Incidence and prevalence of pressure ulcers in hospitals: 10-year cohort study

Pressure ulcers (PUs) are common in patients with serious illnesses and/or physical impairment.^1,2 In the US, 1.3–3 million individuals were estimated to have PUs between 2003 and 2013.³ PUs occur generally in patients with limited mobility due to spinal cord injury, cerebrovascular accident or hip fracture, and with cognitive impairment.^4,5

In addition, morbidity and mortality associated with PUs are serious problems. The most frequently associated diseases in hospitalised patients with PUs include pulmonary diseases (40%), followed by sepsis (26.7%) and neurological diseases (12%).⁶ Moreover, comorbidities increase the mortality risk of patients with PUs to nearly three times that of patients without a PU.⁷ In one study, 60% of patients with a PU had more than three comorbidities.⁷

Pre- and postoperative management are also important because PUs occur more frequently in older patients after surgical intervention.⁸ Lyder et al. reported that 4.5% of hospitalised patients develop PUs during the hospitalisation period. Patients with PUs have an average hospitalisation period of 6.7 days longer than those without a PU. However, patients with grades III and IV PUs not only have longer hospital stay durations but also have higher treatment costs, which can reach up to $43,180 USD in the US.^9,10

Due to the negative outcomes of PUs, such as increased pain and discomfort, high risk of infection, longer hospital stay, increased costs and decreased quality of life for patients, a PU preventive programme is important. Appropriately identifying common comorbidities is paramount to implementing a targeted care plan to avoid PU development, as well as selecting appropriate treatments if a PU develops.⁶

There are few published national level studies on the epidemiology of PUs in Asia, and the incidence or prevalence of PUs in Korea is still not clear. Even though it may not be possible to predict or avoid a PU, it is necessary to investigate, through a large-scale data analysis, which comorbidities correlate with higher incidence and prevalence of PU development and patient survival rate. Thus, in this study we aimed to:

• Determine the rate of incidence and prevalence of PUs in hospitalised patients in South Korea
• Determine the mortality rate of patients with major comorbidities who have a PU
• Describe the common comorbidities of PUs in the South Korean population.

Methods

Hospital system in South Korea

The first health insurance law in South Korea (the Medical Insurance Act) came into force in December 1963. Progressively, insurance coverage was expanded to cover all areas/regions of South Korea, and in 2000 all health insurance societies were integrated into a single insurer, the National Health Insurance Program.

Data source

This study was based on the South Korean population database provided by the South Korean National Health Insurance Service (NHIS). The NHIS is a single insurer managed by the South Korean Government and provides a mandatory universal insurance system for >98% of the South Korean population. The NHIS data include personal and medical information, such as surgical or medical treatment provided, diagnoses, age, sex, and characteristics of medical institutions and clinicians. Hospitals included in this data range in size from 30 to over 2000 beds.

The diagnosis was coded in accordance with the International Classification of Disease, Tenth Revision (ICD-10). All of the procedures, such as PU dressing, and all surgery related to PUs were coded in accordance with the South Korean national code system and used during the reimbursement process. All claim reimbursements were reviewed and regularly evaluated by the Health Insurance and Review Assessment (HIRA) using the digital patient chart. Using the NHIS claims data, which depend on fee-for-service reimbursement, can provide high diagnostic accuracy; in one validation study of rheumatoid arthritis prevalence, reported diagnostic accuracy was >90%.^10,11,12,13 Since 2015, the database has been accessible to any researchers whose study protocols are approved by an official review committee (Fig 1).

Definition of PU and rates of incidence and prevalence

Patients with a diagnosed PU were defined and identified, as per the following steps. First, patients were identified based on the ICD-10 code for PU (L89) and procedures including PU dressing (debridement of wound, simple dressing and infectious wound dressing).

Secondly, patients with an ICD-10 code for PU (L89) and who underwent a flap operation (including island, muscle, musculocutaneous or myocutaneous, fascia, and fasciocutaneous flaps) were identified (Box 1). In patients with PUs who developed comorbidities, several major comorbidities were identified using the ICD-10 codes, and their mortality was tracked. Inclusion and exclusion criteria are shown in Box 2.

