Content uploaded by Heidi Sveistrup
Author content
All content in this area was uploaded by Heidi Sveistrup
Content may be subject to copyright.
Technology and Disability 18 (2006) 45–55 45
IOS Press
Evaluation of bath grab bar placement for
older adults
Heidi Sveistrupa,b,∗, Donna Lockettb, Nancy Edwardsb,cand Faranak Aminzadehb,d
aSchool of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa, Ottawa, ON, Canada
bCommunity Health Research Unit, Ottawa, ON, Canada
cSchool of Nursing, Faculty of Health Sciences, University of Ottawa, Ottawa, ON, Canada
dRegional Geriatric Assessment Program of Ottawa-Carleton, Ottawa, ON, Canada
Abstract. The patterns of use, perceived usefulness and perceived safety of five different configurations of bathtub grab bars were
evaluated by 103 community-living seniors in Canada. Current bathing activities, fall history, sociodemographic characteristics,
balance measures as well as details about the home bathing environment were recorded. Participants were then videotaped as they
got into, sat down, got up and exited a bath tub using each of the five configurations. The videotapes were used to determine the
pattern of grab bar use for each configuration and participants ranked each configuration for perceived function and safety. The
five configurations corresponded to standards published by the Canadian Standards Association, the US Uniform Accessibility
Standards, a modification of the Ontario Building code (OBC, a Canadian provincial code), a “common configuration” and a
composite configuration.
Although most respondents did not have bathtub grab bars installed in their home, those with home bars reported that they used
the bars on a regular basis. Significant differences in mean ratings of safety, comfort, ease of use, helpfulness, likelihood of use,
and total composite score were detected between configurations with the modified OBC configuration consistently ranked least
favourable. There were no statistically significant relationships between the respondents’ profiles, their preferred configurations
or their patterns of bar use. A series of recommendations and suggestions for future research are made.
Keywords: Aging, assistive devices, bathing
1. Introduction
International studies have documented that approxi-
mately one-third of community-living older adults re-
port at least one fall each year [13,26]. Self reports in-
dicate that 25% to 77% of falls occur inside the home,
with bathrooms being one of the most common loca-
tions [17,22]. A recent study reported that 55% of bath-
room falls reported by a sample of 550 community-
living older adults in Canada occurred while bathing,
with 70% of these falls occurring during unsuccessful
∗Address for correspondence: Heidi Sveistrup, School of Reha-
bilitation Sciences, Faculty of Health Sciences, University of Ot-
tawa, 451 Smyth Road, Ottawa, ON, Canada K1H 8M5. Tel.:
+1 613 562 5800 (8016); Fax: +1 613 562 5428; E-mail: Heidi.
Sveistrup@uottawa.ca.
transfers [2]. Moreover, almost one-third of the study
participants were restricted in their bathing practices
because they had difficulty with bath transfers includ-
ing getting into or out of the tub and sitting into or
getting up from the bottom of the tub.
Bathroom aids, including bath grab bars, can com-
pensate for the effects of age-related functional limita-
tions such as impaired balance, poor coordination, lim-
ited range of motion, and reduced muscular strength,
allowing for safe and independent bathing among el-
derly persons [4,31]. Bathroom aids are among the as-
sistive devices most commonly owned by community
living seniors [11,12,24,28,30,33]. In a randomized
controlled trial of over 1500 Canadian older adults, the
installation of grab bars and use of raised toilet seats
were the most frequently reported home safety modifi-
cations made by the participants in the year preceding
ISSN 1055-4181/06/$17.00 2006 – IOS Press and the authors. All rights reserved
46 H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults
the study [9]. Moreover, when safety modifications
to bathrooms are made, the average senior’s home re-
ceived two grab bars [25]. Even seniors who do not
own bathroom aids identify the need for these devices.
North American studies have identified grab bars as the
device for which there is the greatest need [18], as a
high priority basic safety feature for all dwellings [7]
and as a device that would be used if accessible [1].
Findings from an earlier study by our team [3] re-
ported that 87% of seniors who had access to bath grab
bars used them on a regular basis. Grab bar use was
greatest among those with two or more grab bars that
were easily graspable. Despite high reportedutilization
patterns, data from qualitative descriptions of falls re-
lated to bath transfers suggested that in all but one fall,
seniors had not been using their grab bar to facilitate the
transfer. The most common reasons cited by seniors for
not using bath grab bars included feeling that the bars
were awkward to use or that they were unsafe. Logistic
regression analysis identified ease of grab bar use as
a significant predictor of actual use [10]. Moreover,
in the absence of appropriate bath grab bars, seniors
reported using potentially hazardous contacts to facili-
tate transfers while bathing including soap dishes, bath
rims, and shower curtain rods. In some circumstances
seniors engaged in hazardous practices in their attempt
to use the bath grab bars such as reaching across bath-
tubs to grab bars on the sidewall of the tub which would
leave them vulnerable to losing their balance, slipping
and/or falling.
