**The 15-point Borg scale 6**

**6. APPENDICES**

**APPENDIX ONE: Meeting to approve the fire-fighting simulation list for Phase Two **
**MEETING: Project Management Team: **

**Date:** 27/2/12

**Location:** Board Room Head Office (FRNSW).

**Present:** Chair: Alison Donohoe (FRNSW), Darren Husdell (FRNSW), Jim Hamilton
(FRNSW), Ken Murphy (FRNSW), Geoffrey Parkes (FRNSW), Brendan Mott
(FRNSW), Megan Smith (FRNSW), Nigel Taylor (UOW).

**Summary:**

(1) Introductions and welcome (AD).

(2) NT gave a very brief overview of the PAT review to date:

During focus groups performed at 11 stations across NSW, 50 physically demanding tasks were identified. 106 FF participated in the focus groups, from 11 stations that were

nominated by JH (DRO) & MB (DGMO) to give a good cross section of the organisation's population, considering gender, age, experience etc. The project management team and subject matter experts then looked at the 50 tasks to determine overlap and duplication etc.

and the task list was subsequently reduced to 30 tasks for inclusion in the survey. The survey went out to the organisation and we received approximately 250 paper based

responses and 750 electronic responses. The survey amongst other things asked staff to rank tasks according to frequency, critical importance, and difficulty involved. The results of the survey were then analysed using a filtration process which was detailed by NT utilising the Executive Summary for this phase of the research. The results of the filtration process identified 15 tasks for detailed task analysis. The 15 trade tasks were tabled as Appendix A for approval by all members of the Project Management team.

A minor amendment to the wording requested by JH, “Ladder use (10.5m) 1-person, under run and stabilise” to “Ladder use (10.5m) 1-person, under run”. JH expressed that this is required as the person footing the ladder is also assisting with the stabilisation.

The agreed task list is as follows:

1. Rolling out uncharged hose lines: 70 mm 2. Hydrant: Locating and connecting

3. Coupling and uncoupling hoses 4. Drag 70-mm charged hose: horizontal 5. Stair climb with PPE, BA and Hose 6. Prolonged use of 38-mm hose

7. Prolonged use of charged hose: 70-mm (two people) 8. Fire attack: prolonged crawl, kneel, crouch and squat 9. Ladder use (10.5 m) 1-person, under run

10. Rescue FF with PPE and BA: 1 person 11. Using spreaders and shears

12. Using sledge hammer to gain entry

13. Carry: ventilation fan (up stairs): 2 people

14. Hazmat: walking, manual handling (encapsulated)

15. Bush: drag charged hose (hilly, sloped and uneven)

This list was endorsed by the committee as the 15 tasks that should be used as the basis for the development of the physical employment standard.

NT provided overview of analysis performed on the tasks to date including explanations of the photos taken on the field testing, and that will appear in the report for phase 2 of the project. NT outlined that his team were able to borrow from the Department of Defence physiological monitoring devices which allowed the field studies to collect essential data.

The limited access to this equipment was the reason for commencing task analysis prior to final task list endorsement. The expectation that not all 15 identified tasks will be in the final standard was discussed.

It was acknowledged that a tiered approach to retained firefighter PATs would be

considered based on job demands at a various locations. The FRNSW Resource Allocation Model may be able to be utilised in this regard. It was discussed that DRO Jim Smith had expressed out of session that he would discuss this with the Senior Planner ORU LLC, plus a risk assessment would be conducted on each station to facilitate this process.

**ACTION: NT to provide report detailing final endorsed task list developed during phase 1**
of the project.

(3) It was unanimously agreed to have the wording “Trade” removed from in front of

“task” throughout the report. The title on the report is also to be amended to “The essential, physically demanding tasks of contemporary firefighting”.

**ACTION: NT to make necessary amendment to report. **

(4) NT: In the next phase UOW will utilise the data obtained during the task analysis to develop screening tests. Once these tests are developed FF will be involved in completing the screening test to receive feedback on appropriateness.