Box 1.Definition of a pressure ulcer case

• ICD-10 code of L89 + HIRA code of pressure ulcer dressing procedure* or
• ICD-10 code of L89 + HIRA code of flap operation**

*Korean Health Insurance and Review Assessment (HIRA) code of pressure ulcer dressing, including debridement of wound, simple dressing and infectious wound dressing (SC027, M0111, M0121);**HIRA code of flap operation, including island, muscle, musculocutaneous or myocutaneous, fascia and fasciocutaneous flaps (S0164, S0165, S0166, S0167, S0168)

Box 2.Study process and population

• Start of follow-up of the Korean population from 2002
• Washing out period from 2002–2005
• Including new individuals aged ≥20 years
• Including pressure ulcer diagnosed cases based on definition criteria
• Start of follow-up of new onset pressure ulcer cases within the Korean population from 1 January 2006 to 31 December 2015
• Inclusion/exclusion criteria
• Include
• Individuals aged ≥20 years old, yearly
• Individuals with a new episode of pressure ulcer
• Exclude
• Diagnosed cases of pressure ulcer at the time of death

The incidence rate is the number of new cases per 10,000 people in a given year. Prevalence rate is the number of cases per 10,000 people which already existed in a given year.

Study population

We initially included hospitalised patients aged ≥20 years who were registered in the South Korean NHIS in each year between 2002 and 2015.

To investigate incidence rate of PU from 2006 onwards, we had to extract recurrent cases who were diagnosed with PUs and readmitted the following year. Therefore, we have a four-year washout period from 2002–2005 in order to identify only new cases from 2006. Finally, this study covered a total of:

• 36,195,121 individuals in 2006
• 36,537,562 individuals in 2007
• 37,485,128 individuals in 2008
• 37,965,386 individuals in 2009
• 38,427,247 individuals in 2010
• 38,877,086 individuals in 2011
• 39,416,713 individuals in 2012
• 39,905,161 individuals in 2013
• 40,430,523 individuals in 2014
• 40,987,643 individuals in 2015.

Patients who were diagnosed with a PU at the time of death were excluded from this study. Patients who died as a consequence of their PU were counted cumulatively from 2006 to the end of 2016 for mortality rate calculation.

Ethical approval

This study was conducted according to the ethical principles of the declaration of Helsinki and was evaluated and approved by the South Korean NHIS (No. NHIS-2018-1-290) and the institutional ethical review board of Hanyang University Guri Hospital (No. GURI201803017) (Table 1).

Statistical analysis

The prevalence and incidence of PUs were calculated with standardised age and sex during the 10-year period, based on the 2010 Korean Statistical Information Service (http://kostat.go.kr/portal/eng/index.action). Incidence of PU was defined as new cases of PUs only, whereas the prevalence rate was the total number of PUs, including cases of recurrence during the study period.

Survival of patients with major comorbidities accompanying PUs was analysed using the Kaplan–Meier method for the study period. All statistical analyses were performed using the SAS software (version 9.4, SAS Institute, US).

Results

Prevalence and incidence of PUs in hospital-based diagnosis

The total number of the study population included was 36,195,121 individuals in 2006, which gradually increased to 40,987,643 in 2015. Table 2 presents the prevalence and incidence of patients with a PU who were hospitalised each year between 2006 and 2015. Age- and sex-standardised prevalence rate showed a slightly decreasing trend from 20.2 per 10,000 people in 2006 to 18.9 per 10,000 people in 2015. The standardised incidence rate also decreased from 17.1 per 10,000 people in 2006 to 14.9 per 10,000 people in 2015. Incidence and prevalence by age group are shown in Table 3.

Changes in PU incidence in individuals aged 60–90+ years showed a marked increase, but the change between 2006 and 2015 was not significant for the same age groups. Prevalence showed a trend similar to that for PU incidence. Prevalence markedly increased in individuals >70 years of age but did not show remarkable changes from year 2006 to year 2015, within the same age group (Fig 2).

