EPUAP STATEMENT ON PREVALENCE AND INCIDENCE
MONITORING OF PRESSURE ULCER OCCURRENCE 2005
Tom Defloor, Michael Clark, Anne Witherow, Denis Colin,
Christina Lindholm, Lisette Schoonhoven and Zena Moore

SUPPORTED BY THE EPUAP TRUSTEES:
BALE S., BELLINGERI A., CHERRY G., DASSEN T., DEALEY C., FLETCHER J., FURTADO K.,
GULACSI L., HEYMAN H., HIETANEN H., LUBBERS M., ROMANELLI M. AND VERDU SORIANO J.
This document is based on: Defloor T, Bours G, Schoonhoven L, Clark M. Prevalence and
incidence monitoring. Draft EPUAP Statement on prevalence and incidence monitoring.
EPUAP Review 2002; 4(1): 13–15.

BOTH prevalence and incidence are used to measure disease frequency. While both have been used to record the number of people with pressure ulcers, they provide different perspectives on the scale of the problem.1

Prevalence of pressure ulcers
Prevalence is defined as a cross-sectional count of the number of cases at a specific point in time, or the number of persons with pressure ulcers who exist in a patient population at a particular moment in time (see Table 1).2

Incidence of pressure ulcers
Incidence is defined as the number of persons who develop a new pressure ulcer, within a particular time period in a particular population. Several approaches to measuring incidence have been explored (see table 2).

Characteristics of Prevalence and Incidence
As mentioned earlier, prevalence and incidence are different measures of disease frequency. The characteristics of prevalence and incidence are summarized in Table 3.

Purpose
Before deciding to measure either pressure ulcer prevalence or incidence it is useful to consider the information the different measures can provide.

Prevalence measures the number of patients with pressure ulcers at a certain point or period in time. Thus, it provides an institution with insight into the magnitude of the problem of pressure ulcers at a given point in time, and may be an aid in planning for health resources and facilities. For example, during a prevalence survey it is possible to record how many devices (e.g., alternating mattresses) are being used at that specific moment. Given that many prevalence surveys also collect information upon aspects of prevention and treatment, such surveys may allow inferences to be made regarding the compliance with prevention and treatment protocols at that specific moment.

Incidence measures the number of persons developing new pressure ulcers during a period in time and thereby provides insight into the nature of patient groups who are at risk of pressure ulcer development. Furthermore, incidence may allow inferences to be made regarding the effectiveness of preventive measures and the compliance with prevention and treatment protocols.

The interpretation of prevalence and incidence date can be challenging When interpreting particular prevalence or incidence data it is important to understand the factors that may influence the apparent size of the pressure ulcer population. Prevalence will be affected by the number of persons with a pressure ulcer present at admission to the current care provider. If this number is high, prevalence proportions may be high too. For example, where patients with pressure ulcers are referred to a specific institution, because of the expertise of the institution in pressure ulcer treatment, this admission practice will influence the prevalence of pressure ulcers. Prevalence is also influenced by discharge practices. For example, a hospital that is able to quickly discharge patients with a pressure ulcer or, even before it is apparent, to a nursing home may have a lower prevalence of pressure ulcers than a hospital that can only discharge patients after the pressure ulcer has healed.

If the prevention and treatment protocols are of low quality or compliance with these protocols is low, then it is likely that both the prevention and treatment of pressure ulcers will be sub-optimal. This may lead to patients experiencing their pressure ulcers for a longer period of time. These patients will then be more likely to be identified during a prevalence survey and hence prevalence may be high. Incidence is also affected by discharge practices. For example, a hospital that discharges patients within a few days, i.e., before pressure ulcers have a chance to develop, is likely to have a lower incidence than a hospital that admits patients for a longer period of time. It is generally assumed (although unproven) that pressure damage may first appear 3 to 5 days after the insult to the skin and soft tissues occurred. In patients with a length of stay of, for example only 3 days, pressure damage may have occurred but not yet be visible. These pressure ulcers would not be registered, resulting in a lower incidence rate (both the incidence density and the cumulative incidence will be lower).

