Risk Assessment Tool For The Predication Of Pressure Ulcer

http://doi.org/10.33698/NRF0139 –  Ashok Kumar, Rajinder Mahal,Triveni Rajappa

Abstract : A useful instrument for the prediction of pressure ulcer requires a highly sensitive with specificity, good predictive value and easy to use in the clinical practice. Objective of the study is to construct risk assessment tool for the prediction of pressure ulcer in patients. Methodological approach was used and tool was constructed and validated on selected patients in selected areas of C.M.C. Hospital, Ludhiana. From related review of literature, experts’ guidance and investigators’ personal experience, a list of risk factors were identified. These identified risk factors were compared with standardized risk assessment scales i.e. Braden Scale, Norton Scale and RAPS (Risk Assessment Pressure Sore) Scale. Content validity was ascertained by expert’s opinion. The final draft of tool consists 15 items. The range of rating score was 15 to 54, with the assumption that higher the scores, the greater the risk of pressure ulcer development. At a cut-off point of ³ 35, the best balance between the sensitivity and specificity was achieved, i.e. sensitivity was 84.61%, specificity 85.11%, predictive value positive 61.11% and predictive value negative 95.24%. Cronbach’s alpha and item- item correlation were used to measure the internal consistency of the tool, Coefficient alpha was 0.84, statistically significant at 0.001 level. Item-item correlations were moderate (between 0.30 to 0.70) among most of the items. In order to estimate equivalence Inter-rater reliability method was used. Correlation coefficient was 0.92 (p<001). Percentage of agreement between raters was 82%.

Key words :

Risk Assessment Tool, Pressure Ulcer

Correspondence at : Ashok Kumar

Lecturer,

MM College of Nursing, Mullana, Ambala

Introduction

A sound skin makes you feel good and look good. Healthy skin is often an indicator of our holistic wellness. Maintenance of glowing and healthy skin needs good personal hygiene, unpolluted environment, correct eating habits, mental peace and happiness. Thus a healthy skin is the result of holistic wellness of body, mind and spirit. 1 Pressure ulcers are also called decubitus ulcers, pressure sores or bed sores. A pressure ulcer is any lesion caused by unrelieved pressure (a compressing downward force on a body area) results in damage to underlying tissue, as defined by the U.S. Public Health Service’s Panel for the Prediction and Prevention of Pressure Ulcers in Adults.2

Studies carried out in various care settings show a prevalence of pressure sores among inpatients ranging between 3.75% and 42%. In National study found an overall prevalence rate 6.7 in hospital patients with pressure sores. Anderson (1991) sampled hospital and community patients in Clwyd South Hospital District and found a prevalence of 17.4% in Hospital patients and 4.9% in Community patients. The incidence (number of new cases appearing over a specified time period) and prevalence (a cross sectional count of the number of cases at a specific point in time) of pressure ulcer is sufficiently high to warrant concern about this problem in acute care, long- term care, and home care settings. The prevalence of pressure ulcers in 148 acute care facilities was reported by Meehan M.(1990) to be less i.e. 9.2%. The incidences in skilled care and nursing home type facilities have been reported to be 23% and 27.5% respectively. 3,4,5 Risk assessment is recommended as the first step in the prevention of pressure sore development in nursing care. Assessment has to be performed on admission, and must be reassessed whenever there is a significant change in the patient’s condition. A risk assessment scale will help the nurses to make a systematic assessment of the patient’s condition and risk of sore development. This can become an impor tant foundation for prevention and quality assurance. The main benefit of using a risk assessment scale may lie in its acting as a reminder to nurses of the possibility of pressure ulcer development. A simple reminder would be easier to use and less time consuming .6

It is largely agreed that prevention plans should include mechanisms for predicting which patients are most likely to develop pressure sores and interventions should be directed to the vulnerable. Risk assessment tools are used to identify those patients at risk and are based on those factors which have been identified as significant in the development of pressure sores. A multiplicity of intrinsic and extrinsic factors is thought to be associated with increased risk. The development of risk calculators which incorporates these factors has been an attempt to identify which subjects are most at risk so that equipment and nursing interventions can be allocated appropriately. 7