Mortality of patients with major comorbidities accompanying PU

A total of 581,299 patients were found to have a PU over the whole study period. The most common comorbidities found in patients with PUs in this study were pneumonia (46,673 cases, 8.0%), cerebral infarction (34,904 cases, 6.0%), sepsis (25,335 cases, 4.3%), femoral neck fracture (24,946 cases, 4.3%) and malignant neoplasm of the bronchus (22,162 cases, 3.8%). The mortality rate of patients with PUs who also had a malignant neoplasm of the bronchus and sepsis was very high, with mortality per 1000 person-years being 3269.8 and 1017.8, respectively. PU patients with pneumonia, cerebral infarction and femoral neck fracture had a relatively low mortality (ranging from 222.4 to 963.8 per 1000 person-years) than those with sepsis (1017.8) and bronchogenic malignancy (3269.8) (Table 4).

The 10-year survival rate of patients with a PU with major comorbidities

The Kaplan–Meier curve shows the survival curve of the major comorbid disease of patients with PUs during the 11-year follow-up period to the end of 2016. Among the diseases with high incidence of PUs, femoral neck fracture shows a gradual inclination in four-year mortality, and >60% of individuals show a two-year survival compared with sepsis and malignant neoplasm of bronchus (Fig 3). As shown in Fig 2, men and women show a different pattern of survival. Femoral neck fracture and cerebral infarction showed a gradual trend in terms of death in women, but PU patients with sepsis and pneumonia showed high mortality. Generally, women showed a gradual mortality trend, even when they had serious comorbid conditions.

Discussion

In this study, the incidence and prevalence of PU in acute hospitals showed a gradual decreasing trend from 2006–2015; however, after the sixth decade, the rates markedly increased until the ninth decade. PU conditions frequently accompany five commonly occurring conditions—pneumonia, cerebral infarction, sepsis, femoral neck fracture and malignant neoplasm of bronchus, in the order of highest incidence. The survival patterns during the 11-year follow-up were also different according to comorbidities. Regarding the relationship between chronic disease and PUs, mortality may be high depending on the accompanying comorbidity of the patient. When patients with a PU also have dementia, Parkinson's disease or anaemia associated with chronic disease, the mortality rate increases up to 1.76, 1.72, and 1.63 times, respectively.¹⁴

Moreover, the management cost of PUs is expensive. Preventing PUs in patients with high-risk comorbidities and early treatments may reduce the cost of medical expenses. Clark et al. estimated that new PUs occur in 4–10% of patients admitted to hospitals in the UK, with the estimated cost of wound care to the National Health Service (NHS) at between £2.3–3.1 billion per year (2005 and 2006, respectively); costs in one NHS trust were estimated to be £9.89 million.

Apart from the financial burden to healthcare, PUs are a significant cause of morbidity and mortality in patients. Moreover, 70% of PUs occur in patients over 70 years of age.^{15,16,17,18,19} Generally, PUs occur in patients with paraplegia, stupor or cognitive impairment, and physical impairment due to spine injury or femur fracture.^1,3 If PUs are to be expected in hospitalised patients, aggressive preventive measures are required because PUs not only can affect the patient's quality of life but can also increase their risk of morbidity and mortality.²⁰

The prevalence of PUs varied across studies. The prevalence rate of PUs in hospitals ranged from 5.8–26% according to studies from Europe, Canada and the US.²¹ Other studies showed that 1.6–59.2% of PUs developed in the intensive care unit.²² A training hospital in China reported a very low prevalence of PUs—approximately 1.8% compared with 18.1% in Singapore and 9.6% in a geriatric hospital in Japan. Some papers reported that while countries such as Germany, the Netherlands and the US had robust data on PU incidence/prevalence in hospitals, other countries, such as the UK, had none.²³ Some reports said that the highest PU prevalence was from the Netherlands, and the lowest was from Finland.²⁴ Prevalence and incidence of PUs are different among studies.

