A Secondary Analysis of Longitudinal Prevalence Data to Determine the Use of Pressure Ulcer Preventive Measures in Dutch Nursing Homes, 2005-2014

Pressure ulcers (PUs) are an important and distressing problem in Dutch nursing homes. A secondary analysis of longitudinal data from the Dutch National Prevalence Measurement of Care Problems (LPZ) — an annual, multicenter, point-prevalence survey — was conducted for the years 2005–2014 to determine the use of specific recommended PU preventive measures from the European Pressure Ulcer Advisory Panel 1998, the National Pressure Ulcer Advisory Panel/European Pressure Ulcer Advisory Panel 2009, and the 2002 and 2011 Dutch PU guidelines.Preventive care was investigated among nursing home residents at risk for PUs and included skin care (moisturization); nutritional and hydration status assessment and optimization; and pressure redistribution involving mattresses, cushions, and heel pressure-relieving strategies and devices. Following abstraction from the study database, data for 3 at-risk groups were distinguished: 1) residents with a Braden score of 17, 18, or 19; 2) residents with a Braden score below 17; and 3) residents with a PU. Data were aggregated at the institutional level. Differences were tested with multiple regression analyses. The mean number of residents over the study period was 5435, the mean age was 82.8 years, and the mean Braden score was 15.3. None of the recommended preventive measures from the guidelines consulted was applied 100% of the time: preventive skin care measures were used in 25.1% to 63.8% of cases and dehydration and/or malnutrition were identified and managed in 27.8% to 65.6% of patients. Pressure redistribution with special types of mattresses was used in 85.2% of patients, cushions in (wheel)chairs were used in 64.8% of patients, and heels were offloaded in 57.8% of patients. The results regarding repositioning for the 3 groups, respectively, showed a maximum use of 9.7%, 30.3%, and 65.6%; the higher the PU risk, the more preventive measures were used. Although the results show a decrease in the percent of category 2 through category 4 PUs from 16.6% to 5.5% and a trend toward increased use of preventive measures (more skin care, attention for dehydration/malnutrition, use of floating heels/heel devices, and pressure-relieving systems when a PU was present), the reason why measurements were suboptimally used remains unclear. Further research to address the application of guidelines in daily practice is needed. 

As the review conducted by Gorecki et al¹ indicates, pressure ulcers (PUs) have long been and remain an important and distressing care problem in nursing homes, causing suffering and loss of quality of life for residents. In addition, PUs are associated with considerable health care costs, as shown in the cost-analysis by Dealey et al.² International PU prevention guidelines^3-6 indicate the use of evidence- and/or practice-based preventive measures and may result in a lower incidence of PUs; the application of PU preventive measures can be facilitated by implementing the recommendations of current evidence-based national and international guidelines.

Appropriate preventive measures include applying a multifactorial approach that consists of PU risk assessment, skin care, nutritional status assessment and optimization, pressure redistribution (mattresses and cushions), and repositioning.^3-6 During the last 20 years, revisions of existing guidelines have led to no real changes in the advised preventive measures because most recommendations have been based mainly on expert opinion instead of evidence-based research.^3-6 Despite possible shortcomings, research has shown guideline recommendations for PU preventive measures provide benefits.^7-10 

Assessment. To assess the risk of developing a PU, a structured instrument such as the Braden scale,¹¹ Waterloo scale,¹² or Norton scale¹³ can be used along with clinical assessment (daily skin inspection). In the Netherlands, the Braden scale is the most widely accepted instrument.^11,14 The scale assesses sensory perception, moisture, activity, nutritional intake, and friction and shear, with item scores ranging from 1 to 3 or 4 points. The maximum scale score is 23 points (no risk at all). A score below 20 indicates patients are at risk of a PU¹⁴; a score below 17 indicates being at medium/high risk of PU development.

Recommended preventive measures. 
Skin appraisal. Proper skin care involves inspecting the skin daily for redness, localized heat, and edema or induration; as well as managing moisture, including use of emollients.^3-6 

Nutrition assessment. Early identification and management of nutritional problems is important^3-6 because many PU patients have a compromised nutritional status.^15,16 Attention to adequate intake of food and liquids and to nonvoluntary weight loss are advised as a part of daily care.