Most building standards reflect the consensus of pro-
ducers and/or are based on the needs of people with
disabilities. However, residential buildings are typi-
cally exempt from accessibility regulations. For exam-
ple, Canadian standards for bath grab bars [8] were de-
signed to reflect a national consensus of producers and
users and originally developed for people with disabili-
ties. However, none of the 514 seniors’ baths examined
in a previous study were equipped with a grab bar con-
figuration that matched the Canadian Standards Asso-
ciation (CSA) standards [2]. There is no evidence that
accessibility standards intended to promote indepen-
dent bathing among people with mobility impairments
meet the functional capabilities of older adults.
These findings highlight the need for a greater under-
standing of optimal bar placement for seniors. The pur-
pose of this study was to determine how seniors would
perceive and how they would use bathtub grab bars for
transfer tasks when provided with the opportunity to
try five distinct configurations.
2. Methods
2.1. Subjects
A convenience sample of community-dwelling se-
niors who could independently enter and exit their bath-
tub, were over the age of 60 years, and obtained a
minimum of 20 on the Folstein Mini-Mental Exam for
cognitive screening1participated in the study.
2.2. Bathing history, balance and mobility
Participants completed a series of questions about
their health and activity levels, current bathing prac-
tice, fall history, sociodemographic characteristics and
home bathing environment. Two clinical measures of
balance were then obtained: timed get-up-and-go [19]
and one-legged stance [5].
2.3. Bath grab bar configurations
A bathtub testing area with interchangeable grab bar
configurations was designed to mimic a small residen-
tial bathroom. The 3.78 by 2.3 metre room included
a bathtub (1.52 ×0.76 ×0.41 metre) and an adjacent
wall-mounted sink with a mirror. Two cameras were
set to record hand positions during testing. The five
grab bar configurations tested are described below (see
Fig. 1). The bath grab bar configurations were installed
with respect to the participant’s home bath. For ex-
ample, the vertical bar in the CSA configuration was
always installed on the end wall corresponding to the
participant’s home faucet location. Testing was com-
pleted with dry bars since the laboratory area was not
setup for wet conditions. The bathtub grab bars were
constructed of brushed satin stainless steel with a non
slip surface of approximately 600 grit (medium tex-
ture). The bathtub surface was covered with a non-slip,
rubber mat.
Configuration 1 (Canadian Standard Association –
CSA) complied with the CSA and consisted of two
120 cm long grab bars. The first bar was placed hor-
izontally on the back wall 18 cm above the back bath
rim. The second bar was placed vertically on the faucet
end wall starting 18 cm above the bath rim.
Configuration 2 (US Uniform Federal Accessibility
Standards – UFAS) complied with the UFAS and con-
sisted of two 61 cm grab bars that were mounted adja-
1To ensure comprehension of tasks and post trial interview data.
H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults 47
OBC UFAS
CSA
C
All
Fig. 1. Pictures illustrating the five configurations tested as they were installed with respect to a right side faucet location.
cent to the outer edges of the tub on the faucet and head
walls. Both bars were mounted 48 cm above the bath
rim or 88 cm above the floor. In addition, two 82 cm
long grab bars located 29 cm from the head and faucet
walls were mounted on the back wall. The lower grab
bar on the back wall was mounted 23 cm above the rim
of the tub. The upper grab bar was mounted at the same
height as the head and faucet wall bars, 48 cm above
the rim or 88 cm above the floor.
Configuration 3 (Ontario Building Code – OBC)
consisted of a single bar and complied with the OBC
for placement. The bar, an “L”-shaped grab bar with
legs separated by 90 degrees, was mounted on the back
wall. The horizontal leg of the “L” was located 17 cm
above and parallel to the rim of the bathtub. The ver-
tical leg of the “L” was located 38 cm from the faucet
end of the bathtub. The OBC specifies that the two legs
of the bar should be 90 cm each. A bar of this length
proved difficult to obtain and a bar with legs of 75 cm
was finally tested.
Configuration 4 (Common Configuration – C) con-
sisted of two bars and matched the configuration most
often seen in a previous study of home bath bars among
514 seniors residing in non-profit apartments [2]. The
first bar was 60 cm long and was mounted at approxi-
mately 45 degrees on the back wall. The bottom end of
the bar was located 23 cm above the rim of the tub. The
top of the bar was located 30 cm from the faucet wall
and the bottom of the bar was approximately 74 cm
from the faucet wall. The second bar was located verti-
cally on the faucet end wall. The bar was 120 cm long
and was located 18 cm above the rim.