50 For dimensional analyses, three symbols will be used: M (mass), L (length) and t (time).

51 Surface area of a sphere = 4 * Pi * radius^{2}

52 Volume of a sphere = 4/3 * Pi * radius^{3}

**APPENDIX TWO: Allometric considerations of mass within occupational standards. **

The aim of this Appendix is to provide a theoretical justification for using power functions
in the normalisation of oxygen consumption to body mass. We will commence this exercise
with a consideration of a uniform geometric shape (the sphere), and explore the dimensional
relationships among 15 spheres, each with a radius 1 cm larger than its predecessor. From
this analysis, we are able to assemble several fundamental facts. For instance, for equal
increments in radius (L)^{50}, their surface areas (L*L or L^{2}) will increase in a curvilinear
manner (Figure A2-1A), such that doubling the radius results in a four-fold increase in
surface area^{51}. Thus, area increases as a square function of radius. The volumes (L*L*L or
L^{3}) of these spheres also increase nonlinearly (Figure A2-1B), so that doubling the radius
produces an eight-fold increase in volume^{52}; volume is a cube function of radius. Therefore,
these spheres are geometrically similar, or isometric objects (Schmidt-Nielsen, 1984).

However, similar relationships can also be used to describe objects of varying size, but with a uniform shape that does not follow these simple geometric (isometric) patterns.

**Figure A2-1: The relationships between area and volume, and the radii of spheres.**

53 Body surface area = 0.202 * mass^{0.425} * height^{0.725} (DuBois and DuBois, 1916).

54 Surface area of a sphere = 4.836 * volume^{0.67}

For example, humans are broadly similar in shape, so one may expect that body surface areas would change in proportion to the square of some linear dimension, as would be the case for an isometric object. However, this is not observed. Instead, human body surface areas, which are almost universally obtained using the DuBois and DuBois (1916)

derivation^{53}, increase approximately 75 cm^{2} for each cm of height gained, if mass remains
constant (Figure A2-2A). This is far from a square function, since a two-fold increase in
height yields only a 1.65-fold change in surface area. Moreover, when considered with
respect to body mass (which is dimensionally equivalent to volume: L^{3}), surface area
increases approximately 100 cm^{2}.kg^{-1} of mass change, if height is held stable (Figure
A2-2B). Thus, doubling the mass increases surface area by a factor of approximately 1.35.

**Figure A2-2: The relationships between human body surface area and height**
when mass is held constant (65 kg; Figure A2-2A), and with mass when
height is held constant (1.55 m; Figure A2-2B).

Thus far, we have considered surface area and volume only with respect to linear
dimensions. So let us now contemplate the inter-relationship between surface area and
volume. This relationship for spheres^{54} is illustrated in Figure A2-3A, for which the surface
area to volume ratio is the smallest (most efficient) of any three-dimensional object. It is
evident that the surface area is not linearly related to volume, unless this relationship is

55 Positive exponents reveal that variable y increases with increments in variable x, whilst negative exponents signify inverse relationships.

56 If humans grew isometrically, then segmental proportions in utero would be retained throughout life, but
this does not occur. Objects and organisms with isometric scaling possess the following characteristics: surface
area will vary as a function of the square of some linear dimension (L^{2} or the second power), and volume will
change as a cube function of that dimension (L^{3} or the third power; Schmidt-Nielsen, 1984).

plotted using logarithmic co-ordinates for both variables, as it is in Figure A2-3B. When
such scales are used, linearity is evident, and the slope (exponent) of this line will be
+0.67. This means that areal increases of a sphere are proportional to the 0.67 power of
volume (volume^{2/3}). Moreover, the relationship between surface area and volume varies as a
function of the size of the sphere, such that when the ratio of these variables is plotted
against spherical volume (Figure A2-3B), it decreases with increments in size, and the slope
of this line will be -0.33 (volume^{-1/3}; Schmidt-Nielsen, 1984)^{55}. Thus, smaller spheres have
greater relative surface areas. These relationships (rules) hold true for all isometric objects
(Schmidt-Nielsen, 1984).

