Pressure ulcer prevalence and prevention rates in Abu Dhabi: an update

Pressure ulcers (PU) are a serious problem for clinicians, causing excess morbidity and mortality among patients, and representing a significant financial burden.^1,2 In addition, PU erode patients' quality of life (QoL) and may cause them considerable pain, particularly when dressings are changed and/or removed.² PU prevention, therefore, is key to improve patient outcomes and reduce health-care costs.

A number of risk factors for PU have been identified, leading to a variety of preventive initiatives, including different body positions,³ alternating pressure air mattresses (APAM)⁴ and Braden risk score analysis,⁵ among others.⁶ The presence of non-blanchable erythema in hospitalised patients is a risk factor for PU,⁷ but PU preventive measures can be delayed until non-blanchable erythema is present without increasing the risk of PU development.⁷ Globally, PU prevalence rates vary considerably from 1.26% in China⁸ to around 54% in acute European settings,⁹ although an average of 18.5% is more likely.^9,10

Background

In the years leading up to 2008, a major increase in PU prevalence was recorded at the author's hospital, Sheikh Khalifa Medical City (SKMC), in Abu Dhabi.¹¹ The SKMC is the largest hospital in the United Arab Emirates, with 586 tertiary beds and 14 outpatient clinics. The increase was related to poor team communication, patient malnutrition and lack of preventive equipment.¹¹ In response, SKMC implemented a PU-prevention strategy, comprising:

• Use of the Braden scale (assessment)
• A strict turning schedule, at least every 2 hours
• Consultation with a dietician to improve nutrition
• Staff and patient education about PU
• Development of a PU-management algorithm
• Incentives/rewards for units achieving zero hospital-acquired PU (HAPU) rates
• Ongoing follow-up of PU requiring the wound care team
• Introduction of a new wound dressing in the intensive care unit (ICU) (Mepilex Border Sacrum, Mölnlycke Healthcare, Sweden) to achieve good coverage of the high-risk sacral area.¹¹

ICU patients are commonly less mobile than other patients and are prone to develop PU over bony prominences. The introduction of Safetac dressings (since 2013) has helped relieve pressure in those high-risk areas, thus reducing the risk of PU development.

After the introduction of these initiatives, the annual PU prevalence audit, conducted at SKMC, indicated that prevalence (including National Pressure Ulcer Advisory Panel (NPUAP) category I PU) had fallen from 9.2% in 2008 to 6.4% in 2013 for all SKMC patients.¹¹ Over the same time period, the number of HAPU had also fallen, from 3.3% to 2.0%.¹¹ A summary of the key data from this study is shown in Table 1.

Since 2013, other initiatives have been introduced to standard PU prevention care: ongoing training in HAPU prevention, including the importance of documentation; ensuring that the PU prevention pathway is being used; patient safety net (PSN) reporting (ie a tracking system for any incident, near miss or actual harm caused); root cause analysis; and educational initiatives for PU prevention and management on differential diagnosis (PU versus moisture-associated skin damage, including incontinence-associated skin lesions).

The present article follows on from the 2008 to 2013 PU prevalence paper,¹¹ by presenting and discussing the recent 2018 PU audit at SKMC. The objective was to ascertain whether this downward trend in PU rates has continued, as it has in other hospitals.⁶ This audit is conducted on an ongoing basis to identify the impact of new prevention initiatives (in this case, introduced around 2013) and optimise quality of care.

Materials and methods

The SKMC has used the International Pressure Ulcer Prevalence (IPUP) Survey for annual assessment of PU prevalence rates within the facility since 2006.¹² The IPUP Survey (Hill-Rom BV, The Netherlands) is an international audit conducted over a 3-day period involving more than 900 hospitals worldwide, against which prevalence rates from any facility can be compared. (A 3-day data collection period is used to accommodate the ≥900 participating patients, although the SKMC undertook the data collection in one day.) The full methods used by the author's team are described in an earlier paper¹¹ and readers are referred there for study methodology detail.

Briefly, following pre-selection of a 24-hour period for data capture, staff at the hospital collected data on PU prevalence as instructed. All at-risk patients, with or without skin damage in PU-prone areas, were checked to validate the audit collection data. Patients with a Braden score of ≤18 were defined as at risk. Afterwards, IPUP provided the hospital with an electronic report consisting of various epidemiological and financial outputs, including organisation prevalence rate, prevalence rates by unit, and costs. SKMC's 2018 IPUP Survey was conducted on 20 February 2018 and prevalence rates were compared with the 2013 IPUP Survey data.¹¹ International benchmarks were used for relevant comparisons.