Sensitivity and specificity are the most commonly used and recommended statistics for evaluating the predictive validity of pressure ulcer risk assessment scale, assuming that a good and useful scale should have both high sensitivity and high specificity. The validity of a risk assessment scale is the degree to which the risk is correctly predicted. Defloor and Grypdonckseem to be more concerned about sensitivity than specificity. Correctly identifying all patients really at risk of developing pressure ulcers is thought to be more important than wrongly identifying patients at risk and needlessly giving preventive care. Most of the scales reported in the literature have not been subjected to testing the sensitivity, specificity or predictive value. Dissatisfaction with the ability of the Norton scores to predict patients at risk as the basis for the development of the Dougles scale, but does not repor t any measures of sensitivity or specificity to substantiate claims that this later score is more valid.8

Investigators’ experience in critical patient care areas of CMC Hospital, Ludhiana, Punjab, where most of the patients are unconscious and bedridden and they are most vulnerable to get pressure ulcers. It was observed that when a pressure ulcer develops hospital costs and duration of hospitalization increases multifold. There was absence of standard tool to identify risk factor for developing pressure ulcer. Hence the investigator got insight to develop risk assessment tool for the prediction of pressure ulcer.

Objectives

  • To construct risk assessment tool for the prediction of pressure ulcer in
  • To test validity and reliability of risk assessment tool for the prediction of pressure ulcer in

Delimitations: The study was delimited to patients with intact skin, patients above the age of 18 years and staying in hospital more than 6 days.

Methodology

A methodological approach was used for the development of Risk Assessment Tool for the prediction of pressure ulcer in patients. The study was conducted in selected areas of C.M.C. Hospital, Ludhiana.

Phase – I

Construction of Risk Assessment Tool

Risk factors identification – From related review of literature, experts’ guidance and investigators’ personal experience, a list of risk factors were identified. These identified risk factors were compared with standardized risk assessment scales i.e. Braden Scale, Norton Scale and RAPS (Risk Assessment Pressure Sore) Scale. Then a set of items and sub items were drawn and framed as a format for the risk assessment tool for prediction of pressure ulcer. The items included in the tool were: Age, Gender, General physical condition, mental status, skin type, Bodily constitution, Activity, Mobility, Nutritional status, Moisture, Body temperature, Sensory perception, Friction and shear.

The risk factors of pressure ulcers were systematically organized as items and sub items, which were further arranged in logical order so important observations can be easily identified and summarized.

Phase – II

Content Validity of the tool

It was done by experts’ opinion. The risk assessment tool was circulated among the seven experts i.e. from different nursing specialties, medicine and surgery to acquire suggestions by keeping in mind relevance of item to the subject of study; items are meaningful and easily understandable, items measures the risk factors causing pressure ulcer and items are observable in hospital setting.

After 1st round of exper ts opinion necessary changes made and tool were further modified as follows:-

  • In risk factor age, the categories increased from two to
  • Since the term mental status is very broad so it was limited to level of
  • The item ‘body constitution’ was not giving clear meaning so it was changed to
  • The item ‘basic nursing care’ was included since it affects the development of pressure

After modification and necessary changes the tool again submitted to same experts for final validation. Followed by 2nd round of experts suggestion the changes made in tool are:

  • The item nutritional status fur ther subdivided in food intake pattern and
  • Written key was also framed to ascertain the objectivity in scoring the domains, in respect of each

The final draft of risk assessment tool for the prediction of pressure ulcer consists of 15 items (risk factors) i.e. age, gender, general physical condition, level of consciousness, skin type, physique, activity, mobility, nutritional status (food intake pattern and hemoglobin), moisture, body temperature, sensory perception, friction & shear, and basic nursing care.