Pressure redistribution. According to PU preventive guidelines published in the last 20 years,^3-6 pressure and shear are the main causal factors of PUs; therefore, preventing/relieving undesired pressure and shear forces is the most important preventive measure. All PU guidelines in this study advised the use of pressure-redistribution systems when the patient is in bed and/or seated utilizing either a passive and/or active system. A passive system involves use of a higher specification foam mattress (eg, a viscoelastic foam mattress) to replace a standard hospital foam mattress. An active system (alternating mattress system) combines redistribution with varying levels of pressure over time. The use of an alternating mattress (active system) is indicated when a viscoelastic foam mattress does not provide sufficient prevention. Combining repositioning with a viscoelastic foam mattress or an alternating air mattress is recommended.^3-6 Repositioning is advised every 3 hours during the day and every 4 hours during the night for individuals at risk, unless contraindicated.^3-6 

Landelijke PrevaIentiemeting Zorgproblemen¹⁷  (LPZ). Every year since 1998, Maastricht University (Maastricht, the Netherlands) has conducted an annual 1-day national prevalence measurement, the LPZ,¹⁷ in Dutch hospitals, nursing homes, and home care organizations that considers relevant care problems such as PUs, malnutrition, falls, and incontinence. For each care problem, the LPZ measures structure, process, and outcome indicators according to Donabedian’s model of quality of care.¹⁸ More about the methodology of the LPZ is available online (https://nl.lpz-um.eu/). 

The aim of this study was to explore which PU preventive measures were used in Dutch nursing homes in 2005–2014 as recommended by guidelines from the European Pressure Ulcer Advisory Panel 1998,⁵ the National Pressure Ulcer Advisory Panel (NPUAP)/European Pressure Ulcer Advisory Panel (EPUAP) 2009,³ and the 2002 and 2011 Dutch PU guidelines⁴ for the specific preventive factors discussed.  

Design. A secondary analysis of data from the LPZ¹⁹ was performed. The study was approved by Maastricht University Medical Centre’s Ethical Committee. For this study, only LPZ nursing home data from 2005 to 2014 that involved PU prevention were analyzed.

Sample. Because participation of the health care institutions in the LPZ is voluntary, data reflect annual differences in the number of participating institutions and the total number of participating residents. To obtain representative results, nursing homes are encouraged to assess all residents of all wards. Data from residents who received palliative care or short-term rehabilitation or whose stay in the nursing home was <30 days were excluded from the study. 

Several cutoff points were used to distinguish between residents at risk and not at risk. The complete data file for the years 2005 to 2014 was divided into 3 categories, based on the risk status of the residents per the original research by Braden¹¹ (risk cutoff of 16) and Halfens et al¹⁴ (cutoff of 20): 1) residents at low risk of a PU (Braden score 17–19), 2) residents at medium/high risk of PUs (Braden score <17), and 3) residents with 1 or more PUs category 2 or higher. All residents not at risk (Braden score >19) were excluded from the analyses because they do not require preventive measures. Stage 1 PUs were excluded because their diagnosis is less reliable; these ulcers frequently are excluded in prevalence reports.^21,22  Data were aggregated at the institutional level because the prevalence rate is an important care factor in the nursing home. 

Instruments. In this study, only data related to PU risk (using the Braden scale¹¹), PU presence/classification, and preventive measures were used. PUs were classified in 4 categories according to the PU grading system of the EPUAP 1998 and NPUAP/EPUAP 2009 guidelines.^3,5  The following PU preventive measures, as recommended by the guidelines, were assessed: skin care (including use of emollients); identification and management of dehydration and/or malnutrition (including monitoring weight loss, body mass index [BMI], and nutritional intake and supplementation when there appeared to be nutritional deficiencies); repositioning regimens (repositioning in bed and chair according to a time schedule); floating heels (offloading or use of heel devices); and pressure-relieving systems (type of mattresses and cushions).^3-6 Questions regarding application of each preventive measure could be answered Yes/No. 

In addition to demographic data that included gender and age, data relevant to the degree of care dependency according to Care Dependency Scale²⁰ (CDS) were considered to assess whether residents became more care-dependent over the years. The CDS comprises the following items, each measured on a 5-point Likert scale (ranging from completely agree to completely disagree): eating/drinking, continence, body posture, mobility, day/night pattern, dressing/undressing, body temperature, hygiene, avoiding danger, communication, contact with others, sense of rules and values, daily activities, recreational activities, and learning ability. The total score varied between 15 and 75 points. The higher the score, the less care-dependent the resident. 

All data were gathered by 2 trained caregivers at each nursing home — 1 from the patient’s ward and 1 from another ward. Data were acquired directly from the patient (Braden scale, prevalence of PU, and when possible information on the preventive measures provided such as pressure-relief strategies) or gathered from the patient documentation/medical records (demographic characteristics and other preventive measures such as identification and management of dehydration and/or malnutrition). 

Statistical analyses. Data were analyzed by Maastricht University using SPSS Statistics, version 23.0 (IBM Corp, Armonk, NY). Mean characteristics of the residents per group and the mean use of preventive measures were calculated, and differences between the years were tested using linear regression analyses, with year as an independent variable. Statistical significance was noted at <.05. 

