I S S U E S A N D IN N O V A T I O N S I N N U R S I N G P R A C T I C E

Pressure area care: an exploration of Greek nurses knowledge and practice

Kalliopi Panagiotopoulou
BA MN RGN

Deputy Director, Department of Nursing Education, Army General Hospital of Athens, Kanelopoulou, Athens, Greece

and Susan M. Kerr

BA MSc RGN HV

Research Fellow, Department of Nursing and Community Health, Caledonian Nursing and Midwifery Research Centre, Glasgow Caledonian University, Glasgow, UK

Submitted for publication 2 January 2002 Accepted for publication 30 July 2002

Correspondence: Kalliopi Panagiotopoulou, Department of Nursing Education, 401 Army General Hospital of Athens, Kanelopoulou 1, 11525, Athens, Greece. E-mail: kalliopipanagiotopoulou@ hotmail.com

P A N A G I O T O P O U L O U K . & K E R R S .M . ( 2 0 0 2 )

Journal of Advanced Nursing

40(3), 285296 Pressure area care: an exploration of Greek nurses knowledge and practice Background. Despite a plethora of information on the prevention of pressure sores, they remain a signicant problem in both hospital and community settings. The need to reduce the incidence of pressure sores has been well documented; unfortunately there is little evidence to suggest improvement. The reasons for this lack of improvement have been explored, but the picture remains unclear. While some studies have suggested that nurses have the appropriate knowledge to prevent pressure sores developing (but do not use their knowledge), others suggest that nurses knowledge of preventive strategies is decient. In Greece, similarly to the United Kingdom (UK), the incidence of pressure sores is high. There is currently no evidence on Greek nurses knowledge and practice and therefore no baseline on which to build, in terms of improving practice. Aim. The purpose of this study was to explore Greek nurses knowledge of risk factors, areas at risk and recommended preventive strategies in relation to pressure area care. In addition, information was sought on nurses current preventive practice and any barriers to good practice. Research methods. The study was exploratory and descriptive, adopting a crosssectional survey approach. The sample was drawn from the population of nurses working in a military hospital near Athens. The data were collected over a 4-week period in June 2000, using a self-completed questionnaire. Results. Although the knowledge-base of many of the nurses was good in relation to risk factors and areas at risk, a signicant proportion were unaware that methods such as massage and donuts are no longer recommended. This lack of knowledge inuenced practice with these methods commonly being used. In relation to barriers to good practice, a signicant proportion of nurses reported that they could not access, read or understand research ndings. This has obvious implications for the implementation of evidence-based practice. Conclusion. The results of this study suggest that the knowledge and practice of participants could be improved. It is of particular concern that methods known to be detrimental were in common use. Finally, there is a need to improve the research skills of Greek nurses in order to provide them with the appropriate knowledge to use research ndings.

2002 Blackwell Science Ltd

285

K. Panagiotopoulou and S.M. Kerr

Keywords: pressure sores, knowledge, practice, risk factors, risk, prevention, research utilization, Greece

Introduction
Despite a plethora of information on the prevention of pressure sores, they remain a signicant problem in both hospital and community settings. The incidence, known to vary from 4 to 16% (depending on the case-mix and classication of sores), represents a signicant burden of suffering for patients and is costly to healthcare providers 1 (Plati et al. 1992, Oot-Giromini 1993, Clark & Watts 1994, National Health Service Centre for Reviews & Dissemination [NHS CRD] 1995, Baltzi-Economopoulou 1997). The need to reduce the incidence of pressure sores has been well documented (Department of Health [DoH] 1992). In considering how this goal might be achieved, nursing care has been highlighted as a major inuence, with good preventive strategies being central (Land 1995). In order to deliver high quality care, it is essential that nurses base their practice on the best available evidence, and if they are to function effectively they must have knowledge of risk factors, areas at risk and preventive strategies (NHS CRD 1995). Knowledge, alone, is insufcient, as nurses must actually use the knowledge they have. Studies that have explored nurses knowledge and practice in relation to pressure area care are reviewed below.

Nurses knowledge
Knowledge of risk factors The number of studies that have explored nurses knowledge of risk factors is relatively small, and comparisons across studies are difcult, as different sampling frames and methods have been used. A study by Maylor (1999) used a cross-sectional survey approach to explore nurses knowledge of risk factors. The study targeted the total population of trained and untrained nurses (excluding midwives, psychiatric nurses and health visitors) working in an NHS trust in Wales (n 625). The nurses were asked to indicate their level of agreement (using a 4-point Likert scale) with a number of statements relating to risk factors (e.g. to state whether a low albumin level is a risk factor). The 18 items included in the list were compiled with the assistance of an expert panel. Results from the 439 respondents (70% response rate) suggested that the average level of agreement with expert opinion was generally good. Perhaps unsurprisingly, registered nurses mean score was signicantly greater than that of unqualied nurses (P < 0001).
286