In risk assessment tool all the items were categorized and rated between 1 and 4 except age, and friction & shear rated between 1 and 3 and gender and basic nursing care rated between 1 and 2. The minimum rating score is 15 and maximum is 54, with the assumption that higher the scores, the greater the risk of pressure ulcer development

Phase – III Pilot study

Pilot Study I : To test the language clarity, relevance and time required for administration, the tool was administered on five patients admitted in male medical ward of C.M.C & Hospital Ludhiana. The chosen subjects were similar in characteristics to those of population under study. The tool was given to staff nurses for scoring. The items were clear and unambiguous and no clarification was required for scoring of items except general physical condition and time required to complete scoring of the tool was 15 – 20 minutes. The written key was modified for categorization and scoring of general physical condition

Pilot study II : It was conducted to ensure feasibility of the study. The data was collected by observing 10 patients from male medical ward of CMC & Hospital and scoring was done on risk assessment tool for prediction of pressure ulcer.

To test the reliability of the tool inter rater reliability method was carried out, correlation coefficient was 0.97 and percentage of agreement between raters was 80%, hence the tool was reliable.

Phase – IV

Try out of the tool

The study was conducted in selected areas of C.M.C. Hospital, Ludhiana i.e. Medical wards, Surgical wards, Orthopedic ward, and Special units (ICU, Neurosurgery) of CMC Hospital, Ludhiana. Patients were selected by using purposive sampling technique (patients with intact skin; patients above the age of 18 years; and staying in hospital for more than 6 days) and were informed that participation in the study is voluntary and were guaranteed that data would be treated anonymously and in aggregated form (for unconscious patients, consent were obtained from significant relatives). Using purposive sampling technique 100 Patients were enrolled in the study after verbal consent allowing both collection of information on their health status and physical examination to assess presence and stage of pressure ulcers. No prescribed treatments for the patients were withheld. Finally 60 patients could be retained as 40 patients have been discharged before final observation.

Prior to data collection formal permission was obtained from the authorities of the hospital. Then the investigator selected the samples by purposive sampling technique. First observation was made within 24 hours of admission, and second observation made within next 24 hours and scoring was done on the tool, the mean score of these two observations considered as risk assessment score. Then third and final observation was made after 6th day of admission, to evaluate the skin of client for grading and checking sites for pressure ulcer development.

Predictive Validity

The predictive validity at each score was calculated by sensitivity, specificity, predictive value positive (PVP) and predictive value negative (PVN) tests (Table 1).

Table 1 reveals sensitivity, specificity, predictive value positive and predictive value negative tests of the risk assessment tool for prediction of pressure ulcer, and these were calculated at each score from 27 to 47. Sensitivity were ranging from 30.76% to 100%, specificity from 36.17% to 100%, predictive value positive from 30.23% to 100% and predictive value negative from 83.93% to 100%. At a cut-off point of > 35, the best balance between the sensitivity and specificity was achieved, i.e. sensitivity was 84.61%, specificity 85.11%, predictive value positive 61.11% and predictive value negative 95.24%.

Since cut -off point of the tool was 35, patients who are having a score > 35, are at risk for development of pressure ulcer.

Reliability

Analysis of the reliability of the risk assessment tool for the prediction of the pressure ulcer consists of internal consistency and equivalence.

Table -1 : Sensitivity, Specificity, Predictive Value Positive (PVP) and Predictive Value Negative (PVN) tests.