Configuration 5 (AB) consisted of a composite of the
grab bars used in the previous four configurations. The
specific bars tested included two vertical bars (left and
right mounting configurations from the CSA configu-
ration); two horizontal bars (bars mounted on the head
wall and faucet wall from the UFAS configuration); one
horizontal bar (mounted on the back wall from the CSA
configuration); and one angled bar mounted on the back
wall (from the C configuration – orientation depended
on the faucet end as reported by the participant).
2.4. Test procedures
Participants were barefoot during all bathtub config-
uration tests. Participants were asked to rise from a
chair outside of the tub, enter the tub, sit on the bot-
tom of the tub, stand up from the bottom of the tub,
exit the tub and sit back down on the chair. Limited
instructions were provided. Participants were told to
use any supports, including the bathtub grab bars, if
they felt they were necessary or if theywould help. Be-
cause of the complexity in setting up the AB and OBC
configurations, all testing began and ended with one of
these two configurations. The order in which these two
configurations were tested was counterbalanced across
participants. Presentation order of the remaining con-
figurations was random.
Participants completed one trial with each bath bar
configuration. After each trial, participants were asked
to rate the configuration on safety, ease of use, helpful-
ness, and comfort. Once the participant completed test-
ing of the five configurations they were asked to rank
48 H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults
Fig. 2. Ratings of grab bar configurations. Significant differences in mean ratings of ease of use, helpfulness and safety were detected between
configurations and attributed to significant paired differences. Ease of use: F
(4,93) =6.2,p!0.007, significant paired comparisons
(p!0.005) = AB-OBC, AB-CSA, C-OBC, C-CSA. Helpful in/out: F(4,92) =5.5,p!0.007, significant paired comparisons (p!0.005) =
AB-OBC, AB-CSA, C-OBC, UFAS-OBC. Helpful up/down: F(4,84) =4.4,p!0.007, significant paired comparisons (p!0.005) =
AB-OBC, AB-CSA, UFAS-OBC. Safety: F(4,84) =4.4,p!0.007, significant paired comparisons (p!0.005) = AB-OBC, AB-CSA,
UFAS-OBC.
order all of the configurations on safety, ease of use,
preference, acceptability, comfort, and helpfulness. A
Composite Rating Score (ranging from 6 to 30) was
computed based on the sum of all rating factors (safety,
comfort, helpfulness getting in/out, helpfulness sitting
down/getting up, ease of use, likelihoodof use). Higher
scores denoted more favourable ratings.
2.5. Data reduction and analysis
Three data coders viewed the video data. Inter-rater
reliability was assessed by having all coders review the
same five videotapes. Concordance for whether bars
were used (range: 84% to 96%) and which bars were
used (range: 80% to 100%) was very good to excel-
lent. Clarification of scoring instruction and discus-
sion of the data sets was followed by reassessment of
inter-rater reliability using five new videotapes. Post-
training concordance for whether bars were used and
which bars were used was excellent (99% to 100%).
Video data were summarised for each configuration
by recording which bar(s) and how many times each
bar was used to: a) get into the bathtub; b) sit in the
bottom of the bathtub; c) get up from the bottom of the
bathtub; and d) get out of the bathtub. Participant use
of any walls, front bath rim, back bath rim, or other
objects for balance or support in the execution of any
of the activities was also coded. Because participants
were asked to enter the bathtub as they would at home,
the faucet and head walls were coded in consideration
of the self-reported location of their faucet at home.
Descriptive statistics were used to summarise: a)
participant profiles; b) home bathing profiles; c) rat-
ings and rankings of the different grab bar configura-
tions; and d) video data. Separate repeated measures
ANOVAS, followed by paired t-tests,were used to com-
pare, between the five configurations, rating scores for
each: comfort, ease of use, helpfulness getting into and
out of the bathtub, helpfulness sitting into and getting
up from the bottom of the bathtub, safety, likelihood
of use and the Composite Rating Score. Bonferroni
corrections were used to compensate for multiple tests,
setting the criteria alpha for each omnibus test at 0.007
(0.05/7) and the criteria for significance for follow-up
tests at 0.005 for each dependent variable (0.05/10).
Cochran Q tests, for multiple related dichotomous
variables, were used to: 1) assess differences in the
prevalence of use of different configurations; and 2)
assess differences in the prevalence of use of walls,
rims, back rims, and total supports used to get into,
sit down, get up, exit, and for all activities together.
Follow-up tests were performed using McNemar tests
for two related samples with Bonferroni corrections
setting the significance criteria at 0.005 (0.05/10).
H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults 49
0
10
20
30
40
50
Ease of Use Helpful Preferred Safety
Ease of Use Helpful Preferred Safety
Percentage
OBC UFAS CSA C
0
10
20
30
40
50
Percentage
OBC UFAS CSA C
A
B
Fig. 3. Percentage of participants that ranked the C configuration and building code standard configurations (OBC, CSA, UFAS) highest (A) or
lowest (B) on each factor evaluated. Between 38 and 59% of participants ranked the AB configuration highest while less than 10% of participants
ranked the AB configuration lowest on each factor evaluated (data not illustrated).
