Clinical and Experimentat Dermatology 1992; 17: 328-331.
Health-care professionals' views of the effectiveness of pressure ulcer treatments. A survey among nursing-home physicians, dermatologists and nursing staff in The Netherlands G.TER RIET, H.VAN HOUTEM AND P.KNIPSCHILD Department of Epidemiology! Health Care Research, University of Limburg, P.O. Box 616, 6200 MD Maastricht, The Netherlands Accepted for publication 2 October 1991
Summary In a survey, 237 nursing-home physicians, 113 dermatologists, and 164 supervisory nurses in nursing homes rated 30 treatments for stage III pressure ulcers by scoring their effectiveness. On average, there seems to be a fair degree of consensus between these groups of health-care workers about the effectiveness of most treatments. The most notable exceptions are ultraviolet light, ultrasound, and laser therapy and myocutaneous flap operations. Here, the dermatologists disagree with both nursinghome physicians and nurses. However, within all specialties there is disagreement about the efficacy of ultraviolet light, ultrasound, and laser therapy and the topical application of zinc oxide, betadine iodine, sodium hypochlorite and a few others.
In Britain, a consensus report on the prevention and treatment of pressure ulcers is likely to be published this year.' In The Netherlands, such a report was published in 1985 (on prevention)' ' and 1986 {on treatment).^-^ These reports may suggest that matters have been settled. However, as Skrabanek has pointed out, 'the very need for consensus stems from a lack of consensus'.** In a survey among members ofthree relevant medical specialties involved in the care of patients with pressure ulcers in The Netherlands, we investigated the extent to which they agreed about the effectiveness of different treatments, many »f which are mentioned in the Dutch con.sensus rcporl.''' \n English summary of that report is given in Appendix 1. Another background to this survey was the preparation for a repeat ofthe only controlled clinical trial of ascorbic acid supplementation for patients with pressure ulcers.' We wanted to assess the prior opinions of specialists in 328
this field with regards to this hypothesis in order to judge whether they would change their opinion in the light of new evidence. In addition, we were looking for a 'free rider' in the trial, i.e. another intervention that we could investigate in the same trial using a factorial design." Because we surmized that the ascorbic acid hypothesis would have low credibility among specialists, we preferred to choose a free rider that would be more controversial while, at the same time, lacking adequate support from controlled trials. In that case the results of this part ofthe trial would be informative to the medical community regardless ofthe outcome. One can compare this strategy to that which experienced clinicians use in selecting diagnostic tests. There, test results are most informative where the pre-test odds of a correct diagnosis of disease are close to 50:50.
Participants and methods We selected 30 treatments as items for the survey. These were chosen partly because of their widespread use and partly because their efficacy has been tested in clinical trials. From the addresses of nursing-home physicians, dermatologists, and supervisory nursing staff in nursing homes in The Netherlands, random samples containing 282, 183, and 225 subjects respectively were drawn. These numbers represent approximately S{)% of all members of these specialties in The Netherlands. The samples ofthe nursing-home physicians and the nurses were drawn in such a way that each nursing-home was represented by one member of each specialty. This was not the case for the dermatologists. In the spring of 1990 they received a postal questionnaire seeking their opinion about the effectiveness of
PROrKSSIONAL VIEWS OK PRKSSURK ULCiKR TREATMENTS different pressure ulcer treatments, presented in alphabetical order. We confronted them with a well defined clinical case which was thought to be representative of a large category of patients. 'I'he following text clarified our survey. 'Of course, in practice interventions are very often combined. However, we ask you to rate them separately. The question which we ask you to answer 30 times is the following: how effective do you rate this intervention, if performed according to the accepted standards, in the treatment of the pressure ulcer in the following patient?' (see Appendix 2). Participants could express their opinion as a score between 0 and 10, where 0 meant harmful, and 1-10 ranged lVom 'not at all effective' to 'very effective'. In addition, 'no opinion' could be indicated. In the analysis (separate for each specialty), we calculated for all interventions the percentage of responders who indicated 'no opinion', who considered the treatment harmful, and the distribution of perceived effectiveness. The mean values in Table 1 are based only on the people w ho responded with a score from 1 to 10. By comparing the means one gets an idea of the agreement between specialties, whereas the distributions show the extent to which agreement e.vists within each specialty.