There was a wide range of incidence rate in several studies. Fogerty et al. reported an incidence rate of 1.5–10.27% in hospitalised patients in the US.²⁵ The National Pressure Ulcers Advisory Panel reported that the incidence of PUs ranged from 0.4–38%, 0–17% and 2.2–23.9% in acute care, home care and long-term care settings, respectively. The prevalence rate also showed a similar range, with the rates being 10–18%, 0–29% and 2.3–28% for acute care, home care and long-term care settings, respectively.^26,27 There are a few studies reporting low PU incidence in hospitalised patients in Western countries. A study conducted in Germany showed a low incidence of PUs, and that the prevalence of PU as a primary diagnosis was 0.06% and 1.19% as a secondary diagnosis. Moreover, the incidence of PUs per 1000 patients in hospitals was very low (0.023% in 2002 and 0.148% in 2013). Given that PU grade for principle diagnosis was mostly grade III or IV and that for additional diagnosis was grade I or II (60%), low grade PUs may likely be excluded in the secondary diagnosis.²⁸

In this study, there was a sudden increase in the incidence and prevalence rate in 2007. The reason for this appears to be that the South Korean Government introduced national health care for older people who were bedbound in 2008, and so many older people were admitted to hospital in 2007 in order to gain a health certificate so that they could qualify for the national financial support. Subsequently, the standardised rate of incidence and prevalence slightly decreased over the following 10 years.

The incidence of PUs was higher in men than in women, and the mortality rate in men was also high. Male patients with a PU tended to have a malignant neoplasm of bronchus, sepsis and malignant neoplasm of the stomach, all of which have a high mortality risk. On the other hand, female patients with a PU tended to have pneumonia, cerebral infarction and femoral neck fracture, which are comorbidities with relatively low mortality risk. In PU patients with low mortality risk, management of PUs should be aimed at treating the ulcers, if the acute phase has passed. Moreover, understanding the comorbid(ities) of the patient with a PU is important as the PUs can greatly affect the patient's mortality. In this study, five major comorbidities, including pneumonia, cerebrovascular disease, sepsis, femoral neck fracture and bronchogenic malignancy, accounted for approximately 26.4% of PU diagnosed cases. In another study, other common comorbidities such as urinary tract infection, heart failure, diabetes, Parkinson's disease and dementia were often related to PUs.²⁹ These comorbidities, in particular lung cancer, pneumonia and sepsis, should be carefully managed to prevent PU development. They could be used as a risk predictor of PU. In large-sized PUs with no improvement after conservative treatment, surgical management should be considered as it has been shown to reduce the morbidity, mortality and consequent economic burden on patients.^30,31,32

Limitations

This study provided nationwide data of incidence, prevalence and mortality rate of PU, but it has several limitations. In this study, the incidence and prevalence rates are low compared with those of studies in Western countries. In South Korea, hospitals reimbursed patients through a fee-for-service system; thus, the few reports of PU dressing or low requests for charge claims cannot be entirely explained in this study. Moreover, given that only patients with the ICD-10 code for PU (L89), diagnosed during hospitalisation in an acute care hospital and who were charged for the PU dressing, or who underwent a PU flap operation were enrolled, there is a high possibility that low-grade PUs were excluded from the additional diagnosis.

Furthermore, in South Korea, many PU patients may stay in nursing homes, which are not hospital care settings. Nursing homes do not provide any statistical outcomes, such as health indicators or diagnoses, at this time. The low incidence and prevalence of PUs reported in this study may be due to the fact that we only focused on patients in hospitals. Furthermore, the definition of PUs was based on the ICD-10 code of claims data. Therefore, PU patients at home or in nursing/care homes were not included.

This study was not able to determine the severity and locations of PUs. Finally, the narrow inclusion criteria may have excluded PUs with additional diagnoses of grade 1 or 2, if they were regarded to be of minimal clinical importance.

Conclusions

This study provided nationwide data on incidence and prevalence of PUs in South Korea. The incidence and prevalence of PUs are rapidly increasing, mainly due to an increasing older population, and some medical conditions were related with the risk of developing a PU. In particular, in patients with a PU who also had pneumonia, sepsis or lung malignancy there was a high mortality rate. Therefore, hospital staff should carefully observe patients' positioning and implement measures to prevent the development of PUs in patients with these comorbidities.

Understanding the impact of PUs on patients and healthcare services, including the financial burden on families and wider society is crucial. To this end, the importance of hospital staff education on prevention and early detection of PUs cannot be overemphasised.

Reflective questions

• What is the pressure ulcer (PU) incidence and prevalence rate in your country?
• How do you manage the prevention of PU in your hospital?
• How many PU patients develop serious comorbid conditions in your hospital?