INTRODUCTION

A pressure ulcer (PU) is a localized injury to the skin and underlying tissue (usually over a bony prominence) as a result of pressure or pressure in combination with shear (European Pressure Ulcer Advisory Panel, National Pressure Injury Advisory Panel, & Pan Pacific Pressure Injury Alliance, 2019). Around the world, PU prevalence in healthcare settings ranges from 0% to 72.5%, with large variations observed between different countries and clinical settings (European Pressure Ulcer Advisory Panel et al., 2019). The cost of preventing PU per patient per day is between $3.06 and $101.39, and the cost of treatment ranges from $1.97 to $543.8 (Demarre et al., 2015). Therefore, the detection of a PU at its earliest stage is imperative to afford timely interventions.

In health facilities, the number of PUs is expected to increase with the advancing age of the older adult population. This is because the risk of PU is increased particularly in individuals aged 65 and over due to the more frequent occurrence of chronic diseases, structural changes in the skin (e.g., turgor and tone reduction, skin dryness), and limitations in movement activity (Gould & Fulton, 2016; Sgonc & Gruber, 2013).

Identifying those who need prevention strategies is the first step in assessing PUs, and this is achieved through a risk assessment (Moore et al., 2014). Currently, there are very few clinically useful tools to assist with early PU detection and prevention (Hettrick, Hill, & Hardigan, 2017). Although the use of a risk assessment tool is recommended by many international PU prevention guidelines, there is no 100% reliable risk assessment tool for PU risk assessment currently available in clinical practice (Moore & Patton, 2019).

While clinical judgment is considered an option in assessing the risk of PU, the lack of a structure that makes the reproducibility of its performance difficult may lead to inconsistent evaluations and interventions. Nonetheless, some form of structure is still warranted to ensure standardization and consistency in assessing PU risk (Mordiffi, Kent, Phillips, & Chi Tho, 2011). However, recent trends, such as the measurement of subepidermal moisture (SEM), offer promise. The SEM Scanner (Bruin Biometrics, LLC, Los Angeles, CA), is a handheld medical device that is an objective and reliable method of assessment (Moore, Patton, Rhodes, & O'Connor, 2017). This device assesses local tissue bioimpedance, and this enables the detection of early tissue damage and pre-stage I PU before the damage becomes visible to the unaided eye (Moore et al., 2017).

Exposure to prolonged, unrelieved pressure and shear forces, most often related to decreased activity and mobility, are known to be primary causes in the development of PU (Moore, Cowman, & Conroy, 2011). Thus, an alternative method to identify patients at risk of developing PU could be through assessing mobility status. Indeed, impaired mobility is increasingly recognized as a strong risk factor for PU development (Lahmann et al., 2015; Moore et al., 2011). In addition, clinical judgment of whether the patient's mobility is impaired is used frequently by nurses to initiate preventive interventions for PU (Fisher, Wells, & Harrison, 2004). However, there is often inconsistency between this assessment and the patients’ actual movements (Källman, Bergstrand, Ek, Engström, & Lindgren, 2015). Thus, there is a possibility to consider using movement assessment as a more accurate alternative way of risk assessment. Focus on movement by using algorithms and taking the necessary measures against PU development may significantly reduce the problems caused by these wounds to individuals, the health system, and the economy as a whole. Although working with the algorithm has been prioritized, there is no national or international standard movement algorithm for detecting PU risk.

Aims

There were two primary aims of this study: (1) to investigate the relationship between activity, mobility, and PU development; and (2) to ascertain the next step for delineating an algorithm based on activity and mobility for detecting PU risk among older adult residents in long-term care.

METHODS

The study design was quantitative, prospective, descriptive, non-experimental, and undertaken among the older adult population. The study site was a long-term care setting in Ireland. Using convenience sampling, all participants in the clinical setting meeting the inclusion criteria were invited to participate. Therefore, a formal sample size calculation was not undertaken; rather, all potential participants were included following consent or assent. The study involved 53 participants.