Another study that found nurses knowledge of risk factors to be good, was undertaken by Bostrom and Kenneth (1992) in the United States of America (USA). A cross-sectional survey approach was used to collect data from a random sample of nurses (n 398) who were members of the Nursing Consortium for Research in Practice. In this case, the nurses were asked to comment on a list of 12 factors (generated from the literature and expert opinion), stating whether each factor was or was not considered to be a risk factor. Data gathered from the 245 respondents (62% response rate) suggested that the level of knowledge was good. Seventy-one percent (n 245) answered 10 of the 12 questions correctly. A factor that may limit the generalizability of these results is that participants were all members of the Nursing Consortium for Research in Practice. It would seem reasonable to assume that this group may have been more knowledgeable than nurses who were not members of the consortium. While the studies discussed suggest that nurses knowledge of risk factors is good, work by Beitz et al. (1999) suggests that it is poor. In this study, a survey approach was used to collect data from a convenience sample of 86 qualied and unqualied nurses working in a community hospital in the USA. Data were collected using the Pressure Ulcer Risk and Treatment Test developed by Hayes et al. (1995). Results from the 86 respondents (100% response rate) suggested that the knowledge was limited in the following areas: importance of pressure relief (45% did not state that this was important); friction (59% did not recognize that friction is a predisposing factor); and age (94% did not agree that a patients level of risk increases with age). A limitation of this work is the fact that a convenience sample was used, which obviously limits the generalizability of the results. Another study that has shown nurses knowledge in relation to risk factors to be poor was conducted by Parker et al. (1998) in a general hospital in England. A survey approach was used to collect data from a purposive sample of 275 nurses selected from a total workforce of 1500. The results from the 255 respondents (87% response rate) suggest that their knowledge of risk factors was limited. With regard to extrinsic factors, the percentage who listed pressure, shear and friction was small (17, 8 and 8%, respectively). In relation to intrinsic factors the results were as follows: nutritional status (39%); body weight (12%); incontinence (12%); neurological factors (4%) and age (7%). The fact that the sample was selected purposively again limits the generalizability of the ndings.

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

Issues and innovations in nursing practice

Pressure area care in Greece

Knowledge of areas at risk Information on nurses knowledge of the areas of the body most at risk of developing pressure sores is very limited. The only study that appears to have explored this was the one conducted by Parker et al. (1998) that was discussed above in relation to nurses knowledge of risk factors. In this case the purposive sample of 255 nurses was asked to identify the two areas of the body that are most at risk of developing sores. Perhaps unsurprisingly, the level of knowledge was found to be good, with 95% of nurses identifying the sacrum and 82% identifying the heels as the most vulnerable areas. This level of information does not allow the reader to gauge knowledge levels in relation to other areas, such as the spine, the elbows and the ischial tuberosities. In addition, no account was taken of areas at risk in different positions (i.e. semi-recumbent, supine and lateral). Knowledge of preventive strategies The number of studies that have explored nurses knowledge of preventive strategies is relatively small. Halfens and Eggink (1995) used a cross-sectional survey approach to gather data from a sample of qualied nurses in the Netherlands. In this case the sample was selected randomly from a group of nurses who received copies of a free weekly nursing journal (it was estimated that 80% of all nurses working in the Netherlands received this journal). The content of the questionnaire was derived from infor2 mation contained in the Dutch Consensus Report (1992). The Consensus Guidelines classify preventive strategies in three ways: methods that are useful and advised for general application in all patients at risk of developing pressure sores; methods expected to be useful in individual cases, but which are not advised for general application; and methods which are not considered useful. Nurses were presented with a list of preventive strategies and asked to comment on how useful they considered each strategy to be. Data gathered from the 373 respondents (76% response rate) suggested that the level of knowledge in relation to methods that are always useful was good. The nurses were particularly knowledgeable about the following: the importance of providing a clean, smooth, dry bottom sheet (997%), maintaining good hygiene (997%) and palpating and inspecting the skin daily (915%). Levels of knowledge were, however, more limited in relation to methods that are sometimes useful. Respondents appeared to have difculty in differentiating between methods that are always useful and those only recommended in individual cases. Finally, knowledge of methods that are not recommended was poor. A substantial percentage stated that methods such as massage,

topical creams and donuts (ring-shaped sitting aids) are always useful (70, 70 and 46%, respectively). Studies exploring nurses knowledge of preventive methods have also been undertaken by Russell (1996) and Hill (1992). Although these UK studies both demonstrated that knowledge was good, the results cannot be generalized as the samples were small (30 and 15, respectively) and convenience sampling was used.

Nurses preventive practice

Preventive strategies involve methods used to assess individual patients level of risk (i.e. risk assessment tools) and also interventions used to prevent pressure sores developing. The number of studies that have explored preventive practice is relatively small. One of the key studies was discussed previously in relation to knowledge of preventive strategies (Halfens & Eggink 1995). Interestingly, despite the fact that nurses were knowledgeable about particular preventive strategies, they did not always appear to translate their knowledge into practice (e.g. where 86% of nurses were aware that nutritional deciency should be treated, only 79% stated that always ensured that a nutritional deciency was in fact treated). On the other hand, poor knowledge about methods that should not be applied was generally translated into practice, with a signicant proportion of nurses reporting that massage, topical creams and donuts were used regularly. The fact that massage is a commonly used preventive strategy was also found by others (Hill 1992, Wilkes et al. 1996, Beitz et al. 1999). A more recent study, undertaken in an NHS hospital in Scotland, used retrospective document analysis (325 case notes) to explore whether there was an association between patients level of risk and specic nursing interventions (Tolmie 2000). The results suggested that almost half (42%) of patients identied as being at risk of developing pressure sores, did not appear to have received care that included appropriate preventive strategies. Although practice appeared to be poor, the author highlighted the fact that to some extent the results may have been due to poor documentation rather than poor practice. Finally, a number of studies have sought to determine whether nurses routinely use risk assessment tools (e.g. Norton 1962, Waterlow 1985, Braden et al. 1987) to predict which patients are at risk of developing pressure sores. Accurate prediction of risk status is important, as it should ensure that nurses intervene, when appropriate. While studies from the UK suggest widespread use of risk assessment scales (Maylor 1999, Tolmie 2000), those from other countries
287

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

K. Panagiotopoulou and S.M. Kerr

report that their use is uncommon (Halfens & Eggink 1995, Mockridge & Anthony 1999).

