Effect of Bathroom Aids and Age on Balance Control During Bathing Transfers

Design

A cross-sectional, repeated-measures design was used. Approval was granted by our institution’s ethics board, and all participants provided written informed consent.

Sample and Recruitment

Fourteen healthy older adults (4 men, 10 women; mean age = 71.36 ± 6.50 yr) and 12 healthy young adults (6 men, 6 women; mean age = 25.67 ± 3.50 yr) formed a convenience sample recruited through community advertisements. Participants were community dwelling, were in self-reported good health, and had normal or corrected-to-normal vision. Exclusion criteria were a self-reported history of vestibular, orthopedic, or neurological conditions affecting walking or upper-body function or cognitive deficits limiting communication.

Experimental Environment

Research was conducted using a custom-designed experimental bathroom consisting of three walls surrounding a bathtub (rim height = 38.1 cm) that was modified to separate its floor from its walls (Figure 1). The bathtub surface and floor immediately outside of the tub were mounted on separate force plates (Model AMTI BP11971197–2000, Advanced Medical Technology, Watertown, MA) to allow force measurements to be collected independently for each surface.

OPEN IN VIEWER

Figure 1.

Schematic showing the orientations of the axis of progression and perpendicular (lateral) axis and the locations of the vertical grab bar (A), horizontal grab bar (B), and bath mat (C).

The bathtub surface was wetted with 0.5 L of a standard soapy water solution (diluted sodium lauryl sulfate; per British Standards Institution slip resistance test standard BS EN 13845) to simulate realistic bathing conditions. The solution was redistributed over the tub surface between trials. Of note, frictional measurements on wet bathtub surfaces are difficult to measure reliably in the field (English, 2003). We therefore focused on ensuring the reproducibility of our method rather than targeting a specific coefficient of friction. For all testing, participants were barefoot. Their hands and the grab bars remained dry throughout testing.

Procedure

Participants were instructed to step into the bathtub, stand quietly for several seconds, and then step out of the tub. For each entry–exit trial, participants were instructed to use a grab bar proactively, touch the wall for support, or not touch anything with their hands (see Bathing Assistance Conditions section). No further instruction or training in transfer techniques was provided, and participants were free to enter and exit using their preferred approach. Three trials were completed for each assistance condition. Conditions were balanced among participants to ensure that any learning, fatigue, or carryover effects did not affect data interpretation. Participants wore an overhead safety harness throughout testing.

Bathing Assistance Conditions

Five assistance conditions were evaluated:

These conditions engaged the upper limbs to support stability when transferring over the bathtub rim, improved the surface friction of the bathtub, or provided no additional assistance. For Conditions 2 and 3, the grab bar was 91 cm in length and 3.8 cm in diameter, with a smooth painted finish.

Data Processing and Analysis

Force plate data were collected at 250 Hz and upsampled to 500 Hz using linear interpolation (MATLAB R2012a, Math Works, Natick, MA). All data were low-pass filtered (fourth order, dual-pass Butterworth filter at 6 Hz) and synchronized using Visual 3D (C-Motion, Germantown, MD).

Bathing entry and exit were evaluated separately. The start of each transfer was identified by a 20-N change in the vertical ground reaction force signal above baseline; the end was defined as the point at which the participant had completed the transfer and the vertical ground reaction force signal stabilized. Any trial that included a fall (i.e., full reliance on the safety harness) was excluded from the analysis.

To quantify postural control, the resultant COP beneath the feet was calculated throughout each transfer, with components analyzed both in the axis of progression, representing the primary direction the person moves to enter and exit the tub, and perpendicular to this axis (see Figure 1). Measures of interest were as follows:

Statistical Analysis

A mixed-model, repeated-measures analysis was performed using SAS 9.4 PROC MIXED (SAS Institute, Cary, NC). The full model included a within-subject factor (assistance condition) and a between-subject factor (age group) and second-order interaction (Assistance Condition × Age Group). Post hoc pairwise comparisons using Tukey’s honestly significant difference adjustments were conducted after identification of a significant main effect. Significance levels were set at p < .05 for all analyses.

Entry and Exit Strategies

Participants entered and exited the bathtub using self-selected strategies. Young and older adults usually entered the bathtub facing forward (the direction of progression). In the wall use condition only, 2 older and 2 young adults entered facing the side wall (perpendicular to the axis of progression).

Most young adults (9 of 12) exited the bathtub while facing the side wall. However, several exceptions were noted. Two participants exited forward in both the vertical and the horizontal grab bar conditions, and 1 participant exited forward for the no-assistance condition. For older adults, strategies were more affected by use of the horizontal grab bar. Like the younger adults, the majority of older participants (9 of 14) exited facing the side wall in all trials. Two other participants followed this strategy in all but the horizontal grab bar condition, when they exited while facing the back wall, and 3 participants exited facing forward in all but the horizontal grab bar condition, in which they faced either the side or back wall.

Impact of Age

Older adults generally adopted a more cautious strategy and demonstrated greater instability than young adults: In the axis of progression, older adults had reduced DispCOP, F (1, 21) > 5.33, p < .03, but greater VarCOP, F (1, 21) > 14.13, p < .002, compared with young adults, evident during both bathtub entry and exit. No significant age-related differences were observed for DispCOP or VarCOP perpendicular to the axis of progression, F (1, 21) < 1.51, p > .23, or for VelCOP in either axis, F (1, 21) < 2.12, p > .16, during either entry or exit.

Impact of Bathing Assistance

Bathtub Entry.

