19 October 1993

We have just completed a study of the energy requirements
of the Canadian Aerobic Fitness Test (CAFT). This is a
cadence-controlled, bench-stepping test of fitness where
the participant ascends two 20.3cm steps forwards and
descends the two steps backwards. There are 6-steps per
ascent-descent cycle. Our study involved testing a
group of younger females (< 35 yrs) and older females
(>65 yrs) at Level 1 (11 cycles per minute), The CAFT
predicts aerobic fitness on the basis of regression
equations. Recently, Shephard has calculated that
the efficiency of a person performing the CAFT is
approximately 14%. He calculated this using (a) the
so-called 'net' metabolic power (subtracting baseline);
and (b) estimating mechanical power based on a point-
mass model.

Our study involved collecting O2 data along with kinematics
of a 12-segment sagittal plane 2D model of the subjects.
Mechanical power was calculated in several ways:

1. Mechanical power assuming that energy can be exchanged
within and between segments

2. Mechanical power assuming only energy exchanges within
segments

3. Mechanical power assuming no exchanges of energy of any
kind

[The above are adapted from Pierrynowski et al. (1980)]

4. Mechanical power of the centre of mass assuming that an
exchange of energy types can occur (KE PE)

5 Mechanical power of the centre of mass assuming no exchanges
of energy types

6 Mechanical power calculated by the following simple equation:

(Mass x (2 x 0.203m) x 9.81)/ time of one cycle
[This is basically Shephard's method]

We also have the metabolic power data in both gross and net
formats.

>From the above 6 mechanical power values (numerator) and 2
metabolic power values (denominators), we can calculate
12 efficiency values (Oh no, not the dreaded "E" word!). We
believe that our "best" estimate of the efficiency of the
CAFT uses #1 above as our mechanical power and the gross
metabolic power, for an efficiency of about 28%. The data
look something like this:

Mechanical Power Gross O2 Power Net O2 Power
Wwb (#1 above) 28% 35%
Ww (#2) 30% 40%
Wn (#3) 38% 50%
CG-exchange not available yet not available yet
CG-no exchange not available yet not available yet
CG-simple calculation 12% 22%

After all this, our query to our learned colleagues is this:

In the absence of any gold standard that we can ascertain, what
is the concensus about the best way to calculate efficiency?
We believe we have something quite valuable to add to the body
of literature in this area. Our data refutes Shephard's assumptions
about the efficiency of the CAFT, and probably has implications for
other activities. However, we would appreciate the input of others
with more experience in this area.

We will be happy to post a synopsis of the responses we receive.

Thank you.

Drew Smith PhD Elaine Aimone MSc LYNDHURST HOSPITAL
Scott Thomas PhD Sara McConnell MSc UNIVERSITY OF TORONTO

************************************************** ************************
* Andrew (Drew) Smith, Ph.D. | INTERNET: awsmith@utcc.utoronto.ca *
* Director of Research | BITNET: awsmith@utorgpu *
* Lyndhurst Spinal Cord Centre | *
* 520 Sutherland Drive | Home Address: *
* Toronto ONT Canada M4G 3V9 | 33 Ponymill Drive *
* (416) 422-5551 x3041,3040 | Scarborough ONT Canada M1V 2X9 *
* (416) 422-5216 FAX | (416) 609-0469 *
* "Damn the torpedoes....full steam ahead!" *
************************************************** ************************