Barriers to good practice

When exploring nurses implementation of research-based ndings (in relation to pressure area care) it is important to be aware of any barriers that may inhibit good practice. Although much has improved since the early 1980s, both within and across countries, many of the barriers highlighted by Hunt (1981) (i.e. nurses do not know about, understand, believe or know how to apply research ndings) still exist. Another barrier to good practice is resources. These can be thought of in two ways: rst facilities that allow staff access to the literature (i.e. adequate library facilities) and second the provision of sufcient human resources and equipment (such as pressure relieving surfaces) to allow nurses to provide adequate care. Although there has been a signicant improvement in library facilities in the UK in the last 10 years, provision varies across the country (Clinical Standards Advisory Group (CSAG) (1998). Nurses in many other countries, including Greece, have a very limited access to library facilities and electronic databases. In Greece, similarly to the UK, the incidence of pressure sores remains high (68%) (Plati et al. 1992). Unfortunately, there is currently no evidence on Greek nurses knowledge and practice and therefore no baseline on which to build in terms of improving practice. The results of studies undertaken elsewhere, although of interest, are not generalisable to the Greek situation as there are a number of cultural and health system differences.

Population and sample The target population was all registered and enrolled nurses working in hospitals in Greece. The sample was selected from the accessible population, that is, from the population of 438 nurses working in a military hospital near Athens. Guided by the literature, and in order to facilitate comparisons across studies, nurses who would be expected to demonstrate either specialist knowledge and/or who would have little or no experience of pressure area care were excluded from the study. In addition, senior nurses with little or no direct patient contact were excluded. The following inclusion and exclusion criteria were therefore applied: Inclusion criteria. Registered and enrolled nurses with roles and responsibilities connected with direct patient care. Exclusion criteria. Registered and enrolled nurses working in intensive care unit (ICU), accident and emergency, theatre or psychiatric wards. Access was negotiated with the Director of Nursing Services. Once this had been agreed, nurses who met the inclusion criteria were identied from the population of nurses working in the hospital. The total number of eligible nurses was 166 (69 registered and 97 enrolled nurses). Instrument The instrument used incorporated previously validated instruments developed by Maylor (1999) and Halfens and Eggink (1995). However, in order to ensure that the study objectives were met, some questions exploring knowledge of areas at risk and barriers to good practice were added. In addition, a small number of questions were excluded as they referred to situations that do not apply in Greece. The main sections of the questionnaire are outlined in Table 1. Face and content validity of the instrument were assessed by a group of experts that included nurse educators, experienced researchers and tissue viability nurses (n 6). This group also assessed issues that might have inuenced the reliability of the instrument, such as the wording and order of questions and layout of the questionnaire. On completion of this process the instrument was translated into Greek by the rst author, with the accuracy of the translated version being ratied by an experienced tissue viability nurse working in Greece. The Greek version of the instrument was then piloted with a small group of nurses in Greece (n 10). Several minor alterations were made to the questionnaire following the pilot study (e.g. the Greek words used to describe the ischial tuberosity were altered as participants stated that the original translation lacked clarity).

The study
This study takes a rst step in gathering information that could improve Greek nurses clinical effectiveness.

Aims
The study aimed to: (1) explore Greek nurses knowledge of risk factors, areas at risk and preventive strategies; (2) explore preventive practice; and (3) identify any barriers that may inhibit good practice.

Research methods
Research approach The study was exploratory and descriptive in nature, adopting a cross-sectional survey approach.
288

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

Issues and innovations in nursing practice Table 1 Outline of questionnaire content Section One Knowledge of risk factors (Maylor 1999) Knowledge of areas at risk of developing pressure sores (PUPG 2000) Knowledge of preventive strategies (Halfens & Eggink 1995) Section Two Perceived practice (Halfens & Eggink 1995) Section Three Barriers to good practice Section Four Continuing professional development Section Five Demographic data Table 2 Sample characteristics

Pressure area care in Greece

Percent Gender (n 116) Female Male Age 2024 2529 3034 3539 Basic training (n 117) Degree (RN) Diploma (EN) Number of years qualied (n 115) <1 years 14 years 5 years Current area of practice (n 116) Medical Surgical

871 129 100 509 300 91 470 530 09 408 583 552 448

Data collection Data were collected during a 4-week period in June 2000. On-site distribution and retrieval of the questionnaires was achieved with the assistance of a colleague working at the study site. Each of the 166 questionnaires was addressed personally and delivered with an information sheet and reply envelope. Nurses who wished to participate in the study were asked to complete and return the questionnaire within a 2-week period. They were assured that their responses would be treated condentially and that participation was voluntary. Following a reminder, the total number of returned questionnaires was 118; the response rate was therefore 71%. Data analysis Data were coded and entered into the statistical software package SPSS (version 8). Descriptive and inferential statistics were prepared. When comparing results between two groups, t-tests were used for outcome measures that could be assumed to be approximately normally distributed (e.g. total knowledge scores). When comparing results across more than two groups, one-way analysis of variance (ANOVA ) was used. Finally, in instances where the data were categorical, the chisquared test was employed. The level of signicance was set at P 005.