The inclusion criteria were the patient was residing in the long-term care, had no PU at the time of recruitment, and had consented to participate or had assent provided by the next of kin.

The exclusion criteria were the patient did not consent to participate or assent was not provided by the next of kin, had a PU at the time of recruitment, and had Parkinson disease.

All patients meeting the inclusion criteria were invited to participate in the study. Ethical approval was obtained from the Research Ethics Committee at the Royal College of Surgeons in Ireland, University of Medicine and Health Sciences (REC1622). The consent request included participants who met the inclusion criteria. If the participant could not consent for him or herself, assent was taken from the next of kin, when available.

Data Collection

Once consent or assent was received (from July 2019 to March 2020), the researcher started data collection. Participants were followed up for 4 days or until a PU developed. The rationale for this time period is that a study carried out by Moda Vitoriano Budri et al. (2020) considered a PU when the individual had at least 3 days of consecutive abnormal SEM assessments (Figure 1).

Study procedure.

Instruments

The Braden Scale is used to assess the risk of PU. Accordingly, a score of 12 points or lower indicates very high risk, 13–14 points high risk, and 15–16 points low risk (Bergstrom, Braden, Laguzza, & Holman, 1987).

The Elderly Mobility Scale (EMS) is a scale for the assessment of mobility (i.e., locomotion, body position balance, and changes in mobility) via a 20-point validated assessment for the older adult (Smith, 1994).

The 6-item Cognitive Impairment Test (6-CIT) is a test for the assessment of cognitive impairment and includes one memory assessment, two mathematical calculations, and three orientation questions (Katzman et al., 1983). The components' values vary from 0 to 28, with higher scores indicating a higher cognitive impairment.

The SEM scanner is placed over an anatomical site, such as the sacrum, trochanters, and heels. The moisture level of the subepidermal tissue is measured using electrode structures placed on the surface of the skin. Low amplitude signals use surface electrical capacitance to assess the level of moisture in the epidermal and subepidermal tissues (Bates-Jensen, McCreath, Pongquan, & Apeles, 2008). SEM measurement was assessed daily over the sacrum, right heel, and left heel. SEM values > 0.5 were considered as abnormal delta.

The EarlySense movement sensor is an automatic motion analysis to facilitate caregivers in identifying patients at risk of developing a PU. In addition, it also measures patient vital signs continuously and it can report patient falls. The system is composed of a monitor that processes and records all data, and a contact-free sensor that is placed under the mattress. Zimlichman et al. (2011) researched to validate the EalySense movement sensor with 114 patients and found the motion score (measured by EarlySense) was significantly different between the PU risk groups as determined by the Norton Scale. Also, the researchers used the Norton Scale as a gold standard to define what cut-off indicates a high risk for developing a PU. The sensitivity and the specificity’s motion score were 85% and 93%, respectively. Therefore, they concluded that mobility, physical condition, and incontinence are highly related to PU development.

The categorization of the EarlySense movement level of participants has been improved using a tool developed by Moda Vitoriano Budri et al. (2020). Using healthy participants’ mean number of movements (i.e., 108.2 movements per hour) as the cut-off point, the authors categorized the movement of the older participants into two groups: (1) “low movers” (i.e., participants who scored < 108.2 movements per hour); or (2) “high movers” (i.e., participants who scored ≥ 108.2 movements per hour).

Visual skin assessment (VSA) was undertaken daily from the sacrum, right heel, and left heel. Once a PU was identified, the international Prevention and Treatment of Pressure Ulcers/Injuries: Clinical Practice Guideline (European Pressure Ulcer Advisory Panel et al., 2019) was applied.

Data Analysis

All data were analyzed using SPSS for Windows 11.5 (SPSS Inc., Chicago, IL, USA). In evaluating the data, frequency, and percentage distribution, the arithmetic average and standard deviation, the chi-square test and the correlation coefficient were calculated.

RESULTS

A total of 53 participants were included in the study. The age of participants varied from 73 to 101 years old, with a mean age of 87.5 years (SD = ± 6.8 years), and 73.6% (n = 39) of the total sample were female. Participants were 100% (n = 53) Irish with a light skin tone. Of the participants, 39.6% (n = 21) had a dementia diagnosis. Of those who had a dementia diagnosis, 20.8% (n = 11) had vascular dementia. More than half of the participants (54.7%, n = 29) had a high specification foam mattress in use (Table 1).

Table 1. Demographics Characteristics of Participants Characteristics Frequency Percent Gender Female 39 73.6 Male 14 26.4 Dementia diagnosis 21 39.6 Type of dementia Alzheimer 9 17.0 Vascular dementia 11 20.8 Lewy body disease 1 1.9 Support surface type High specification foam 29 54.7 Alternating air hybrid 22 41.5 Alternating air dynamic 2 3.8 Mean age 87.5 ± 6.8 (min = 73; max = 101)

The total Braden score showed that 69.8% (n = 37) of participants were at low/mild risk of developing a PU. A total of 67.9% (n = 36) were assessed as being dependent according to the EMS. Most of the participants (79.2%, n = 42) had significant cognitive impairment.

Movement, Mobility, and Activity

The mean movement (mov) score was 102.8 mov/hour (min = 1 mov/hour; max = 365 mov/hour; SD = ± 102.8 mov/hour; Figure 2). Participants’ movement levels showed that 41% (n = 21) were low movers (or prone to immobility) and 59% (n = 32) were high movers (or prone to agitation).

Movement results of participants.

According to Braden mobility subscale, most of the participants (62.2%, n = 33) had no limtations/slightly limited mobility and according to Braden activity subscale 58.4% (n = 31) of participants were chairfast/bedfast.

PU Incidence

Among the 53 participants, 15.1% (n = 8) developed a visual PU (VSA-PU), and all PUs were grade I. The sacrum was the most common anatomical site affected (66.7%; n = 8). Analysis of the SEM delta scores ≥ 0.5 indicated that there was an incidence of 79.2% (n = 42) of early PU damage among the participants (Table 2). In terms of SEM-PU, the right heel was the most common anatomical site affected (73.8%, n = 31).

Table 2. Pressure Ulcer Incidence Frequency Percent VSA-PU 8 15.1 Anatomical location of VSA-PU Sacrum 8 66.7 Right heel 3 25.0 Left heel 1 8.3 SEM-PU 42 79.2 Anatomical location of SEM-PU Sacrum 17 40.4 Right heel 31 73.8 Left heel 19 45.2 Note. VSA-PU = visual skin assessment-pressure ulcer; SEM-PU = subepidermal moisture-pressure ulcer.

Analysis of the SEM delta scores according to the anatomical location indicated that the overall mean SEM score was lower in the sacrum than the heels. Over the study follow-up, both heels of all participants had persistent mean SEM delta scores above ≥ 0.5 (Table 3).

Table 3. Mean SEM Scores for Anatomical Location Location/day Frequency Min. Max. Mean SD Sacrum SEM score – Day 1 53 0.1 1.5 0.46 0.30 SEM score – Day 2 53 0.1 1.4 0.47 0.29 SEM score – Day 3 53 0.0 1.7 0.46 0.31 SEM score – Day 4 52 0.0 1.2 0.36 0.27 Right heel SEM score – Day 1 52 0.1 1.8 0.74 0.43 SEM score – Day 2 52 0.0 1.9 0.72 0.37 SEM score – Day 3 52 0.0 1.7 0.62 0.39 SEM score – Day 4 52 0.0 1.8 0.64 0.38 Left heel SEM score – Day 1 52 0.0 1.9 0.72 0.52 SEM score – Day 2 52 0.0 1.5 0.61 0.33 SEM score – Day 3 52 0.0 1.7 0.57 0.35 SEM score – Day 4 52 0.0 1.5 0.55 0.31 Note. SEM = subepidermal moisture. SEM-PU, VSA-PU, and Movement Category

Of the eight PUs observed using VSA, 62.5% (n = 5) occurred among the low movers, and 37.5% (n = 3) occurred among the high movers. From the 42 SEM-PUs observed, 62% (n = 26) occurred among the low movers, and 38% (n = 16) occurred among the high movers (Table 4).

