Saudi Arabian adult intensive care unit pressure ulcer incidence and risk factors: a prospective cohort study

Abstract

The purpose of this study was to identify pressure ulcer (PU) incidence and risk factors that are associated with PU development in patients in two adult intensive care units (ICU) in Saudi Arabia. A prospective cohort study design was used. A total of 84 participants were screened second daily basis until discharge or death, over a consecutive 30‐day period, out of which 33 participants with new PUs were identified giving a cumulative hospital‐acquired PU incidence of 39·3% (33/84 participants). The incidence of medical devices‐related PUs was 8·3% (7/84). Age, length of stay in the ICU, history of cardiovascular disease and kidney disease, infrequent repositioning, time of operation, emergency admission, mechanical ventilation and lower Braden Scale scores independently predicted the development of a PU. According to binary logistic regression analyses, age, longer stay in ICU and infrequent repositioning were significant predictors of all stages of PUs, while the length of stay in the ICU and infrequent repositioning were associated with the development of stages II–IV PUs. In conclusion, PU incidence rate was higher than that reported in other international studies. This indicates that urgent attention is required for PU prevention strategies in this setting.

Introduction

Pressure ulcers (PUs) are one of the most common problems faced globally in health care settings 1, 2, 3. PUs have an impact on the rising health care costs and on the patient's health because of the increase in both morbidity and mortality 4. Studies that examine PU development have become of increasing interest in the drive to improve patient outcomes. PUs are predictable and preventable phenomena, thus making this complication one of the key indicators to measure quality of nursing care and patient safety in the health care setting 5, 6.

The National Pressure Ulcer Advisory Panel (NPUAP) and the European Pressure Ulcer Advisory Panel (EPUAP) 7 defined a PU as a lesion or a trauma to the skin and the underlying tissue resulting from unrelieved pressure, shear, friction, moisture or a combination of all these, usually over a bony prominence. PU staging refers to a recognised and established system to classify the level of tissue damage or depth of injury observed 7. Use of the NPUAP/EPUAP PU‐staging system enables the objective of assessment of the depth of tissue injury by the health care workers.

PU prevalence data vary widely across both settings and country. Prevalence rates have been reported at 3–26% in North

America 8, 9; 8·1–49% across Europe 3, 10; 3–50% in Australia 11, 12; 7–44·4% in the Middle East 13, 14; 2·1–31·3% in Asia 15 and 9·7–51·6% in Africa 16. Furthermore, PU incidence data also vary by clinical setting: in long‐term care, from 2·3% to 23·9%; in acute care, from 0·4% to 38·6%; in home care, from 0% to 17% and in rehabilitative care, from 0% to 6% 13, 17. However, data highlight that intensive care units (ICUs) have the highest PU incidence rates in health care settings, which have been reported as high as 50% 18. The high rates in the ICUs can be attributed to the high acuity of patients, the nature of their critical illness and the highly invasive nature of the interventions and therapies that critically ill patients receive 4, 19.

Identifying patients at risk for PU development is essential for the effective implementation of PU prevention programmes and usage of resources. Tayyib et al. 6 ascertained 28 factors associated with accelerated PU development in critically ill patients; however, the most frequently reported risk factors were older age, longer ICU stay, history of cardiovascular and diabetes disease and infrequent repositioning.

There is a paucity of research examining the extent of hospital‐acquired pressure ulcers (HAPUs) in the Middle East countries with only two studies identified. One Saudi Arabian (SA) study 13 reported acute care PU prevalence of 44·4% and incidence of 38·6%. A second Jordanian study reported an overall PU prevalence of 12% in the health care setting and 29% in the intensive care setting 14. However, prevalence data do not adequately reflect the magnitude of the problem; it provides a snapshot of the problem for quality assurance purposes. Incidence data, on other hand, provide an accurate figure and a picture of the extent of the problem in the health care setting over a period of time 20.

In Saudi Arabia, no reported baseline data exist on PU incidence in the ICU. This is the first study conducted in the Middle East, specifically Saudi Arabia, to identify the PU incidence in the intensive care setting. Therefore, this study aimed to: (i) describe characteristics of the ICU patients in Saudi Arabia, (ii) identify the incidence of PUs in the ICUs in Saudi Arabia and (iii) determine the risk factors associated with PU development in this population. This study also provides a benchmark for PU incidence in the ICU and risk factors studies in Saudi Arabia and a comparison with other international studies.