329
three specialties, especially by the dermatologists. .Mso, the dermatologists agreed among each other that these therapies were ineffective, while in the other specialties opposing beliefs were more prevalent. Surprisingly, the topical wound care measures were not judged to be effective by most respondents. However, in all specialties there was most disagreement on zinc oxide, betadine iodine, Du(Klerm@, silversulphadiazine, sodium hypochlorite, and Opsite@. A high percentage of respondents indicating no opinion suggest that many did not know what to think of therapies such as the use of autologous growth factors and the topical application of ketanserin or insulin. It may also indicate that many had never heard of them. In the category o( miscellaneous treatments there was striking consensus that patient education was effective. Whether myocutaneous flap operations are effective in patients with the kind of pressure described, appeared to be a matter of dispute in all specialties. Discussion
The data apply strictly to the opinions ofthe specialties. Of course, these opinions should not be taken as proof of true efficacy. Also, the opinions apply to patients with pressure ulcers such as the one described in .Appendix 1. Results In our opinion, presenting a well-defined clinical case .^fter two reminders, the response among nursing-home representative of a large but specific category of patient is physicians, dermatologists, and nurses was 237 (84",,), the only way to collect this kind of data in an unambiguous way. Thus, instead of limiting it, the generalized 113 (62",,), and 164 (73"/,,), respectively. Table 1 shows the scores for each specialty. We report applicability is enhanced. On the other hand, the small these scores under livt- headings, namely measures non-response somewhat limits the generality of the involving body support, nutritional measures, application results. We are not aware of similar studies in other of physical energy, topical wound care, and miscellaneous countries. other measures. In the Dutch consensus report^-^ specific mention is There seems to be a high degree of consensus both made of 14 ofthe 30 therapies in Table I. Ironically, in among and within the three specialties that the measures seven of them a large degree of within-specialty disagreeinvolving body support are effective treatments. For ment can be .seen. On the measures involving body example, in all specialties, over 90*\, considered position- support (except, perhaps, flotation pads), patient eduing of patients every 2 h lo be effective (or even very cation and treatment with complamin iontophoresis effective; score > 7). Also, a fair degree of consensus was holds a good deal of consensus between and within the found for air-tluidi/ed beds and water mattresses with specialties. The dissent is not surprising. Why should mean scores of approximately 7. The effectiveness of there be agreement if unequivocal evidence is lacking.'' flotation pads appeared more controversial. Of course, there may be useful evidence which clinical In the area oi nutritional measures, on average, the three decision makers (usually physicians) are unaware of That specialties agreed to a targe extent. However, a protein- is to say, some of the disagreements may be unnecessary. enriched diei (in ihis patient with a normal serum A potentially useful way to confront clinical decision albumin) turned out to he a source of disagreement with makers with the evidence from clinical trials might be the some 40*^i, of respondents rating it with a 7 or higher and publication of review articles. some 30";, with a 4 or lower in all specialties. The other Where u.seful evidence is lacking, a survey .such as the measures under this heading were not given much credit. one presented here could be helpful in selecting therapies The different applications of physical energy were not for format testing in a new clinical trial. The analogy with rated very positively, i.e. no mean scores over 4 4 in any of diagnostic strategies was explained in the introduction of the groups. Fairly common therapies such as ultraviolet this article. We selected ultrasound therapy (little evilight and ultrasound therapy were given low credit by all dence and a great deal of di.sagreement) as a treatment in a
G.TER RIET, H.VAN HOUTEM AND P.KNIPSCHILD
330
o .a Z .S
g o o o o
Al
I O p
If)
O "n in
p
00
o
rt
X r— Al
-5
-1-1
O ^ O - ^ O i n - ^ O O
"o b
oo r- ' o CO
VI iri
rt Ul W O
no —I >^ p ro sb o fs rs rs
p
rs rs
"5 y O-2
o o o o
11 o -a f^ Tt- OC oo
C u
u o
p o —p 4- ^ rs rs
ro -— p CO r-- r-
*j
"
fe
>1
C
I
rt
fe-£
S T3 "bO
'11 ^s
B
^ 3 CO >^ «.y . O S rt
iali
-H -5 ":S E .i u u OH
-a o
o, u
c -S « •= . £ •§
=
£ o 5 -j= •I"? S
11 i 5 ^J S
u •u ^ C
a
g
&•
?S C
u
u
n
.^ o |fc
p5 -o. •
PROFESSIONAL VIEWS OF PRESSURF. ULCER TREATMENTS
331
randomized trial, in which, for other reasons, the effect of vitamin C will also be tested. This trial started in January 199L Data colleetion will be complete at the end of 1992.
