Independent Municipal and Allied Trade Union and Another v City of Cape Town (LC521/03) [2005] ZALC 10; [2005] 10 BLLR 1084 (LC); (2005) 26 ILJ 1404 (LC) (18 July 2005)

72 Reportability

Brief Summary

Labour Law — Employment discrimination — Blanket ban on employment of diabetics as firefighters — City of Cape Town refusing to appoint applicant based on insulin-dependent diabetes — Applicant alleging unfair discrimination under Employment Equity Act — Court finding that blanket ban constitutes unfair discrimination as it fails to consider individual assessments of medical fitness — Respondent's justification based on inherent job requirements not sufficient to uphold blanket exclusion.

REPORTABLE
In the matter between:                                                         Case No. C 521/2003
IMATU                                                                                    First Applicant
STUART MURDOCH                                                            Second Applicant
and
CITY OF CAPE TOWN                                                          Respondent
JUDGEMENT
MURPHY AJ
1. This case concerns the difficult but important question of whether the City  
of Cape Town’s imposition of a blanket ban on the employment of diabetics as  
firefighters amounts to unfair employment discrimination.
2. The respondent, the City of Cape Town, refused to appoint the second  
applicant, Mr. Stuart Murdoch (“Murdoch”) to the position of firefighter on the  
grounds that he is an insulin dependent diabetic. The first applicant (the  
union) alleges that the respondent directly discriminated against Murdoch on  
the grounds of disability, alternatively, on an arbitrary ground or an analogous  
unlisted ground, being his medical condition as an insulin dependent diabetic.  
Section 6(1) of the Employment Equity Act, Act No. 55 of 1988 (“the EEA”)  
prohibits any person from unfairly discriminating, directly or indirectly, against  
an employee, in any employment policy or practice, on one or more grounds,  
including disability. Section 6(2)(b) provides that it is not unfair discrimination  
to distinguish, exclude or prefer any person on the basis of an inherent  
requirement of a job.
3. The respondent denies that the imposition of the blanket ban constitutes  
unfair discrimination and asserts that the ban is fair and justified on the basis  
of the inherent requirements of the job of a firefighter. 
4. Section 11 of the EEA provides that whenever unfair discrimination is  
alleged in terms of this Act, the employer against whom the allegation is made  
must establish that it is fair. 
The background

5. Murdoch is a 31­year­old male currently employed by the respondent as a  
law enforcement officer in the Directorate: Protection Services. He has  
occupied this position since 1 July 1997 when he was employed as such by  
the South Peninsula Municipality, which has subsequently been incorporated  
into the respondent. He is an insulin dependent diabetic with Type 1 diabetes.  
He has been a volunteer reservist firefighter at Fish Hoek since 1991 when he  
was still in high school, initially for the Fish Hoek Municipality, and thereafter  
for its legal successors. As a law enforcement officer Murdoch is not required  
to render services as a firefighter. He has done this on a purely voluntary  
basis, though during times of big bush fires in the South Peninsula, he has  
been permitted to perform firefighting duties during his working hours.
6. The Fish Hoek Municipality was a municipality established in terms of the  
Municipal Ordinance 20 of 1974. In terms of the Local Government Transition  
Act 209 of 1993 it became a part of the South Peninsula Transitional  
Metropolitan Substructure on 1 February 1995, the South Peninsula  
Municipality on 28 May 1996, and then, in terms of the Municipal Structures  
Act of 1998, a part of the respondent on 6 December 2000. The   restructuring 
of respondent and its predecessor, the Cape Town City Council, and  the 35  
other local authorities in the Cape metropolitan area is an ongoing process.  The 
economic rationalization and organization of the respondent’s services,  
including fire and protection services, is taking place in a broader context of  
local government restructuring nationally.
7. On or about 28 January 1998, Murdoch requested an internal transfer from  
law enforcement to Fire Services, which also resides under the Directorate of  
Protection Services. At that stage, there was no vacancy. Vacancies arose for  
firefighters in the Department of Fire and Emergency Services towards the

firefighters in the Department of Fire and Emergency Services towards the  
end of 2002. Murdoch applied again for a transfer within the Directorate:  
Protection Services, from law enforcement to the position of firefighter on 2  
December 2002. 
8. As a part of the application process, Murdoch undertook physical tests at  
the Epping Fire Station and an assessment at a recruitment interview, which  
he passed in January or February 2003. Murdoch then underwent medical  
testing on 18 February 2003. 
9. Dr John Woolley, an occupational health medical practitioner employed in  
the respondent’s Department of Occupational Risk Management to assess  
the medical fitness of employees to perform the tasks assigned to them,  
examined Murdoch and found him to be medically unsuitable for the position  
of firefighter.
10. Woolley assessed Murdoch as part of a large group of new applicants. He  
took a brief medical history and had regard to a self evaluation form filled in by  
2