Materials and methods

Study design

This multicentric prospective observational cohort study was completed over a 4 weeks consecutive period between July and August 2013 in the ICUs of two major metropolitan hospitals in Saudi Arabia.

Participants and setting

All patients admitted to the ICU during the data collection period, who were aged 18 years or more, were included in the study.

The research setting was two hospitals operated by the Ministry of Health, Saudi Arabia. Both facilities are major tertiary care hospitals, each with more than 400 beds and ambulatory services. These two facilities each provide services to patients from the western region of Saudi Arabia and from the neighbouring regions, and each site has 24 intensive care beds. Patient admission diagnoses in both ICUs include cardiovascular illness, respiratory disease, cancer, renal dysfunction, sepsis and multi‐trauma injury such as head injury. A diverse number of complex treatments and advanced technological equipment are provided for the ICU patients to support their body system functions during critical illness. According to the portfolios of both ICUs, the average length of patient stay in 2011 was 8–9 days 21, 22.

Data collection

A data extraction form was specifically designed for this study. Data were collected on three levels: baseline ICU survey, baseline patient survey and second daily patient's skin inspection and data collection. Data collection processes were piloted with validated instruments where available including the Sequential Organ Failure Assessment (SOFA) score 23, Braden Scale score 24 and PU‐staging scales 7.

Baseline ICU data

Current system‐level data present at the time of commencement of data collection at each ICU were collected, which included total bed capacity, ratio of ventilator/non‐ventilator beds, annual number of patient admissions, average length of patient stay, number of patient mechanical ventilation days, nurse/patient ratios, existence of specific practices regarding skin care and the types of strategies used to prevent PU development.

Baseline patient data

Patient demographic data were collected on admission for all patients that included age, sex, nationality, body mass index (BMI) and clinical demographic data consisting of diagnosis on admission, comorbidities, emergency or elective admission, length of time in the operating theatre or emergency department prior to ICU admission in hours if applicable, Braden Scale score 24, presence or absence of PU on admission (yes/no), mechanical ventilation (yes/no) duration in days if applicable, ICU length of stay in days and ICU outcome (discharge to ward or death).

Second daily patient data collection

Every second day, the patients' skin was assessed using standard physical examination techniques 25. The data collected included the presence or absence of a PU. If a PU present, its grading and site, patients' ventilation status, frequency of patient repositioning and SOFA score were also recorded.

Established data collection tools

Braden Scale score

The Braden Scale score 24 is a well‐recognised, widely used and validated score for predicting the risk of PU development 26. Accordingly, it consists of six elements: sensory perception, moisture, activity, mobility, nutrition and friction and shear. The total score ranges from 6 to 23, with a higher score indicating a lower risk of developing a PU and vice versa. Using this scale, all the patients were divided into four subcategories: at mild risk (score 15–18), moderate risk (score 13–14), high risk (score 10–12) and very high risk (score ≤ 9).

PU staging

PUs were identified as either community‐acquired PUs (CAPU) or HAPU, and if they related to equipment, as medical device‐related PUs (MDRPU). CAPUs were defined as PU diagnosed on admission, or within the first 24 hours of the patient's admission to ICU. Furthermore, PUs were classified according to the NPUAP and the EPUAP 7 definitions in relation to stage, type or site of injury (e.g. skin or mucosal ulcers). Additionally, the PU site was identified on the data collection form by drawing a circle over the relevant area in the body figure.

SOFA

SOFA was designed to determine the extent of a person's organ function or rate of failure in the ICU 27. The score is based on the description of six‐organ dysfunction/failure, namely, the respiratory, cardiovascular, hepatic, coagulation, renal and neurological systems. Each organ is graded from 0 (normal) to 4 (the most abnormal), providing a daily score of 0–24 points. The purpose of the SOFA score is to create a reliable, simple and continuous score that can be easily applied in all institutions. Furthermore, the SOFA score provides an indication of patient acuity on a daily basis (i.e. when measured).

Procedure

Permission to access the ICUs was received from the medical and nursing directors. All ICU nurses and doctors were informed in writing and in person about the study. All the patients who met the study inclusion criteria were included in the study. The researcher, who was trained in data collection tools, collected all data to ensure reliability.

While a comprehensive skin examination and assessment was performed second daily on each participant, owing to the variation in admission times, data were collected daily over a consecutive 30‐day period in the ICU. Data were collected on a second daily basis until patient discharge from the ICU or death. The baseline ICU data were collected at the time of commencement of the study and patient baseline data collection was completed on admission.