Table 3 Continuing professional development Percent Read a nursing journal/s (n 117) More than once a month 154 Once every 13 months 274 Once every 46 months 214 Less than once every 6 months 188 I have not read a nursing journal in the last 12 months 171 Read a research-based article on pressure area care (n 117) Less than 1 month ago 12 13 months ago 205 46 months ago 171 More than 6 months ago 368 Never 137 Attended in service training on pressure area care (n 117) Less than 1 year ago 411 12 years ago 256 More than 2 years ago 120 Never attended such training 214

Results
The results are presented below. As would be expected when collecting data using a self-completed questionnaire, the data set is not 100% complete. The level of response for each question is given.

was 287 years (SD 374). Fifty-three percent were enrolled nurses and 583%, had been qualied for more than 5 years. In relation to continuing professional development (Table 3), 496% of the nurses reported that they had read a research-based article on pressure area care in the previous 6 months while 667% had attended in-service training on pressure area care in the previous 2 years.

Sample characteristics and continuing professional development

The characteristics of respondents are presented in Table 2. As indicated, the majority were female (871%) and the mean age

Knowledge of risk factors

The nurses knowledge of risk factors is summarized in Table 4. The question asked was as follows: In your opinion
289

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

K. Panagiotopoulou and S.M. Kerr Table 4 Knowledge of risk factors (answers judged as correct in terms of expert opinion are indicated in bold) Strongly agree (%) 907 684 00 605 179 157 829 63 289 325 00 168 322 339 144 333 43 648 556 414 539 675 220 287 Agree (%) 93 282 120 316 339 463 171 107 474 513 63 460 466 443 333 474 200 269 444 379 443 299 263 539 Disagree (%) 00 26 370 79 415 343 00 500 184 137 559 292 195 183 441 175 574 65 00 164 17 17 432 157 Strongly disagree (%) 00 09 509 00 71 37 00 330 53 26 378 80 17 35 81 18 183 19 00 43 00 09 51 17

Risk factors High pressure over a bony prominence, for a long duration (n 118) Incontinence (n 117) High blood albumin levels (n 108) Friction (n 114) Shearing forces (n 112) High pressure over a bony prominence, for a short duration (n 110) Immobility (n 117) Analgesics (n 112) Low pressure over a bony prominence for a long duration (n 114) Hospital mattresses (standard mattress) (n 117) Hospital easy chairs (n 111) Patients usual home mattresses (n 113) Immobile patient propped up in bed (n 118) Immobile patient sitting up in a chair (n 115) Patients usual home chair (n 111) Patient age of 75 years (n 114) Low pressure over a bony prominence for short duration (n 115) Low blood albumin levels (n 108) Poor nutritional status (n 117) Body weight (above average) (n 116) Body weight (below average) (n 115) Concurrent diseases (n 117) Confused mental status (n 114) Medications (n 115)

which of the following can be an important contributory factor to the development of pressure sores? Respondents were asked to indicate their answer on a 4-point Likert Scale (strongly agree, agree, disagree, strongly disagree). The answers judged as correct in terms of expert opinion (Maylor 1999) are highlighted in bold for ease of identication. As indicated, the level of agreement with expert opinion was particularly high for the following risk factors: high pressure over a bony prominence for a long duration (907%), incontinence (684%), friction (605%) and low blood albumin (648%). Level of agreement with expert opinion was limited in the following areas: shearing forces (179%), analgesics (107%), hospital easy chairs (0%) and confused mental status (263%). Knowledge score (risk factors) Following guidelines produced by Maylor (1999), a total knowledge score was calculated for each respondent. Taking each of the 24 response-items individually, four points were awarded for complete consensus with expert opinion (e.g. expert response strongly agree; participant
290

response strongly agree), three points for partial agreement (e.g. expert response strongly agree; participant response agree), two points when there was a weak discrepancy (e.g. expert response strongly agree; participant response disagree) and nally one point was awarded when there was a strong discrepancy (e.g. expert response strongly agree; participant response strongly disagree). The knowledge score therefore ranged from 24 (lowest possible score) to 96 (highest possible score). This score was then transformed to produce a percentage agreement with expert opinion. The average level of agreement with expert opinion was 71%, with a range of 486847%.

Knowledge of areas at risk

Knowledge of areas of the body at risk of developing pressure sores in the semi-recumbent, supine and lateral positions is summarized in Table 5. Respondents were provided with a picture of a patient in each of the three positions and were asked to indicate which areas of the body they

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

Issues and innovations in nursing practice Table 5 Knowledge of areas at risk (answers judged as correct by the Pressure Ulcer Prevention Consensus Group are indicated in bold) Areas at risk (n 117) Occiput Back of head Ear Shoulder blades Shoulders Elbows Spine Hip Sacrum Ischial tuberosity Knee Heels Toes Ankle Semi-recumbent Yes (%) 385 632 145 675 145 521 470 436 829 829 43 949 26 179 Supine Yes (%) 470 761 94 855 239 778 761 393 923 778 26 974 17 111 Lateral Yes (%) 94 17 932 282 769 590 26 778 26 513 376 410 205 872