3. Results
3.1. Participant characteristics
A total of 103 participants took part in the study.
Their mean age was 70.4 +/−6.2 years and ranged
from 60 to 83 years. The majority were female
(n= 64). Relative to their peers, participants rated
their health as very good (n= 63), good (n= 33)
or fair/poor (n=7) and their activity levels as more
active (n= 74), about the same (n= 25) or less ac-
tive (n=4) than their peers. Although thirty-one par-
ticipants self-reported problems with balance, only 11
used a cane for walking.
There were 27 participants who reported a fall in the
previous year. Of these, two took place in the bathroom
with both participants reporting that they were not using
grab bars at the time of the fall. One fall occurred
as the individual was getting out of the tub. In the
second fall, the individual’s hand slipped on the bath
rim being used for support while standing up from the
bottom of the bathtub. Difficulties in getting into or
out of the tub were reported by 16 individuals. Sitting
or standing in the tub was difficult for 42 individuals.
Two individuals reported restricted bathing routines. A
minority of participants (n= 26) had one bath grab bar
installed in their home with 11 participants reporting
two or more grab bars.
Three participants expressed difficulty with sitting in
the bottom of the bathtub and thus were instructed to
simply step into and then exit from the bathtub. Six ad-
ditional participants were not asked to sit into the bot-
tom of the bathtub for at least one configuration because
the interviewers deemed that their level of mobility was
insufficient to safely perform the task.
3.2. Ratings of grab bar configurations
A significant omnibus test (F(4,80) =9.6,p!
0.007) between configurations on the Composite Rat-
ing Score is attributed to significantly higher ratings for
the AB compared to all other configuration. Further,
50 H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults
Fig. 4. Percentage of participants who used at least one grab bar while getting in/out of bathtub (A) and sitting down/standing up from bottom of
bathtub (B) for each configuration. Significant differences in the use of bars were detected between configurations and attributed to significant
paired differences. Get in: Q(4) =27.5, p!0.01, significant paired comparisons (p!0.005) = AB-OBC, OBC-UFAS, OBC-CSA, OBC-C.
Get out: Q(4) =39.0, p!0.01, significant paired comparisons (p!0.005) = AB-OBC, OBC-UFAS, OBC-CSA, OBC-C.
the Composite Rating Scores for the C and UFAS con-
figurations were each significantly higher than that for
the OBC configuration (see Fig. 2 for comparisons on
categories of ease of use, helpfulness with getting in/out
and sitting in/getting up, as well as for safety). Follow-
up comparisons amongst the C and the three standards
association configurations (UFAS, CSA, OBC) resulted
in several significant paired differences. Specifically,
the UFAS rated significantly higher than the OBC on
the rating factors help with getting in/out of bathtub,
help with sitting in/getting up from bottom of bathtub,
and safety. The C configuration was rated significantly
higher than the CSA on rating factors ease of use and
likely to use. Finally, the C was rated significantly
higher than the OBC on rating factors ease of use and
help with getting in/out of bathtub.
3.3. Rankings of grab bar configurations
The AB configuration was consistently ranked the
highest in terms of acceptance, comfort, ease of use,
overall helpfulness, safety and overall preference. The
C configuration was also ranked highest by many in
terms of: acceptance (32.4%), comfort (21.6%), ease of
use (24.5%), helpfulness (25.5%) and preferred group-
ing (31.4%). However, it was ranked highest in terms
of safety by only 14.9% of respondents. Figure 3 illus-
trates the relative ranking of the C and the three stan-
dards association configurations (UFAS, CSA, OBC).
The OBC configuration was consistently ranked least
favourable on each factor with almost half of respon-
dents ranking it the least acceptable, least comfortable,
most difficult to use, least helpful, least safe and least
preferred. The CSA configuration was also not widely
H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults 51
endorsed on any ranking factor. The UFAS configura-
tion was ranked as the third least favourable configura-
tion on all factors although by usually less than 20% of
respondents.
3.4. Ideal bars
A total of 93 participants2were asked to identify
their ideal grab bar configuration. Of these, 13 indi-
viduals reported that they did not need or could not en-
vision an ideal configuration. Overall, 19 participants
indicated that they would ideally include one bar; 37
identified two bars; 21 identified three bars; and 3 iden-
tified four bars. Of the 80 respondents who reported
that they desired at least one grab bar at home, the
majority (n= 69) identified one bar oriented either
angled or horizontal on the back wall of the bathtub.
Vertical faucet wall bars were the second most com-
monly identified orientation and location for ideal bars
(n= 37).