Availulile i:\idence together with measures that seem io};ica! and not harmtul liave been summarized in the report. In addition, a description ol grades of severity is given. Basically, four grades arc distinguished: (i) local redness (not removable by pressure); (Ii) formation of a blister; (iii) superficial lesion, and {iv) deep lesion. Grades III and l\' are both subdivided into grades 'a' and 'b' to Acknowledgments indicate lack or presence of necrosis, respectively. Wiih regards lo treaiment three levels of usefulness are disThis work was supported by a grant of The Netherlands tinguished: useliil, potentially useful in individual cases (not advised Organization for Scientific Research (NWO). tr)r general use), and not useful. The continuation o) pre\entive We would like to thank .Annemie Mordant for her help measures (frequent positioning, patient instruction and establishing with the data management and Andre Knols for his an individualized nursing plan) is considered useful in every treatadvice on the description of the clinical case. ment, Ueiailed description ol' ihe consensus on cleaning and covering of ulcers is beyond the scope of this article. However, Table 2 shows [ how the report juilges 14 trealments that were mentioned speciHcally when dealing with grade Ilia ulcers, References Irrespeciivc of gradu, ilie repori designates a niimlx-r ol' lopical 1. (,iualit} assurance in liiiig-tcrm care of elderly people: Guidelines treatments as 'not useful' e,g, preparations likely lo induce allergic tiir care. Report of a joint workshop of the Royal College of reactions, mercury chrome, complamin iontophoresis and irradiation Physicians and the British Geriatrics Society. In press. with ultraviolet light. 2, Centraat Begeleidingsorgaan voor de Intercollegiate Toetsing. Consensus hijeenhimst: decuhititsprevenlie. Utrecht: CBO, 1985: .'iO-2. As regards trealnient evaluation, much is expected from meticulous }. Bakkcr 11, tlonscn.susbijirenkonisi preventie van decubitus. Ned registration of risk assessments, grading, and prescribetl therapies. Tijdsihr Ceneestd I'JSS; 129: 1280 -t, 4. Centraal BcgeUidtngsorgaan voor de Intercotlegiale Toetstng. Consensus bijeenkomst: hehandelmg decuhttus. Utrecht: CBO, 1986- 21 Appendix 2 88-89, 93 95. 5. Bakker H. Oinscnsus-hijccnkomsi behandcling van dccubitus, It concerns a 76-year-old female nursing-home patient. Her right AW Tijduhr r.ettceshd 1987; 131: 1.107 1309, buttock shows a .1-day-old pressure ulcer. It has the following 6. Skrabanek P. Nonscnsus consensus. Lancet 1990; 335: 1446-1447. characteristics: stage Ilia (ulcer reaching into the subcutaneous fat, no 7. Taylor 'r\'. Rimtnt-r S, Day It, Butcher J, Dymock IW. .\scorbic necrotic tissue and on palpation no signs of deep-tissue involvement); acid supplcmentarion in the treatment of pressure-sores. Lancet surface area: 12 cm'; little or no granulation tissue; no scab formation; the skin surrounding the ulcer is red. General information: the medical 8. Pocock SJ. Clinical Trials: ii Practical .Approach. Chichester: John reason f()r her admission is a paralysis of the left side of her body Wiley & Sons, 1989: 139 41. (hemiplegia); body-weight: 7.1 kg; length: 1 62 m (5-3 ft); she has been bed-ridden for the last week due to influenza; otherwise healthy with godtl appetite; no loss of sensibility on the right side of her body; nn Appendix 1 peripheral vascular problems; no incontinence; no diabetes mellitus; Summary of'The Netherlands Consensus Report on the treatment of no intake of soporific medication; body temperature: 3S-8 C.; mobility: she uses a wheelchair for distances > 100 m (90 yd); .serum albumin pressure ulcers.""'' The report nolt-s that relatively liltle evidence from contnilleil concentration: 42 g/I; hemoglobin: S 4 mmol/1 clinical research on the etiectivencss of several treatments exists.