**Figure A2-3: The relationships between the surface areas and volumes of**
spheres plotted using linear (A) and logarithmic co-ordinates (B).

Let us now return to humans, and consider how anatomical and physiological characteristics may change with increments in body surface area or volume. Whilst humans come in

different shapes and sizes, we are all of a similar configuration (generic shape). But we do
not possess isometric body shapes. In fact, as humans grow, we do so allometrically. That
is, we retain the same general shape, but not the same segmental proportions^{56}. The classical
example of this is evident for the head, which grows much faster during the first decade of

57 Normalising involves dividing the index of interest (e.g. oxygen consumption) by some variable that is
tightly correlated with that index (e.g. surface area or mass). Thus, the absolute oxygen consumption (L.min^{-1})
is converted to a relative (specific) oxygen consumption (mL.m^{-2}.min^{-1} or mL.kg^{-1}.min^{-1}). In these examples, it
is assumed that the relationship between the index of interest and the chosen divisor is always linear.

58 The word “specific” designates any quantity normalised to (divided by) body mass (Royal Society, 1975).

life than do the other body segments. Because of this allometric growth, the isometric
relationships of spheres (Figures A2-1 and A2-3) do not obtain in humans with the same
rigidity. However, an awareness of these relationships is critical to understanding and
evaluating how absolute values may differ among individuals, simply on the basis of size
variations. To address these variations, one may normalise^{57} the size of various anatomical
structures (brain, heart, lungs) or physiological functions (stroke volume, blood volume,
oxygen consumption) to some index of body size (body mass or surface area), so that
individuals of varying size can be more readily compared. Whenever body mass, which is
an analogue of volume, is the denominator of choice, the resultant output is known as a
specific^{58} variable (e.g. specific oxygen consumption; Royal Society, 1975).

In the resting state, the absolute oxygen consumption of any individual is function of body size, with perhaps the most useful anthropometric indices being the body surface area (Sarrus and Rameaux, 1839; Rubner, 1883; Seltzer, 1940) and body mass (Richet, 1889;

Kleiber, 1932, 1947, 1961). These relationships hold true across all mammals. For

instance, the absolute resting oxygen consumption of men and women differs, but when it is
normalised to either body mass or surface area (mL.kg^{-1}.min^{-1} or mL.m^{-2}.min^{-1}), this

difference becomes minimal. Thus, the absolute values are correctly interpreted to mean that the resting oxygen consumption of men and women differed mainly because of

variations in body size. Therefore, differences between resting men and women are merely gender-related, since men tend to be larger, but they are not gender-dependent.

During exercise, body mass has long been known to correlate better with oxygen

consumption than body surface area (Seltzer, 1940). Thus, exercising oxygen consumption data are often normalised to body mass. However, this normalisation is based upon a linear (arithmetic) assumption, such that, across the entire range of body sizes, the simple division of body mass into the absolute oxygen consumption will always permit one to compare the relative impact of a given physical activity upon different individuals, with the affect of body mass now being completely removed (body mass-independence). Notwithstanding its popular use, the significance of the difference between the absolute and specific oxygen consumption derived in this manner is often misunderstood, and the following discussion provides a more complete treatment of this topic, with a view to facilitating an

understanding of the data presented within this, and subsequent reports.

Whilst this linear mass normalisation is widely used and accepted, it does not mean that it is appropriate. For instance, it fails to account for all of the inter-individual variability in

59 Less than 25% of the variation in the resting, absolute oxygen consumption can be explained on the basis of variations in either body surface area (r=0.505) or body mass (r=0.412). However, during moderate

exercise, the predictive power of body mass is increased, and it can now explain about 60% of this variation, whilst during heavy exercise, it can account for about 75% of this variability (Seltzer, 1940). Nonetheless, there remains considerable unexplained variability, so the relationship is imperfect.

60 Data from 20 individuals (Taylor et al., 2012b): coefficient of variation for mass (kg) = 15.2; coefficients
of variation for oxygen consumption (L.min^{-1}): rest = 27.5, steady-state walking (4.8 km.h^{-1}) = 18.4.

61 Absolute oxygen consumption = a * mass^{0.75} (Kleiber, 1932).

Specific oxygen consumption = absolute value / mass or a * mass^{0.75} / mass.

Thus: specific oxygen consumption = a * mass^{-0.25}.

62 The coefficient of variability for the resting metabolic rate normalised to body mass was about 80% for animals ranging in mass from 150 g to 679 kg (Kleiber, 1932).

63 The coefficient of variability for the resting metabolic rate normalised to body surface area was about 34%

for animals ranging in mass from 150 g to 679 kg (Kleiber, 1932).

oxygen consumption^{59}. Moreover, the coefficient of variation for oxygen consumption will
often exceed that for body mass^{60}. This means that a wider range of exercising observations
may be found within the oxygen consumption than within the mass data, such that a simple
one-to-one relationship between oxygen consumption and body mass does not exist. These
facts have also long been known (Kleiber, 1947; Tanner, 1949; Taylor et al., 1981;

Schmidt-Nielsen, 1984; Åstrand and Rodahl, 1986), and, if ignored, can lead to specious data interpretation when normalising oxygen consumption during both resting and

exercising states, and particularly if one considers masses beyond the normal adult range Tanner (1949).

In the case of maximal exercise (e.g. peak aerobic power), one can observe a positive
relationship between the peak absolute oxygen consumption and body mass, but a negative
relationship is simultaneously evident between peak specific oxygen consumption (mL.kg^{}

-1.min^{-1}) and body mass (Taylor et al., 1981; Schmidt-Nielsen, 1984; Åstrand and Rodahl,
1986; Nevill et al., 1992; Bilzon et al., 2001a). Indeed, the same relationship was

demonstrated within resting animals almost a century earlier (Richet, 1889), it is predictable
on a first-principles basis^{61} and it follows the dimensional characteristics of spheres (Figure
A2-3B). Thus, normalising maximal exercise data for mass will disadvantage larger

individuals, whilst potentially inflating data for some smaller people (Åstrand and Rodahl, 1986; Nevill et al., 1992). These outcomes are artefacts of this form of normalisation.

Therefore, the injudicious division of mass into oxygen consumption can be invalid in many circumstances.

There is no doubt that normalising for body mass may help to explain some of the
variability among individuals in either the absolute resting and exercising oxygen
consumption. However, a significant amount of this variation will remain unexplained^{62}
(Kleiber, 1932). Normalising for body surface area will dramatically improve this state^{63} at
rest, and it was suggested that the surface area relationship may be associated with the need
to balance metabolic heat production against heat loss, with the latter being a function of
body surface area (Rubner, 1883). This area-specific procedure has been adopted as a
clinical convention for resting individuals, but it too is imperfect, whilst normalising for
body mass remains the method of preference within disciplines associated with exercise.

However, neither of these denominators is correct.

64 Also known as the “3/4-power law”.

65 The coefficient of variability for resting metabolic rate normalised to the 0.75 power of body mass was 7%

for ten groups of mammals (Kleiber, 1932).

66 Resting oxygen consumption (mL.s^{-1}) = 0.188 * mass^{0.75} (Kleiber, 1961).

67 Peak aerobic power (mL.s^{-1}) = 1.94 * mass^{0.79} (Taylor et al., 1981).

68 Biologists continue to debate the veracity of the 3/4-power law. For a recent discussion, see Glazier (2008).

Firstly, it is an inherent assumption of this normalising procedure that the linear function describing the relationships between absolute oxygen consumption and either body mass or surface area pass through the origin. That is, at zero body mass, the specific oxygen consumption of an individual will also be zero. Of course, this must hold true. However, the natural extension of this assumption is that this linear relationship, which has almost invariably been derived from experiments conducted using adults, will remain valid across the entire range of body masses. This is not correct. Indeed, when body mass or surface area standards for a variety of physiological functions are applied to individuals falling on either side of the mean obtained from the population sample used to construct the standard (e.g. cardiac function, oxygen consumption, plasma volume (Tanner, 1949)), then those individuals appear to deviate from normal purely on the basis of the difference between their size and that of the sample mean. This artefact increases as individuals approach the

extremes of body size (i.e. the confidence intervals widen; Schmidt-Nielsen, 1984), and so the extrapolation of such regression relationships beyond the range of primary observations is fallacious (Tanner, 1949; Schmidt-Nielsen, 1984).

