Comparison of risk assessment scales for pressure ulcers in critically ill patients

Abstracts

OBJETIVO: Comparar as escalas de risco para úlcera por pressão de Norton, Braden e Waterlow entre pacientes em estado crítico. MÉTODOS: Estudo exploratório e longitudial abrangendo 42 pacientesque foram avaliados, por 15 dias ou pelo menos 10 dias consecutivos, em três Unidades de Terapia Intensiva de Fortaleza-Brasil, no período de março a julho de 2009. Cada paciente foi avaliado, simultaneamente, por três enfermeiros, sendo cada um responsável pela aplicação de apenas uma das escalas. RESULTADOS: Houve uma incidência de 59,5% de lesões e um aumento na pontuação das escalas de Norton (p=0,028) e Braden (p=0,004), entre os 1º-15º dias, e de Waterlow (p=0,005) entre os 1º-10º. Quando comparadas a Norton e Braden, os escores de Waterlow aumentaram constantemente (p<0,001). Os pacientes com alto e altíssimo risco, 2% e 92%, respectivamente, desenvolveram lesões (p=0,005). CONCLUSÃO: A escala de Waterlow apresentou maiores escores na avaliação do risco para úlcera por pressão em relação às escalas de Norton e Braden.

Úlcera por pressão; Medição de risco; Escalas

ORIGINAL ARTICLE

Comparison of risk assessment scales for pressure ulcers in critically ill patients*

Comparación de escalas de evaluación de riesgo para úlcera por presión en pacientes en estado crítico

Thiago Moura de Araújo^I; Márcio Flávio Moura de Araújo^I; Joselany Áfio Caetano^II

^INursing MSc. Professor at the Nursing Graduation Course at Universidade Federal do Maranhão – UFMA – Imperatriz (MA), Brazil. Student of the Nursing Post-graduation Program (PhD), Universidade Federal do Ceará – UFC – Fortaleza (CE), Brazil

^IINursing PhD. Professor at the Nursing Post-graduation program, Universidade Federal do Ceará – UFC – Fortaleza (CE), Brazil

Corresponding Author

ABSTRACT

OBJECTIVE: To compare the Norton, Braden and Waterlow scales of risk for pressure ulcer in critically ill patients.

METHODS: An exploratory and longitudinal study covering 42 patients who were evaluated for 15 days or at least 10 consecutive days, in three intensive care units within Fortaleza, Brazil, from March to July 2009. Each patient was evaluated, simultaneously, by three nurses, each of whom was responsible for the application of only one of the scales.

RESULTS: There was a 59.5% incidence of ulcers and an increase in score of Norton (p = 0.028) and Braden (p = 0.004) scales, between the 1^st -15^th day, and the Waterlow (p = 0.005) between the 1^st -10^th. When compared to Norton and Braden, the Waterlow scores steadily increased (p <0.001). Patients with high and very high risk, 2% and 92%, respectively, developed ulcers (p = 0.005).

CONCLUSION: The Waterlow scale showed higher scores in the risk assessment for pressure ulcers in relation to the Norton and Braden scales.

Descriptors: Pressure ulcer; Risk assessment; Scales

RESUMEN

OBJETIVO: Comparar las escalas de riesgo para úlcera por presión de Norton, Braden y Waterlow entre pacientes en estado crítico.

MÉTODOS: Estudio exploratorio y longitudinal que abarcó a 42 pacientes que fueron evaluados, por 15 días o por lo menos 10 días consecutivos, en tres Unidades de Cuidados Intensivos de Fortaleza-Brasil, en el período de marzo a julio del 2009. Cada paciente fue evaluado, simultáneamente, por tres enfermeros, siendo cada uno responsable por la aplicación de apenas una de las escalas.