Pressure area care in Greece

methods in one of three ways: methods that are useful and are advised for general application in all patients at risk; methods expected to be useful in individual cases, but which are not advised for general application in patients at risk; and methods that are not useful. The results are presented in Table 6. Correct answers are highlighted in bold for ease of identication. As indicated, the level of agreement with the Consensus Guidelines in relation to methods considered as always useful was generally high. Nurses appeared to have difculty in identifying methods not advised for general application, but which may be useful in individual cases. For example, air mattresses were incorrectly considered by a majority (675%) to be always useful. Finally, level of agreement with the consensus guidelines on methods considered to be never useful was low. Only 69% of the nurses stated correctly that massage was never useful, while more than half (569%) considered that it was always useful. Knowledge score (preventive strategies) In order to judge levels of knowledge in relation to preventive strategies, a knowledge score was calculated. Following the guidelines provided by Halfens and Eggink (1995), a score of 1 was awarded for every correct response and 0 for every incorrect response. As there were 20 response-options, the possible knowledge scores ranged from 0 (lowest possible score) to 20 (highest possible score). Following this process the average level of agreement with expert opinion was 50% with a range of 3075%.

considered were at risk from a choice of 14 alternatives (e.g. sacrum, heels, buttocks). Answers judged as correct in terms of the Pressure Ulcer Prevention Consensus Group (PUPG 2000) are highlighted in bold for ease of identication. As indicated, levels of agreement with the consensus guidelines were particularly high for heels in the semi-recumbent (949%) and supine (974%) positions. The level of agreement with the consensus guidelines was also high for the sacrum in the semi-recumbent (829% and supine (923%) positions. Of note is the fact that the toes were rarely identied as an area at risk (semi-recumbent 26%; supine 17%). Knowledge score (areas at risk) In order to judge levels of knowledge of areas at risk in each of the three positions, a total score was calculated. A score of 1 was awarded for each correct response and 0 was awarded for each incorrect response. As there were 14 response-options for each position, the possible score ranged from 0 (lowest possible score) to 42 (highest possible score). Following this process, the average level of agreement with expert opinion was 705% with a range of 452905%.

Preventive practice
Results in relation to preventive practice are summarized in Table 7. Again, this section of the questionnaire was derived from that developed by Halfens and Eggink (1995). The questions asked nurses to comment on their practice in terms of: methods used in their ward (in principle) for every patient at risk; methods used in individual cases (but not for every patient at risk); and methods never applied. Again, correct answers are highlighted in bold for ease of identication. As indicated (Table 7), the level of agreement with the consensus guidelines (Halfens & Eggink 1995) with respect to the methods used in principal for every patient at risk of developing pressure sores was generally high. However, 202% of the sample did not use a risk assessment tool and 214% reported that they did not attempt to prevent or treat nutritional deciency for every patient at risk. Again, similar to the results in relation to knowledge of preventive strategies, many nurses did not appear to be able to differentiate between methods that were generally applicable
291

Knowledge of preventive strategies

The questions contained in this section were derived from the questionnaire developed by Halfens and Eggink (1995). The latter is based on guidelines on preventive methods contained in the Dutch Consensus Report (1992). It should be noted that the Dutch guidelines are largely consistent with those produced by the USA [Agency for Health Care Policy Research (AHCPR) 1992], the most commonly used international guidelines. The Dutch guidelines grade preventive

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

K. Panagiotopoulou and S.M. Kerr Table 6 Nurses knowledge of preventive strategies (correct answers are indicated in bold) Always useful (%) Methods judged as always useful Provide clean, smooth and dry bottom sheet (n 118) Maintain good hygiene (n 118) Prevent or treat nutritional deciency (n 117) Palpate and inspect the skin daily (n 117) Prevent maceration (softening of the skin) (n 116) Involve patient in prevention(n 116) Reposition at least once every 3 hours (n 118) Assess decubitus risk with a risk assessment tool (n 116) Methods judged as sometimes useful Use foam mattresses and/or pillows (n 113) Use water mattresses and/or pillows (n 115) Use air mattresses and/or pillows (n 118) Use gel mattresses and/or pillows (n 113) Pressure relieving material with adhesive to stabilize heels or elbows (n 117) Cream with topical medication to prevent or treat dry skin (n 116) Paramedical treatment (e.g. physiotherapy) for mobility and reactivity (n 116) Involve family and friends in prevention (n 114) Methods judged as never useful Use donut (n 115) Insert catheter (to prevent maceration due to incontinence) (n 118) Massage (n 116) Cream with topical medication to prevent blood ow disturbance caused by pressure (n 118) Sometimes useful (%) Never useful (%) Dont know (%)

983 941 863 735 974 690 932 784 301 278 675 115 350 784 707 658 217 492 569 415

17 51 137 248 26 302 59 190 345 409 205 221 530 216 284 272 383 458 353 314

00 08 00 17 00 09 08 09 212 70 34 124 85 0 0 61 391 42 69 93

00 00 00 00 00 00 00 17 142 243 85 540 34 0 09 09 09 08 09 178

and methods only useful in individual cases. This was particularly true of use of topical medications (353%) and air mattresses (328%). Regarding methods considered never useful, just less than a third (314%) reported that they always used massage, a strategy considered to indicate poor practice in pressure sore prevention. Practice score (preventive practice) In order to judge nurses level of practice a score was calculated following the scoring guidelines produced by Halfens and Eggink (1995). A score of 1 was awarded for practice that was in accordance with the guidelines and 0 was awarded when practice did not accord with what is recommended. As there were 20 response-options, possible scores ranged from 0 (lowest possible score) to 20 (highest possible score). Following this process the average percentage agreement with expert opinion was 545% with a range of 2580%.

inuence levels of knowledge. The results demonstrated that there was a statistically signicant difference in relation to qualications, with registered nurses exhibiting greater knowledge and reporting better preventive strategies than enrolled nurses (knowledge RN mean 105; EN mean 96; t 230; P 002; practice RN mean 114; EN mean 105, t 2163; P 0033). A more detailed analysis using the chi-squared test highlighted that a signicantly greater proportion of enrolled nurses were not aware that donuts and topical creams should not be used (v2 1406, d.f. 3, P 0003; v2 94, d.f. 3, P 0024, respectively). This lack of knowledge was reected in their practice as enrolled nurses were more likely to report that they used donuts and topical creams than registered nurses (v2 894, d.f. 2, P 0011; v2 1210, d.f. 2, P 0002, respectively). Overall, the reported preventive practice of registered nurses was signicantly better than that of their enrolled nurse colleagues (t 2163, P 0033).

Factors that may have inuenced knowledge and practice

One of the reasons that the sample characteristics and information on continuing professional development (CPD) were collected was to allow exploration of whether factors such as age, gender and in-service training appeared to
292

Perceived barriers to good practice

The results for barriers to good practice are summarized in Table 8. Nurses were asked to select barriers that they considered applied personally or in their work environment

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

Issues and innovations in nursing practice Table 7 Preventive practice (correct answers are indicated in bold) Always applied (%) Methods that should always be applied Provide clean, smooth and dry undersheet (n 118) Maintain good hygiene (n 118) Prevent or treat nutritional deciency (n 117) Palpate and inspect the skin daily (n 117) Prevent maceration (softening of the skin) (n 118) Involve patient in prevention (n 118) Reposition at least once every 3 hours (n 118) Assess ulcer risk with a risk assessment tool (n 114) Methods that should be applied in individual cases Use foam mattresses and/or pillows (n 108) Use water mattresses and/or pillows (n 109) Use air mattresses and/or pillows (n 116) Use gel mattresses and/or pillows (n 105) Pressure relieving material with adhesive to stabilize heels or elbows (n 116) Cream with topical medication to prevent or treat dry skin (n 116) Paramedical treatment (e.g. physiotherapy) for mobility and reactivity (n 117) Involve family and friends in prevention (n 118) Methods that are never useful Use donut (n 118) Insert catheter (to prevent maceration due to incontinence) (n 118) Massage (n 118) Cream with topical medication to prevent blood ow disturbance caused by pressure (n 117)

Pressure area care in Greece

Sometimes applied (%)

Never applied (%)

1000 966 786 778 975 720 847 772 269 119 517 67 216 629 547 610 153 483 314 222

00 34 214 214 25 271 153 202 500 284 328 248 586 353 444 373 390 492 559 479

00 00 00 09 00 08 00 26 231 596 155 686 198 17 09 17 458 25 127 299

from a list of 17 items (the most commonly cited barriers only are tabulated). The most frequently identied barriers were as follows: of lack of staff/manpower (949%), lack of equipment (788%) and overcrowding in the ward (common in Greek hospitals) (791%). More than half of the respondents (60%) reported that research ndings were not user friendly, 462% that they could not access relevant literature and 313% that they lacked knowledge of relevant literature. Finally, chi-squared analyses demonstrated that there was no statistical difference in barriers to good practice reported by registered and enrolled nurses.

Discussion
Nurses knowledge
The study sought to explore Greek nurses knowledge of risk factors, areas at risk and preventive strategies. In relation to knowledge of risk factors, the average level of agreement with expert opinion was 71%. Previous studies have suggested that a level of knowledge (i.e. agreement with expert opinion) greater than 70% is good (McLay 1998). Results

in this study therefore concur with previous studies that found nurses knowledge of risk factors to be good (Bostrom & Kenneth 1992, Maylor 1999). The level of knowledge obtained by this sample of Greek nurses was greater than that demonstrated by those who participated in Maylors UK study (574%). Nurses level of knowledge of areas at risk was also good, with the average level of agreement with expert opinion being 705%. This data adds valuable information to the knowledge base, as we could identify no previous studies exploring nurses knowledge of areas at risk in a comprehensive manner. As discussed previously, the information gathered in the study by Parker et al. (1998) was very limited. In relation to overall knowledge of preventive strategies, nurses level of agreement with expert opinion was only 50%. To some extent this could be attributed to the fact that, while there is a wealth of research-based evidence in relation to risk factors and areas at risk (AHCPR 1992), the evidencebase for preventive strategies is more limited (AHCPR 1992, NHS CRD 1995). As the instrument used to assess knowledge of preventive strategies was adapted from the one developed by Halfens and Eggink (1995), the Greek nurses knowledge was compared to the level exhibited by
293

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

K. Panagiotopoulou and S.M. Kerr Table 8 Barriers to Good practice Agree/strongly agree (%) 949 791 788 695 635 600 462 313 260 209 202

Barriers to good practice Lack of staff/manpower (n 118) Overcrowded ward (n 115) Lack of equipment (n 118) Lack of co-operation with other health professionals (n 118) Lack of disposable material (n 118) Research ndings are not user friendly (n 115) Lack of access to relevant literature (n 117) Lack of knowledge of relevant literature (n 115) Unfamiliarity with an existent risk assessment tool (n 115) Lack of understanding of relevant literature (n 115) Unfamiliarity with an existent pressure sore grading scale (n 114)

methods that are not recommended reected level of knowledge, that is massage, creams and catheterization were used on a regular basis. These results suggest an overall inability to provide tailored, individualized care. Our experience suggests that in the Greek situation, where there are no formal protocols to guide practice in terms of quality and cost-effectiveness, nurses generally err on what they consider is the safe side, often in circumstances where this is not warranted. Comparisons with studies other than that of Halfens and Eggink (1995) are necessarily limited, as different methodological approaches have been used. However, in relation to methods that should not be used, the ndings on misuse of massage have been demonstrated elsewhere (Hill 1992, Wilkes et al. 1996, Beitz et al. 1999).