Cut -off point (Total Score) Sensitivity (%) Specificity (%) PVP (%) PVN (%)
> 27 100 36.17 30.23 100
>28 100 46.81 34.21 100
>29 100 48.94 35.13 100
>30 100 57.45 39.39 100
>31 100 65.96 44.83 100
>32 92.31 68.08 44.44 96.97
>33 84.61 80.85 55.00 95.00
>34 84.61 82.99 57.89 95.12
>35 84.61 85.11 61.11 95.24
>36 84.61 87.23 64.70 95.35
>37 84.61 91.49 73.33 95.55
>38 84.61 100 100 95.92
>39 61.54 100 100 90.38
>40 38.46 100 100 85.45
>41 30.76 100 100 83.93
Maximum Score = 54 Minimum Score = 15      

Internal Consistency

Coefficient alpha and item-item correlation (Table 2) were used to measure the internal consistency of the risk assessment tool for the prediction of pressure ulcer. Coefficient alpha was 0.84, which was statistically significant at 0.001 level.

Table 2 reveals correlations between items on the risk assessment tool. For age there was moderate correlation with other items (ranged from 0.33 to 0.51) and were significant at 0.01 level except with general physical condition (0.13) level of consciousness (0.14), skin type (0.18) and physique (0.14) were having weak correlation. For gender there was a moderate correlation with other items except skin type (0.12), food intake pattern (0.25) and body temperature (0.26). For general physical condition there were moderate correlations with all the other items.

Table – 2 : Spearman Rank Correlation Coefficient between Items on the Risk Assessment Tool for the Prediction of Pressure Ulcer

AgeItems                              1        2        3         4         5         6       7       8        9        1 0     1 1     1 2     1 3 1 4 1 5

  1. Gender 35**
  2. Generalphysical condition 0.13 67**
  3. Levelof consciousness 14 0.49** 0.49**
  4. Skin type 0.18 0.58** 0.42** 0.40**
  5. Physique 0.14 0.12 0.35** 0.42** 0.49**
  6. Activity 0.51** 0.33** 0.61** 0.26* 0.52** 0.39**
  7. Mobility 0.46** 0.67** 0.63** 0.34** 0.28* 0.31* 0.10
  8. Food intake pattern 0.37** 0.25* 0.34** 0.24 0.37** 0.32**0.41* *0.39**
  9. Haemoglobin 0.33** 0.39** 0.51** 0.55** 0.30* 0.28* 0.43**0.49* *0.55**
  10. Moisture 0.39** 0.36** 0.49** 0.53** 0.31* 0.47**0.32**0.50* *0.30*0.32**
  11. Body temperature 0.45** 0.26* 0.39** 0.30* 0.29* 0.22 0.28* 0.30* 0.29* 0.33**0.35**
  12. Sensory perception 0.35** 0.41** 0.35** 0.39** 0.53** 0.64**0.30* 0.39* *0.48**0.54**0.32**0.30*
  13. Friction & Shear 0.42** 0.62** 0.48** 0.36** 0.39** 0.70**0.39**0.44* *0.38**0.51**0.50**0.47**0.46**
  1. *p<0.05; **p<0.01
    Basic nursing care 0.37** 0.49** 0.60** 0.28* 0.32** 0.34**0.39**0.41* *0.34**0.53**0.40**0.38**0.31* 0.30*

For level of consciousness, moderate correlation was achieved with all the items except activity (0.26), food intake pattern (0.24) and basic nursing care (0.28). Skin type was having moderate correlation with all the items. Physique was having moderate correlation with other items except food intake pattern (0.28) and body temperature (0.22). For activity moderate correlation was achieved except with mobility (0.10) and body temperature (0.28) having weak correlation. Mobility, food intake pattern, haemoglobin, moisture, body temperature, sensory perception, friction and shear and basic nursing care were having moderate correlations with all the items.

The item – item correlation should be moderate, between 0.30 to 0.70, so for present tool item-item correlations were moderate among most of the items. Hence it can be inferred that the tool was having good internal consistency, so it was reliable.

Equivalence

In order to estimate equivalence Inter- rater reliability method was used. Two investigators had rated the score on risk assessment tool independently. Correlation coefficient was 0.92 (p<001). Percentage of agreement between raters was 82%.