Ethical considerations

Ethical approval to conduct this study was obtained from the Unit of Ethics of the relevant hospitals, Saudi Arabia, and the Queensland University of Technology (QUT) Human Research Ethics Committees, Australia. This study received approval with a waiver of informed consent, in accordance with the National Health and Medical Research Council (NHMRC) guidelines, because measuring PU incidence in the ICU does not involve an intervention or a change to standard care. Furthermore, informed consent can be impractical to obtain when patients are critically ill, mechanically ventilated and sedated 28. As this study aimed to identify the incidence of PU development in two ICUs, complete patient numbers were crucial to accurately calculate the incidence.

Statistics

Descriptive statistical methods were performed using SPSS (version 21; SPSS, Chicago, IL). The level of statistical significance was set at a P‐value less than or equal to 0·05 (P ≤ 0·05) 29. All the demographic and clinical characteristics were analysed using descriptive statistics (frequencies and means where appropriate). Incidence was calculated as cumulative incidence, defined as the proportion of the participants who develop new PUs within a specific time 20. Bivariate analyses using chi‐square test, independent t‐test and Mann–Whitney U‐test were performed to identify the variables associated with PU development. A binary logistic regression model was used to develop the best model for predicting PU occurrence in the ICU. It also explored the direction of the relationship between the predictive variables and PU occurrence.

Results

Baseline ICU characteristics

Table 1 shows the main characteristics of both ICUs. For both ICUs, the staffing ratio was one nurse to two, or at times three, mechanically ventilated patients and one charge nurse per shift for each of the three 8‐hour shifts per day. In both hospitals, the registered nurse incharge of the ICU was not always supernumerary and was sometimes allocated to provide care for two patients. Registered nurses (RNs) delivered complete patient care. Furthermore, there were no dedicated clinical instructors or nurse educators in the units, nor any dedicated respiratory therapists and physiotherapists.

Table 1.

Baseline intensive care unit (ICU) characteristics

Variable	ICU 1	ICU 2
Total bed capacity	24 beds	24 beds
Ratio of ventilator/non‐ventilator beds	6:1
Annual number of patient admissions	2000–2500 patients annually	2000–2300 patients annually
Average length of patient stay	9 days	10 days
Nurse/patient ratio	1:2 to 1:3	1:2 to 1:3
ICU patient skin care
Risk assessment	No PU risk assessment scale used	Same as per ICU 1
Skin assessment	Comprehensive physical assessment of the patient's skin is undertaken and documented within 24 hours of admission and every 8 hours.	Same as per ICU 1
Skin care	Bed‐bath once a day at 06:00 hours using antiseptic soap containing 2% hydrogen peroxide.	Same as per ICU 1
Nutrition	Nutrition plan for each patient will be provided by clinical nutritionist	Same as per ICU 1
Repositioning	Patients' position alternated from right side, back and then left side or vice versa.	Same as per ICU 1
Support surface	All patients managed on reactive support surface (i.e. air mattress)	Same as per ICU 1
Education and training	During the orientation for new ICU staff	Same as per ICU 1
Documentation of pressure ulcers	Pressure ulcer documentation tool especially designed for the hospital	Same as per ICU 1

Both study facilities had practices related to the prevention of PU development (see Table 1). These practices included physical examination of the patient's skin each shift (however, no risk assessment tool was used), use of support surfaces to manage patient load and pressure and a two‐hourly repositioning policy. Additionally, routine skin hygiene practice for ICU patients in both the units was a bed‐bath at 6:00 am using an antiseptic soap containing 2% hydrogen peroxide without application of a skin moisturiser.

Characteristics of the study population

Of a total of 90 patients admitted to the ICUs during the 30‐day study period, 84 participants were included in this study. Six participants were excluded as they had CAPUs on admission. Participant mean age was 52·8 years, with a range of 18–99 years. Almost two‐thirds of the participants were men (n = 56, 66·6%). Majority of the individuals in the study sample were non‐Saudi nationals (n = 46, 54·8%). Table 2 provides an overview of the participants' demographic characteristics and clinical features. About 85·7% of the participants were at high risk for PU development with a mean Braden Scale score of 10 (SD 2·12)

Table 2.