RESULTADOS: Hubo una incidencia del 59,5% de lesiones y un aumento en la puntuación de las escalas de Norton (p=0,028) y Braden (p=0,004), entre los 1º-15º días, y de Waterlow (p=0,005) entre los 1º-10º. Cuando fueron comparadas la Norton y Braden, los scores de Waterlow aumentaron constantemente (p<0,001). Los pacientes con alto y altísimo riesgo, 2% y 92%, respectivamente, desarrollaron lesiones (p=0,005).

CONCLUSIÓN: La escala de Waterlow presentó mayores scores en la evaluación del riesgo para úlcera por presión en relación a las escalas de Norton y Braden.

Descritores: Úlcera por pressión; Medición de riesgo; Escalas

INTRODUCTION

Pressure ulcers (PU) are understood as skin lesionsdue to bony prominences compressions that are notrelieved, with clinical, social and economic severe and expensive consequences. In the world, billions of dollarsare spent on direct or/and indirect treatment of a totallypreventable health problem. Easily identifiable, PUfinancial costs are constantly mentioned in publicationswhich do not consider the social damages which are,most of the time, irreversible.

Even with the health care services modernization, PU prevalence remains as a problem to be solved,especially among inpatients, because it culminates in alonger hospitalization period, in the necessity ofspecialized professionals and products, and also in theincrease of the infection risk^(1-2).

PU epidemiological prospect in developed countriesis not different from developing ones. In the UnitedStates of America, for example, it presents a prevalenceof 15% and increasing costs of 50% due to patients'hospitalization extension⁽³⁾. In Brazil, some publicationshave detected prevalence that ranges from 27% to 39.4%among patients hospitalized for this problem^(4-5). In Europe, countries like England, Germany, Sweden, Italyand the Netherlands the percentages are of 7.9%, 8.3%,20%, 23% and 24.2%, respectively. In other countries,like Portugal, there are serious problems reporting PU cases⁽⁶⁾. In Asia, in nations like South Korea, PU prevalence ranges from 10.5% to 45.5%⁽⁷⁾.

An early and regular stratification of the risk ofdeveloping a PU is essential to adopt adequate preventivemeasures on an appropriate therapeutic strategyimplementation. It includes methods to reducepredisposing factors and to optimize the patients' generalhealth. But, in most cases, the therapeutic strategy stilldepends on the health care professionals and on costlyinputs provision. Thus, it's notorious the necessity ofstudies which evaluate instruments, techniques andproducts used to prevent PU, which can be adapted toseveral health service scenarios. In this regard, it isremarkable the use of indirect methods like risk assessment scales for PU, because if they present reliablepsychometric indices, they will be economical, practicaland effective^(1-8).

Nowadays, there is an average of 40 risk assessmentscales for PU. And studies about them, on most literature review, have considered experts beliefs or existinginstruments adaptations. However, they don't report thevalue assigned to the risk factors neither the use ofadequate statistics techniques⁽²⁾. In this regard, authorshave created assessment instruments capable ofpredicting the PU occurrence. Among these instruments,the best known are Norton, Waterlow and Braden scales⁽¹¹⁾. These scales have already been evaluatedseparately^{(4, 12-13)}, in pairs⁽¹⁴⁾ and together^{(8, 15-18)}. Only fivepapers were found in which these scales were analyzedtogether: two which were developed in Great Britainand the others in Germany, the Netherlands and China.

In the bibliographical research carried out until thismanuscript's preparation, it was not identified apublication which compares Norton, Braden andWaterlow scales among Brazilian studies.

Due to space limitations on health publications oryet the execution time of several nursing care activitiesat Intensive Care Units (ICU), it is important to guaranteethe nurse access to practical instruments that are capableto predict the risks of developing PU in critically illpatients. Thus, this paper objective is to compare theNorton and Waterlow scales for evaluating PUdevelopment risks in Brazilian critically ill patients.