Factors that may have inuenced knowledge and practice

nurses in Holland. Dutch nurses average level of knowledge was less than that of Greek nurses in relation to methods that are always useful (77% compared to 86%) and it was considerably greater for methods that are sometimes useful (54% compared to 31%) and never useful (27% compared to 15%). This suggests that Greek nurses had more difculty than Dutch nurses in differentiating between methods that are always useful and those not advised for general application. Further comparisons indicated that the percentage of nurses who were regularly using outdated, ineffective (sometimes detrimental) methods such as massage and donuts was considerably greater in Greece than Holland. In relation to inuences on the nurses knowledge of risk factors and areas at risk, no statistically signicant differences were found for the demographic variables. The number of years qualied did not appear to inuence level of knowledge. A similar nding has been highlighted in previous research (Maylor 1999, Mockridge & Anthony 1999). In relation to continuing professional development, exposure to in-service education on pressure area care did not appear to increase the nurses level of knowledge in this study. This nding contradicts that of others, as Maylor (1999) found that nurses who had been exposed to in-service education on pressure area care demonstrated signicantly greater levels of knowledge. When exploring inuences on knowledge and practice of preventive strategies in this study, there were statistically signicant differences in relation to qualications, with registered nurses exhibiting a greater level of knowledge and practice. Maylor (1999), perhaps not surprisingly, found that registered nurses level of knowledge was greater than that of nursing care assistants.

Nurses perceived practice

Another key aim of the study was to explore nurses reported practice in relation to pressure area care. Their average level of agreement with what is suggested as good preventive practice was 545%. Again, this relatively low score may be attributed to the rather limited evidence-base on the effectiveness of preventive strategies (AHCPR 1992, NHS CRD 1995). Comparisons between these ndings and those obtained in Holland (Halfens & Eggink 1995) were again undertaken. While Greek nurses practice in relation to methods that should always be applied was considerably better (85% compared to 60%), it was worse than that of Dutch nurses for methods that are sometimes useful (40% compared to 44%), and considerably worse for those that should not be applied (22% compared to 40%). Again therefore Greek nurses were not good at differentiating between methods advisable for general application and those only useful in a small number of cases. Practice in relation to
294

Barriers to good practice

The nal aim of the study was to determine whether there were any barriers to good practice in pressure area care. The most commonly cited factors were work-related, and this nding has been reported elsewhere. In particular, it has often been reported that the work environment is unsupportive of good practice. First, the utilization of research ndings is not always supported at managerial levels and, second, stafng levels are often so low that it is difcult to provide the desired level of nursing care (Wilkes et al. 1996, Carroll et al.

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

Issues and innovations in nursing practice

Pressure area care in Greece

1997, Walsh 1997, CSAG 1998, Kajermo et al. 1998). More than three-quarters (79%) of the respondents reported that overcrowding was a constraint on good practice. This is a common problem in all Greek hospitals and our ndings echo 3 those of Plati et al. 1992) in a civilian hospital in Greece. Personal barriers, such as lack of awareness and lack of understanding of the relevant literature, have also been reported by others (Funk et al. 1991, Hill 1992, Russell 1996, Wilkes et al. 1996, Carroll et al. 1997, Walsh 1997, CSAG 1998, Kajermo et al. 1998). Research ndings have often been criticised for not being user friendly (Hunt 1981, CSAG 1998).

Study limitations
The main limitation centres on the use of the self-report questionnaire. As would be expected, there was a level of missing data. In addition, although nurses were asked not to confer with each other and not to consult textbooks when completing the questionnaire, we have no way of knowing if they complied with this request. In addition, of course, we do not know if their reported practice reected their actual practice. Finally, another limitation is arguably the fact that nurses who were considered to have specialist knowledge of pressure area care (i.e. those working in intensive care) were excluded from the study. This was done in order to facilitate comparisons across studies. However we believe that future work should include nurses who are considered to have specialist knowledge and that subgroup comparisons (e.g. between those working in ITU and in general medical/ surgical wards) would be informative. In relation to the generalizability of the ndings, we cannot comment with certainty on the level of knowledge and practice of nonresponders. However, as the response rate was high (71%) and the demographic characteristics (including numbers of enrolled and registered nurses) of the responders and nonresponders were shown not to be signicantly different, there is no reason to assume that the results cannot be generalized to all nurses working at the study site who met the inclusion criteria. In relation to generalization of the results to the target population (i.e. all hospitals in Greece), this cannot be done with any certainty as we consider that the population at the study site (nurses working in a military hospital) is more knowledgeable about pressure area care.

knowledge and practice of pressure area care. Although some practices were questionable, levels of knowledge were generally good. There is an obvious need to improve the research skills of Greek nurses in order to provide them with the appropriate knowledge to understand and implement research ndings. In addition, nurse clinicians should be encouraged to develop clinical indicators to improve the quality of nursing practice for the prevention of pressure sores in the Greek population. This could be done by using clinical experts in the area of pressure sores as well as undertaking audits of patient records. Finally, there is a need for further Greek-based research studies. The rst author (who is a Major in the Greek Army) intends to replicate this study in other military hospitals in Greece (n 3) and will encourage colleagues working in civilian hospitals, where knowledge is likely to be more limited, to do likewise. In addition, there is a need for observational studies to determine actual rather than perceived practice.