During bathtub entry, assistance condition significantly affected VarCOP (Figure 2), F (4, 80) = 5.76, p = .005, but not VelCOP (Figure 3) or DispCOP (Figure 4), F (4, 80) < 1.1, ps > 0.36, in the axis of progression. In this axis, pairwise comparisons revealed that compared with all other conditions, use of the vertical grab bar reduced VarCOP (p < .05), whereas use of the horizontal grab bar increased it (p < .005). Perpendicular to the axis of progression, a significant main effect of assistance condition was found for VarCOP, F (4, 80) = 7.58, p = .007. Pairwise comparisons revealed lower VarCOP when using a bath mat than in the no-assistance or horizontal grab bar conditions (p < .03). Wall use also reduced VarCOP in this direction compared with the no-assistance condition (p = .0015). A similar trend was observed for the vertical grab bar condition, but it did not reach significance (p = .052; Figure 2). In addition to the reduction in VarCOP, bath mat use, compared with all other assistive conditions, also resulted in reduced VelCOP perpendicular to the axis of progression (p < .04; Figure 3).

OPEN IN VIEWER

Figure 2.

Variability of the COP in the axis of progression and perpendicular axis for older adults and young adults during bathtub entry and exit.

OPEN IN VIEWER

Figure 3.

Velocity of the COP in the axis of progression and perpendicular axis for older adults and young adults during bathtub entry and exit.

OPEN IN VIEWER

Figure 4.

Displacement range of the COP in the axis of progression and perpendicular axis for older adults and young adults during bathtub entry and exit.

Friction-Enhanced Tub Surface

Bath mats are a common bathroom aid recommended by occupational therapists (Cumming et al., 2001) and used by community-dwelling older adults to support bathing (Aminzadeh et al., 2000). Our results indicate that in addition to reducing the risk of slips, bath mat use can lead to slightly improved postural control perpendicular to the direction of progression as people step into and out of the bathtub.

During a bathing transfer, when a person is facing the direction of progression, as was most commonly observed in our study during entry, loss of balance in the perpendicular axis will likely require a lateral step for recovery in the absence of a graspable support. Even on higher friction surfaces, these lateral steps are challenging and often poorly executed by older adults (Maki, Edmondstone, & McIlroy, 2000), increasing the risk of a potentially injurious lateral fall. This difficulty may be exacerbated during bathing transfers because (1) the slippery surface decreases the effectiveness of compensatory stepping and (2) foot trajectories must be executed to avoid a substantial obstacle (the bathtub rim). The use of a friction-enhancing device, such as a bath mat, is important to reduce the risk of a lateral slip and may contribute to the likelihood of a successful stepping response if balance loss occurs.

Interestingly, the bath mat did not appear to provide significant benefit in the axis of progression and was essentially equivalent to the no-assistance condition. This result is despite significant momentum generated in the axis of progression during bathtub entry and exit, with high risk of slipping underfoot because of reduced friction (Siegmund et al., 2010). The apparent lack of influence of the bath mat in this direction suggests that in the axis of progression, the inherent threats to balance associated with obstacle crossing may dominate over the slipping risk. For this reason, it appears that simply increasing friction in the bathtub does not lead to significant improvements in postural control in the axis of progression.

Engaging Upper Limbs to Support Balance

We investigated three conditions that engaged the upper limbs to support balance during bathing transfers. In our scenario, use of the wall or a vertically mounted grab bar led to similar improvements in stability. Both conditions provide tactile input from the upper limb throughout the transfer. Numerous studies have shown that this kind of feedback improves postural stability (Baccini et al., 2007; Reid et al., 2011). In contrast, when using the horizontal grab bar mounted on the far wall of the tub, participants released contact with the grab bar for a portion of the task, reducing its ability to contribute sensory feedback.

Past research has shown that grab bar use rates are relatively low (Aminzadeh & Edwards, 1998; Aminzadeh et al., 2000). The wall use condition we tested thus reflects a common bathing transfer practice for older adults. Although we found that wall use provided many similar benefits to the vertical grab bar to support balance control during an uneventful transfer, it offers much less potential to anchor the body to prevent a fall if loss of balance occurs. Guitard et al. (2011) investigated the use of different grab bar configurations after a balance perturbation and found that older adults required grab bars to regain balance almost 60% of the time, compared with 14% of the time for young adults. Such findings emphasize performance differences between a fully graspable handhold that can provide substantial multidirectional reaction forces compared with a wall.

Limitations and Future Directions

Because this study is the first to quantitatively assess balance control and the influence of assistive devices during independent bathing transfers, we included only a limited number of assistance conditions. Although these conditions do not represent all options recommended by clinicians or situations people may encounter, they are common scenarios. Participants in this study used dry grab bars with dry hands, both of which are more representative of bathtub entry than exit. Future research should evaluate a greater number of grab bar locations and alternative devices, as well as the influence of wet hands and grab bars on the safety of these transfers.

Note that we did not control for entry and exit strategy. Participants were encouraged to enter and exit the bathtub using their preferred approach to allow them to enter using the strategy they perceived to be safest and most comfortable for each assistive device condition. Although this factor may increase the variability of our findings, it also provides greater external validity. Inspection of the data confirmed that the differences in strategy did not lead to outliers for any of the outcome variables.

Finally, this study investigated only uneventful balance transfers, not fall recovery responses. Bathroom assists are expected to contribute both to preventing an initial loss of balance and to assisting in recovery after balance is lost. Only the former function was evaluated in this study. Further research is required into the contributions of each device to supporting effective balance recovery responses.

Implications for Occupational Therapy Practice

Findings from this study can support clinical recommendations as follows: