PREVALENCE OF PRESSURE ULCERS IN CANADIAN HEALTHCARE SETTINGS
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
PREVALENCE OF PRESSURE
ULCERS IN CANADIAN
HEALTHCARE SETTINGS
– M. Gail Woodbury, BScPT, MSc, PhD; and Pamela E. Houghton, PhD, BScPT
Al t h ough stati s ti cs regarding the number of pre s sure across Canada was 26.0% (95% Co n f i d ence Interva l ,
ulcers in the US and ot h er cou n tries are available, little 25.2% to 26.8%). The Canadian prevalen ce estimates dif-
information is known about the nu m ber of indivi duals in fered among the healthcare settings and were higher than
Canada who have pre s su re ulcers. Su ch information is those reported in the US and the Netherlands. Al t h ou gh
important to assess the scope and healthcare costs of pre s- additional studies are need ed, the data suggest that pre s-
sure ulcers and devel op pu blic pol i cies. To obtain estimat- su re ulcers are a significant concern in all healthcare set-
ed pre s sure ulcer preva l ence ra tes in Canada, existing tings in Canada.
data (ga t h ered between 1990 and 2003) from different
healthcare set ti n gs across the cou n try were obtained fro m KEYWO R D S : pre s sure ulcers, ep i d em i ology, prevalen ce ,
peer- reviewed published studies and from unpublished Canada, healthcare settings
studies provided by indivi duals and pre s sure ulcer sup-
port surface manu f a ctu rers. Methods used to ga t h er and O s to my/Wound Ma n a gement 2004;50(10):22–38
report prevalen ce data in each study were criti c a lly
linicians working in wound care apprec i a te
C
appra i sed using a mod i f i ed version of pu blished criteria.
Retro s pe ctive chart audit studies that did not involve h ow life for indivi duals with wounds is dis-
direct patient assessment were excl u d ed. The data includ- rupted by care , cost issu e s , and the pain asso-
ed info rmation from 18 acute care faci l i ties involving ciated with tre a tment. Although managing pre s sure
4,831 patients, 23 non-acute care faci l i ties with 3,390 u l cers is of ten a passion for wound care specialists,
patients, 19 mixed healthcare settings with 4,200 the majority of the pop u l a tion is unaw a re of the
patients, and five co m munity care agencies that surveyed challenges invo lved and many healthcare profession-
1,681 patients. Es ti m a tes of pressu re ulcer preva l en ce als place little emphasis on identifying and tre a ting
were 25.1% (95% Co n f i d en ce Interval, 23.8% to 26.3%) skin ulcers. The nu m ber of indivi duals seeking
for acute care settings, 29.9% (95% Co n f i d en ce In terva l , wound care servi ces continues to grow, wh i ch su g-
28.3% to 31.4%) in non-acute care set ti n gs, 22.1% (95% gests that pre s sure ulcers are a relatively common
Co n f i d en ce Interval, 20.9% to 23.4%) in mixed health h e a l t h c a re con cern and an escalating probl em.
settings, and 15.1% (95% Co n f i d en ce In terva l , 13.4% to Statistics are ava i l a ble rega rding the number of pre s-
16.8%) in co m munity care . The overa ll estimate of the sure ulcers in the US1 as well as for other co u n tries
preva l en ce of pre s sure ulcers in all healthcare institutions of the world.2 However, little information is available
Dr. Woodbu ry is an Inve s tiga to r, Lawson Health Research Institute, Pro gram in Rehabilitation and Geri a tric Care, London, Ontario,
Canada; and Adjunct Assistant Professor, Depa rtment of Epidemiol o gy and Biostati s ti cs , Univers i ty of Western Ontario, London,
On t a ri o, Canada. Dr. Hou gh ton is As so ciate Professor, Pro gram in Physical Therapy, Univers i ty of Western On t a rio; and an
Investigator, Lawson Health Research Insti tu te. Pl e a se address co rrespondence to: Dr. M. Gail Woodbu ry, Lawson Health Research
Institute, Pro gram in Rehabilitation and Geri a tric Care, St. Jo seph’s Healthcare London, Park wood Hospital Site, 801 Commissioners
Road East, London, On t a ri o, Canada N6C 5J1; email: Gail.Woodbu ry@sjhc.london.on.ca or gwood bur@uwo.ca.