3.5. Recorded patterns of grab bar use
All but one participant used at least one grab bar
during some part of the testing. The individual who
chose not to use any grab bars for any activities did not
use any supports, including walls, the bath rim, or back
rim, to facilitate getting into, sitting down, getting up
from, or getting out of the bathtub.
Over 10% of respondents did not use any bars at all
to get in (n= 13, 12.7%) or out (n= 16, 15.7%) of the
bathtub. Of the 13 participants who did not use a bar
to get in, one participant was observed to use a wall to
facilitate with the transfer. Of the 16 participants who
did not use a bar to get out of the bathtub, four used a
wall during the transfer.
Fewer than 10% of respondents chose not to use a
bar to facilitate sitting down in (n=6, 5.9%) or getting
up from (n=8, 7.8%) the bottom of the bathtub. Of
the six who used no bars to sit down, five used the bath
rim or back rim to help in getting down. Similarly, of
the eight who used no bars to get up, six used the bath
rim or back rim to get up.
Significant differences were noted in the prevalence
with which grab bars in different configurations were
used overall (Q(4) = 16.6,p!0.01). This significant
omnibus test is attributed to a significant difference in
2We added the questionnaire item asking participants to identify
an ideal grab bar configuration after having tested an initial 10 people.
prevalence of use of grab bars when using the CSA as
compared to the OBC configuration (p!0.005). In-
deed, the configuration most commonly used for any
activity was the CSA configuration, used to get in, sit,
get up, or out of the bathtub by 99 participants. Al-
though not statistically significant, the next most com-
monly used configuration was the C configuration, used
by 97 participants. The UFAS and OBC configura-
tions were used with the lowest prevalence (93.1% and
85.3%, respectively).
Significant differences noted in the prevalence with
which grab bars in different configurationswere used to
enter (Q(4) = 27.5,p!0.01) and exit (Q(4) = 39.0,
p!0.01) the bathtub are attributed to a significantly
lower prevalence of use when using the OBC, as com-
pared to the All Bars, UFAS, CSA, and C configura-
tions (p!s!0.005; Fig. 4A). There were no significant
differences in the prevalence of use of different con-
figurations to sit in and get up from the bottom of the
bathtub (Fig. 4B).
3.6. Relationship between self-reported and video
coded data
Correlations between the number of activities (range
from 1 to 4) for which different configurations were
used and self-reported Composite Rating Scores were
non-significant for each configuration: OBC (r=
0.14), UFAS (r=0.16), CSA (r=0.01), C (r=
0.12), All Bars (r=0.28).
3.7. Relationship between participant profiles,
preferred configurations and use of grab bars
After correcting for multiple tests, participant char-
acteristics were not predictive of configuration rank-
ings on any factor (p>0.004 for all comparisons).
Moreover, participant characteristics were not signif-
icantly associated with grab bar use for bath entry or
exit (p>0.005 for all comparisons).
3.8. Use of other supports
All but one of the 102 participants for whom video
data were available, relied on supports other than the
grab bars to either enter, sit in, get up from, or exit
the bathtub. The use of supports other than grab bars
was slightly less common in the UFAS (n= 97), CSA
(n= 93), C (n= 94) and All Bar (n= 93) than the
OBC (n= 98) configurations. Overall, 43 individuals
relied on a wall during the task primarily to get into
52 H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults
Fig. 5. Location of the circular mounting plates used to hold the different bars used. Note the consistent grouping of the mounting plates within
a reasonably well-defined area. The inverted “L/T-shaped” reinforced area illustrated would correspond to a bath with a faucet on the right side
as one enters the tub.
and out of the bathtub. Only 2 participants used the
wall for support when getting up from the bottom of
the bathtub. None of the participants used a wall as
support when sitting into the bathtub.
Several participants (n= 12) used the bath rim or
back rim to get into or out of the bathtub while all but
one participant used the bath rim to sit into (n= 101)
or get up from (n= 100) the bottom of the bathtub.
Finally, 29 participants used the back rim to either sit
into or get up from the bottom of the bathtub.
4. Discussion
The purpose of this research study was to assess dif-
ferent locations and configurations of bathtub grab bars
and to identify the safest and most useful options based
on observational data as well as self-reports of seniors.
The participating seniors were relatively healthy with
the majority able to complete all components of the
testing protocol. Demographic and health profiles, in-
cluding the prevalence of falls, reflected the general
community-dwelling population of seniors living in the
Ottawa/Carleton region and were comparable to esti-
mates provided in other international community-based
studies [6,14–17,20,21,23,27,29,32]. Despite being
generally healthy, almost one-third of our participants
reported balance problems and falls in the previous year
and over one-third reported difficulties with bath trans-
fers. Of interest, objective evaluations of balance and
mobility matched the self-rated scores indicating that
individuals were able to appropriately evaluate their in-
dividual competency levels. Further, although not an
objective of the current study, we did identify that one
third of the study participants who reported a balance
problem did not use a mobility aid. The lack of mobility
aid use may be a function of the participant characteris-
tics. Most participants in the study rated their health as
good/very good and their activity levels as better than
peers suggesting the balance problems reported were
not perceived as significant.