Demographic and clinical characteristics of study's participants

Variable	Total participants (n = 84)	Participants with PU (n = 33)	Test	P‐value
Male (number, %)	56 (66·7)	18 (54·5)	3·59*	0·58
Nationality (number, %)			1·9*	0·168
Saudi	38 (45·2)	18 (54·5)
Pakistani	6 (7·1)	3 (9·1)
Indian	4 (4·8)	1 (3)
Burmese	5 (6)	2 (6·1)
Bangladeshi	4 (4·8)	1 (3)
Yemeni	3 (3·6)	2 (6·1)
Indonesian	2 (2·4)	0
Thai	2 (2·4)	0
Malaysian	1 (1·2)	0
Chinese	1 (1·2)	0
Egyptian	3 (3·6)	2 (6·1)
Algerian	1 (1·2)	1 (3)
Sudanese	2 (2·4)	1 (3)
Somali	2 (2·4)	0
Nigerian	3 (3·6)	1 (3)
Ethiopian	2 (2·4)	0
Mauritanian	4 (4·8)	1 (3)
Burkinabe	1(1·2)	0
Age (years) (mean, SD, range)	(52, 20·1, 18–99)	(65·45, 20·21, 27–99)	5·33†	<0·001§
BMI (mean, SD)	(26·18, 3·86)	(26·67, 3·89)	0·926†	0·357
Braden Scale (mean, SD)	(10, 2·1)	(9·33, 1·63)	537·5‡	0·004§
Admission via emergency department (yes)	64 (76·2)	29 (87·9)	4·093*	0·043§
Length of time in emergency department (minute) (mean, SD)	(81·40, 54·89)	(81·55, 50·62)	653·0‡	0·081
Admission post‐operation (yes)	11 (13·1)	0
Length of time in operating theatre (minute) (mean, SD)	(50·9, 24·16)	0
Comorbidities (number, %)
Hypertension	46 (54·8)	20 (60·6)
Insulin‐dependent diabetes	43 (51·2)	18 (54·5)
Peripheral vascular disease	11 (13·1)	6 (18·2)
Kidney disease	9 (10·7)	6 (18·2)
Respiratory disease	19 (22·6)	7 (21·2)
Heart disease	20 (23·8)	12 (36·4)	4·722*	0·03§
Nil	7 (8·3)	3 (9·1)
Diagnosis (number, %)
Trauma	17 (20·2)	6 (18·2)
Medical‐related illness	50 (59·5)	16 (48·5)
Post‐surgery	10 (11·9)	6 (18·2)
Sepsis/infectious disease	7 (8·3)	5 (15·2)
ICU length of stay (days) (mean, SD)	(9·1, 6·66)	(13·3, 8·36)	397·0‡	<0·001§
SOFA score (mean, SD)	(7·8, 3·01)	(8·75, 3·16)	1·9†	0·06
Mechanical ventilation (yes)	64 (76·2)	31 (93·9)	5·86*	0·015§
Average hours of patient repositioned hours (mean, SD)	(3·8, 1·3)	(4·96, 1·28)	7·304†	<0·001§
Average days before PU development (SD, median, range)	N/A	10·09 (4·62, 9, 5–23)

ICU, intensive care unit; PU, pressure ulcer; SOFA, Sequential Organ Failure Assessment.

^*Pearson χ ²‐test.

^† t‐test.

^‡Mann–Whitney U‐test.

^§Significance at P < 0.05.

Incidence rate of PUs in ICUs

Of the 84 patients, 33 patients with new PUs were identified giving a HAPU incidence of 39·3% (33/84 patients). Table 2 shows the demographic and clinical characteristics of participants with PUs. A total of 41 HAPUs were recorded in 33 patients. The common areas of PU development were the sacrum (24·3%) and the heel (29·2%), with a greater tendency for the PUs to develop in the heels. Stages I (23/41) and II (15/41) PUs were the most often recorded. Only three PUs were stage III, and no stage IV PU was recorded (see Table 3). The overall incidence of MDRPU was 8·3% (7/84). Of the 41 HAPUs, 8 (20%) were related to medical devices, and the most common site was the ear (37·5%).

Table 3.

PU incidence of surveyed participants (n = 84)

Variable	HAPU
Number of participants with PUs	33
Number of PUs per participants
One (number, %)	19
Two (number, %)	4
Three (number, %)	2
Four (number, %)	0
Description of PUs
Patients with skin ulcers
Suspected deep injury	0
PU stage I (number, %)	23
PU stage II (number, %)	15
PU stages III & IV (number, %)	3
Unstageable (number, %)	0
MDRPUs (number, %)	8 (19·5)
Total number of PUs	41
Skin ulcer location (number, %)
Occiput	1
Ear	2
Elbow	1
Shoulder	2
Ischium	1
Sacrum	10 (24·3)
Buttock	4
Heel	12 (29·2)
MDRPU location (number, %)
Nare	1
Lip	1
Neck	2
Ear	3
Leg	1

HAPU, hospital‐acquired pressure ulcer; MDRPU, medical device‐related pressure ulcer; PUs, pressure ulcers.