22 OstomyWound Managementabout the nu m ber of i n d ivi duals in Ca n ada wh o in regions ac ross Ca n ad a . Prevalen ce refers to the
h ave pre s su re ulcers. Na ti onal esti m a tes for the nu m- proporti on of a group (pati ents not ulcers) that has a
ber of pre s su re ulcers in va rious healthcare settings pre s sure ulcer at a given single point in time or ti m e
in regions ac ross Ca n ada are non ex i s tent. Wi t h o ut period during wh i ch the cases are co u n ted. A cross-
this informati on , estimating costs to the Ca n adian s ecti onal stu dy is the appropri a te design for deter-
h e a l t h c a re sys tem assoc i a ted with managing ch ronic mining the number of patients with pre s sure ulcers
wounds is not possible. f rom the number of patients assessed.
Pre s su re ulcers are not nati onally recognized as an
important healthcare probl em in Ca n ad a . Cu rrently, Methods
little national or provincial funding is ava i l a ble to Data collection. Bet ween Ja nu a ry 2003 and
provide coord i n a ted healthcare delivery programs November 2003, all available data from preva l ence
for their preventi on and management or to prom o te s tudies con du cted bet ween 1990 and 2003 were col-
the development of edu c a tional programs for health- lected. Several sources for locating studies on the
c a re profe s s i onals. Furt h ermore , nationally funded prevalen ce of pre s sure ulcers in all healthcare set-
grants to su pport re s e a rch programs for identifying ti n gs and in the gen eral pop u l a ti on were inve s tigat-
the underlying cause of chronic wounds and estab- ed, including peer- reviewed publ i s h ed reports,
lishing new interventions and innovative healthcare unpublished stu d i e s , and wound care com p a ny data-
delivery models are ra re . To focus national attention bases. For studies in wh i ch preva l ence could be sepa-
and re s o u rces on this serious and growing healthcare rated by facility or facility type, each facility or facili-
probl em in Ca n ad a , national aw a reness abo ut pre s- ty type was tre a ted as a distinct stu dy.
su re ulcers must be raised. Gathering facts and stati s- Sys tematic com p uter and manual searches of
tical data that describe the ex tent of the problem in l i brary databases PubMed (Medline ®) and CINA H L ®
Ca n ada is essential to the su ccess of lobbying health- were con ducted using the keywords ulcers, Canada,
c a re administrators and govern m ent officials and for and prevalence to locate studies invo lving all health-
informing the general Ca n adian pop u l a tion about c a re settings and pop u l a ti ons publ i s h ed in peer-
the extent of the probl em. reviewed journals. Few published arti cles describing
Recently, the Na tional Pre s sure Ul cer Advisory the prevalen ce of pre s su re ulcers in Ca n ada were
Panel (NPUAP) completed a large stu dy describing fo u n d . Af ter locating the arti cl e s , all referen ces were
the prevalen ce of pressu re ulcers in the US.3 In reviewed and re s e a rchers in this field were con t acted
Ca n ada, many nati onal or ganizations that
su pport other com m on disease con d i tions
Ostomy/Wound Management 2004;50(10):22–38
su ch as diabete s , c a rd i ovascular disease,
and cancer have inve s ted significant
human and financial re s o u rces to develop KEY POINTS
and maintain large nati onal registries. • The problem of pressure ulcers spans the continuum of
Cl e a rly, ga t h ering this information is no healthcare settings and affects a wide variety of healthcare
small task. It takes ye a rs to orga n i ze , col- professionals.
lect, and co llate the data. • To obtain nationwide pressure ulcer prevalence estimates, the
G iven the enormity of the task, the authors obtained published and unpublished prevalence data
obtained in Canada between 1990 and 2003 using actual skin
authors believed an important first step
assessments.
was to sys tematically search and identify • The overall prevalence rate was high (26%) with higher rates
ex i s ting data on the prevalence of pre s sure (29%) in non-acute and lower rates (15%) in community care
u l cers in Ca n ada. Specifically, the goal of settings.
this project was to determine, f rom cur- • The differences between these findings and those reported
rent available information, the prevalen ce from other countries warrant further examination because
they may be the result of study methodology or patient/care
of pre s sure ulcers in different care set ti n gs
differences.
October 2004 Vol. 50 Issue 10 23TABLE 1 term mixed health settings
QUESTIONS FOR CRITICA L LY APPRAISING STUDIES OF refers to preva l ence stu d-
PREVALENCE OF PRESSURE ULCERS IN A HEALTH SETTING ies in settings that con s i s t
of a mixtu re of ac ute ,
A.Are the study methods valid? non - ac ute and/or com-
1. Is the sample random or the whole population surveyed? mu n i ty care healthcare
2. Is the study design prospective? Is a physical examination performed? delivery models; the
3. Is the sample size adequate (>300 subjects)? prevalen ce esti m a te is
4. Are objective, suitable,standard methods used for measurement of pressure ulcers? ex pre s s ed overall rather
5. Is the outcome measured in an unbiased fashion? than bro ken down by spe-
6. Is the response rate adequate? Are the refusers described? cific setting type.
B. What is the interpretation of the results?
The methodo l ogy used
7. Are the estimates of prevalence given with confidence intervals?
to co llect prevalen ce
8. Are the estimates of prevalence given in detail by subgroups?
i n formation in all stu d i e s ,
C. What is the applicability of the results?
9. Are the study subjects and the setting described in detail and similar to those of
p u bl i s h ed and unpub-
interest to you? lished, was critically
a ppraised using a mod i-
Each question is scored 0 (no) or 1 (yes) to yield a Methodological Score ranging from 0-9.
f i ed version of recom-
m en ded cri teria for eva lu-
to loc a te peer- reviewed arti cles that might have been ating preva l en ce studies. The original critical
4
missed. a ppraisal work by Lon ey et al4 relates to pati ents with
Ma ny members of the Ca n adian Association for dementia and consists of a series of questi ons that
Wound Ca re (CAWC) re s pon ded to a general web s i te a re more appropri a te for health probl ems that can be
request for information and provi ded unpubl i s h ed evalu a ted using pop u l a tion su rveys than for ch ron i c
Ca n adian stu dy data. Approximately 50 people were wo u n d s , wh i ch are generally eva lu a ted in healthcare
contacted; of ten, the search for studies requ i red con- f ac i l i ty settings. Hen ce , it was necessary to modify
tact with several people before the correct con t act the questions to reflect the most com m on stu dy situ-
was made. ations in wh i ch pati ents in healthcare facilities were
Several wound care companies have large databas- evalu a ted, of ten by physical examinati on over a rela-
es of prevalen ce and/or inciden ce studies performed tively short time. The critical appraisal questi on s
as a service for their customers . One company, used in the pre s ent report are listed in Table 1.
Ki n etic Con cepts, In c . (KCI Medical Ca n ada, In c . , The aut h ors indepen den t ly determined a method-
Mi s s i s s a u ga, Ontario) ex pended great ef fort to con- ological score for each stu dy by assigning each of the
tact individual consu m ers to address proprietary and nine nu m bered questions a score of one (1) point if
con f i dentiality issues in order to share this va luable a ppropri a te methods were used. The final method-
information. The re sults of i n d ivi dual studies con- ological score for each stu dy was obt a i n ed by con-
du cted in Ontario and Quebec were made available. s en sus, with higher score studies repre s enting more
In addition, aggrega te data (without facility names a ppropri a te and rigorous re s e a rch methods and less
and without pati ent information) were provi ded by po tential bias in the re su l t s . Therefore, prevalen ce
KCI and Hi ll - Rom Ca n ada (Mi s s i s s a u ga, Ontario). esti m a tes from studies with higher methodological
Definiti ons and critical appraisal. Because of s cores can be accepted with more con f i den ce .
regional differences and recent ch a n ges in terminolo- Questi ons 2, 4, and 5 of the critical appraisal (see
gy, the term non-acute care has been used in this Table 1) were the most vital. Studies were con s i dered
report to inclu de the fo ll owing types of c a re setting: to lack validity if: 1) re s ponses indicated that skin
su b ac ute care, ch ronic care , com p l ex continuing u l cers were co u n ted by methods other than direct
c a re , long-term care (LTC ) , and nu rsing hom e . The physical skin assessment, 2) outcome measu res used
24 OstomyWound ManagementTABLE 2 non - ac ute care
OUTLINE OF THE RECEIVED STUDIES were received.