METHODS

This exploratory and longitudinal research was carriedout at three Intensive Care Units of a Heath Institute that is a reference on urgency and emergence care inBrazilian Northwest, which is located in Fortaleza City,in Brazil. The study was carried out from March to Junein 2009, with all the patients accepted during this time atthe three adult Intensive Care Units of the Institution. Criteria for selecting the study subjects were as follows:being at least 18 years old; not presenting a PU at theadmission moment; being at the ICU for at most 48hours before data collection. On the other hand, the criteria for excluding subjects from the study were asfollow: being a hemodynamically instable patient; havinga brain death diagnosis; and having a prognosis of ICUdischarge in less than 15 days.

At the researched institution, there are 31 ICU beds, from which eight are for pediatric patients. So, only 23beds were eligible to compose the study sample.According to the Institutional Nucleus ofEpidemiological Surveillance, in 2008, 432 adult patientswere accounted at these three ICU. Thus, to calculate the sample extent, it was chosen the "PU incidence atICU" as an outcome. The adopted value was 25.6%,according to a study carried out with ICU patients inSão Paulo – Brazil⁽¹⁹⁾. On the sample calculus, the statisticalformula for longitudinal studies was used before andafter a 95% confidence interval. After the calculus process, it was identified a sample of 42 patients tocompose the study population.

During the four months of research, there were 83inpatients. Among them 11 were accepted with PU; 02were younger than 18 years old; 10 died; and 18 wereremoved before being there for 10 follow-up days.

During data collection, four instruments were used:the first one was a questionnaire consisting of social,demographic and clinical data; the three others referredto the translated and adapted versions of Norton, Bradenand Waterlow assessment scales for PU risk.

The Norton scale assesses five parameters for PUrisk: physical condition; level of consciousness; activity;mobility; incontinence. Each parameter was scored from1 to 4. The four parameters sum resulted in a score whichvaried from 5 to 20 points, understood like this: <14 (risk); and < 12 (high risk). Moreover, the smaller the final sum is, higher is the risk of developing PU⁽²⁰⁾.

The Braden Scale assesses the sensorial perception,humidity, activity, mobility, nutrition, friction and shear.The maximum score is 23 points; and the smaller thescore is, higher will be the PU risk. To the analyses, theBraden scores were dichotomized into two categories:low risk (score < 16) and high risk (score <16)⁽²¹⁾.

The Waterlow scale assesses seven main topics:weight/height relation (BMI), visual evaluation risk skinareas, gender/age, continence, mobility, appetite, andmedications. Besides these four items that score specialrisk factors, there are: undernourished tissue, neurologicdeficit, surgery time over two hours and trauma belowthe lumbar spine. In this case, a high score indicates ahigh risk of developing PU. Patients in study weredivided into three groups, according to their evaluations:at risk (score from 10 to 14); high risk (score from 15to 19) and very high risk of developing PU (score >20)⁽⁴⁾.

Each patient was assessed once a day for 15 days orfor at least 10 successive days, by three nurses at thesame time. During the assessment, each one of theseprofessionals was responsible for only one scale ofassessing PU risk. Evaluations were done only once aday, always during the morning, during the patients' bathor dressing treatment.

The data collecting period was based on a previousstudy which informed that the critical period fordeveloping PU injuries occurs until 14 days after thepatient's hospital admission⁽¹⁸⁾. In the cases that was impossible to conclude 15 visits to the patients becausehe/she died or was removed, this individual onlycomposed the research sample when it was possible tovisit him/her at least 10 times.

It was recorded the day on which the PU appearedin those patients who have developed this injury, as wellas, the ulcer stage and location. These patients still beingassessed until the 15^th day in order to identify any newinjury and to keep up with his/her inpatient time untilthe end (discharge, death or hospital transference) forstatistics aims. PU stage was classified according to theEuropean Pressure Ulcer Advisor Panel prevention and treatments guidelines⁽²²⁾. It's worth saying that duringthis data collection, beyond the physical test, the patient'smedical records has been checked and that some information has been gotten from the patient's family.