References
Agency for Health Care Policy Research (1992) Clinical Practice Guidelines: Pressure Ulcers in Adults: Prediction and Prevention. Number 3, Rockville MD, AHCPR, Publication no. 920047. United States Center for Research and Dissemination, Rockville, MD, USA. Baltzi-Economopoulou D.E. (1997) Assessment of the Cost of Pressure Sore Treatment. PhD Thesis Department of Nursing Studies, University of Athens, Greece. Beitz J.M., Fey J. & OBrein D. (1999) Perceived need for education vs. actual knowledge of pressure ulcer care in a hospital setting nursing staff. Dermatology Nursing 11, 125136, 138. Bostrom J. & Kenneth H. (1992) Staff nurses knowledge and perceptions about prevention of pressure sores. Dermatology Nursing 4, 365368. Braden B.J., Bergstrom N., Laguzza A. & Holman V. (1987) The Braden scale for predicting pressure sore risk. Nursing Research 36, 205210. Carroll D.L., Greenwood R., Lynch K.E., Sullivan J.K., Ready C.H. & Fitzmaurice J.B. (1997) Barriers and facilitators to the utilisation of nursing research. Clinical Nurse Specialist 11, 207212. CBO (1992) Herziening consensus decubitus (Revision of the pressure sore consensus). Centraal Begeleidingsorgaan voor de Intercollegiale Toetsing, Utrecht. Clark M. & Watts S. (1994) The incidence of pressure sores within a National Health Service Trust hospital during 1991. Journal of Advanced Nursing 20, 3336. Clinical Standards Advisory Group (1998) Report on Clinical Effectiveness Using Stroke Care as an Example. The Stationery Ofce, CSAG, London. Department of Health (1992) The Health of the Nation: a Consultative Document for Health in England. HMSO, London. Funk S., Champagne M., Wiese R. & Tournquist E. (1991) Barriers: the barriers to research utilisation scale. Applied Nursing Research 4, 9095. 295

Conclusions and recommendations for practice

Despite the limitations, it is considered that this study takes a valuable rst step in providing information on Greek nurses

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296

K. Panagiotopoulou and S.M. Kerr Halfens R.J.G. & Eggink M. (1995) Knowledge, beliefs and use of nursing methods in preventing pressure sores in Dutch Hospitals. International Journal of Nursing Studies 32, 1626. Hayes P., Wolf Z. & McHugh M. (1995) Effect of teaching plan on a nursing staffs knowledge of pressure risk, assessment, and treatment. Journal of Nursing Staff Development 10, 207213. Hill L. (1992) The question of pressure: how much do nurses know about preventing pressure sores and do they put that they know into practice? Journal of Wound Care 88, 7682. Hunt J. (1981) Indicators for nursing practice: the use of research findings. Journal of Advanced Nursing 6, 189194. Kajermo K.N., Nordstrom G., Krusebrant A. & Bjorvell H. (1998) Barriers to and facilitators of research utilisation, as perceived by a group of registered nurses in Sweden. Journal of Advanced Nursing 27, 798807. Land L. (1995) A review of pressure damage prevention strategies. Journal of Advanced Nursing 22, 329337. Maylor M. (1999) Pressure sore survey Part 2: nurses knowledge. Journal of Wound Care 8, 4952. McLay L. (1998) To Determine the Level of Knowledge that Registered Nurses have of Diabetes Mellitus in a Teaching Hospital. Master Dissertation. University of Glasgow, Nursing & Midwifery School, Glasgow. Mockridge J. & Anthony D. (1999) Nurses knowledge about pressure sore treatment and healing. Nursing Standard 13, 66, 69 66, 70, 72. National Health Service Centre for Reviews and Dissemination (NHS CRD) (1995) The prevention and treatment of pressure sores. 4 Effective Health Care, NHS CRD, York. Norton D., McLaren R. & Exton-Smith A.N. (1962) An Investigation of Geriatric Nursing Problems. London Hospital National Corporation for the care of old people. London Hosptal National Corporation, London. Oot-Giromini B.A. (1993) Pressure ulcer prevalence, incidence and associated risk factors in the community. Decubitus 6, 2432. Parker K., Morgan L., Clayton J., Gerrish K. & Nolan M. (1998) Knowledge and practice in pressure area care. Professional Nurse 11, 301305. Plati C., Lanara B., Katostaras F., Portokalaki A. & Brokalaki H. (1992) Contributing factors to development and severity of pressure sores. (Greek) Nursing Times 54, 3641. Pressure Ulcer Prevention Consensus Group (2000) Pressure Ulcer Prevention Guidelines. PUPCG, Glasgow. Russell L. (1996) Knowledge and practice in pressure area care. Professional Nurse 11, 301306. Tolmie L. (2000) Pressure Sores: an Investigation into the Clinical Nursing Management of the Prevention and Management of Pressure Sores within an Acute Hospital Trust. MSc Thesis. University of Glasgow, Nursing & Midwifery School, Glasgow. Walsh M. (1997) How nurses perceive barriers to research implementation. Nursing Standard 11, 3439. Waterlow J. (1985) A risk assessment card. Nursing Times 27, 4955. Wilkes M.L., Bostock C., Lovitt L. & Dennis G. (1996) Nurses knowledge of pressure ulcer management in elderly people. British Journal of Nursing 5, 858864.

296

2002 Blackwell Science Ltd, Journal of Advanced Nursing, 40(3), 285296