Acute care Non-acute Community Mixed Across 65
care care health h e a l t h c a re
12
f ac i l i ties/insti-
Number of studies received 23 4 6
Number of facilities 18 23 5 19 tutions, 14,102
Total number of patients 4,831 3,390 1,681 4,200 patients were
Years studies conducted 1990-2002 1990-2003 1990-2003 1990-2003 evalu a ted in
Sample size: prevalen ce
Mean 439 206 420 700 s tu d i e s ,
Minimum - maximum 58-1,525 65-768 29-1,466 202-2,384 demon s trating
Methodological score 0-9 a wi de range of
Mean 6.3 6.5 5.6 6.6 sample size s
Range 2 to 8 2 to 7 3.5 to 6.5 6 to 7 ( bet ween 29
Number of excluded studies 1 5 0 0 and 2,384). No
(ScoreTABLE 3
PUBLISHED CANADIAN STUDIES
References Subjects Design/method Facility Prevalenc Data Metho
type e source d score
Foster et al, N=2,384 from three Patients assessed over 1 day, one skin Overall 25.7 Clinical 6.5
19927 teaching hospitals, one care committee nurse/unit as survey- Acute care 27
community hospital, two or, surveyors trained Extended care 30
long-term care facilities, Community 13
two community health agencies
agencies in Ontario
Harrison et al, N=738 from acute care Patients assessed within 12-hour peri- Acute care 29.7 Clinical 8
19968 740-bed facility in od (skin breakdown, risk); conducted (95% CI
Ontario midweek to reflect accurate mix of 26.4-33.0)
admissions and case types, in
September to avoid seasonal fluctua-
tions. Education workshop; survey
team trained, validated; 10% reliability
checked
Fisher et al, 19969 N=1,020 from two acute Patients assessed within 12-hour peri- Acute care 23.9 Clinical 8
care hospitals in Ontario od (skin breakdown, risk) by survey
teams of RNs. Study conducted mid-
week to reflect mix of new admis-
sions and long stay operative cases.
ET therapists available for difficult to
classify ulcers
McNaughton & N=210 and 202 from one Pre and post intervention. 2-week peri- Chronic care Year 1:32.4 Clinical 5
Brazil, 199510 facility in Ontario od prevalence: Survey of all patients Year 2: 22.3
by nurses to locate ulcers; ulcer physi-
cal assessment using standard form
Nicolle et al, N= 198 and 259 from two Prospective surveillance for 2-year Long-term 2.8 and 3.5 Clinical 2
199412 long-term care facilities in period. Patients with decubiti identi- care
Ontario fied at the beginning of the study peri-
od; surveillance and data collection
including microbiological studies by
study nurse who visited the facility at
least twice per week. Residents with
ulcers followed until ulcer healed, dis-
charge, or death, or participation ter-
mination
D'hoore et al, N=13,555 from long-term Retrospective analysis of administrative Long-term 4.0 Database 2
199713 care facilities in Quebec, data set for year 1993-1994. care
except psychiatric or “Required nursing time measurement
totally private centers tool” identified nursing actions
required. Pressure ulcer existence
inferred from evaluation of required
nursing actions, treatment of pressure
sores; distinction between Stage I and
Stage II versus Stage II and Stage IV
Davis & Caseby, N= 95 and 92 from two Patients assessed by KCI member, one Long-term 36.8 and Clinical 7
200111 long-term care facilities in facility nurse, one healthcare care 53.2
Ontario aide/team, on 1 day, standard proce-
dures for presence and number of
ulcers
28 OstomyWound Managementsitu ati on. S pecifically,
the Quebec LTC data-
base was analyzed
retro s pectively to
determine the rela-
tionship between
pre s en ce of pre s sure
sores and nursing
workload.13 The use
of a retro s pective
analysis rather than a
cro s s - s ecti onal stu dy
design provi ded an
imperfect situ a ti on
Figure 1
Prevalence by methodological score in non-acute care. for determining
prevalen ce because
methods (eg, standard definitions of pre s su re ulcer specific data on the pre s en ce of pressu re ulcers were
pre s en ce and staging) to ensure that assessments not recorded. The pre s en ce of pre s su re ulcers was
were done reliably and without bias.8,9 These two inferred from an eva lu a tion of required nu rsing
studies produ ced prevalen ce esti m a tes of 29.7% and acti ons for the tre a tm ent of pressure ulcers . The ret-
23.9%, respectively. ro s pective stu dy design and non s t a n d a rd measu re-
The on ly Ca n adian stu dy estimating the preva- ment of outcome adversely affected the methodo l og-
len ce of pressu res ulcers in a ch ronic care facility in ical score obt a i n ed in the critical appraisal. These
Ontario was con du cted before 1995, with pre and t wo latter studies had poor methodo l ogy scores and
post wound care pro tocol evaluati on s . 10 The stu dy ’s were exclu ded from the su m m a ry data.