Health-care professionals' views of the effectiveness of pressure ulcer treatments. A survey among nursing-home physicians, dermatologists and nursing staff in The Netherlands G.TER RIET, H.VAN HOUTEM AND P.KNIPSCHILD Department of Epidemiology! Health Care Research, University of Limburg, P.O. Box 616, 6200 MD Maastricht, The Netherlands Accepted for publication 2 October 1991
Summary In a survey, 237 nursing-home physicians, 113 dermatologists, and 164 supervisory nurses in nursing homes rated 30 treatments for stage III pressure ulcers by scoring their effectiveness. On average, there seems to be a fair degree of consensus between these groups of health-care workers about the effectiveness of most treatments. The most notable exceptions are ultraviolet light, ultrasound, and laser therapy and myocutaneous flap operations. Here, the dermatologists disagree with both nursinghome physicians and nurses. However, within all specialties there is disagreement about the efficacy of ultraviolet light, ultrasound, and laser therapy and the topical application of zinc oxide, betadine iodine, sodium hypochlorite and a few others.
In Britain, a consensus report on the prevention and treatment of pressure ulcers is likely to be published this year.' In The Netherlands, such a report was published in 1985 (on prevention)' ' and 1986 {on treatment).^-^ These reports may suggest that matters have been settled. However, as Skrabanek has pointed out, 'the very need for consensus stems from a lack of consensus'.** In a survey among members ofthree relevant medical specialties involved in the care of patients with pressure ulcers in The Netherlands, we investigated the extent to which they agreed about the effectiveness of different treatments, many »f which are mentioned in the Dutch con.sensus rcporl.''' \n English summary of that report is given in Appendix 1. Another background to this survey was the preparation for a repeat ofthe only controlled clinical trial of ascorbic acid supplementation for patients with pressure ulcers.' We wanted to assess the prior opinions of specialists in 328
this field with regards to this hypothesis in order to judge whether they would change their opinion in the light of new evidence. In addition, we were looking for a 'free rider' in the trial, i.e. another intervention that we could investigate in the same trial using a factorial design." Because we surmized that the ascorbic acid hypothesis would have low credibility among specialists, we preferred to choose a free rider that would be more controversial while, at the same time, lacking adequate support from controlled trials. In that case the results of this part ofthe trial would be informative to the medical community regardless ofthe outcome. One can compare this strategy to that which experienced clinicians use in selecting diagnostic tests. There, test results are most informative where the pre-test odds of a correct diagnosis of disease are close to 50:50.
Participants and methods We selected 30 treatments as items for the survey. These were chosen partly because of their widespread use and partly because their efficacy has been tested in clinical trials. From the addresses of nursing-home physicians, dermatologists, and supervisory nursing staff in nursing homes in The Netherlands, random samples containing 282, 183, and 225 subjects respectively were drawn. These numbers represent approximately S{)% of all members of these specialties in The Netherlands. The samples ofthe nursing-home physicians and the nurses were drawn in such a way that each nursing-home was represented by one member of each specialty. This was not the case for the dermatologists. In the spring of 1990 they received a postal questionnaire seeking their opinion about the effectiveness of
PROrKSSIONAL VIEWS OK PRKSSURK ULCiKR TREATMENTS different pressure ulcer treatments, presented in alphabetical order. We confronted them with a well defined clinical case which was thought to be representative of a large category of patients. 'I'he following text clarified our survey. 'Of course, in practice interventions are very often combined. However, we ask you to rate them separately. The question which we ask you to answer 30 times is the following: how effective do you rate this intervention, if performed according to the accepted standards, in the treatment of the pressure ulcer in the following patient?' (see Appendix 2). Participants could express their opinion as a score between 0 and 10, where 0 meant harmful, and 1-10 ranged lVom 'not at all effective' to 'very effective'. In addition, 'no opinion' could be indicated. In the analysis (separate for each specialty), we calculated for all interventions the percentage of responders who indicated 'no opinion', who considered the treatment harmful, and the distribution of perceived effectiveness. The mean values in Table 1 are based only on the people w ho responded with a score from 1 to 10. By comparing the means one gets an idea of the agreement between specialties, whereas the distributions show the extent to which agreement e.vists within each specialty.