Data was entered twice and it was storage on anExcel program data basis. SPSS 13.0 software was usedto formulate the percentage and absolute frequency, inaddition to the measures of central tendency.Kolmogorov-Sminorv test was used for datahomogeneity assessment and a 95% trust level wasapplied. Within each scale group and also among thescales, all the values were compared, for the 15assessments, using the Friedman non-parametric test,aiming to found statistics differences between the scoresof patients with and without injuries over time; multiplecomparisons using the Conover test were also carriedout over the research period. The correlation analysis of the three scales averages was accomplished using thePearson correlation coefficient.

After the Institutional Ethics Committee approval,the study was carried out according to protocol number86145/08. To unconscious patients, the permission wasgiven by their family or guardians.

RESULTS

Patients were assessed using the three scales ofassessing PU risks, daily, for 15 or, at least, 10 days. Thefollow-up days average was 14.2 (SD±3,6). During thedata collection 32 (76.2%) patients were observed for15 days; 07 (16.7%) for 10 days; 02 (4.7%) 13 days; andonly 01 (2.4%) for 11 days.

The investigated sample consisted of 34 young men(81%), as follows: 31% aged from 18 to 25 years old,and 26.2% aged from 36 to 46 years old. The age averageand median were 35.3 and 33.3 years old, respectively.

A substantial proportion of patients admitted at theICU who were included on this study have come fromthe anesthetic recovery room, and from the EmergencyUnit, 25 (59.5%) and 14 patients (33.3%), respectively.During the admission process, the following clinicalsituations were verified as the most common ones: neurological dysfunction, with traumatic brain injuryprevalence (61.9%), followed by surgeries (26.2%),mostly exploratory laparotomies and neurosurgeries. Onthe other patients, it was detected the use of mechanicalventilation (78.6%), vasoactive drugs (31%) and drugsfor sedation (69%). Most patients (85.7%) did notpresent any pre-existing diseases and among those whopresented comorbidity, there was a prevalence of arterialhypertension and diabetes mellitus.

Out of the 42 assessed patients, 25 developed PUevidencing an incidence of 59.5%. PU were detected inthe patients between the 2^nd and the 14^th follow-up days,with a time of PU appearance average of 9.6 (SD±3.3) after the hospitalization.

From the total of 47 injuries identified, 23 (48.9%)were stage-I pressure ulcers and 24 (51.1%), stage-II.The number of injuries per patients states as follows:06 patients with 01 PU; 16 with 02 PU; and 3 with 03PU. Among those patients who developed two PU,four injuries were sacral and heel; nine were sacral andoccipital; and three were heel and occipital pressureulcers. Altogether, patients have had 47 PU at differentlocations and stages, with an average of 1.88 PU(SD±0.7) for patient. In regards to pressure ulcersstaging, 23 (48.9%) stage-I and 24 (51.1%) stage-IIpressure ulcers were observed.

Patients assessed with the Norton scale presented adaily average score that varied from 8.8 to 9.1 (SD±6.7),showing a discrete increasing, but statistically insignificantduring the 1^st – 15^th hospitalization days (p=0.028)(Table 1). These patients' assessment detected a moderaterisk in two patients (4.8%) and a high risk in 40 patients(95.2%). Among those who did not presented risk, itwas not detected the PU presence, but in those with a high UP risk, 62.5% developed injuries, neverthelessthere was not statistical significance on this relation(p=0.099).

Patients assessed with Braden scale presented anaverage score that varied from 11.6 to 12.5 (SD±6.7),showing a significant statistical increase between the 1^st15^th and the 5^th-15^th assessment days (p=0.004) (Table1). Based on this table, the subject's pressure ulcers riskwas discriminated as follows: low (01 patient), moderate(34 patients) and high (07 patients). Among the patientswho showed moderate and high risks, 76% and 20%developed during the study period, although there wasnot a statistic significance on this relation (p=0.070).