The type of clinician collecting the data remained consistent throughout the audit; training on data collection was provided by Hill-Rom to maintain consistency. No patients were specifically excluded from the audit. However, those who refused, were being discharged or were out of the room during data collection were excluded, as were those with severe clinical conditions, such as multiorgan failure, preventing safe PU identification.

Procedures

Demographics assessed included age and skin tone (dark, medium or light, a subjective assessment left to the survey team's perception, as variations of skin tone exist within ethnic groups). The category and quantity of PU, the type of PU risk assessment performed (when applicable), the risk assessment score and time between admission into the health-care facility and assessment of risk were recorded and reported. All data were self-reported by each participating facility. If incomplete records were generated, the fields present were incorporated into the data set. An in-depth description of study data collection methods has previously been published.¹³

Incidence data were also collected for all patients at SKMC. A HAPU was documented as soon as it occurred; monthly incidence rates were then collated.

Financial implications of HAPU

It was not possible to calculate the total financial cost of HAPU. However, partial costs were determined by correlating the hospital length of stay and the associated treatment costs, for each reported (HAPU) ICD-10 code. Estimations of treatment costs took into account multidisciplinary consultations, dressing usage, nutrition and nursing time. These estimates and weightings for total costs were devised using methods supplied by Hill-Rom.

Results

For prevalence data collection, the total number of patients audited for PU at SKMC was a little higher in 2018 (n=441) than in 2013 (n=406), but their demographic and baseline characteristics were generally comparable (Table 2). Among the patients surveyed, many were either completely immobile (2013, 33% versus 2018, 32%) or independent (2013, 40% versus 2018, 41%). There were fewer infants/children (10.6% versus 19.1%), more 30–39 year olds (12.5% versus 6.3%) but fewer 80–90 year olds (5.2% versus 8.3%) in 2013 versus 2018. Heels were elevated (a preventive measure used on all at-risk patients in the hospital) in around 43% of at-risk patients in both years.

Prevalence rates

For the entire patient cohort, the PU prevalence rate changed from 6.4% in 2013 to 10.4% in 2018 (Table 3), while the rate for HAPUs fell from 2.0% in 2013 to 1.8% in 2018 (including category I PU). This was a point prevalence survey and the primary injury source (i.e. hospital-acquired or not) was not collected for 5/26 patients with PUs (19%) in 2013 and 11 of the 68 PUs (16%) surveyed in 2018 (the total number of PUs in the 26 patients surveyed in 2013 is not known).

Most of the patients with PU (>70%), including those that were hospital acquired, were at very high risk/high risk/moderate risk on the Braden scale at admission, in both years, although 25% of those developing HAPU had been assessed as low risk at this time point. At admission, around 40% of patients in both years were at high risk of developing a PU; by the time patients had had their final risk assessment (Braden scale), just 21% of those in the 2013 cohort were judged to be at high risk of developing a new PU, compared with 44% of 2018 patients. Among patients with a HAPU, the risk of developing a new PU was high or moderate in 73% of patients in 2013 (data not shown) versus 100% in 2018 (Fig 1).

Among the PU patients in 2013 versus 2018 (Table 3), the most common sites for occurrence were: the sacrum/coccyx (54.2% versus 45.6%), trochanter (14.6% versus 2.9%), buttocks (11.3% versus 13.2% and heel (9.4% versus 11.8%). Most of the PU and HAPU (2013 versus 2018) were category II (31.1%/4.7% versus 23.5%/4.4%) (Fig 2). Around 8% of HAPU in 2013 were unstageable compared with 4.4% in 2018; deep tissue injury occurred in 14.6% of HAPU in 2013 but in 0% of HAPU in 2018. In 2018, 30% of stage IV HAPU and 40% of unstageable HAPU were on the sacrum.

Almost all 2018 patients (99%) had their skin assessed and PU risk documented within 24 hours of admission. Documentation of the final risk assessment was slightly lower (93%). The adequacy of documentation of all PU fell slightly between 2013 (97%) and 2018 (86%). For HAPU, adequate documentation was present in 100% of cases in 2013 compared with 90% in 2018.

Most patients who developed a HAPU were female (62%, n=8) and aged either 20–29 years old (25%, n=8) or 70–79 years old (25%, n=8). All (100%) HAPU patients were lying on three layers of linen (fitted sheet, draw sheet, incontinence pad), which is considered to be a preventive depth, on their side (88%, n=8) and 75% had a lengthy hospital stay (>30 days). In 2018, 15 HAPU were detected in ICU patients compared with 26 HAPU in ICU patients in 2013.