Secondly, Kleiber (1932) found that normalising using body mass raised to the 0.75 power^{64}
provided a far superior explanation for variations in resting metabolism^{65}. That is, the
relationship was not arithmetically linear (one-to-one), but only became linear when graphed
on logarithmic scales (e.g. Figure A2-3). Thus, for one to compare the mass-independent
resting oxygen consumption of individuals of different sizes, one must derive specific
oxygen consumption as a power^{66}, and not as a linear function. Moreover, one must

absolutely base this relationship upon data obtained across the widest possible physiological range. The methods used by Kleiber (1932: mice to cattle) satisfy both of these criteria.

Some 50 years after this relationship was established for body mass and resting oxygen
consumption, Taylor et al. (1981) undertook an evaluation of its efficacy during maximal
exercise. Peak aerobic power was measured across a very wide range of body masses in
animals (7.2 g to 263 kg), and it too was found to be proportional to the 0.75 power of
body mass^{67}. Not surprisingly, subsequent confirmations of this power function have been
provided within maximally exercising humans (Åstrand and Rodahl, 1986; Nevill et al.,
1992), although the exponents have not always been 0.75. When such normalising is
applied, the bias that is inherent within the linear normalisation procedure disappears
(Åstrand and Rodahl, 1986). Indeed, it appears that, while the 0.75 power function is
appropriate across mammalian species (Kleiber 1932; Taylor et al. 1981; Schmidt-Nielsen,
1984), within a species, and during resting and exercising states, the exponent may be
closer to 0.67 (Heusner, 1982; Schmidt-Nielsen, 1984)^{68}. Therefore, whilst it is well
established that size is important, it is absolutely critical that we apply the correct scaling

69 For dimensional analyses, three symbols will be used: M (mass), L (length) and t (time).

70 Elastic criteria, which dictate the relationships between body mass and muscle dimensions, require the diameter of a muscle to conform to the 0.38 power of body mass (Schmidt-Nielsen, 1984).

71 Power = work / time or Power = energy use / time

function (power), rather than the more convenient (linear) function. Accordingly, this
convention has been adopted herein (mL.kg^{-0.67}.min^{-1}).

Let us now explore why these relationships should exist. Schmidt-Nielsen (1984: Pp. 83-86)
presented the case for this in considerable detail. Whilst a full reiteration of this is beyond
the scope of the current work, some key features are noted below for the reader, and these
are presented using the nomenclature of dimensional analysis^{69}.

Muscle force = tensile stress * cross-sectional area * shortening distance Power = [tensile stress] * cross-sectional area * [shortening distance / time]

Power = [M * L^{-1} * t^{-2}] * L^{2} * [L / t] Power = M * L^{2} * t^{-3}

However, instead of the first dimensional equation being simplified to the second (as shown above), the two parenthetical terms within this equation can be discounted, since they behave as physiological constants. In the first instance, the maximal tensile stress developed by skeletal muscle is a characteristic that is constant across species. It is independent of the size of an animal. Instead, it is determined by the actin and myosin filaments themselves, which are similar across species, as are the number of cross-bridges (Schmidt-Nielsen, 1984). Thus, muscle force is wholly dependent upon the cross-sectional area of the myocyte generating the force. Furthermore, the length and speed of muscle shortening will vary minimally across species (Schmidt-Nielsen, 1984). Thus, these terms become constants (k), and the equation for skeletal muscle power may be re-written as:

Power = k_{1} * L^{2} * k_{2}

This simplification is well known to muscle physiologists. However, it may be stated

another way. Maximal muscle power is a function of muscle diameter squared (L^{2}), and it is
proportional to body mass to the power 0.38 (Schmidt-Nielsen, 1984)^{70}.