Patients assessed with the Waterlow scale obtained an average score which varied from 22.9 to 24.8 (SD±16.1), with a significant statistical increase between the 1^st-10^th days (p=0.003) (Table 1). Based on thisframework, the subjects were discriminated as follows:at risk (3), at high risk (7) and at very high risk (32).Differently from Norton and Braden scales, amongpatients at high risk and at very high risk of developingpressure ulcers, according to Waterlow, 2% and 92%,respectively, developed PU during the study (p=0.005).

During the patient's monitoring, it was found anincrease on Norton and Braden scores until the 10^th assessment with a discrete decline on the 15^th day(p<0.001) over the Waterlow average which increasedcontinuously among the analyzed patients (Table 2).

After performing the average correlation tests forthe three scales, simultaneously, it was observed betweenNorton and Braden (r=0.711, p<0.001); Norton andWaterlow (r=-0.535, p<0.001), Braden and Waterlow(r=-0.426, p=0.005) statistically significant correlations.The r and p values presented above has evidenced thatNorton scale is directly proportional to the Waterlowscale; and that, in the other hand, the Braden scale is inversely proportional to the Waterlow scale.

DISCUSSION

In the current paper, the sample was composed,mostly, by male adults. These patients' profiles weredifferent from those in large part of other analyzedstudies, which have applied the same assessment scalesfor pressure ulcer risk in countries like German, theNetherlands, Turkey and Brazil, where patients' profileinclude elderly female patients who were in bed andpresented association of chronic degenerative diseases^{(12, 23-25)}.

The fact that our study was carried out at a specializedservice of emergency care on traumatology andneurology, which is normally associated to trafficaccidents involving young male individuals, has perhapsdetermined such difference. Even though, nurses at theseinstitutions need to identify and understand each medicalspecialty and the PU risk correlations, in order to act onthe intrinsic and extrinsic factors related to PU.

In this research, the average of the monitoring dayswas 14.2 (SD± 3.6). And most of the studied populationhas developed PU, and 64% presented two injuries. Inregards to detected PU incidence (59.5%) in Brazil, theproblem's estimative at chronic care units was diversified,but some identified publications has shown a lowerincidence in comparison to what was found during thisresearch, like 5.9%; 39.7%; 26.8% and 11.8%, respectively^(25-28). Thereby, it is urgent and necessary thatpreventive care attitudes can be performed; even in orderto reverse other morbidities that have probably beencaused by these injuries.

In regards to the chronological aspects, it is knownthat PU do not develop in a precise time, due to eachpatient clinical conditions specificities; however, theliterature points out that it can appear after 24 hours ofhospitalization, or 10 – 15 days after the patientsadmission, depending on the given inputs and aids⁽²⁹⁾.

In other publications about chronic patients the PUincidence happened in 8, 4 and 10 days, respectively⁽³⁰³¹⁾. So, since the patient admission, the nursing care is animportant predictive factor to the PU outcome, becausethe concern in examining the patient's skin, giving thepreventive inputs and monitoring them with instrumentswhich show effective preventive potential, both negativeand positive, can be the guarantee of PU absence duringthe hospitalization.

Risk assessment scales are important tools to nurses,because they indicate vulnerable points, reinforce theconstant assessment necessity and stimulate prevention.However, it is necessary, beyond the nurse abilities, theadoption of various efficient instruments to this workenvironment.

In this manuscript, both the Norton and the Bradenscales showed a significant statistical increase in theirscores until the 10^th hospitalization day. A similar facthas happened with the Waterlow scale, but thisproceeded during all the research time. However, Nortonand Braden are negative scales and Waterlow is a positiveone, in other words, the detected points increase indicatesa reduction in the subjects' vulnerability to PU, accordingto Norton and Braden; and an increase in PU risk, according to the Waterlow scale. Besides, it was foundthat Norton scale is directly proportional to the Bradenone; and that Braden scale is inversely proportional tothe Waterlow one. In fact, in the research sequence, theWaterlow scale, perhaps because of the increased score,evidenced more statistically significant PU cases than theother two instruments. The interference of this in other health service which adopted only Norton and Bradenscales might result in underestimate cases of vulnerablesubjects to PU development.