If the prevention protocol is of low quality or the compliance with good protocols is poor, then preventive care may not be optimal and therefore incidence may be higher. As patients with an existing ulcer but who develop additional pressure ulcers may be included in incidence studies, then adherence to treatment protocols may also influence incidence. Where a pressure ulcer heals quickly due to staff compliance with a high quality treatment protocol, it is possible that a pressure ulcer may then re-occur on the previously injured site and this may then be counted as a ‘new’ ulcer. This illustrates the complexity of determining which pressure ulcers, and which patients, to include in any incidence monitoring project.

Both prevalence and incidence are influenced by the case mix of the institution. While variations may arise it is likely that where two institutions provide identical preventative care then the centre with more patients at high risk of developing pressure ulcers may have a higher incidence of pressure ulcers. In the previous example the prevalence of pressure ulcers within the centre with more high risk patients may also be higher but this indicator will be susceptible to the admission of patients with pre-existing pressure ulcers.

Practical issues related to the collection of prevalence and incidence data
Measuring incidence rates requires a longitudinal design and in consequence such studies are likely to be more labour intensive, and hence more costly than would be a point prevalence survey. The frequency of patient observation to record incidence of new pressure ulcers may depend upon the care setting, but it is likely that in acute care daily observation of the skin would be required. Regardless of whether incidence or prevalence is to be recorded the accuracy of the submitted data needs to be assessed. Despite the fact that many studies have been performed in various countries to record incidence and (primarily) prevalence, comparison between this data are extremely re- stricted given factors such as the use of different pressure ulcer classification systems, incomparable patient groups, small samples and differences in data sources.2–4 Therefore, data must always be examined in light of the specific study methodology.2 Appendix 1 gives some practical suggestions for measuring prevalence and incidence. The selection of either prevalence or incidence data as a means of illustrating the occurrence of pressure ulcers should be made following a detailed consideration of the strengths and limitations of both epidemiological measures.

The European Pressure Ulcer Advisory Panel considers that measuring pressure ulcer incidence is the most appropriate approach if the goal is to understand how the introduction of new protocols and interventions has affected the number of patients with pressure ulcers or to predict pressure ulcers or develop and evaluate risk assessment scales. Where the goal is to identify the current size and characteristics of the pressure ulcer affected population, then prevalence may be more appropriate.

Reference list
1. Rothman K. and Greenland S. Modern Epidemiology. Second Edition. Philadelphia: Lippincott Williams & Wilkins, 1998.
2. National Pressure Ulcer Advisory Panel. Pressure Ulcers in America: Prevalence, Incidence, and Implications for the Future. Cuddigan J, Ayello EA, Sussman C, editors. 2001. Reston, VA, NPUAP.
3. Allman RM. Pressure ulcer prevalence, incidence, risk factors, and impact. Clin Geriatr Med 1997; 13(3): 421– 436.
4. van-Rijswijk L. Epidemiology. In: Morison M, editor. The Prevention and Treatment of Pressure Ulcers. London: Mosby, 2001: 7–15.
5. Defloor T, Bours G, Schoonhoven L, and Clark M. Prevalence and incidence monitoring. Draft EPUAP Statement on prevalence and incidence monitoring. EPUAP Review 2002; 4(1): 13–15.
6. Bours, G. J., Halfens, R. J., Lubbers, M. and Haalboom, J. R. (1999). The development of a national registration form to measure the prevalence of pressure ulcers in The Netherlands. Ostomy/Wound Management, 45, 28–8, 40.

a. Purpose
b. Duration.
c. Frequency.
d. Size of population.
e. Inclusion and exclusion criteria.
f. Ethical approval
g. Data collection forms.
h. Education.
i. Data collection and reliability
j. Data handling and analysis.
k. Cost.
l. Awareness.
m. Bench-marking.
n. Assessing quality.

3) Reporting of pressure ulcer incidence and
prevalence data.

a. General issues.
b. Specific issues related to the reporting of Incidence data.

1) Pressure ulcer definitions.

• It is recommended to use the EPUAP classification system for assessing the severity of pressure ulcers.
• Grade 1 pressure ulcers should be recorded as ‘warning signals’, but not included in the calculation of either prevalence or incidence rates.
• Blue/black heels would be recorded as grade 4 pressure ulcers.
• The incidence or prevalence of pressure ulcer lesions (grade 2, and a combined total of grade 3 or grade 4) should be reported.
• Clinical incidents would be defined as those individuals who develop grade 3 or grade 4 pressure ulcers.