4.1. Characteristics of bars and bar use at home
Most study participants had no grab bars installed in
their home bathrooms. The small proportion with grab
bars installed in their homes generally indicated that
they were used on a regular basis. Although, findings
from our previous research [2] suggest that a minimum
of two bath grab bars are needed, only 11 participants
in the present study reported having two or more bath
grab bars in their home. We did not, as part of this
study, address the question of why grab bars were not
installed.
Although the placements and orientations of the
home bathtub grab bar configurations were varied, the
most commonly reported were bath bars on the back
wall. However, 20% of respondents with grab bars in-
dicated that they had an inverted u-shaped bar on the
bath rim. Personal communication with occupational
therapists working in community practice suggests that
the inverted u-shaped bar mounted on the bathtub rim
may be a safety hazard predisposing people to falls
from tripping. Thus, in planning the study we decided
H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults 53
not to add the horseshoe rim-mounted configuration
to the series evaluated in the current study. However,
given the possibility of safety risk associated and the
high prevalence of the bar in the older adults’ homes,
follow-up studies of this bar are warranted.
4.2. Preferred configurations
Ratings on all factors were consistently higher for all
configurations compared with the OBC configuration.
This is not surprising since the OBC configuration con-
sists of an L-shaped bar on the back wall with no bar
on either the head or faucet wall. Thus the bar configu-
ration would not be helpful for entering and exiting the
tub.
The relatively low rating of the CSA configuration
is more difficult to explain since it includes a vertical
bar on the faucet wall. One pattern of entry into the
tub is for a participant to grab the faucet wall bar, step
into the tub and then quickly grab the back wall bar
before stepping into the tub with the second foot. If
this is the preferred entry pattern, it is possible that the
horizontally oriented bar on the back wall in the CSA
configuration was placed too low to serve this purpose.
This interpretation remains to be confirmed.
4.3. Video data
The limited relationship between the Composite Rat-
ing Scores and the number of activities forwhich a con-
figuration was used was surprisingand suggests that the
self-report or perceived ratings on the factors tested may
not be reflective of actual grab bar use. Thus, although
participants may have positively ranked a given con-
figuration, they may not have used the bars within the
configuration when they actually performed the bath-
tub test. Future studies will address the concordance
between perceptions of bar use and actual bar use.
Results of video data showed that all but one partici-
pant used at least one grab bar at some point during the
testing with the All Bar, CSA, C, and UFAS configu-
rations most commonly used. The OBC configuration
was least likely to be used. There were no significant
differences between configurations in the prevalence of
use of grab bars to sit down into or get up from the bot-
tom of the tub. This is not surprising since all config-
urations had at least one bar mounted on the back wall
of the bathtub. However, there was a significant differ-
ence between configurations for entering and exiting
the tub with the OBC configuration significantly less
used than all of the other configurations tested. This
too is not surprising since the OBC configuration does
not include a bar to facilitate entry/exit on either the
faucet or head wall of the tub.
Most participants used supports, other than grab bars,
during the testing. Regardless of configuration tested,
the majority of participants (99%) used the bathtub rim
when sitting into or getting up from the bottom of the
tub. A high prevalence of use of bath edges to sit down
and get up from the bottom of the bathtub is alarming
and highlights the need for additional safety features
such as non-slip surfaces for bath edges.
4.4. Characteristics of seniors for whom bath grab
bars may be most useful
There were no differences in the characteristics of
participants who ranked each configuration high ver-
sus low on the different factors. This suggests that the
rank order obtained may reflect perceived preference
regardless of health, demographic and fall history for
seniors with similar profiles to our study participants.
Similarly, there were no differences in health, demo-
graphic and fall characteristics between grab bar users
and non-users suggesting use of grab bars would be
helpful for individuals with profiles similar to those
of our study participants – that is, healthy, ambulatory
community-living seniors.
4.5. Ideal bath grab bar configurations
When asked to identify an ideal configuration, the
most prevalent configuration consisted of two bars. In
the ideal configuration, one vertically oriented bar was
located on the faucet wall while a second bar, oriented
either horizontally or on an angle was located on the
back wall. This is consistent with patterns of grab
bar use observed from video data and with supports
required to get into/out of the bathtub (faucet wall) and
up/down from the bottom of the bathtub (bath edges).