There are several criticisms concerned to the Norton, Braden, Waterlow and Gosnell risk scales, because some of them underestimate and other overestimate the assessment of at-risk patients. One of these severalcriticism lies on the fact that these scales present inversescore order (ascending and descending) or yet havedifferent cut-off point for the PU risk assessment. Thisfact makes it more difficult to compare the results ofresearches which assess the use of these instruments^{(7, 9)}.

The directly correlation detected between Nortonand Braden scales was a discovery confirmed by asystematic bibliographic revision that, beyond this,evidenced that Braden scale presents better specificityand sensibility balance to prevent and predict injuries.On the other hand, this publication's authors point outthat Waterlow scale is a great instrument for sensibility(50.6%) and specificity (60.1%). In the end, the authors'critical judgment has showed Braden and Norton scalesas presenting better results, fact that is different fromwhat was found out in our study⁽³²⁾.

Another foreign publication has identified theWaterlow scales as the one with better sensibility whencompared to Braden and Norton scales, and the bestspecificity was given to Norton scale⁽¹⁶⁾. Some researches that point out some problems on the predictable factorsof these scales were also identified. They questionwhether these instruments provide an accurateassessment and if they really help on the clinical practice^{(8, 14, 33)}. For example, a research carried out in Great Britainwith 110 nurses has identified a percentage of 72.6%of wrong classification of the PU risk developmentwhen using Waterlow scale. Besides, only 12% of thenurses made a precise score for patients using thismethod⁽³⁴⁾. Thus, beyond the use of these instruments,the heath care professional must rely on his/her clinicalexperience and knowledge to attribute or not a subject'svulnerability to develop PU.

CONCLUSION

The current paper presents some limitations. Oneof them is due to the fact that heel injuries have notbeen divided into left and right heels. The establishedtime to the patients monitoring on 10-15 days made itimpossible to follow-up the injuries outcome, as wellas the patients healing or the enlargement of theirhospitalization period. In addition, despite of conductinga sample calculation on which the number was achieved,when compared to other papers, the sample was small.Even though, the study found out important facts like abetter performance of the Waterlow scale assessing thePU risks if compared to Norton and Braden scales in apopulation composed mainly by young male. Inaddition, the results reinforced the existing directcorrelation between Braden and Norton scales, highlighted by other publications.

Scales score comparison carried out separately andin group showed that even with a score increase manypatients developed PU, even knowing that these increaseindicates a vulnerability reduction, detected during theassessing period when using Norton and Braden scales.

So, in spite of being valid, practical and efficientinstruments, it is necessary to base the subjects' assessmentmainly on the health professional clinical knowledge andexperience. Mainly because these instruments were madefor populations that are different from the Brazilian onesand it is always possible that their limitations are noticedin the hospital everyday life. Thus, it is suggested thatsome new researches are carried out with better delimitation and a more representative sample, in orderto clarify these instruments validity in other scenarioslike nursing homes, medical wards and Brazilian homesto determine the better choice for the health professionals to predict injuries risks.