• 2) Preparation.
  a. Purpose
• All care settings should define their prevalence of pressure ulcers at least 4 times each year.
• Incidence data should only be collected where
  information is required to solve a problem (prevalence too high when compared with existing data or other care providers);
  to evaluate the effectiveness of preventive care;
  to evaluate risk assessment tools.
  Always bear in mind why pressure ulcer occurrence data is being collected – this question will influence both the selection of method (prevalence or incidence) and the variables that need to be collected
  

  b. Duration.
  Incidence data collection should be time-limited with data being collected for 4 to 8 weeks only. This 1 to 2 month incidence monitoring should occur whenever prevalence or incident data indicates that there may be a problem. Longer-term incidence recording may not be beneficial due to a probable reduction in data quality as time passes. While it may be superficially attractive to undertake long-term incidence recording using electronic patient records the reliability of this data source with the potential for underreporting of pressure ulcers is apparent.
  
  Follow-up
  Patients included in data collection should be observed during the total follow-up time. If a patient develops a pressure
  ulcer, data collection continues until the end of follow- up time, in order to collect information about the course
  of the pressure ulcer. However, if this patient develops pressure ulcer on another location this is not counted in
  the incidence measurement again, i.e. a patient can only be counted once.
  
  c. Frequency.
  When considering acute care, it may be considered to be a general rule that the less frequently patients are observed, the less reliable the collected incidence data becomes. Incidence should be recorded at least daily in acute care (and probably more frequently in areas such as intensive care). In non-acute care incidence should be recorded on each occasion that the client is seen by a health professional in combination with family/carer education where they are asked to call the relevant health professional if they see a break in the skin.
  
  d. Size of population.
  It is unlikely that reliable incidence data will be gathered across an entire facility. In acute care incidence should be recorded at ward level with further input required to clarify the appropriate unit of data collection in non-acute care settings.
  
  Where incidence is being used to evaluate the effectiveness of preventive interventions or to develop or evaluate risk factors and risk assessment scales then the population to be included in the incidence monitoring should be dictated by power analysis.
  
  e. Inclusion and exclusion criteria.
  Clearly define the population to be surveyed before starting data collection. All patients within the surveyed population should be included while monitoring incidence or prevalence. When measuring incidence, patients with an established pressure ulcer at the start of data collection will be excluded. There should be few exclusion criteria as the ability to make general comments from the data will be diminished if multiple exclusion criteria are used.
  
  f. Ethical approval
  Formal research ethics approval appears unnecessary where prevalence data are collected or where incidence is being collected to help understand areas with high pressure ulcer prevalence. Research ethics approval will be required where incidence is being used to evaluate the effectiveness of interventions.
  
  g. Data collection forms.
  The data collection form needs to include the outcome measure – pressure ulcer development – along with data gathered to enable standardisation of the incidence or prevalence rates – for example age, risk, continence. The identification of preventive care interventions would also be appropriate although these would have to be grouped generically. In general the greater the number of data items collected the higher the costs (financial and time) of the incidence monitoring and the lower the reliability of the data.
  
  In incidence surveys, some of the information on the form will be collected once – age, sex. Some will be collected
  daily – pressure ulcer development; other element will be collected on a regular (defined as when change in condition occurs) basis – risk, continence. Information upon admission and discharge locations, the anatomical location of the pressure ulcer and significant events such as ‘has the patient been to the operating theatre or interventional x-ray’ should be included.
  
  h. Education.
  Prior to incidence monitoring there is a need for training of staff to:
  • classify pressure ulcers. It is essential that the assessors are able to distinguish a pressure ulcer lesion from other types of wounds, for instance incontinence damage, to prevent misclassification. Therefore observers should be trained in the classification of pressure ulcers (the use of the PUCLAS software may be useful). Interrater reliability should be checked and reported.
  • complete the data collection forms appropriately.
  
  i. Data collection and reliability
  Inspect the pressure areas of each individual patient. Using medical or nursing records is not a very reliable method of data collection. Caregivers are not always aware of the existence of pressure ulcers. Use a transparent device for facilitating the assessment of grade 1 (non-blanchable erythema) (see figure 1).