Table 4. SEM PU and VSA and Movement Category Movement level Total

Low mover

% (n)

High mover

% (n)

SEM-PU Yes 62 (26) 38 (16) 100 (42) No 63.7 (7) 36.3 (4) 100 (11) VSA-PU Yes 62.5 (5) 37.5 (3) 100 (8) No 62.2 (28) 37.8 (17) 100 (45) Note. VSA-PU = visual skin assessment-pressure ulcer; SEM-PU = subepidermal moisture-pressure ulcer. Braden Risk Status and Movement Level

More than half of the participants who were considered to be at no risk/mild risk were classified as low movers, using the movement sensor (Table 5). No statistically significant difference between Braden Risk Status and Mobility Level was found (p = .405).

Table 5. Cross-Tabulation Between Braden Risk Status and Mobility Level Braden risk status Movement level Total p; (x2)

Low

% (n)

High

% (n)

No risk 50 (5) 50 (5) 100 (10) .405 Mild risk 59.2 (16) 40.8 (11) 100 (27) Moderate risk 87.5 (7) 12.5 (1) 100 (8) High risk 62.5 (5) 97.5 (3) 100 (8)

Correlation analysis was undertaken and the correlation between Braden Activity and Movement Level was r = 0.15, Braden Mobility and Movement Level was r = .24, and EMS and Movement Level was r = .05.

DISCUSSION

The present study was designed to determine whether an algorithm based on activity and mobility could be developed to detect PU risk among older adult residents in long-term care.

The incidence rate of VSA-PU was 15.1% (n = 8) and analysis of the SEM delta scores indicated that there was an incidence of 87.5% (n = 42) of early PU damage. The sacrum was the most common anatomical site where PU developed. It is interesting to note that all eight participants who developed a VSA-PU also had a SEM-PU. These results are consistent with other studies and suggest that when there is a change in SEM measurements this corresponds with changes in PU development. Thus, SEM measurement may be a useful method for detecting early non-visible PU development (Bates-Jensen, McCreath, Kono, Apeles, & Alessi, 2007; Bates-Jensen, McCreath, Nakagami, & Patlan, 2018; Bates-Jensen, McCreath, & Patlan, 2017; Bates-Jensen, McCreath, & Pongquan, 2009; Bates-Jensen et al., 2008; Bates-Jensen, Reilly, Hilliard, Patton, & Moore, 2020; Gefen & Gershon, 2018; Gershon, 2020; Guihan et al., 2012; Harrow & Mayrovitz, 2014; Kim, Park, Ko, & Jo, 2018; Moda Vitoriano Budri et al., 2020; O'Brien, Moore, Patton, & O'Connor, 2018; Okonkwo et al., 2020; Park, Kim, & Ko, 2018; Raizman, MacNeil, & Rappl, 2018; Smith, 2019). SEM is described as this accumulation of fluid below the epidermis and changes in SEM due to interstitial fluid accumulation and localized edema are accepted as biomarkers in the development of PUs (Okonkwo et al., 2020). Although the development of PU is usually determined by VSA, which is based on evidence of changes in the skin surface, this method cannot detect what is happening below the skin surface (Gefen et al., 2020; Moore et al., 2017; Oliveira, Moore, O'Connor, & Patton, D., 2017). Therefore, SEM measurement is a clinically relevant tool to predict early-stage PUs.