The aggregated demographic characteristics of the study population over the years are shown in Table 1. During the study period (2005–2014), patients were mostly female (mean 73.6%, range 69.8%–76.8 %) with a mean age of 82.8 (range 81.8–83.9) years; P <.01) and a mean Braden score of 15.3 (range 14.9–15.7; P <.0001); the latter data differed significantly among the years. However, the differences do not indicate a specific trend and are very small; the maximum difference in age is 2.1 years, and the maximum difference in Braden score is 0.8. No significant difference was found regarding gender and care dependency. 

Although these demographic characteristics do not show large differences between the years, the PU prevalence shows an almost linear decrease from 16.6% to 7.1% during the first years (2005–2009) and stabilization from 2010 to 2014 (from 7.0% to 4.0%). 

Table 2 shows the results regarding the use of the 4 recommended measures for PU prevention for each at-risk group. The extent to which these measures were used increased with an increase of the PU risk.  

Preventive skin care. Over the study years, a statistically significant difference was found for all 3 at-risk groups with regard to skin care practice. Although structured skin assessment increased over the years, by 2014 ~60% of the residents in the low-risk group, 45% in the medium/high risk group, and 40% in the group with PUs did not receive preventive skin care.

Identification and management of dehydration and/or malnutrition. Identification and management of dehydration and/or malnutrition differed among the years in all 3 at-risk groups. Preventive measures involving hydration and nutrition status increased significantly only among residents with a PU (from 39.3% to 58.6%, P <.00001). However, not all residents in all groups received structured assessment of possible dehydration and/or malnutrition. 

Repositioning schedule. The use of repositioning differed among years;  a significant decline in the use of repositioning occurred in the at-risk group with a Braden score below 17 starting in 2006. In 2014, only 18.4% of residents of this group received repositioning according to a time schedule.  

Use of floating heels or heel devices. Floating the heels or using heel devices increased over the years from 20.3% to 31% (P <.0001) in the medium/high risk group as well as the group with a PU. In the group of residents at low risk for PUs, the use of these heel pressure offloading strategies remained very low and was not statistically significant  (7.9% to 11.3%). 

Pressure-relieving systems. Table 3 shows the use of pressure-redistributing bed systems. The data show that over the years no significant differences were found regarding the total percentage of use of pressure-redistributing mattresses/overlays. However, pressure-redistributing systems were used more often when the PU risk was higher (77.9% versus 71.5% in the high-risk group and 66.5% in the low-risk group). 

Since 2012, the results show a decline in the use of alternating air mattresses in all groups in favor of the use of static air overlay mattresses. The use of foam mattresses in general was stable over the years, but they were used less frequently when the PU risk is higher. 

Pressure-redistributing cushions. The total use of pressure-redistributing cushions was constant over the years. Since 2012, their use mostly seemed to trend upward in all 3 groups (group 1 from 20.8% to 33.4%, group 2 from 31.8% to 47.9%, and group 3 from 30.8% to 64.8%) (see Table 4), as noted by  the increased use of air cushions (group 1 from 5.7% to 21.4%, group 2 from 10.4% to 31.5%, and group 3 from 16.7% to 50.0%). The use of foam cushions declined during the study period (especially in the low-risk and medium/high-risk groups, respectively) from 23.3% to 9.4% (P <.0001) and from 24.6% to 10.9% (P <.0001).  

This study shows PU preventive measures are used more often when the risk for PU development is greater or when PUs are already present. However, not all at-risk residents received the recommended PU preventive measures as advised in the guidelines, such as skin care or a pressure-redistributing mattress.^3-6 Over the years, data show an increase in the application of appropriate skin care and the use of static air overlays and preventive air cushions in all 3 risk groups, as well as an increase in the use of heel offloading or heel devices in groups 2 and 3. The use of repositioning and alternating air mattresses in group 2, as well as the use of foam cushions in groups 1 and 2, decreased. Overall, the use of pressure-relieving systems was fairly constant. Looking at the decrease of PU development during the study period, it is possible to conclude use of pressure-relieving systems may result in less PU development.  

Use of appropriate skin care showed a statistically significant increase over the years for all groups but never reached a percentage higher than 60%. Because adequate skin care is very important, 100% use was expected by the study group. Attention to malnutrition and dehydration increased significantly over the study years in the group of residents with a PU but never reached a percentage higher than 65.6%. This is in line with the results of a prevalence study¹⁹ about prevention measures for residents at risk in Austrian nursing homes that showed attention to nutritional status occurred in only 53.5% of residents at risk of developing a PU.

