Recommendations on Standards for the Design of Medical Diagnostic Equipment for Adults with Disabilities, Advisory Committee Final Report

IDeA Center, University at Buffalo

Memorandum:

Date: 4/23/2013

From: Clive D’Souza and Edward Steinfeld, IDeA Center, University at Buffalo

To: Rex Pace, Technical Assistance Coordinator, U.S. Access Board

RE: Revised Knee Clearance Depth and Footrest Clearance dimensions for sizing of mammography equipment

Introduction

This memo provides a revised description and analyses examining knee clearance dimensions in mammography equipment (as requested in your email dated: 4/15/2013). The analyses pertain to:

1. Abdomen depth (AD)
2. Footrest clearance height and depth

1. Abdomen depth (AD)

Abdomen depth was analyzed as the overall clearance depth beneath the mammography machine, and measured from the anterior-most aspect of the abdomen to the anterior-most aspect of the occupied wheelchair (e.g., to the footrest or toe-tip). The data were analyzed using both, univariate and multivariate methods.

Univariate analysis
The univariate analysis examines AD alone and is appropriate when no other clearance dimension is constrained or bounded (e.g., knee clearance height, foot clearance height or depth). Table 1 and Figure 1 show the percentiles and maximum observed values for male and female wheelchair users. Within device type, female users had slightly less AD compared to male users. The 95th percentile value for female users of manual and power wheelchairs was approximately 30.5 in. and would accommodate about 95% of female users assuming no other clearance dimensions is restricted or bounded. Based on the current study sample, accommodating the entire user population would require a clearance depth between 33 in. – 37 in.

Table 1: Percentile and maximum values for Abdomen Depth (in.) obtained from univariate analyses

Figure 1: Boxplot showing the distribution for abdomen depth (inches) across wheelchair type and occupant gender. The dotted line (30.5 in) depicts the 95th percentile value for female wheelchair users, both manual and power.

Note on Multivariate Analysis

The above univariate analysis is only applicable in the case of one critical dimension (i.e., abdomen to anterior-most point depth) and all other knee clearance dimensions are unrestricted or unbounded. If more than one dimension is bounded (e.g., combinations of knee clearance height or depth, foot clearance height or depth, occupied width, etc.) then a multivariate analysis which involves simultaneous consideration of multiple dimensions is more appropriate for performing clearance assessments. As the number of dimensions in the multivariate analysis increases it is possible that the resulting space requirements are larger in comparison to the univariate case where each dimension is analyzed separately.

2. Analysis of footrest clearance height and depth

Any base support or flange for the mammography machine at floor level (inside the clear floor space width) would obstruct some wheelchair users from approaching the mammography machine. The purpose of this analysis was to determine suitable dimensions for the base flange depth and thickness at floor level to minimize obstructions in forward approach for such users.

Over 30% of manual wheelchair users in this sample did not use footrests. Detailed footrest measurements (i.e., foot rest height and corner points on the top surface) were collected from 158 manual wheelchair users and 108 power wheelchair users. Footrests were present and recorded on the right-side for 101 (64%) manual wheelchair users and 105 (97%) of power wheelchair users. Front castor and footrest measurements were available for 67 power wheelchair users (excluding front-wheel drives for this analysis, due to time constraints). Three dimensions were computed on the right-side for the sub-set of wheelchairs with footrests:

1. Abdomen-Castor edge depth: measured as the horizontal distance from the abdomen to the distal (forward-most) edge of the front castor measured at castor hub height

2. Abdomen-Footrest distal depth: measured as the horizontal distance from the abdomen to the distal (forward-most) edge of the footrest

3. Footrest height: measured as the height from the floor to the top surface of the footrest at the proximal (near) outside corner

Both depth dimensions are referenced to the abdomen assuming that the wheelchair occupant makes contact with the breast-platform at the abdomen. The range and median values measured on the right-side for these three dimensions are presented in table 2. Critical values are highlighted in yellow. For each dimension, the more conservative estimate across manual and power wheelchairs is recommended. Based on the data in Table 2 we recommend that:

• The total knee clearance depth should preferably be no less than 30 in. to minimize contact with the footrest (i.e., 95th percentile value for Abdomen-Footrest Distal depth). The abdomen-anterior-most point depth presented in the previous section is a preferred dimension as the feet/toes often go past the footrest.

• Any flange or base support should preferably be no deeper than the overall clearance depth (e.g., 32 in. or 30 in.) minus the 95th percentile value for abdomen-castor edge depth. For example, for power wheelchair users, recommended flange depth = 30 in. – 19.3 in. = 10.7 in.

• Any flange or base support thickness should preferably be less than the 5th percentile value of footrest height for manual wheelchair users minus the thickness of the footrest (assuming 0.5 in.), i.e., approx. 1.7-0.5 = 1.2 in.

Table 2: Observed ranges and percentile values for Abdomen-Castor edge depth, Abdomen-Footrest edge depth, and Footrest Height, measured on the right-side for manual and power wheelchair users (Note: combined data for male and female). Critical values are highlighted in yellow.

[Click image above to view HTML version]

Figures 2 and 3 provide scatter-plots showing the relationship between Abdomen-Castor edge depth (and castor hub height) and Abdomen-Footrest distal depth (at footrest height) referenced from the abdomen point measured on the right side for manual wheelchair users (fig. 2) and power wheelchair users (fig. 3) respectively.

Assuming a 30 in. knee clearance depth, sample profiles of flanges based on the footrest measurements for manual and power wheelchair users are also shown. Although the flange depth and thickness is limited (e.g., 1-1.5 in.) the data suggests that the flange thickness could be increased closer to the base of the machine e.g., a triangular profile.

Further, it can be assumed that individuals that did not have a footrest (over 30% of manual wheelchair users) have some lower extremity mobility and would be able to lift and place their feet on top of the flange/base.

Figure 2: Scatter-plot showing observed values footrest height (on the vertical axis) and Castor-Footrest Depth (on the horizontal axis) measured on the right-side for manual wheelchair users referenced to the abdomen point. The grey curves show sample profiles of castor-wheels.

Figure 3: Scatter-plot showing observed values footrest height (on the vertical axis) and Castor-Footrest Depth (on the horizontal axis) measured on the right-side for power wheelchair users referenced to the abdomen point. The grey curves show sample profiles of castors-wheels.

Conclusions

Design criteria based on the univariate analysis of abdomen depth is sufficient if all other clearance dimensions are unbounded (i.e., unrestricted) and abdomen depth is the only relevant dimension. The 95th percentile value for female users of manual and power wheelchairs was approximately 30.5 in. and would provide adequate clearance depth such that approximately 95% of female users can approach the machine up to the abdomen.

Analysis of the footrest clearance height, abdomen-castor depth and abdomen-footrest depth suggests a flange of depth 10-10.7 in. and thickness 1-1.2 in. maximum. However, a graphical plot of the data suggests that the flange thickness could be increased closer to the base of the machine e.g., with a gradient or triangular profile.