Among the 441 patients assessed for PU risk, around 50% were judged to be at risk and most of these (98%) had been subject to a PU prevention strategy within the previous 24 hours. Around 95% had been given nutritional support, 96% moisture management and 94% had been routinely repositioned.

Incidence rates

The incidence rate of PU acquired elsewhere (i.e. PU present at admission) was higher in 2018 than in 2013 (2018: 1.1% versus 2013: 0.9%). This is probably because more high-risk patients were admitted to hospital in 2018. However, the incidence of HAPU reduced in 2018 when compared with 2013 (0.1% versus 0.3%).

Financial implications

The financial implications of HAPU were estimated for the eight patients who developed these ulcers (Table 4). For patients with HAPU, the weighted average cost (per case) was estimated at US$8035.32, giving a total estimated cost for the eight patients of US$64,282.54 (Table 4). When these costs were extrapolated to annual values (determined on the basis of average length of stay [ALOS]), estimated annual costs were US$1,830,082.32 for these eight patients (Table 4).

Discussion

PU prevalence acts as a marker for quality of nursing care.¹¹ The SKMC has focused on the prevention of PU since 2006 and audits PU prevalence annually to determine how effective these efforts have been. Its 2018 PU prevalence audit described here (n=441), comparing data with those obtained during the 2013 audit (n=406), aimed to determine whether the authors' preventive efforts have continued to impact positively on PU prevalence.¹¹

The results showed an improvement, albeit small, in PU prevalence in 2018 compared with 2013, as well as a larger reduction in incidence. The overall prevalence of PU at SKMC in 2018 was 10.4% and 1.8% for HAPU, specifically, compared with 6.4% and 2.0%, respectively, in 2013. This is reinforced by the annual incidence data rates, which showed an improvement: 0.3% for 2013 versus 0.1% for 2018.

Focusing on the increased PU prevalence in 2018, based on our experience at the SKMC, it is likely that the following factors play a role: increase in the geriatric population in whom PU are most common; more non-nationals in Abu Dhabi, whose medical insurance does not cover homecare services; no regular community medical services; lack of education about PU among the patient, carers and family; unavailability of appropriate pressure relieving equipment, such as air mattresses; and lack of knowledge about the part played by nutrition in increasing the risk for PU.

The very low prevalence rate for HAPU (1.8%) suggests that our PU prevention strategies¹¹ continue to be effective, although we acknowledge that this small reduction in percentage could be due to fluctuations in sample size and acuity/increased status of patients. Nevertheless, it is noteworthy that this low percentage (2%) remained consistent over the five years analysed.

These prevalence data are in line with the IPUP acute care PU prevalence benchmarks of 9.0% (n=9482 patients) and 3.1% for HAPU (n=3265 patients), from an overall sample of 104,952 patients.¹⁴ Generally, international benchmarks for PU prevalence show a clear downward trend, while there does not appear to be a clear pattern to PU prevalence at SKMC (data not shown); however, the low prevalence of HAPU indicates that patients at risk received a high standard of care.

Over the last 3–5 years, a number of studies have documented the prevalence and prevention of PU in the acute setting. In a 2018 paper, reporting on a cross-sectional study (n=225), Borsting et al.¹⁵ found that, among their middle-aged (≥52 years) and elderly hospitalised patients, overall PU prevalence was 14.9% and commented that adding age, weight and diabetes status to their risk assessment might improve the identification of patients more likely at risk of HAPU.¹⁵ Likewise, Koivunen et al.¹⁶ in an acute general hospital in Finland, found that among 229 patients, on one day in 2013, overall PU prevalence was 8.7%.¹⁶ The patients' age and length of hospital stay were positively associated with PU risk; older patients had a higher risk of PU than younger ones and the longer the hospital stay, the higher the PU risk.¹⁵ Some of these findings reflect what we have also found during our 2013 and 2018 audits, although, in relation to age, we found that patients aged 70–79 years of age as well as those aged 20–29 years were more at-risk than other age groups, as they are more likely to experience traumatic injuries.

Jull et al.¹⁷ assessed 8274 patients over a 3-year period, among whom 517 patients developed a HAPU, giving an overall prevalence of 6.2% (95% CI: 5.7 to 6.8%). HAPU were found in patients of all ages, but were more common in children up to 14 years (17.4%) and in elderly adults aged ≥75 years (38.7%). The authors suggested that a monthly random sampling of patients within clinical units could be used to evaluate PU prevalence change and to monitor PU prevention improvement, as an alternative to cross-sectional prevalence surveys which are conducted only infrequently.¹⁷

PU prevalence in ICU versus non-ICU patients, 2012–2014, was assessed by Coyer et al.,¹⁸ who showed that HAPU prevalence (excluding category I PU) in the ICU was far higher than in non-ICU wards (11% versus 3%). Patients in the ICU were almost four times more likely to develop a HAPU; significant differences between ICU and non-ICU patients were found in HAPU prevalence by stage and location, reflecting a possible association between critical illness and skin injury,¹⁸ although another reason may be that patients in the ICU tend to be less mobile and not change body position as much as more mobile patients, perhaps due to a lower level of consciousness.