Results

The risk assessment tool for the prediction of pressure ulcer was developed in this study. The final draft of the tool consists 15 items i.e. age, gender, general physical condition, level of consciousness, skin type, physique, activity, mobility, nutritional status (Food intake pattern & Hemoglobin), moisture, body temperature, sensory perception, friction & shear, and basic nursing care. The items in the tool were categorized and rated between 1 and 4 except age, and friction & shear rated between 1 and 3 and gender and basic nursing care rated between 1 and 2. The minimum rating score is 15 and maximum is 54, with the assumption that higher the scores, the greater the risk of pressure ulcer development.

At a cut –off point of > 35, the best balance between sensitivity and specificity i.e. 84.61% and 85.11% respectively were achieved. Tool was reliable i.e. coefficient alpha was 0.84, item – item correlation was moderate for most of the items, correlation coefficient was 0.92 and percentage agreement between raters was 82%.

Discussion

Pressure sore development constitutes a major problem, which causes excessive pain and suffering in affected patients. Problems with pressure sores are also associated with significant costs for society. Identification of patients at risk for pressure sore development is perhaps the most important issue in pressure sore prevention. The investigator developed a risk assessment tool for the prediction of pressure ulcer in patients, which is composed of 15 variables including age, gender,  general physical condition, level of consciousness, skin type, bodily constitution, activity, mobility, nutritional status, moisture, body temperature, sensory perception, and friction & shear, and basic nursing care. Similarly Nor ton et al (1970)9 presented a risk assessment scale ( Norton Scale) for prediction of pressure sore development among elderly patients and included five variables i.e. general physical condition, mental status, activity, mobility and incontinence. The Braden Scale developed by Braden and Bergstrom (1987)10, the scale is composed of six subscales that reflect sensory perception, skin moisture, activity, mobility, friction and shear and nutritional status. Lindgren M. et al (2002) 11 also developed a risk assessment pressure sore (RAPS) scale, includes 12 variables i.e. general physical condition, activity, mobility, food intake, fluid intake, moisture, sensory perception, friction & shear, skin type, bodily constitution, body temperature and serum albumin.

Reliability of risk assessment tool consists of internal consistency (coefficient alpha & Item-item correlation) and equivalence (correlation coefficient & percentage agreement). Coefficient alpha was 0.84 (p<0.001) and item-item correlation was moderate (0.30-0.70) except age, gender level of consciousness and physique were weakly correlated (< 0.30) with cer tain items. Correlation coefficient was 0.92 (p<0.001) and percentage agreement between raters was 82%. These findings are in accordance to Lindgren M. et al (2002)11 developed RAPS scale, coefficient alpha was 0.80, corrected item-item correlation was above 0.30, correlation coefficient was 0.83 and percentage of agreement among nurses was 70%.

Predictive validity at each score was calculated by sensitivity, specificity, predictive value positive and predictive value negative tests. At a cut-off point of >35, the best balance between the sensitivity and specificity was achieved for patients admitted in CMC hospital, Ludhiana, i.e. sensitivity was 84.61%, specificity 85.11%, predictive value positive 61.11% and predictive value negative 95.24%. Similarly Lindgren M. et al (2002)11 presented predictive validity of RAPS, at a cut of point of 31 the best balance was achieved for medical patients i.e. sensitivity was 75%, specificity 70%, predictive value positive 19.2% predictive value negative 96.7%. Predictive validity of the Norton scale has been examined in different settings, a sensitivity ranging from 63% to 100%, specificity from 26% to 89%, predictive value positive from 9% to 70% and predictive value negative from 35% to 93% have been reported (Goldstone & Goldstone, 1982; Dealey, 1989; Vardman C, 1991). 12,13,14

The predictive validity of the Braden Scale has been examined in different studies with various populations, in these studies sensitivity range between 38% & 100%, specificity between 60- 92%, predictive value positive and predictive value negative between 54% & 90% ( Barnes & Payton, 1993; Vanden et al, 1996; Halfen et al. 2000).15,16,17