Risk factors

Table 2 shows that age (t(82) = 5·33, P < 0·001), length of stay in the ICU (U = 397·0, z = 1723·0, P < 0·001), history of cardiovascular disease (χ ² = 4·722, P = 0·03), infrequent repositioning (t(48·7) = 7·308, P < 0·001), emergency admission (χ ² = 4·093, P = 0·043), mechanical ventilation status (χ ² = 5·86, P = 0·015) and a lower Braden Scale score (U = 537·5, z = 1098·5, P = 0·004) were significantly associated with all stages of PU development. Mechanical ventilation (χ ² = 5·707, P = 0·017), length of stay in the ICU (U = 397·0, z = 1723·0, P <0·001) and infrequent repositioning (t(82) = −4·562, P < 0·001) were positively associated with the development of stages II–IV PUs.

All the factors mentioned above were entered into a binary logistic regression model as exploratory variables for all stages of PU development (Table 4). Age (OR: 1·254; 95% CI: 1·054–1·492; P = 0·011), longer stay in the ICU (OR: 1·831; 95% CI: 1·014–3·309; P = 0·045) and infrequent repositioning (OR: 250·04; 95% CI: 5·230–11 954·16; P = 0·005) were significant predictors of all stages of PUs. Length of stay in the ICU (OR: 1·23; 95% CI: 1·087–1·392; P = 0·001) and infrequent repositioning (OR: 2·96; 95% CI: 1·23–7·153; P = 0·015) were associated with the development of stages II–IV PUs.

Table 4.

Risk factors of pressure ulcer (PU) development using binary logistic regression^{* , †}

Independent variable	B	SE	Wald	Significance	Exp(B)	95% CI for Exp(B)
						Lower	Upper
(A) For all PU stages
AGE	0·226	0·089	6·521	0·011	1·254	1·054	1·492
Infrequent repositioning	5·522	1·973	7·831	0·005	250·043	5·230	11 954·166
ICU_length of stay	0·605	0·302	4·019	0·045	1·831	1·014	3·309
Constant	−32·969	12·746	6·691	0·010	0·000
(B) For participants with stages II–IV
Infrequent repositioning	1·087	0·449	5·861	0·015	2·966	1·230	7·153
ICU_length of stay	0·207	0·063	10·762	0·001	1·230	1·087	1·392
Constant	16·511	7157·897	0·000	0·998	1 4810 627·253

ICU, intensive care unit.

^*(A) Nagelkerke R ² = 0·914; (B) Nagelkerke R ² = 0·683.

^†Hosmer and Lemeshow test: (A) χ ² = 2·395, df = 8, P = 0·966; (B) χ ² = 1·781, df = 8, P = 0·987.

Discussion

Characteristics of the study population

Results of this study showed an increased percentage of participants from outside Saudi Arabia. This is typical of the patient population in this research setting as Makkah is a city of significant spiritual significance with many Muslims travelling to Makkah for religious pilgrimage. Almost 72% (33/46) of non‐Saudi participants were older than 40 years, which is expected, as many Muslims delay the pilgrimage journey until they become older 30. The reasons behind this could be financial, social or spiritual beliefs.

Incidence of PUs

In this study, the cumulative incidence of HAPUs in the adult ICUs was 39·3%. This HAPU rate was higher than other international studies' findings of 1·8% 31, 7·5–14·3% 32 and 12% 33. A possible explanation of the higher PU incidence rate in the ICUs is the length of stay in the ICU (mean 9·3 days), which is longer than other time periods reported in international studies 2, 15, 34, 35. In the SA context, the longer length of stay may be related to patients' prognosis, patients' treatment and the fact that in Saudi Arabia, withholding or withdrawal of treatment is not practised because of religious and cultural beliefs. Furthermore, the incidence rate may be high because of low nurse/patient ratios (i.e. 1:2 or 1:3) and the consequent high nursing workload leading to infrequent repositioning of these high‐risk patients, thus accelerating PU development. We found that the mean time to reposition patients in this study was 4·96 hours, although it is acknowledged that these data were collected retrospectively from the patients chart and may, therefore, not present an accurate picture of clinical care. Although low nurse/patient ratios exist in other countries, ICU patient care delivery is organised differently. For example, in the USA, patient care in the ICU is provided by a dedicated team including RNs (care coordinator), nursing assistants (who provide hygiene cares), respiratory therapists, nutritional consultants and ICU physicians 36, 37. Conversely, in Australia and the UK, the nurse/patient ratio for mechanically ventilated patients is 1:1; however, the RNs provide and coordinate all patient care 38. A number of studies have reported that the high nurse/patient ratio was significantly associated with high quality, safety and positive patient outcome 39, 40, 41.