Interestingly, the percentage of PU found on the trochanter fell from 14.6% in 2013 to 2.9% in 2018—a significant reduction and one that highlights the efficacy of our preventive efforts in that area. The 2011 paper by Bales and Duvendack⁶ is impressive in that it reports a zero prevalence of PU in a US hospital setting in December 2008. One of the areas that had previously been identified as problematic was under-reporting of HAPU, therefore, what these authors began to do was risk assess on a weekly basis, with patients scoring <19 on the Braden scale having daily skin assessment (looking for skin breakdown), being turned and re-positioned every two hours, with a certified wound ostomy and continence nurse reviewing all documentation—any deficiencies in documentation were used as an educational opportunity for nursing personnel. They also changed the mattresses from foam to air and—somewhat surprisingly—played a tune every two hours reminding staff to turn and toilet their patients.⁶

As already highlighted, risk stratification on admission is the key to identifying patients who are more likely to develop a PU while hospitalised, and which can act as a trigger for health professionals to begin using a prevention care plan.¹⁹ At admission, around 70% of all PU patients in this study were at very high risk/high risk/moderate risk on the Braden scale of developing a HAPU, although 25% of those who went on to develop a HAPU had been assessed as ‘low risk’ at this point in time. It is possible that those HAPU patients, identified initially as low risk, actually got worse in hospital (hence, become at higher risk) due to deterioration of their condition or a complicated postoperative period, both of which would impact on risk assessment. Proven risk factors for HAPU include: older age (>65 years), gender, low level of consciousness,²⁰ low weight,¹⁵ diabetes,¹⁵ low mobility, incontinence,²¹ lengthy hospital stay (>30 days), and total Braden scale score.^5,22 Our data showed that patients at risk received a good standard of care including a strong PU prevention care plan. The audit found that, excluding cases of category I PU, 86% and 90% of PU and HAPU patients, respectively, had adequate documentation of their PU and care plan, although 10–12% of patients had no data collected for the various reasons discussed in the methods section. Future audits will need to be aware of these reasons and try to reduce the proportion of patients for whom there is no data.

Among all PU patients, 99% had their skin assessment documented within 24 hours of admission (100% of all HAPU patients) and 98% overall had their PU risk evaluated and documented within 24 hours. A recent study showed that staff nurses often perform poorly on documenting HAPU appearance, staging and treatment²³ and our data indicated a fall in adequate documentation between 2013 (100%) and 2018 (90%). Nurses do not always have the right level of knowledge about PU prevention, so regular training courses and review of up-to-date PU prevention guidelines can be helpful in ensuring nurses have all the information they need.²⁴

New strategies for PU prevention are always welcome and need not be complex. For example, in the study by Shieh et al.²⁵ patients who met certain PU risk criteria had a sheet of bright pink paper (saying in a large font ‘SKIN AT RISK’) hung at the head of their bed to trigger a specific set of PU preventive measures. This simple tool led to a dramatic reduction in incidence of PU, from 1.2 to 0.4 HAPU/1000 patient days over 4 years.²⁵

Limitations

Our PU prevention plan has some clear limitations—it is not transferable to the community, at least not without a complete overhaul of systems and procedures which is extremely unlikely. Also, the variables that are used for data capture have changed over time, limiting our ability to compare some results from one year to the next. Finally, because the data were not risk-adjusted, we cannot make at-risk correlations.

Conclusion

This 2018 PU prevalence audit indicated that the levels of prevalence obtained are in line with international benchmarks, as measured against data from the IPUP Survey. In fact, our HAPU prevalence rate is lower than the benchmark (1.8% versus 3.1%), which suggests that our facility's PU prevention strategies have been highly effective. There is still a considerable way to go before we reach zero HAPU. However, with ongoing improvements in prevention care plans, we can be optimistic about a downward prevalence trend.

Reflective questions

• How might patients become actively involved in pressure ulcer (PU) prevention programmes and what may stop them doing so?
• Are there any features of PU that consistently affect outcome? How could you deal with these?
• Think about motivating your nursing staff to take an active role in a PU prevention program that is based on a government—rather than hospital-based—care plan