329
three specialties, especially by the dermatologists. .Mso, the dermatologists agreed among each other that these therapies were ineffective, while in the other specialties opposing beliefs were more prevalent. Surprisingly, the topical wound care measures were not judged to be effective by most respondents. However, in all specialties there was most disagreement on zinc oxide, betadine iodine, Du(Klerm@, silversulphadiazine, sodium hypochlorite, and Opsite@. A high percentage of respondents indicating no opinion suggest that many did not know what to think of therapies such as the use of autologous growth factors and the topical application of ketanserin or insulin. It may also indicate that many had never heard of them. In the category o( miscellaneous treatments there was striking consensus that patient education was effective. Whether myocutaneous flap operations are effective in patients with the kind of pressure described, appeared to be a matter of dispute in all specialties. Discussion
The data apply strictly to the opinions ofthe specialties. Of course, these opinions should not be taken as proof of true efficacy. Also, the opinions apply to patients with pressure ulcers such as the one described in .Appendix 1. Results In our opinion, presenting a well-defined clinical case .^fter two reminders, the response among nursing-home representative of a large but specific category of patient is physicians, dermatologists, and nurses was 237 (84",,), the only way to collect this kind of data in an unambiguous way. Thus, instead of limiting it, the generalized 113 (62",,), and 164 (73"/,,), respectively. Table 1 shows the scores for each specialty. We report applicability is enhanced. On the other hand, the small these scores under livt- headings, namely measures non-response somewhat limits the generality of the involving body support, nutritional measures, application results. We are not aware of similar studies in other of physical energy, topical wound care, and miscellaneous countries. other measures. In the Dutch consensus report^-^ specific mention is There seems to be a high degree of consensus both made of 14 ofthe 30 therapies in Table I. Ironically, in among and within the three specialties that the measures seven of them a large degree of within-specialty disagreeinvolving body support are effective treatments. For ment can be .seen. On the measures involving body example, in all specialties, over 90*\, considered position- support (except, perhaps, flotation pads), patient eduing of patients every 2 h lo be effective (or even very cation and treatment with complamin iontophoresis effective; score > 7). Also, a fair degree of consensus was holds a good deal of consensus between and within the found for air-tluidi/ed beds and water mattresses with specialties. The dissent is not surprising. Why should mean scores of approximately 7. The effectiveness of there be agreement if unequivocal evidence is lacking.'' flotation pads appeared more controversial. Of course, there may be useful evidence which clinical In the area oi nutritional measures, on average, the three decision makers (usually physicians) are unaware of That specialties agreed to a targe extent. However, a protein- is to say, some of the disagreements may be unnecessary. enriched diei (in ihis patient with a normal serum A potentially useful way to confront clinical decision albumin) turned out to he a source of disagreement with makers with the evidence from clinical trials might be the some 40*^i, of respondents rating it with a 7 or higher and publication of review articles. some 30";, with a 4 or lower in all specialties. The other Where u.seful evidence is lacking, a survey .such as the measures under this heading were not given much credit. one presented here could be helpful in selecting therapies The different applications of physical energy were not for format testing in a new clinical trial. The analogy with rated very positively, i.e. no mean scores over 4 4 in any of diagnostic strategies was explained in the introduction of the groups. Fairly common therapies such as ultraviolet this article. We selected ultrasound therapy (little evilight and ultrasound therapy were given low credit by all dence and a great deal of di.sagreement) as a treatment in a
G.TER RIET, H.VAN HOUTEM AND P.KNIPSCHILD
330
o .a Z .S
g o o o o
Al
I O p
If)
O "n in
p
00
o
rt
X r— Al
-5
-1-1
O ^ O - ^ O i n - ^ O O
"o b
oo r- ' o CO
VI iri
rt Ul W O
no —I >^ p ro sb o fs rs rs
p
rs rs
"5 y O-2
o o o o
11 o -a f^ Tt- OC oo
C u
u o
p o —p 4- ^ rs rs
ro -— p CO r-- r-
*j
"
fe
>1
C
I
rt
fe-£
S T3 "bO
'11 ^s
B
^ 3 CO >^ «.y . O S rt
iali
-H -5 ":S E .i u u OH
-a o
o, u
c -S « •= . £ •§
=
£ o 5 -j= •I"? S
11 i 5 ^J S
u •u ^ C
a
g
&•
?S C
u
u
n
.^ o |fc
p5 -o. •
PROFESSIONAL VIEWS OF PRESSURF. ULCER TREATMENTS
331
randomized trial, in which, for other reasons, the effect of vitamin C will also be tested. This trial started in January 199L Data colleetion will be complete at the end of 1992.