REFERENCES

• 1. Papanikolaou P, Lyne P, Anthony D. Risk assessment scalesfor pressure ulcers: a methodological review. Int J NursStud. 2007;44(2):285-96.
• 2. Anthony D, Papanikolaou P, Parboteeah S, Saleh M. Dorisk assessment scales for pressure ulcers work? J TissueViability. 2010;19(4):132-6.
• 3. Pelham F, Keith M, Smith A, Williams DV, Powell G. Pressure ulcer prevalence and cost in the US population. JAm Med Dir Assoc. 2007;8(3):B20.
• 4. Rocha AB, Barros SM. Avaliação de risco de úlcera por pressão:propriedades de medida da versão em português da escala deWaterlow. Acta Paul Enferm. 2007;20(2):143-50.
• 5. Souza DM, Santos VL. Risk factors for pressure ulcerdevelopment in institutionalized elderly. Rev LatinoamEnferm. 2007;15(5):958-64.
• 6. Meesterberends E, Haffens R, Lohrmann C, de Wit R. Pressure ulcer guideline development and disseminationin Europe. J Clin Nurs. 2010;19(11-12):1405-503.
• 7. Kim EK, Lee SM, Lee E, Eom MR. Comparison of thepredictive validity among pressure ulcer risk assessmentscales for surgical ICU patients. Aust J Adv Nurs.2008;26(4):87-94.
• 8. Gould D, Goldstone L, Gammon J, Kelly D, Maidwell A.Establishing the validity of pressure ulcer risk assessmentscales: a novel approach using illustrated patient scenarios.Int J Nurs Stud. 2002;39(2):215- 28.
• 9. Dealey C. Cuidando de feridas: um guia para asenfermeiras. 3a ed. São Paulo: Atheneu; 2008.
• 10. Pedley GE. Comparison of pressure ulcer grading scales:a study of clinical utility and inter-rater reliability. Int J Nurs Stud. 2004;41(2):129-40.
• 11. Maklebust J. Interrupting the pressure ulcer cycle. NursClin North Am. 1999;34(4):861-71, vi.
• 12. Sayar S, Turgut S, Dogan H, Ekici A, Yurtsever S, DemirkanF, et al. Incidence of pressure ulcers in intensive care unitpatients at risk according to the Waterlow scale and factorsinfluencing the development of pressure ulcers. J ClinNurs. 2007;18(5):765-74.
• 13. Gomes FS, Bastos MAR, Matozinhos FP, Temponi HR,Velásquez-Meléndez G.Fatores associados à úlcera por pressãoem pacientes internados nos centros de terapia intensiva deadultos. Rev Esc Enferm USP. 2010;44(4):1070-6.
• 14. Kwong E, Pang S, Wong T, Ho J, Shao-Ling X, Li-jun T.Predicting pressure ulcer risk with the modified Braden,Braden, and Norton scales in acute care hospitals inMainland China. Appl Nurs Res. 2005;18(2):122-8.
• 15. Gould D, Goldstone L, Kelly D, Gammon J. Examiningthe validity of pressure ulcer risk assessment scales: areplication study. Int J Nurs Stud. 2004;41(3):331-9.
• 16. Balzer K, Pohl C, Dassen T, Halfens R. The Norton, Waterlow, Braden, and Care Dependency Scales: comparingtheir validity when identifying patients' pressure sore risk.J Wound Ostomy Continence Nurs. 2007;34(4):389-98.
• 17. Schoonhoven L, Haalboom JR, Bousema MT, Algra A,Grobbee DE, Grypdonck MH, Buskens E; prePURSEstudy group. The prevention and pressure ulcer risk scoreevaluation study. Prospective cohort study of routine useof risk assessment scales for prediction of pressure ulcers.BMJ. 2002;325(7368):797.
• 18. Pang SM, Wong TK. Predicting pressure sore risk withthe Norton, Braden, and Waterlow scales in a Hong Kong rehabilitation hospital. Nurs Res. 1998;47(3):147-53.
