forum and the CMS. These entities list stage 3, stage 4, and unstageable PIs as reportable severe events or HACs.¹ Concerning financial compensation to facilities, the CMS helps pay for the treatment of MS-DRG–related codes for stage 3 and stage 4 PIs that are POA.⁸ The MS-DRG is not available for stage 1, stage 2, or deep tissue PIs, nor any PI that is an HAPI or HAC.⁸ The increased reporting of PIs POA may allow for an additional increase in revenue potential for the facility by the discovery of higher-acuity injuries when they present. The inability to capture a PI that is POA is problematic for 2 reasons. One reason is the missed earning potential for the treatment of PIs, especially stage 3 and 4 PIs, and the second and more important reason is the missed opportunity to perform proper wound care that would prevent the PI from progressing during the inpatient stay.

Although in this study there was not enough statistical evidence to suggest that the number of PIs determined to be POA among all PIs was significantly different between preimplementation and postimplementation at any of the facilities (P >.05 for all facilities), the number of coded PIs POA was greater in the postimplementation phase, with significant increases between specific years in the 200- and 400-bed facilities. This may be attributed to a more streamlined implementation across these 2 facilities, because the initial efforts of the program began at the 700-bed facility. These results should be considered within the aforementioned context that identifying PIs can result in early intervention and treatment. Prevention requires ongoing assessment of risk, and use of the app can aid in identification of PI POA as well as in monitoring the PI throughout the inpatient stay. In this study, more stage 3 and 4 PIs POA were captured after implementation of the app.

The implementation and use of the WMS app fosters the creation of a more collaborative structure among the health care workers involved within the wound management team, a so-called Woundology Department that is led by an AWCT program supervisor (level 1) who is in charge of the oversight of AWCT super users (level 2) and AWCT staff (level 3). Such structured oversight mimics the general functionality and hierarchy of many hospital departments with a similar workflow (ie, radiology department). The PI visual depiction is documented in a managed system in which the proper coding and analysis becomes more streamlined and accurate as it progresses through each stage of the departmental team’s ­responsibilities.

A limitation of the data collection is that due to the de-identification of the data, it is possible that more than 1 PI was reported for the same patient both within 1 year and across the 2 time periods studied. The study authors assumed independence and used the chi-square test for comparisons because this study is focused on the number of PIs determined to be POA regardless of whether more than 1 PI was reported for the same patient. In addition, patient days were not available under this protocol, so ADC was used to determine a greater opportunity for any PI. The ADC is an average census of the patients currently in the facility; a difference in average monthly ADC for each year does not necessarily indicate that the number of patients admitted into the facility differed. The ADC was the best method of assessing whether the increase in identification of PI was due to a larger patient population, and future studies should capture the number of patient days and rates of PI. With respect to the findings in the ADC, although the differences may be significant, this statistical significance may not be practically meaningful or be outside the expectations of changes in the average monthly ADC of health care facilities in general.

The data in this study are limited to patients presenting to an acute care hospital. There was no analysis of any demographic of the authors’ clinical data because the purpose was to determine trends with respect to coding PIs. Additionally, the relative experience of the users between the facilities was not analyzed.

Research has shown that potential risk factors for PI may be specifically dependent on the population studied, for example, intensive care settings or following spinal cord injury.^11,12 The authors of the current study suggest that the data may be more applicable to the larger general population of patients presenting to the hospital with a wide range of comorbid conditions, allowing for an analysis of data comparable to hospital admissions as a whole, rather than a specific patient population. However, more research is required to determine if these trends are applicable using the WMS app in various settings (ie, rehabilitation facilities, home health facilities). Additionally, limitations exist with the retrospective use of large databases such as the varying degrees of detail and accuracy between datasets.¹³ In the current study, an attempt was made to control for these limitations through more diligent data collection and review based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision coding upon admission, thereby providing more structure to the heterogeneous nature of large database review.¹⁴

The results presented herein may have been influenced by variables other than the use of the WMS app. However, many facilities already have wound care teams, and such teams existed at all 3 facilities studied herein before the implementation of the app. The use of the app in addition to the existing practices of the wound care team allowed for a more streamlined process with physician input for identifying PIs POA and providing appropriate patient care.

The use of the WMS app had a positive effect on the quality and standard of patient skin assessment, despite an overall increase in coded PIs. The WMS app provides for more transparent and systematic coding of PIs. This study shows that the system was successfully implemented at all 3 facilities and that its use resulted in enhanced wound care overall.