methodological score of 5 was affected by the rela- Unpublished studies. Thirty - t h ree unpubl i s h ed
tively small sample size and by the fact that patients s tudies were received from 19 peop l e . Of these, 30
with ulcers were identified
based on pati ent survey TABLE 4
re s ponses obt a i n ed by the PREVALENCE ESTIMATE:ACUTE CARE*
nu rses before direct phys i-
Study Year Sample Methodologic Prevalenc Prevalenc
cal ex a m i n a ti on. size al score 0-9 e e
The stu dy of pre s su re
Ontario 1a 1990 1,525 6.5 27
u l cers in LTC with the
Ontario 2 1993 738 8 29.7 13.6
highest met h odo l ogical
Ontario 3 1994 1,020 8 23.9
score (7) produ ced high Nova Scotia 1 1995 233 5 26.2 9.4
e s ti m a tes of preva l en ce — British Columbia 1a 1997 58 5 15.5
36.8% and 56.2%.11 A lower Newfoundland 1 2002 203 5 4.9
e s ti m a te was obtained in Ontario 4 1998 135 6 26.7 17.0
a n o t h er LTC stu dy in wh i ch British Columbia 2 2000 250 5 34.8
u l cers were assessed using KCI 4 2002 133 7 18.8
non-standard measu re s . A 12
KCI 5 2000 112 7 34.8
t h i rd stu dy in LTC , wh i ch KCI 11 2002 424 7 16.3
was not con du cted primari- *Sample size, methodological score, and prevalence estimates of individual published,
ly to determine preva l en ce , unpublished and KCI clinical studies
c re a ted an unfavora bl e These studies were numbered within each province; letters were applied to studies repre-
sented in different settings.
30 OstomyWound Managementwere accepted for use in TABLE 5
this report. In clu ded with P R E VALENCE ESTIMAT E: NON-ACUTE CARE*
the unpubl i s h ed stu d i e s Study Year Sample Methodologic Prevalenc Prevalenc
were 11 indivi dual stu d i e s size al score 0-9 e e
provided by a company — Ontario 5a 2000 95 7 36.8
t h ree repre s en ting ac ute Ontario 5b 2000 92 7 53.3
c a re and eight repre s en t- Ontario 6 1993 210 5 32.4
ing non - ac ute care . All Ontario 1b 1990 768 6.5 30.0
unpublished studies were Nova Scotia 2 1995 206 6 31.6 13.1
critically evalu a ted and the Newfoundland 2 2002 143 5 4.2
re sults com bined with the British Columbia 3 2000 136 5 13.2
p u bl i s h ed stu d i e s . British Columbia 4 2000 120 5 16.7
British Columbia 1b 1997 229 7 10.9
Af ter excluding the
British Columbia 1c 1997 143 7 5.6
s tudies with scores of 2 or
KCI 1 2002 142 7 39.4
less and those with nega- KCI 2 2003 65 7 29.2
tive answers to the three KCI 3 2002 157 7 20.4
key methodological ques- KCI 6 1999 92 7 53.3
ti on s , the mean met h od- KCI 7 2001 142 7 26.1
ological scores for the pub- KCI 8 2002 180 7 43.3
lished and unpubl i s h ed KCI 9 2001 231 7 41.6
s tudies ac ross the fo u r KCI 10 2002 239 7 51.0
h e a l t h c a re set ti n gs ra n ged *Sample size, methodological score, and prevalence estimates of individual published,
from 5.6 to 6.6 (maximum unpublished and KCI clinical studies
= 9) (see Ta ble 2). These studies were numbered within each province; letters were applied to studies repre-
sented in different settings.
Prevalen ce estimates.
Esti m a tes of prevalen ce
f rom studies with poorer TABLE 6
methodological scores PREVALENCE ESTIMATE: COMMUNITY CA R E *
ten ded to be lower than
Study Year Sample Methodologic Prevalenc Prevalenc
esti m a tes obt a i n ed from size al score 0-9 e e
s tudies in wh i ch bias was
Ontario 1c 1990 91 6.5 13.2
contro lled (see Figure 1).
British Columbia 1d 1997 29 6 24.1
Therefore , studies with Saskatchewan 1 2001 95 3.5 15.8
s cores 2 were exclu ded. Manitoba 1 2003 1,466 6.5 15.0
The prevalen ce esti-
*Sample size, methodological score, and prevalence estimates of individual published,
mates of the indivi du a l unpublished and KCI clinical studies
p u bl i s h ed, unpublished, These studies were numbered within each province; letters were applied to studies repre-
and KCI studies were sented in different settings.
su m m a ri zed for acute
c a re , non-ac ute care, com mu n i ty, and mixed health CI, 23.8% to 26.3%) (see Figure 2). The 95% con f i-
s et tings (see Tables 4 to 7, re s pectively). The overall den ce limits for each esti m a te were narrow (less than
mean prevalen ce for the healthcare settings ra n ged t wo percen t a ge points from the esti m a te ) , reflecti n g
f rom 15.1% (95% CI, 13.4% to 16.8%) in com mu n i- the large sample sizes that re su l ted from the com bi-
ty care , to 29.9% (95% CI, 29.3% to 31.4%) in non - nation of s tu d i e s . Conversely, a large range of values
ac ute care, with mixed health setting at 22.1% (95% was noted bet ween the minimum and maximum
CI, 20.9% to 23.4%) and ac ute care at 25.1% (95% esti m a tes reported for the pressu re ulcer prevalen ce
32 OstomyWound ManagementTABLE 7
PREVALENCE ESTIMATE: MIXED HEALTH SETTINGS* of each i n d ividual stu dy.