Murdoch. He then had regard to the job functions of a firefighter at the  
respondent but also generally as set out in various documents to which I shall  
refer more fully later. He looked also at some medical literature and  
occupational health guidelines and discussed the matter with colleagues, both  
fire protection officials and medical experts, including Professor Bonnici,  
Murdoch’s specialist physician, and concluded that in view of his insulin usage  
there was always a risk of Murdoch suffering a debilitating hypoglycaemic  
attack, which could prove disastrous on the fire ground. While accepting that  
Murdoch was an optimally controlled diabetic, and that much of the literature  
supported the idea that well controlled diabetics, individually assessed, should  
not be debarred, he felt a blanket ban was still justified. He moreover  
acknowledged that Murdoch was fit, well exercised and had good vision. As  
an occupational health practitioner he was evidently principally influenced by  
his apprehension that the increased risk of hypoglycaemia experienced by  
diabetics posed a real hazard for a firefighter wearing fire protective ensemble  
that is encapsulating and insulated, resulting in significant fluid loss and  
dehydration, especially in view of the unpredictable emergency conditions  
they work in for prolonged periods of extreme physical exertion, often without  
rest, meals or access to medication or hydration. In addition, because  
firefighters need to function as an integral component of a team, he feared  
sudden incapacitation through hypoglycaemia could pose a risk to other team  
members.
11. Woolley reasoned therefore that the appointment of Murdoch as a  
firefighter, given the occupational requirements of the job, would have  
represented an unacceptable safety risk to Murdoch himself, to other  
employees, to the general public and to the respondent by reason of his  
medical condition as an insulin dependent diabetic. He accordingly advised

medical condition as an insulin dependent diabetic. He accordingly advised  
the respondent that Murdoch’s condition as an insulin dependent diabetic  
rendered him medically unsuitable for that position in the respondent’s fire  
protection service. 
12.   At   the   time   this   dispute   was   referred,   existing   firefighters   who   were  
already in the respondent’s employ were not medically examined for diabetes.  
Nor is there a written policy or collective agreement dealing with the question.  
Neither   the   respondent   nor   any   of   its   predecessors   had   ever   assessed  
Murdoch’s medical fitness as a reservist firefighter in the light of his insulin  
dependent diabetes as there were no criteria pertaining to the recruitment of  
reservist fire fighters. During his cross­examination Woolley was referred to a  
document   prepared   by   the   respondent’s   occupational   risk   management  
department,   which   served   as   a   pre­placement   medical   job   specification   in  
respect  of  firefighters.  The  status  of  this  document  is  uncertain,   but  it  was  
notable that under the heading of specific medical  exclusions no reference  
was made to diabetes, asthma or obesity. The exclusion was limited to certain  
phobias and eye conditions. This compelled Woolley to the concession that  
the respondent had no explicit policy in relation to the other conditions or prior  
3

to this case a blanket ban in respect of Type 1 diabetes.
13. When the union’s representative, Chris Hagen (“Hagen”) objected to the  
rejection of Murdoch’s application for a transfer, the respondent referred  
Murdoch to Dr S E Carstens on 1 March 2003 for a second opinion. Dr  
Carstens is employed as a Principal Medical Specialist at the University of  
Stellenbosch in the Department of Community Health, where he has been  
Head of the Division of Occupational and Environmental Health since 2000. 
14. Carstens consulted with Murdoch on 19 March 2003 and evaluated his  
medical suitability for the position of firefighter. He had regard to Murdoch’s  
medical history, his medical condition,  Woolley’s opinion and information  
provided by him, including a letter dated 28 January 2003 from Murdoch’s  
treating doctor, Prof Francois Bonnici. Relying on these, his own specialist  
knowledge of occupational medicine and experience in the field, Carstens  
concurred with Woolley that Murdoch was not suitable for work as an active  
firefighter based on the risk that he represented to the respondent with his  
condition as an insulin dependent diabetic. He too accepted that Murdoch was  
physically fit and optimally controlled. However, because he supported the  
blanket ban for the same reasons as Woolley, he did not perform a medical  
examination on Murdoch, instead merely recording the following note in  
manuscript between transverse lines across the examination form:
Examination not done. Decision regarding “non fit for work as active fire fighter” not  
based on state of physical/clinical fitness but on the risk that insulin dependent  
diabetes carries regarding hypoglycaemia.
15. On 14 April 2003, Hagen sent an e­mail message to the respondent's  
senior human resources practitioner, David White­Phillips, requesting full  
reasons why Murdoch's application had been rejected, pointing out that  
Murdoch had been a reservist firefighter for more than 10 years and had

Murdoch had been a reservist firefighter for more than 10 years and had  
passed all the tests but was rejected solely on grounds which the union  
considered to be discriminatory.
16.   After   some   correspondence   passed   between   the   union   and   the  
respondent, the union received an e­mail response from White­Phillips on 18  
June   2003   purporting   to   be   "the   final   response"   regarding   the   non­
appointment   of   Murdoch.   White­Philips   stated   that   it   is   an   inherent  
requirement for firefighters to be physically fit to carry out their functions and  
hence   part   of   their   assessment   to   determine   physical   fitness   is   a   medical  
examination.   The   respondent's   occupational   medical   officer   had   found   that  
Murdoch was not physically fit to perform firefighter duties as he is an insulin  
dependent   diabetic,   a   condition   which   the   respondent   felt   could   place  
Murdoch, the respondent and the public at high risk during firefighting. Some  
five   months   later,   on   14   November   2003,   Murdoch   was   ordered   to   cease  
active firefighting duty and to commence watch room duties at the Fish Hoek  
Fire Station.
4

17. The dispute was referred to the Commission for Conciliation, Mediation  
and Arbitration (the CCMA) on 28 August 2003 in terms of section 10 of the  
EEA. The dispute could not be resolved at conciliation and thus was referred  
to this court for adjudication. 
18. Murdoch’s disappointment at the decision to exclude him is predicated on  
his   view   that   the   blanket   ban   unjustifiably   applies   outdated,   prejudiced  
stereotyping   to   his   individual   situation.   An   individual   assessment   of   his  
condition and experience, he contended, demonstrated that he is as capable  
as any other firefighter to perform the essential tasks of the job.
19.   Murdoch   has   actively   participated   in   many   aspects   of   firefighting.   In  
rigorous and prolonged bush fires, such as the fires that raged in the southern  
Cape Peninsula in 2000, he was actively on duty for shifts up to 34 hours.  
During those times, he has never put himself or others at risk as a result of his  
insulin dependence.
20. In his 13 years  of  active firefighting, Murdoch  has never had a severe  
hypoglycaemic episode (“a severe hypo”), defined by the medical experts as  
an episode where third party intervention is needed. Murdoch’s last severe  
hypo was at the age of 10 or 11, within the first year of being diagnosed with  
Type 1 diabetes.
21.   Murdoch   is   in   optimal   control   of   his   diabetes,   as   confirmed   by   his  
consultant physician for the past 22 years, Prof Bonnici, a pre­eminent expert  
in   the   fields   of   endocrinology   and   diabetes.   I   propose   to   examine   Prof  
Bonnici’s testimony fully later, however, the gist was that, because of his hypo  
awareness,   control,   motivation   and   education,   Murdoch   is   able   to   fulfil   the  
duties   of   a   firefighter.   He   also   submitted   that   blanket   exclusions   from  
hazardous   jobs   are   based   on   outdated   and   prejudiced   assumptions   about