5. Conclusions and recommendations
The use of bars by all but one individual regardless of
participant health status, demographics or fall history
suggests that grab bars should be universally available
for individuals over 65 years of age. Universal acces-
sibility for younger individuals may also be beneficial
and future studies will address this consideration. Fur-
ther, self-reports of ideal configurations and data from
video-taping suggests that a minimum of two grab bars
54 H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults
be required to constitute universal access. Based on
the data from this study, one bar should be mounted on
the head or faucet wall to assist in entering/exiting the
bathtub. A vertically oriented bar is useful for both en-
tering and exiting the tub. A horizontally oriented bar
is useful for entering the tub; however it is less helpful
when a person is exiting the tub unless the bar extends
past the rim of the tub into the “clear” space. The sec-
ond horizontal or angled bar to assist in sitting/standing
from the bottom of the tub should be located onthe back
wall of the tub. The CSA and Common configurations
meet these recommendations (see Fig. 1).
In addition, non-slip surfaces on the bath front and
back rims should be incorporated as standard safety
features for bathing among seniors. Further study is
needed to determine whether grab bars in the bath-
tub rim would be beneficial. Finally, new construc-
tion should include a “U-shaped” reinforced area on
the back wall as well as two “L/T-shaped” areas on
the faucet and head wall to permit easy installation of
appropriate configuration (see Fig. 5).
This study did not examine reasons why participants
did not use assistive devices such as grab bars and/or
canes when balance assessments suggested they may be
useful. Future research should address the social norms
and other barriers that discourage seniors/adults from
installing assistive devices such as grab bars. Similarly,
given the high percentage of bathtub grab bar use by
the study participants, it is important to determine why
individuals do not install grab bars in their own homes.
In a separate study of bath grab bar use following an
experimentally induced loss of balance, we are com-
pleting follow-up interviews with study participants to
determine whether study participation influences home
installation and use of bathtub grab bars. Further, al-
though we did not determine the safety, ease of use,
or helpfulness of a rim-mounted bathtub grab bar, this
configuration was mentioned as useful by participants
and requires further investigation. Finally, future stud-
ies of individual behaviour would be enhanced by in-
corporating both self-reports as well as more objective
videotaped performance measures.
Acknowledgements
This Project was partially funded by Canada Mort-
gage and Housing Corporation (CMHC), but the views
expressed are the personal views of the author(s)and do
not represent the official views of CMHC. Additional
funding was provided by the Community Health Re-
search Unit and the Faculty of Health Sciences, Univer-
sity of Ottawa. HS is a Career Scientist with the Min-
istry of Health & Longterm Care, Ontario. NE holds a
Nursing Chair from the Canadian Health Services Re-
search Foundation, the Canadian Institutes of Health
Research and the Ontario Ministry of Health and Long
term Care. The opinions expressed in this publication
are those of the authors. Publication does not imply en-
dorsement of these views by the participating partners
and funders of the Community Health Research Unit
nor the CMHC.
References
[1] F. Aminzadeh and N. Edwards, Exploring seniors’ views on
the use of assistive devices in fall prevention, Public Health
Nursing 15 (1998), 297–304.
[2] F. Aminzadeh, N. Edwards, D. Lockett and R. Nair, Utilization
of bathroom safety devices, patterns of bathing and toileting
and bathroom falls in a sample of community living older
adults, Technology and Disability 13 (2000), 95–103.
[3] F. Aminzadeh, N. Edwards, D. Lockett, R. Nair and Y. Ro-
bitaille, Factors influencing the use of bathroom safety de-
vices among community living seniors: A case for research
in primary health care, Final report submitted to the Canadian
Nurses Foundation, (1999), 1–58.
[4] L.A.Axtell and Y.L. Yausda, Assistive devices and home mod-
ifications in geriatric rehabilitation, Geriatric Rehabilitation 9
(1993), 803–821.
[5] R.W. Bohannon, P.A. Larkin, A.C. Cook, J. Gear and J. Singer,
Decrease in timed balance test scores with aging, Physical
Therapy 64 (1984), 1067–1070.
[6] A.J. Campbell, M.J. Borrie and G.F. Spears, Risk factors for
falls in a community-based prospective study of people 70
years and older, Journal of Gerontology44 (1989), M112–117.
[7] Canadian Mortgage and Housing Corporation, Maintaining
seniors’ independence: A guide to home adaptations. Ottawa:
author, 1995.
[8] Canadian Standard Association (CSA), Barrier-Free Design
(No.CAN/CSA-B651-95). Rexdale, ON: Canadian Standards
Association, 1995.
[9] N. Edwards, N. Birkett, M. Murphy and R. Nair, Fall preven-
tion: Results of baseline interviews, Paper presented at the
Canadian Gerontology Conference, Montreal, Quebec, 1993.
[10] N.Edwards, D. Lockett, F. Aminzadeh and R. Nair, Predictors
of bath grab-bar use among community-living older adults,
Canadian Journal on Aging 22 (2003), 217–227.