Study Year Sample Methodologic Prevalenc Prevalenc Because the con f i den ce
size al score 0-9 e e limits for different health-
Ontario 1d 1990 2384 6.5 25.7 11.0 c a re set ti n gs do not over-
Ontario 7 2001 406 7 13.1 9.4 l a p, the esti m a tes in these
Ontario 8 1998 310 7 21.6 13.5 set ti n gs are sign i f i c a n t ly
British Columbia 5 1996 202 6 10.9 - different. Wh en all data
Nova Scotia 3 1995 439 6 28.7 11.2 a re com bined to report
British Columbia 1e 1997 459 6 10.7 6.7 overa ll preva l en ce (rega rd-
*Sample size, methodological score and prevalence estimates of individual published, less of s et ting and avoi d i n g
unpublished and KCI clinical studies repre s enting stu dy data
These studies were numbered within each province; letters were applied to studies repre-
sented in different settings. more than on ce ) , the mean
preva l en ce is higher at
26% (95% CI, 25.2-
26.8%), based on 10,911 su bj ects.
Aggregate data prev a l ence
estimates. KCI provi ded ye a rly
preva l en ce esti m a tes for 61 ac ute
c a re facilities from studies con-
ducted from 1997 to 2003.
Because similar re s e a rch meth-
ods were used to co llect these
preva l en ce esti m a te s , they can be
com p a red over time (see Figure
3). Esti m a tes of pre s sure ulcer
preva l en ce were found to
decrease gradually from 42% in
Figure 2 1997 to 29% in 2002.
Estimates of pressure ulcer prevalence in various healthcare settings (95% Confidence Re sults of preva l ence stu d i e s
Interval bars). con du cted by Hill - Rom in 35
acute and non - ac ute Ca n ad i a n
f ac i l i ties — ie, mixed healthcare
s et tings — from 2001 to 2003
produ ced a preva l en ce esti m a te of
14.9% based on 6,828 pati en t s .
Discussion
This proj ect provi ded rel i a bl e
esti m a tes of pressu re ulcer preva-
l en ce for four healthcare settings
in Ca n ada. The esti m a te s , wh i ch
h ave narrow con f i dence intervals,
a re based on individual stu d i e s
that were critically appraised and
found to be methodo l ogically
sound, yielding large combi n ed
Figure 3 samples from ac ross the co u n try
Acute care prevalence: KCI Studies 1997-2002.
34 OstomyWound Managementfor most healthcare facility types.
The preva l en ce esti m a tes obt a i n ed and a de s i red
n a rrow 95% CI width of 10% can be used to ascertain
the sample size requ i rem ents for con ducting a futu re
preva l en ce stu dy in each of the healthcare setti n gs,
using the formula su gge s ted by Baumgarten. 5 (The
acceptance of a 95% CI wi der than 10% re sults is a
trade - off in precision for a smaller sample size
requirement.) The requ i red sample size esti m a te s
a re : 289 acute care subjects, 322 non - ac ute care su b-
jects, 197 com munity care su bjects, and 265 mixed
h e a l t h c a re su bjects. In other words, in set ti n gs in
wh i ch the prevalen ce is anti c i p a ted to be above 25%,
a pprox i m a tely 300 su bjects are needed, while fewer
su bjects are needed if the prevalence is anti c i p a ted to
be as low as 15%. These sample size calculati ons lend
su pport to the methodo l ogical criterion su gge s ted by
Lon ey et al4 that appropri a te sample size for preva-
len ce studies is gre a ter than 300 su bjects to all ow for
the possibility that the prevalen ce is as high as 25%.
Com p a ring esti m a tes with those from other co u n-
tries su ggests that the esti m a te for the nu m ber of
patients with pre s sure ulcers in acute care (25.1%) is
higher than two previ o u s ly reported pre s sure ulcer
prevalen ce esti m a tes from across the US.1,3 The
NPUAP pre s sure ulcer preva l en ce esti m a te ranged
bet ween 10% and 17.1%.3 Wh i t tington et al1 reported
a 15.1% preva l en ce of pre s sure ulcers from a series
of studies con du cted by KCI in acute care facilities
ac ross the US. The differen ces bet ween Ca n adian and
US esti m a tes of preva l en ce of pre s sure ulcers in
ac ute care facilities might be due to different
methodologies em p l oyed and the time period over
wh i ch the data were collected. However, US and
Ca n adian esti m a tes generated by KCI in 1999 using
i dentical methods found a pre s sure ulcer prevalen ce
esti m a te in Ca n ada of 27% (see Figure 3) that was
con s i derably higher than the 15.1% va lue reported in
US.1 Therefore, these esti m a tes may repre s ent true
d i f feren ces bet ween the Ca n adian and US healthcare
s ys tems. Ot h er possible explanations inclu de differ-
en ces in the sample sizes and pati ent prof i l e s . The US
KCI esti m a te reported by Wh i t tington was limited to
adult pati ents in medical-surgical and intensive care
units, while samples su rveyed in Ca n adian healthcare
instituti ons in this report inclu ded more units.
The stu dy prevalen ce esti m a te for pati ents in non -ac ute care facilities in Ca n ada (29%) is close to the upper va lue in
the ra n ge reported by the NPUAP, 2.3% to 28%,3 and to another
recently publ i s h ed preva l ence esti m a te (28%).14 Horn et al14 used a
retro s pective co h ort sample of 2,420 pati ents who were at risk of
developing pre s su re ulcers as indicated by Braden scores of 17 or
less. One would have expected the current esti m a tes for pressu re
u l cer prevalen ce in all pati ents within Ca n adian non - ac ute care fac i l-
ities to be mu ch higher than esti m a tes from the US that examined
on ly pati ents who had alre ady been identified as at risk of develop-
ing pressu re ulcers . Di f feren ces bet ween current esti m a tes in Ca n ad a
and US pre s su re ulcer prevalen ce esti m a tes may be due to the use of
d i f ferent methodologies — the US stu dy used retro s pective analys i s
ra t h er than direct skin ex a m i n a ti on from wh i ch the Ca n adian esti-
m a tes in this report were derived. The aut h ors found that stu d i e s
that used retro s pective data analysis received lower methodo l ogical
s cores and tended to produ ce lower prevalen ce esti m a te s ; therefore,
these studies may be more likely to underesti m a te true prevalen ce of
pre s sure ulcers .
Di f feren ces bet ween Ca n adian and US esti m a tes of pre s sure ulcer
prevalen ce also may be explained by the fact that the term non-acute
c a re encompasses a po tentially diverse pop u l a ti on. The de s c riptors
for healthcare facilities with non - ac ute pati ents have ch a n ged over
the past several ye a rs and are not uniform in different regions ac ro s s
Ca n ada. They inclu de LTC , nu rsing hom e s , com p l ex con tinuing care ,
skilled nu rsing facilities, reh a bi l i t a tion, and geriatrics. Because of the
varying terminology, the re sults were com bi n ed into a generic cate-
gory: non-ac ute care . It is possible that the true prevalen ce in any
particular subgroup within this classificati on may be masked by this
h e a l t h c a re set ti n g’s diversity.
The national estimate of pressure ulcer prevalence in Canada (26%)
is slightly higher than the estimate reported from a national study done
in the Netherlands and considerably higher than the international
aggregate estimate for 2003 provided by Hill-Rom. The epidemiological
study performed in the Netherlands examined 16,344 patients and pro-
duced an overall estimate of 23.1% for all the health settings.2 The inter-
national estimate from Hill-Rom was 15.5%, based on 61,427 surveyed
patients in 461 facilities of all types. The majority of these Hill-Rom
international studies were conducted in the US.
Limitations
The divers i ty of the non - ac ute care pop u l a tion that was com bi n ed
for the pre s ent project may have re su l ted in an esti m a te for this
gen eric sample that is not acc u ra te for any of the su b groups, (eg,
LTC facilities, nu rsing hom e s , com p l ex con tinuing care , skilled nurs-
ing facilities, rehabilitation, and geriatrics).
Most stu d i e s , p u bl i s h ed and unpublished, reported insu f f i c i ent
information to answer all the methodo l ogical questions abo ut thes tu d i e s . Some gaps in informati on were filled by References
additi onal contacts with proj ect aut h ors but when 1. Whittington K, Patrick M, Roberts JL. A national
the information was not ava i l a ble or aut h ors could study of pressure ulcer prevalence and incidence in
acute care hospitals. J Wound Ostomy Continence
not be contacted, it was assu m ed that rigorous met h-
Nurs. 2000;27(4):209-215.
ods were not performed or that certain re sults were 2. Bours GJ, Halfens RJ, Abu-Saad HH, Grol RT.