A key finding of this study was related to activity, mobility, and movement level of participants and PU development. Activity is determined by the patient's ability to move, and it may be temporarily affected by trauma, illness, or exacerbation of a health condition. This activity is linked to the mobility of the individual and is normally an intrinsic capability of the person (Allman, Goode, Patrick, Burst, & Bartolucci, 1995; Moore et al., 2011; Okuwa et al., 2006). In this study, more than half of the participants were assessed as being in the chairfast category according to the Braden Scale.

In the context of PUs, mobility is described as the physical ability to make postural changes, embodying both the concept of frequency and magnitude of movement (Coleman et al., 2013). In terms of structured assessment tools use, the mobility status of participants assessment was undertaken using subscales of the Braden Scale and EMS. According to the Braden Scale, most participants (62.2%, n = 33) had no limitations/slightly limited mobility, while the EMS indicated most of the participants (67.9%, n = 36) were in the dependent category.

Specific devices to monitor the patient's mobility that counts the frequency or calculates the magnitude of movement are also used in practice (Barbenel, 1990; Källman et al., 2015; Kotowski, Davis, Wiggermann, & Williamson, 2013; Zimlichman et al., 2011). In this study, participants’ movement level was measured with the EarlySense movement sensor. This device was used for the first time by Moda Vitoriano Budri et al. (2020) for detecting the relationship between movement and PU development in older adult populations. Moda Vitoriano Budri et al. (2020) found that of the 19 PUs observed using VSA, 53% (n = 10) developed among those classified as being low movers, and 47% (n = 9) developed among those classified as being high movers. In the same study, of the 118 SEM-PU observed, 51% (n = 60) occurred among those classified as being low movers, and 49% (n = 58) occurred among those classified as being high movers. It is somewhat surprising that an equal number of individuals (low and high movers) developed a VSA-PU or an SEM-PU (Moda Vitoriano Budri et al., 2020). The findings from the current study corroborate those of Moda Vitoriano Budri et al. (2020). A relatively similar number of participants developed a VSA-PU or an SEM-PU regardless of whether they were high or low movers. Yet, almost half of the participants were considered not to be at risk according to the Braden Scale. However, the movement level of the participants was analyzed objectively with the EarlySense sensor, meaning that movement assessed subjectively, like when using assessment tools such as the Braden Scale, may not give an accurate picture of the individual's actual movements. This is a finding that have been reiterated in the wider literature (Källman et al., 2015; Moda Vitoriano Budri et al., 2020).

Moda Vitoriano Budri et al. (2020) were the first to describe that abnormally high movements lead to PU development, and the authors delineated theoretical cohort-based recommendations for the prevention of PUs among the two cohorts of patients (low movers and high movers). Findings from Moda Vitoriano Budri et al. (2020) and this current research have identified a lack of an established algorithm to enable the determination of PU risk by assessing the activity and mobility level of individuals, thus emphasizing the need for further investigation (Moda Vitoriano Budri et al., 2020). The present study provides additional evidence that PUs develop in both low and high movers, enhancing our understanding of the role of activity and mobility in PU. Taken together, these findings suggest further work needs to be done because in terms of current detection systems, it is not yet clear how to confidently classify an individual as being in the normal, low, or high movement category. The sensor used actually detects every single movement without describing whether this is a healthy movement or not, and the categorization was based on the number of movements, rather than on the protective nature of the movement. If research is to be moved forward, more accurate methods for the assessment patients’ mobility status need to be developed.

On the other hand, there are studies where mobility has been quantitatively measured using devices that count the frequency or calculate the magnitude of the movement. Study authors explored the relationship between movement and PU occurrence by investigating different populations’ movement patterns, generally during sleep (Barbenel, 1990). The literature acknowledges different techniques for the measurement of movement in bed where, traditionally, systems and devices originally used to investigate sleep patterns were adapted to investigate the influence of mobility on PU occurrence (Barbenel, 1990).