Figure 1. Transparent device

In prevalence surveys, assessing the skin of the surveyed patients should be carried out by two observers working independently; ideally one of the observers should not be a staff member of the unit where the patient is located. This has proven to be a reliable method.6

In incidence surveys, it would be ideal if two independent observers were to assess the skin of each patient, but this is unlikely to be practical. The reliability of data could be quantified through unannounced visits from a third party (specialist nurse for example) with ward nurses aware that at least one visit will take place within the 4 weeks of data collection. If the third-party sees a random sample of patients then inter-observer reliability could be calculated. The use of preventive interventions should be recorded during prevalence and incidence surveys; it is important to note which devices are in place at the bedside or chair of each surveyed patient to have the possibility to evaluate the compliance with the pressure ulcer protocol (prevalence and incidence) or to evaluate the effectiveness of preventive or therapeutic measures (incidence). Data on position changes should be gathered from questioning patients and/ or the nurse. Not all preventive measures used are reported in records while not all preventive measures mentioned in the medical and/or nursing records are used as prescribed.

j. Data handling and analysis.
Data handling and analysis should be decided upon before collecting data. It may be helpful to plan a short pilot study to test handling and processing of data prior to the first time incidence is collected within a health care facility.

k. Cost.
The cost of incidence recording may be high as the number of variables rises. Incidence should only be measured where and when it is necessary otherwise use prevalence surveys to monitor pressure ulcer occurrence.

l. Awareness.
There may be a Hawthorne effect upon informing staff that incidence is going to be recorded. This may be monitored by comparing each ward’s incidence data with the previously collected prevalence data.

m. Bench-marking.
This is attractive to many health care professionals and institutions. However, without standardization of the data this is meaningless and even with standardization interpretation can be challenging.

n. Assessing quality.
There are currently no agreed levels of pressure ulcer incidence that mark good or bad quality care. It was recommended that in time EPUAP should conduct a pilot study of pressure ulcer incidence to establish baseline data.

---

3) Reporting of pressure ulcer incidence and prevalence data.
a. General issues.
In all cases the population surveyed should be fully described in any report or publication; this facilitates comparison with other pressure ulcer epidemiological data. Among the items that may be described are patient ages, gender, vulnerability to developing pressure ulcers, mobility, activity, expected length of stay and care location (acute, non-acute, and specific populations such as intensive care). Pressure ulcer incidence and prevalence data should be based upon the number of patients with pressure ulcers. If any individual has more than one pressure ulcer, that person is counted only once. The data should be reported in two formats; the first including all pressure damage (including areas of non-broken skin), the second excluding grade 1 pressure ulcers and so reporting only areas where the skin was broken (pressure ulcer lesions: grade 2, 3 or 4). Comparison of results (be they incidence or prevalence) between different care providers or within a single provider over time should be done with caution (if at all). In any comparison patient characteristics and case mix should be taken into account.

b. Specific issues related to the reporting of Incidence data.
It is possible that those pressure ulcers that develop during the first few days that a patient is being monitored were the result of excessive tissue loading prior to the entry into their current care location. Always remember that both prevalence and incidence are calculated upon the number of people who have or develop pressure ulcers and not upon the number of pressure ulcers they may develop! So a person who develops several pressure ulcers over a period of time would only be counted once in calculations of prevalence or incidence.

---

4) Quality indicators.
a. Prevalence indicators
• Number of patients with pressure ulcers

• Protocol
* The number of patients at risk, receiving adequate preventive measures
Patients at risk are those patients with pressure ulcer (grade 1, 2, 3 or 4), when non-blanchable erythema is used as risk assessment method. When using a risk assessment scale, patients at risk are those patients with a risk score according to the risk assessment scale or with existing pressure ulcers.

b. Incidence-indicators
The starting point is the cumulative incidence. Incidence density is calculated by dividing the number of patients by the total period of follow-up (see earlier).
• Number of patients with pressure ulcers
Patients with pressure ulcers at admission will not be included as long the pressure ulcer(s) is (are) not healed.

• Protocol
* Number of patients receiving permanent (24 at 24 hours) adequate preventive measures
Patients at risk are those patients with pressure ulcer (grade 1, 2, 3 or 4), when non-blanchable erythema is used as risk assessment method. When using a risk assessment scale, patients at risk are those patients with a risk score according to the risk assessment scale or with existing pressure ulcers.