Availulile i:\idence together with measures that seem io};ica! and not harmtul liave been summarized in the report. In addition, a description ol grades of severity is given. Basically, four grades arc distinguished: (i) local redness (not removable by pressure); (Ii) formation of a blister; (iii) superficial lesion, and {iv) deep lesion. Grades III and l\' are both subdivided into grades 'a' and 'b' to Acknowledgments indicate lack or presence of necrosis, respectively. Wiih regards lo treaiment three levels of usefulness are disThis work was supported by a grant of The Netherlands tinguished: useliil, potentially useful in individual cases (not advised Organization for Scientific Research (NWO). tr)r general use), and not useful. The continuation o) pre\entive We would like to thank .Annemie Mordant for her help measures (frequent positioning, patient instruction and establishing with the data management and Andre Knols for his an individualized nursing plan) is considered useful in every treatadvice on the description of the clinical case. ment, Ueiailed description ol' ihe consensus on cleaning and covering of ulcers is beyond the scope of this article. However, Table 2 shows [ how the report juilges 14 trealments that were mentioned speciHcally when dealing with grade Ilia ulcers, References Irrespeciivc of gradu, ilie repori designates a niimlx-r ol' lopical 1. (,iualit} assurance in liiiig-tcrm care of elderly people: Guidelines treatments as 'not useful' e,g, preparations likely lo induce allergic tiir care. Report of a joint workshop of the Royal College of reactions, mercury chrome, complamin iontophoresis and irradiation Physicians and the British Geriatrics Society. In press. with ultraviolet light. 2, Centraat Begeleidingsorgaan voor de Intercollegiate Toetsing. Consensus hijeenhimst: decuhititsprevenlie. Utrecht: CBO, 1985: .'iO-2. As regards trealnient evaluation, much is expected from meticulous }. Bakkcr 11, tlonscn.susbijirenkonisi preventie van decubitus. Ned registration of risk assessments, grading, and prescribetl therapies. Tijdsihr Ceneestd I'JSS; 129: 1280 -t, 4. Centraal BcgeUidtngsorgaan voor de Intercotlegiale Toetstng. Consensus bijeenkomst: hehandelmg decuhttus. Utrecht: CBO, 1986- 21 Appendix 2 88-89, 93 95. 5. Bakker H. Oinscnsus-hijccnkomsi behandcling van dccubitus, It concerns a 76-year-old female nursing-home patient. Her right AW Tijduhr r.ettceshd 1987; 131: 1.107 1309, buttock shows a .1-day-old pressure ulcer. It has the following 6. Skrabanek P. Nonscnsus consensus. Lancet 1990; 335: 1446-1447. characteristics: stage Ilia (ulcer reaching into the subcutaneous fat, no 7. Taylor 'r\'. Rimtnt-r S, Day It, Butcher J, Dymock IW. .\scorbic necrotic tissue and on palpation no signs of deep-tissue involvement); acid supplcmentarion in the treatment of pressure-sores. Lancet surface area: 12 cm'; little or no granulation tissue; no scab formation; the skin surrounding the ulcer is red. General information: the medical 8. Pocock SJ. Clinical Trials: ii Practical .Approach. Chichester: John reason f()r her admission is a paralysis of the left side of her body Wiley & Sons, 1989: 139 41. (hemiplegia); body-weight: 7.1 kg; length: 1 62 m (5-3 ft); she has been bed-ridden for the last week due to influenza; otherwise healthy with godtl appetite; no loss of sensibility on the right side of her body; nn Appendix 1 peripheral vascular problems; no incontinence; no diabetes mellitus; Summary of'The Netherlands Consensus Report on the treatment of no intake of soporific medication; body temperature: 3S-8 C.; mobility: she uses a wheelchair for distances > 100 m (90 yd); .serum albumin pressure ulcers.""'' The report nolt-s that relatively liltle evidence from contnilleil concentration: 42 g/I; hemoglobin: S 4 mmol/1 clinical research on the etiectivencss of several treatments exists.