hazardous   jobs   are   based   on   outdated   and   prejudiced   assumptions   about  
Type 1 diabetes and diabetes treated with insulin. In his expert opinion, each  
case   should   be   treated   on   its   own   merits   based   on   an   individual   medical  
assessment.
22. Both Murdoch and Tim Hoy, (a UK firefighter who also suffers with Type 1  
insulin dependent diabetes, yet has been a firefighter since 1986), testified  
that it is easy for them to access fast acting glucose while on duty, as and  
when needed to counter a hypo. They testified that they control their insulin  
levels before and during shift times and according to the exigencies of specific  
situations.   This   is   done   by   regularly   testing   their   blood   glucose   levels   –   a  
simple test that takes about 30 seconds – by injecting themselves with insulin  
when needed – that takes about 4 seconds, and making sure that they eat  
glucose when necessary. Murdoch and Hoy demonstrated in an experiment  
conducted before the court that they could access fast acting glucose within  
seconds   when   dressed   in   protective   gear.   Murdoch   also   testified   that   he  
5

habitually keeps a bag with a packed lunch on a fire tender when actively  
attending to fires. This is a precautionary measure in the unlikely event that  
the utility vehicle that distributes meals and rations to all  firefighters during  
incidents does not arrive. In his 13 years of firefighting, including the extensive  
fires in 2000, that has never happened.
The essential functions and duties of firefighters
23. Before turning to an assessment of the medical evidence on diabetes and  
the   respondent’s   risk­averse   policy,   it   is   necessary   to   reflect   upon   the  
essential functions, duties and requirements of firefighters. Evidence in this  
regard was adduced from Murdoch, Tim Hoy and Donald Sparks. Although  
the witnesses emphasized different aspects of the job, they were broadly in  
agreement about what the job actually requires and basically confirmed the  
information   contained   in   the   documentary   evidence   to   which   they   were  
referred.
24.   Mr.   Donald   Sparks   (“Sparks”)   has   had   40   years   experience   in  the   fire  
service where he rose to the highest level until his retirement from the position  
of Acting Chief Fire Officer in April 2005. He is a registered fire engineer and  
holds   an   MA   (cum   laude)   in   Public   Administration.   In   his   testimony   he  
confirmed that the catalogue of essential job tasks and descriptions contained  
in   NFPA   1582:   The   Standard   on   Comprehensive   Occupational   Medical  
Program   for   Fire   Departments   2003   Edition   of   the   USA   National   Fire  
Protection Association  (“NFPA 1582”) accurately reflect the job tasks required  
of firemen in Cape Town. The list is comprehensive and includes: 
• Performing   firefighting   tasks   (e.g.   hoseline   operations,  
extensive crawling, lifting and carrying heavy objects, ventilating  
roof or walls using power or hand tools, forcible entry) rescue  
operations,   and   other   emergency   response   actions   under

operations,   and   other   emergency   response   actions   under  
stressful   conditions   while   wearing   personal   protective  
ensembles   (PPE)   and   self­contained   breathing   apparatus  
(SCBA),   including   working   in   extremely   hot   or   cold  
environments for prolonged time periods.
• Wearing   an   SCBA,   which   includes   a   demand   valve­type  
positive   pressure   face   piece   or   HEPA   filter   masks,   which  
requires the ability to tolerate increased respiratory workloads.
• Exposure   to   toxic   fumes,   irritants,   particulates,   biological  
(infectious)   and   non­biological   hazards,   and/or   heated   gases  
despite the use of PPE including SCBA.
• Climbing 6 or more flights of stairs while wearing PPE weighing  
at least 50lb or more and carrying equipment/tools weighing an  
6

additional 20 to 40lb.
• Wearing   fire   protection   ensemble   that   is   encapsulating   and  
insulated. Wearing this clothing will result in significant fluid loss  
that   frequently   progresses   to   clinical   dehydration   and   can  
elevate core temperature to levels exceeding 39C.
• Searching,   finding   and   rescue­dragging   or   carrying   victims  
ranging   from   newborns   up   to   adults   weighing   over   200   lb   to  
safety despite hazardous conditions and low visibility.
• Advancing water­filled hose lines up to 2.5 inches in diameter  
from   fire   apparatus   to   occupancy   (approximately   150   ft);   can  
involve negotiating multiple flights of stairs, ladders and other  
obstacles.
• Climbing ladders, operating from heights, walking or crawling in  
the dark along narrow and uneven surfaces, and operating in  
proximity to electrical power lines and/or other hazards.
• Unpredictable emergency requirements for prolonged periods of  
extreme   physical   exertion   without   benefit   of   warm­up,  
scheduled   rest   periods,   meals,   access   to   medication   or  
hydration.
• Operating fire apparatus or other vehicle in an emergency mode  
with emergency lights and sirens.
• Critical, time­sensitive, complex problem solving during physical  
exertion   in   stressful,   hazardous   environments   (including   hot,  
dark   tightly   enclosed   spaces),   further   aggravated   by   fatigue,  
flashing lights, sirens and other distractions.
• Ability   to   communicate   (give   and   comprehend   verbal   orders)  
while   wearing   PPE   and   SCBA   under   conditions   of   high  
background noise, poor visibility, and drenching from hose lines  
and/or fixed protection systems (sprinklers).
• Functioning as an integral component of a team where sudden  
incapacitation of a  member  can result in  mission failure or  in  
risk of injury or death to civilians or other team members.