Interestingly, our findings show a low incidence rate of MDRPUs, but this accounted for one‐fifth of the HAPUs in this study. The MDRPUs in the ICU remain an under‐reported phenomenon; however, our findings are similar to other reported MDRPUs rates 42, 43, 44. The majority of MDRPUs were related to poor positioning or fixation of respiratory equipment. Occurrence of these ulcers could be prevented with implementation of preventive strategies such as regular assessment of the area underneath and around medical devices and regular repositioning or securement of devices 45. Using protective dressings to secure and stabilise devices also could reduce the risk for MDRPU development.

Of the PUs identified in this study, 56% were found to be stage I, while 36·5% were stage II. Consistent with findings from previous research 15, 46 the most common anatomical areas for PU development in the present study were the sacrum and the heels. The majority of the patients were positioned in a semi‐fowler's position with no heel elevation or off‐loading of pressure, leading to increased pressure points on the heels and the sacrum. This was compounded by the reduced frequency of patient repositioning. Santamaria et al. 47 found that the prophylactic use of a soft silicone multi‐layered foam dressing is effective in the prevention of sacral and heel PUs in trauma and critically ill patients.

Risk factors

The present study indicates that age, length of stay in the ICU, history of cardiovascular disease and kidney disease, infrequent repositioning, time of operation, emergency admission, mechanical ventilation and lower Braden Scale scores were the significant factors contributing to PU development. This finding corroborates that of other reported international findings 6, 34, 48. However, there are conflicting findings in the literature regarding age as a PU risk factor. Many studies report that ICU patients aged above 60 years show significant association with PU development, which is similar to our study findings 34, 48, 49. Hoshowsky and Schramm 50 state that patients who are aged above 40 years were at high risk for PU development. This is in contrast to studies that demonstrate that age is not a predisposing factor for PU development 46, 51, 52. However, comparison between studies is limited by small sample sizes, shorter stays in the ICU 51, differing nurse/patient ratios 46 and purposive sample age 52. Thus, further research is needed to examine different age groups as predictors of PU development and include them in the risk assessment scale.

In contrast, the present study did not find an association between PU development and other factors previously identified such as low BMI 53 and time spent in the emergency department 48. A possible explanation for this is related to the fact that those studies had some limitations such as small sample size, retrospective design and excluding stage I PU. Therefore, further prospective longitudinal studies are required to confirm the association between those factors and PU occurrence in the ICUs.

Limitations

Our study is limited by the time frame of data collection. A longer data collection period may have provided different data. Besides the frequency of repositioning, we did not collect data on the processes of care measures and, therefore, we cannot confirm that the PU prevention measures were carried out according to the hospital/unit policy. Furthermore, data on patient repositioning time frames were recorded retrospectively from patient notes.

Conclusion

This study has reported the PU incidence in two ICUs in SA hospitals. It is argued that measuring PU incidence rates is essential for obtaining an accurate picture of the scope of the problem and for evaluating the quality of care and monitoring patient outcomes. The data presented in this study provide baseline information of PU incidence rates in the ICU in Saudi Arabia, thus significantly adding new information to this area. The PU incidence rate in the ICUs in Saudi Arabia is high (39·5%), which indicates that more attention is required for PU prevention. Furthermore, urgent implementation of evidence‐based PU prevention protocols is needed to alleviate this pressing problem. This study also reports that MDRPUs are a continuing problem in the ICUs. A set of prevention strategies to prevent the potential MDRPUs for highly dependent patients is recommended to directly address this problem. Furthermore, identification of the risk factors that accelerate PU development in the ICU in Saudi Arabia is essential to determine the appropriate prevention strategies and the appropriate usage of available resources.

Findings from this study indicate strongly that implementation of a comprehensive PU prevention programme could prevent the majority of PU development in the ICU in Saudi Arabia. Randomised controlled trials are needed to develop and determine optimal PU prevention and management strategies. Moreover, continued measurement and evaluation of PU incidence, including MDRPUs, and evaluation of the risk factors for PU development are recommended in SA hospitals to monitor and promote best practice in skin care for highly dependent patients.