[11] N.I. Edwards and D.A. Jones, Ownership and use of assis-
tive devices amongst older people in the community, Age and
Ageing 27 (1998), 463–468.
[12] L. George, V.E. Binns, A.D. Clayden and G.P. Mulley, Aids
and adaptations for the elderly at home: Underprovided, un-
derused, and undermaintained, British Medical Journal 296
(1988), 1365–1366.
[13] l.D. Gillespie, W.J. Gillespie, R. Cumming and S.E. Lamb,
Interventions for preventing falls in the elderly, The Cochrane
Database of Systemic Reviews 1(2000).
[14] W.C. Graafmans, M.E. Ooms, M.A. Hofstee, P.D. Bezemer,
L.M. Bouter and P. Lips, Falls in the elderly: A prospective
H. Sveistrup et al. / Evaluation of bath grab bar placement for older adults 55
study of risk factors and risk profiles, American Journal of
Epidemiology 143 (1996), 1129–1136.
[15] W.A. Hale, M.J. Delaney and W.C. McGaghie, Characteris-
tics and predictors of falls in elderly patients, The Journal of
Family Practice 34 (1992), 577–581.
[16] D. Lockett, N. Edwards and F. Aminzadeh, Risk factors for
falls in bathroom among community dwelling seniors, Pre-
sented at the Ontario Public Heath Association Annual Meet-
ing, Toronto, Ontario (Poster), November, 1999.
[17] S.R. Lord , J.A. Ward, P. Williams and K.J. Anstey, An
epidemiological study of falls in older community-dwelling
women: The Ranwick falls and fracture study, Australian
Journal of Public Health 17 (1993), 240–245.
[18] W.C. Mann, K. Hurren and M. Tomita, Comparison of assis-
tive device use and needs of home-based older persons with
different impairments, The American Journal of Occupational
Therapy 47 (1993), 980–987.
[19] S. Mathias, U. Nayak and B. Isaacs, Balance in elderly pa-
tients: The Get Up and Go test, Archives of Physical Medicine
Rehabilitation 76 (1986), 387–389.
[20] M.C. Nevitt, S.R. Cummings and E.S. Hudes, Risk factors for
injurious falls: A prospective study, Journal of Gerontology
46 (1991), M164–170.
[21] M.C. Nevitt, S.R. Cummings, S. Kidd and D. Black, Risk
factors for recurrent nonsyncopal falls, Journal of American
Medical Association 261 (1989), 2663–2668.
[22] J. O’Loughlin, The incidence and risk factors for falls and fall-
related injury among elderly persons living in the community,
Ph.D. Dissertation, McGill University, 1991.
[23] J. O’Loughlin, Y. Robitaille, J.F. Bovin and S. Suissa, Inci-
dence and risk factors for falls and injurious falls among the
community living elderly, American Journal of Epidemiology
137 (1993), 342–354.
[24] M.G. Parker and M. Thorslund, The use of technical aids
among community-based elderly, The American Journal of
Occupational Therapy 45 (1990), 712–718.
[25] B. Plautz, D. Beck, C. Selmar and M. Radersky, Modifying the
environment: a community-based injury-reductionprogram
for elderly residents, American Journal of Prevention: Re-
search Linkages Between Academia and Public Health Prac-
tice 12 (1996), 33–38.
[26] E. Rawski, Review of the literature on falls among the elderly,
Image: Journal of Nursing Scholarship 30 (1998), 47–52.
[27] A.S. Robbins, L.Z. Rubenstein, K.R. Josephson, B.L. Schul-
man, D. Osterweil and G. Fine, Predictors of falls among
elderly people: Results of two population-based studies,
Archives of Internal Medicine 149 (1989), 1628–1633.
[28] U. Sonn and G. Grimby, Assistive devices in an elderly popu-
lation studied at 70 and 76 years of age, Disability and Reha-
bilitation 16 (1994), 85–92.
[29] G.S. Sorock and D.M. Labiner, Peripheral neuromuscular dys-
function and falls in an elderly cohort, American Journal of
Epidemiology 136 (1992), 583–591.
[30] G. Thornely, M.A. Chamberlain and V. Wright, Evaluation of
aids and equipment for the bath and toilet, British Journal of
Occupational Therapy 40 (1977), 243–246.
[31] R. Tideiksaar, Falling in Old Age: Prevention and Manage-
ment, (2nd ed.), New York: Springer Publishing Company,
1997.
[32] M.E. Tinetti, M. Speechley and S.F. Gitner, Risk factors for
falls among elderly persons living in the community, England
Journal of Medicine 319 (1988), 1701–1707.
[33] F. Trickey,D. Maltais, M.A. Gosselin and Y.Robitaille, Adapt-
ing older persons’ homes to promote independence, Physical
& Occupational Therapy in Geriatrics 12 (1993), 1–14.