risk of injury or death to civilians or other team members.
25.   Sparks   highlighted   the   hostile   working   environment,   the   ever­present  
danger of injury and death, unpleasant working conditions, abnormal working  
hours and sustained stress levels. Firefighters spend extensive time outside  
exposed to the elements, tolerating extreme fluctuations in temperature while  
wearing equipment that significantly impairs body­cooling mechanisms. They  
7

work   in   wet   and   muddy   areas   performing   a   variety   of   tasks   on   slippery,  
hazardous surfaces such as on rooftops or from ladders. They are at constant  
risk of traumatic or thermal injuries, and face exposure to carcinogenic dusts  
such   as   asbestos,   toxic   substances   like   hydrogen   cyanide,   acids,   carbon  
monoxide,   or   organic   solvents,   either   through   inhalation   or   skin   contact.  
Accordingly, they are often required to rely on their senses of sight, hearing,  
smell and touch to help determine the nature of an emergency, to maintain  
personal  safety,  and to  make  critical  decisions  in  a  confused, chaotic, and  
potentially   life­threatening   environment   throughout   the   duration   of   an  
operation.
26. In order to cope firefighters are required to possess mental, sensorial and  
motor skills  sufficient to perform  these tasks under  such difficult conditions  
safely   and   effectively.   This   includes   the   ability   to   be   stable   with   regard   to  
consciousness and to have the functional capacity to respond appropriately to  
emergency   situations.   Equally   essential   is   the   ability   to   maintain   mental  
alertness and reliable judgement necessary to perform firefighting functions  
without posing a threat to self or others; and at the same time have acuity of  
senses   and   ability   of   expression   to   communicate   accurately   while   using  
equipment.   At   the   same   time   firefighting   requires   a   high   level   of   aerobic  
fitness, strength and good vision.
27.   The   operational   incidents   to   which   firefighters   are   exposed   vary  
considerably.   In   Cape   Town   they   include   ship   fires   where   firefighters   are  
exposed to extreme heat, poor visibility and stress in ship holds. Bush fires  
are also common in the peninsula often raging out of control for days, thus  
involving long shift work. A unique feature of firefighting in Cape Town are

involving long shift work. A unique feature of firefighting in Cape Town are  
informal settlement fires where numerous dwellings burn down and firefighters  
are tasked with the gruesome and stressful task of recovering burnt bodies.  
Other kinds of fires are those commonly found around the world, including:  
factory fires, with exposure to chemicals and explosive materials; hazardous  
material incidents involving spills of acids, radio­active substances, flammable  
solids, gases and corrosives; accidents; collapsed buildings; bomb blasts and  
ordinary house or commercial premises fires.
28. It was accepted by all the witnesses that Murdoch had the necessary state  
of physical fitness to perform the tasks of the job. He keeps fit by attending  
gym regularly, by cycling and jogging. Over 13 years as a firefighter he has  
accumulated   extensive   experience   in   bush   fires,   house   fires   and   vehicle  
accidents.   He   has   however   never   attended   a   ship   fire   and   has   limited  
experience in factory fires. Sparks in his testimony elaborated particularly on  
the hazardous nature of ship fire work and the potential danger such would  
pose to a diabetic, even though ship fires are not a common occurrence, there  
having been only about 10 or 12 during the past 3 years. Nevertheless, the  
exertion   expended   in  industrial   fires,   ship  fires   and   high­rise  buildings   was  
much greater and might more easily precipitate a severe hypo. These fires in  
8