Even though repositioning is an integral component of PU prevention and treatment and is widely recommended, this study shows it is not widely used in daily practice,²³ similarly noted in a prevalence study by Moore et al²⁴ where only 9% of older residents in Irish long-term care facilities were repositioned. Contrary to the current study  findings, these authors found no difference in the use of repositioning strategies between residents with and without a PU. In a descriptive, comparative study, Breimaier et al²⁵ found that ~40% of Austrian nursing home residents at risk for a PU were provided regular repositioning; the same study showed 60% of residents were provided heel offloading, compared to the maximum of 55% among at-risk persons with a PU in the current study. 

Pressure-relieving mattresses were used for 70% of residents in all the groups studied; this percentage increased by approximately 10% after a PU developed. In a prevalence study²⁴ involving nursing homes in Ireland, 50% had a pressure-redistribution device in bed, and in a descriptive, comparative study²⁵ conducted in Austrian nursing homes more than 80% of persons at risk for PUs were provided a pressure-redistributing device.²⁵ 

Possible reasons for insufficient implementation of recommended PU preventive measures. The main question that remains is why PU preventive measures are not fully implemented as advised in the guidelines. It is generally known that national and international guidelines often are challenging to read because they are extremely comprehensive, written in complicated language, and difficult to translate into daily practice. This may cause barriers to their implementation. Lack of awareness, knowledge, and skills and unfamiliarity with the recommended measures also may deter nurses from following the guidelines. A qualitative study by Meesterberends et al²⁶ indicated most nurses in Dutch nursing homes were not aware of the guideline recommendations regarding PU prevention, although the mean knowledge score regarding preventive measures was 71.3%. A longitudinal study by Demarré et al²⁷ conducted in Belgian elder care facilities found nurses’ knowledge about PU preventive measures was low and that the attitude of the nursing staff was a predictor of the application of prevention recommendations for residents at risk. 

Because the current study implies the implementation of evidence-based PU preventive measures remains a challenge, the question also arises whether the current guideline recommendations are tailored sufficiently to the daily practice of PU prevention and care. PU guidelines should better explain which (combinations of) preventive measures should be used in specific patient situations. It is the opinion of the study group that this would allow PU preventive measures to be tailored to individual patients to a greater degree, facilitating their application in daily patient care. 

Several literature and systematic reviews^28-33 underscore that the implementation of national and international guidelines in daily practice is often insufficient but might be facilitated by easy-to-use tools such as checklists, alerts, and apps. Active implementation strategies, as presented in implementation studies,^28-34 have been shown to be more effective than passive strategies in overcoming implementation barriers and also encourage better implementation of PU guidelines. Adequate and more continuous coaching and daily practice guidance should be provided instead of only providing hard-to-read, massive guideline documents.

Secondary analyses of existing data inherently come with limitations. Patient demographic data showed that during the period 2005–2014 the mean age and mean Braden score differed significantly among the years. However, these differences were very small but statistically significant due to the large sample size. Whether such small differences are clinically  relevant is unclear. In addition, because the methodology of the LPZ measurement involves a cross-sectional design, it is not possible to draw causal conclusions. Finally, although most of the participating nursing homes are involved annually in the national LPZ survey, not every nursing home participates every year. 

A longitudinal study of national data found PU preventive measures are not provided to all persons deemed at risk, although application of guideline recommendations for PU prevention is increasing, especially in patients at high PU risk or who already have a PU. This study shows that guideline availability is only 1 step in the provision of evidence-based care. More research into the reasons for the lack of provision of preventive strategies and solutions for their implementation is needed.

Dr. van Leen is an elderly care physician, Martin’s Geriatric & Wound Consultancy, Rotterdam, The Netherlands; and a researcher, Faculty of Health, Medicine and Life Sciences, Caphri/Department of General Practice, Maastricht University, Maastricht, The Netherlands. Prof Schols is an educator, old age medicine, Faculty of Health, Medicine, and Life Sciences, Caphri/Department of General Practice, Masstricht University. Prof Hovius is a plastic surgeon and Head, Orthopedic Ward, Department of Plastic Surgery, Erasmus Medical Center, Rotterdam, the Netherlands. Dr. Halfens is a project leader for Dutch National Prevalence Measurement of Care Problems (LPZ), Faculty of Health, Medicine, and Life Sciences, Caphri/Deptartment of Health Service Research, Maastricht University.

Please address correspondence to Martin WF van Leen, MD, PhD, Martin’s Geriatric & Wound Consultancy, Helmbloem 15, 3068AC Rotterdam, The Netherlands; email: martinwfvl@gmail.com.