The nursing administrators can make effort to involve the risk assessment tool with the treatment chart of the patients. Through in-service education programme all bedside nurses can be trained to use risk assessment for all the in-patients. The tool can be used in different clinical areas to assess the at risk patient and thereby helps in prevention of pressure ulcers. It helps as a reminder for nurses of the possibility of development of pressure ulcers so early preventive measures can be used for those who are at risk. It would make it possible to identify those patients who really need immediate preventive measures. Risk assessment tool helps to reduce cost and duration of hospitalization of critically ill patients. The present study would generate scientific literature for trained as well as student nurses and serve a baseline in developing more valid risk assessment tool for the prediction of pressure ulcer in different settings. The study can be utilized in clinical to assess prevalence of pressure ulcer in various units. Based on the present study more risk factors of pressure ulcer can be identified.

Hence it is recommended that tool can be implemented on a large sample and for a longer duration. Feasibility of tool can be assessed at different settings for calculating sensitivity and specificity to establish appropriate cut-off points. Comparative studies can be conducted to establish predictive validity by using present risk assessment tool versus standardized tool (Norton Scale, Braden Scale etc). Standards or protocols on preventive measures for pressure ulcer among the high risk patients can be developed based on the identified risk factors. Present risk assessment tool can be modified by identifying some more risk factors.

References

  1. Jose Healthy Skin – Indicator of Holistic Wellness. Health Action 2001; 14:2-3.
  2. Kozier B, Erb G, Berman A, Snyder Fundamentals of Nursing, 7th Edition, Singapore: Pearson Education 2004; 856-9.
  3. David JA, Chapman RG, Chapman FJ, Locket An Investigation of the Current Methods used in Nursing for the Care of Patients with Established Pressure Sores. Nursing Practice Research Unit. University of Surrey Guildford, 1983.
  4. Anderson An investigation into causation, prevention and management of pressure ulcers in hospital and community patients. Macmillan Magazines 1991;35-41.
  5. Meehan Multisite pressure ulcer prevalence surgery. Decubitus 1990; 3: 14-7.
  6. Ek AC, Nordstrom G, Lindgreen Quality indicators for patients at risk for pressure sore development. Quality indicators in Nursing Care 2001; 9:11-27.
  7. Edwards The rationale for the use of risk calculators in pressure sore prevention, and the evidence of the reliability and validity of published scales. Journal of Advanced Nursing 1994; 20: 288-96.
  8. Pritchard Calculating the risk. Nursing Times 1986; 82(8): 59-61.
  9. Norton D, Mcharen R, Exton – Smith An Investigation of Geriatric Poblems in Hospital, 3rd Edition. London: Churchill Livingstone 1979.
  10. ergstrom N, Braden B, Languzza A, Holman The Braden scale for predicting pressure sore risk. Nursing Research 1987; 36 (4): 205-10.
  11. Lindgren M, Unosson M, Krantz AM, Ek AC. A risk assessment scale for prediction of pressure sore development: reliability and Journal of Advanced Nursing 2002; 38 (2): 190-9.
  12. Goldstone CA, Goldstone The Norton Score: An early warning of pressure sores. Journal of Advanced Nursing 1982; 7: 419-26.
  13. Risk assessment of pressure sores: a comparative study of Norton &Waterlow scores. Nursing Standard 1989; 3:11-2.
  14. Vardman C, Norton V Nursing Times 1991; 87: 74-8.
  15. Barnes D,  Payton Clinical application of the Braden Scale in the acute care setting. Dermatology Nursing 1993; 5: 386-8.
  16. Vanden BT, Montoye C, Satwicz M, Durkee LK, Boylan Predictive validity of Barden Scale and Nurse’s perception in identifying pressure ulcer risk. Applied Nursing Research 1996; 9: 80-6.
  17. Halfen RJG, Van AT, Bal RH. Validity & Reliability of the Braden Scale and the influence of risk factors: a multi-centre prospective Intervention Journal of Nursing Studies 2000; 37:313-9.