One technique used to monitor the mobility of patients while in bed was to connect the sensor to the bed frame so that the bed frame's movement would reflect the patient’s movement (Barbenel, 1990). Sherwin, Exton-Smith, and Haines (1961) designed an interesting system based on an inertia switch connected to an electrical impulse counter. Every movement performed by the patient in bed caused the switch attached to the bed frame to generate an electrical impulse, with each impulse counted as a movement. A problem with Sherwin et al.'s (1961) “apparatus” was that the vibration of the bed mesh or the mattress itself provided multiple signals for large movements and small movements which were probably not distinguishable from the patient self-movement. Consequently, this likely impacted the accuracy of the number of movements counted (Barbenel, 1990). Later studies suggested that measurements using the bed frame were less than ideal systems for this type of research (Barbenel, 1990; Barbenel, Ferguson-Pell, & Kennedy, 1986).

In clinical practice, identifying individuals with impaired mobility, identifying individuals with abnormally high movements (i.e., a state of agitation), and using a combination of SEM assessment and VSA may help to reduce the development of PUs. Despite this positive outcome, the sensitivity and specificity of SEM measurement has been explored in three studies (Bates-Jensen et al., 2018; O'Brien et al., 2018; Okonkwo et al., 2020). Specificity scores ranged from 32.9% (Okonkwo et al., 2020) to 83% (O'Brien et al., 2018). Okonkwo et al. (2020) acknowledged the specificity limitation, reporting that SEM biocapacitance measures can complement visual skin and tissue assessment. This study and the current literature corroborates the fact that SEM assessment is more sensitive than the traditional visual assessment. More importantly, all these data are showing that what is being measured with the aid of technology using different devices for measuring SEM is fully agreeing with what is being visually assessed. SEM seems to have a higher power in identifying tissue damage compared to VSA but nonetheless do not contradict VSA findings. Therefore, PU prevention strategies employed affect the sensitivity of detection, and this should be taken into consideration. Further, it is important to know that the patient may be experiencing the early development of a PU so you can intervene to stop the knock-on effect of inflammation on further cell deformation.

A number of important limitations need to be considered for this study. The current investigation was limited to an older adult population and participants were Irish, with light skin tone. Therefore, generalization to other populations should be carefully considered. Another limitation was related to the movement sensor. Small movements or big movements could not be distinguished by the movement sensor. Thus, further work is needed in this area to provide this much needed information for clinical practice. Linking Evidence to ActionFocus on movement by using algorithms and taking the necessary measures against pressure ulcer development may significantly reduce the problems caused by these wounds to individuals, the health system, and the economy as a whole. Sub-epidermal moisture as a biomarker successfully predicts early-stage PUs. Classification of high and low movers was made based on movement frequency, and distinguishing normal from abnormal movement is still a challenge in practice, as current devices do not adequately focus on the protective nature of the movement. CONCLUSION

The aim of this study was to provide data for the development of an algorithm related to activity and mobility for the detection of PU risk among older adult residents in long-term care. The study has enhanced our understanding of the relationship between activity, mobility, and PU development. Outcomes showed that all participants who developed VSA-PU had also an SEM-PU. The results of this research support the idea that SEM as a biomarker successfully predicts early-stage PUs. The most obvious finding to emerge from this study was that VSA-PU or SEM-PU developed in individuals classified as being low movers and high movers. Yet, almost half of the participants were considered not at risk according to the Braden Scale. However, the sensor used in this study captured all movements and did not distinguish between healthy movements and unhealthy movements. However, SEM assessment enabled the determination of the individuals’ responses to movements or lack of movements. Classification of high and low movers was made based on the frequency of movement. and distinguishing normal from abnormal movement is still a challenge in practice, as current devices do not adequately focus on the protective nature of movement. Thus, there still exists a clinical need to fill this theory–practice gap. This is of particular importance given the high prevalence and incidence of PUs.