not obt a i n ed. Prevalence, prevention, and treatment of pressure
The data obt a i n ed do not repre s ent all data co l- ulcers: descriptive study in 89 institutions in the
lected from Canadian healthcare settings. No esti- Netherlands. Res Nurs Health. 2002;25(2):99-110.
mates of prevalen ce in acute and non - ac ute care 3. National Pressure Advisory Panel, Cuddingan J, Ayello
EA, Sussman C, eds. Pressure Ulcers in America:
came from the Prairie Provinces; therefore, one mu s t
Prevalence, Incidence, and Implications for the Future.
a s sume that the preva l en ce of pre s sure ulcers in these Reston, Va.: NPUAP;2001.
t wo settings would be similar in these provinces. The 4. Loney PL, Chambers LW, Bennett KJ, Roberts JG,
information received is on ly a sample from relatively Stratford PW. Critical appraisal of the health research
few insti tuti ons across Ca n ada. It is possible that literature: prevalence or incidence of a health prob-
f ac i l i ties electing to do prevalen ce studies do so when lem. Chronic Dis Can. 1998;19(4):170-176.
5. Baumgarten M. Designing prevalence and incidence
they su s pect a po tential problem and this might con-
studies. Adv Wound Care. 1998; 11(6):287-293.
tri bute to the higher prevalen ce reported in relation 6. Hamilton L, Cleverly S. A skin care resource nurse
to other co u n tries. program: is it effective? Perspectives. 1994;18(1):10-14.
All pressure ulcer prevalence estimates obtained for 7. Foster C, Frisch SR, Forler Y, Jago M. Prevalence of
this project are within the healthcare sector; no nation- pressure ulcers in Canadian institutions. CAET
al estimate was secured for the general population. Journal. 1992;11(2):23-32.
8. Harrison MB, Wells G, Fisher A, Prince M. Practice
Several national databases, such as the National
guidelines for the prediction and prevention of pres-
Population Health Survey (NPHS), were reviewed in sure ulcers: evaluating the evidence. Appl Nurs Res.
vain to find a statistic for the population. 1996;9(1):9-17.
9. Fisher A, Denis N, Harrison MB, et al. Quality man-
Conclusion agement in skin care: understanding the problem of
This project provi des preva l ence esti m a tes for pressure ulcers. Canadian Journal of Quality in
Healthcare. 1996;13(1):4-11.
pre s sure ulcers in various Ca n adian healthcare set-
10. McNaughton V, Brazil K. Wound and skin team.
ti n gs of 15% to 30%, and an overall esti m a te of 26%. Impact on pressure ulcer prevalence in chronic care. J
These esti m a tes seem to be higher than esti m a tes Gerontol Nurs. 1995;21(2):45-49.
f rom the US and the Netherlands, perhaps because of 11. Davis CM, Caseby NG. Prevalence and incidence
the trend in the Ca n adian healthcare sys tem to limit studies of pressure ulcers in two long-term care facili-
hospital admission and redu ce length of s t ay; there- ties in Canada. Ostomy Wound Manage.
2001;47(11):28-34.
by, re su l ting in sicker pati ents within the sys tem.
12. Nicolle LE, Orr P, Duckworth H, et al. Prospective
This informati on will be useful to clinicians, study of decubitus ulcers in two long-term care facili-
re s e a rchers, and policy makers in Ca n ada and other ties. Can J Infect Control. 1994;9(2):35-38.
co u n tries to advoc a te for the needs of pati ents with 13. D’Hoore W, Guisset AL, Tilquin C. Increased nursing-
ch ronic pre s su re ulcers. Ad d i ti onal information is time requirements due to pressure sores in long-term-
needed abo ut the preva l en ce of other types of care residents in Quebec. Clinical Performance and
Quality Healthcare. 1997;5(4):189-194.
wo u n d s . In a recent sys tematic revi ew of preva l en ce
14. Horn SD, Bender SA, Bergstrom N, et al. Description
of lower limb ulcers, the preva l en ce of venous leg of the National Pressure Ulcer Long-Term Care Study.
u l cers was determined to be 0.12% to 1.1% based on J Am Geriatr Soc. 2002;50(11):1816-1825.
the studies that employed clinical validation of 15. Graham ID, Harrison MB, Nelson EA, Lorimer K,
u l cers.15 More studies are needed to esti m a te the Fisher A. Prevalence of lower-limb ulceration: a sys-
nu m ber people in the general pop u l a ti on with tematic review of prevalence studies. Adv Skin Wound
Care. 2003;16(6):305-316.
ch ronic wounds of any cause. - OWM
38 OstomyWound ManagementYou can also read