• 19. Cardoso MC, Caliri MH, Hass VJ. Prevalência de úlceras de pressão em pacientes críticos internados em um hospitaluniversitário. REME Rev Min Enferm. 2004;8(2):316-20.
• 20. Lopes MI, Santos MR. A prática de cuidados deenfermagem na prevenção das úlceras de pressão.Referência. 2002;8(1):63-9.
• 21. Paranhos WY, Santos VL. Avaliação de risco para úlcerasde pressão por meio da Escala de Braden, na línguaportuguesa. Rev Esc Enferm USP. 1999;33(N Esp):191-206.
• ²²
  European Pressure Ulcer Advisory Panel (EPUAP)[homepage on the Internet]. Oxford: European PressureUlcer Advisory Panel. Pressure Ulcer Prevention andTreatment Guidelines: c1998 - [updated 2008 Sep 29; cited2006 Oct 10]. Available from: http://www.epuap.org/gltreatment.html
  » link
• 23. Louro M, Ferreira M, Póvoa P. Avaliação de protocolo deprevenção e tratamento de úlceras de pressão. Rev Bras TerIntensiva. 2007;19(3):337-41.
• 24. Tannen A, Dassen T, Bours G, Halfens R. A comparisonof pressure ulcer prevalence concerted data collection inthe Netherlands and Germany. Int J Nurs Stud. 2004;41(6):607-12.
• 25. Moro A, Maurici A, Valle JB, Zaclikevis VR, KleinubingJúnior H. Avaliação dos pacientes portadores de lesão porpressão internados em hospital geral. Rev Assoc Med Bras.2007;53(4):300-4.
• 26. Rogenski NM, Santos VL. Estudo sobre a incidência deúlceras por pressão em um hospital universitário. RevLatinoam Enferm. 2005;13(4):474-80.
• 27. Sousa CA, Santos I, Silva LD. Aplicando recomendaçõesda Escala de Braden e prevenindo úlceras por pressão:evidências do cuidar em enfermagem. Rev Bras Enferm.2006;59(3):279-84.
• 28. Carvalho MP, Lüdtke EB, Oliveira V, Fonseca PG, Rosales GG, Ferreira ALD, et al. Perfil dos pacientes com úlceras depressão internados no Hospital Universitário São Franciscode Paula (HUSFP). Rev Saúde UCPEL. 2007;1(1):32-8.
• 29. Nijs N, Toppets A, Defloor T, Bernaerts K, Milisen K,Van Den Berghe G. Incidence and risk factors for pressurein the intensive care unit. J Clin Nurs. 2008;18(9):1258-66.
• 30. Feijó CA, Leite Júnior FO, Martins AC, Furtado JúniorAH, Cruz LL, Meneses FA. Gravidade dos pacientesadmitidos à unidade de terapia intensiva de um hospitaluniversitário brasileiro. Rev Bras Ter Intensiva. 2006;18(1):18-21.
• 31. Abelha FJ, Castro MA, Landeiro NM, Neves AM, Santos CC. Mortalidade e o tempo de internação em uma unidadede terapia intensiva cirúrgica. Rev Bras Anestesiol.2006;56(1):34-45.
• 32. Pancorbo-Hidalgo PL, Garcia-Fernandez FP, Lopez-MedinaIM, Alvarez-Nieto C. Risk assessment scales for pressureulcer prevention: a systematic review. J Adv Nurs. 2006;54(1):94-110.
• 33. Moreno-Pina J, Richart-Martínez M, Adolf Guirao-Goris J, Duarte-Climents G. Análisis de las escalas de valoración del riesgo de desarrollar una úlcera por presión. EnfermClin. 2007;17(4):186-97.
• 34. Kelly J. Inter-rater reliability and Waterlow's pressure ulcerrisk assessment tool. Nurs Stand. 2005;19(32):86-7, 90-2.

*

Artigo extraído da Dissertação de Mestrado "Acurácia de escalas para avaliação de risco para úlcera por pressão em pacientes críticos" defendida em 2009 no Programa de Pós-graduação em Enfermagem, Universidade Federal do Ceará.- UFC - Fortaleza (CE), Brasil.

Publication Dates

• Publication in this collection
  02 Dec 2011
• Date of issue
  2011

History

• Accepted
  26 May 2011
• Received
  14 Mar 2011