Diameter is proportional to body mass^{0.38}

L^{2} is therefore proportional (body mass^{0.38})^{2} or body mass^{0.75}
Power = k_{1} * M^{0.75} * k_{2}

From this derivation, it can be seen that muscular power is related to body mass with an exponent of 0.75 (Schmidt-Nielsen, 1984).

Since exercise involves the extensive activation of skeletal muscles, then one can apply this generalisation to the entire, exercising musculoskeletal system and the consumption of oxygen to fuel that exercise (Schmidt-Nielsen, 1984). Thus, metabolic rate during exercise should be normalised to the 0.75 power of body mass, just as it was at rest. This is because it is an accepted convention to approximate metabolic rate from measures of oxygen

consumption (Kleiber, 1947), because this oxygen is used in the liberation of stored
chemical energy. This energy, in turn, enables the performance of work. Since both the
absolute and specific units for oxygen consumption are time derivates, then we are actually
obtaining an approximation of metabolic power^{71}, which has the same dimensional units
developed above for muscular power:

Force = mass * acceleration Force = M * L * t^{-2}

Work = force * displacement Work = M * L * t^{-2} * L = M * L^{2} * t^{-2}
Power = work / time Power = M * L^{2} * t^{-2} / t = M * L^{2} * t^{-3}.
From this treatment, one may conclude that variables related to power must be scaled using
a power function of body mass. Across mammalian species, the exponent would be 0.75
(Kleiber, 1947; Schmidt-Nielsen, 1984), but within a species, the exponent approximates
0.67 (Heusner, 1982; Åstrand and Rodahl, 1986; Nevill et al., 1992). As an extension of
this, Schmidt-Nielsen (1984) further demonstrated that variables related to frequency (e.g.
heart rate, breathing frequency) should be scaled to the -0.25 power of body mass.

**APPENDIX THREE: Meeting to report on, and approve the completion Phase Two**
**research activities **

**Date:** 21/5/12

**Location:** Board Room Head Office (FRNSW).

**Present:** Alison Donohoe (FRNSW), Darren Husdell (FRNSW), Megan Smith (FRNSW),
Brendan Mott (FRNSW), Jim Hamilton (FRNSW), Jim Smith (FRNSW), Nigel
Taylor (U0W), Lee Barlow (FRNSW)

**Apologies:** Gray Parks (FRNSW).

**Summary:**

(1) Previous Minutes of 27 February 2012 were accepted by all (AD).

(2) BM gave a general overview of the project to date and the purpose of this project management team meeting. The UOW research team is seeking the endorsement of Phase 2 of the research, specifically the 9 recommendations arising from the simulations conducted across the state. The UOW research team is also seeking approval from the project

management team to progress to phase 3 of the research project.

(3) NT discussed the Phase 2 report detailing and highlighting areas of importance for the project management team.

NT reinforced that the research and reports need to be robust and scientifically valid to withstand legal challenges and the UOW study is structured to provide this level of protection.

NT noted that, at present, the “Shuttle Run” is a valid field-based test for assessing

cardiovascular fitness, but it is not necessarily a defensible test for the physical screening of firefighters. NT advised it is likely that his team would be making a recommendation to replace this with a more appropriate test in the FRNSW physical employment standard.

NT also stated that there is the possibility that some of the existing PAT test components could be included in the new physical employment standard, however, this would need to be investigated in the next phase (phase 3) of the research.

NT provided the details of the data collected during the phase 2 simulations, and the methodology used to determine the physically demanding tasks that impose meaningful levels of physiological strain upon firefighters.

All 9 of the recommendations leading to the list of criterion firefighting tasks were discussed in detail. The criterion tasks were broken down into 4 classes detailed in the Executive Summary.

AD called for the endorsement of the criterion task list by all members of the project management team present. All agreed.

AD also called for the UOW research team to be provided with approval to progress to phase 3 of the project (development of physical screening tests). All agreed.