particular involved intensive, prolonged firefighting under severe conditions. 
29.   Sparks   was   therefore   convinced   that   the   volatility   of   the   firefighting  
situation   coupled   with   the   unpredictability   of   the   possible   hypoglycaemic  
effects   of   decreased   blood   glucose   levels   brought   on   by   insulin   injections  
would   yield   an   unacceptable   risk   of   employee   failure   in   critical   emergency  
operations. Carstens and Woolley, as explained earlier, shared his disquiet. In  
particular, they felt that it would be difficult for Murdoch to maintain proper  
control of his blood glucose level during the unpredictable strenuous exercise  
involved in a sustained period of active firefighting. 
The nature, effect and treatment of diabetes: the medical evidence
30.   In   order   to   assess   the   probability   of   such   risks   materialising,   or   the  
rationality of the risk assessment analysis offered by Sparks, Carstens and Mr  
Henry   Rowen   (‘Rowen’),   the   respondent’s   expert   witness   on   risk  
management,   it   is   important   first   to   consider   the   expert   testimony   on   the  
nature, effect and treatment of diabetes. In this respect, as I have said, the  
court had the benefit of the evidence of South Africa’s leading authority on  
diabetes, Prof Bonnici, who is also an accredited expert internationally.
31. Of the three medical experts who testified only Prof Bonnici has expert  
knowledge   of   the   advances   recently   made   in   the   field   of   diabetes.   Drs  
Carstens   and   Woolley   are   both   experts   in   occupational   medicine   and  
understandably approached the issue from that perspective. In assessing the  
occupational   risk   posed   by   diabetes   they   were   compelled   to   rely   on   the  
specialist   literature   at   their   disposal,   much   of   it,   it   would   seem,   somewhat  
dated. Prof Bonnici, on the other hand, as I have already said, is a leader in

dated. Prof Bonnici, on the other hand, as I have already said, is a leader in  
the field and proved through his erudition and entertaining delivery to be a  
most impressive witness.
32.   Prof   Bonnici   is   the   immediate   past   Head   of   the   Division   of   Diabetic  
Medicine in the Department of Medicine, University of Cape Town and Groote  
Schuur Hospital: a position he held over 20 years. He holds the qualifications  
MBChB,   MMed,   FCP   (SA)   and   ADE.   He   has   published   and   lectured  
extensively on endocrinology and diabetes. He is also the recipient of awards  
for   outstanding   service   and   leadership   from   the   South   African   Diabetes  
Association, the International Society for Paediatric Adolescent Diabetes and  
the   International   Diabetes   Federation   (“IDF”).   He   was   the   head   of   the  
Paediatric and Adolescent Endocrine and Diabetes Unit, Groote Schuur and  
Red Cross War Memorial Children’s Hospital, and has a Diabetes Unit named  
after him at the University of Cape Town. He has conducted clinical research  
in drug development on an ongoing basis in Endocrinology and Metabolism,  
and   Type   1   and   Type   2   diabetes.   He   has   been   actively   involved   in   the  
development   and   evaluation   of   diabetes   education   among   health   care  
workers across the spectrum. He was the vice­president of the International  
9

Diabetes Federation; an organisation, which has 145 country affiliates, from  
1994 to 2000. He has been a member of the WHO Expert Advisory Panel  
(Diabetes) since 1997. He was president of the lay South African Diabetes  
Association from 1981 to 2004, and is presently their medical adviser. He is  
the   founder   member   and   immediate   past   president   of   the   Pan   African  
Diabetes   Study   Group,   an   advisory   council   member   of   the   International  
Society for Paediatric and Adolescent Diabetes, a member of the Consultative  
Section   on   Diabetes   of   the   International   Diabetes   Federation   and   has   an  
impressive array of publications and conference presentations. As luck would  
have it, he has also been Stuart Murdoch’s specialist physician for almost 20  
years.
33. Prof Bonnici's evidence has value for more than one reason. Firstly, he  
was   ideally   placed   to   pronounce   upon   the   nature   and   effects   of   diabetes.  
Secondly,   his   knowledge   and   involvement   in   clinical   research   in   drug  
development   allowed   him   to   discourse   on   the   implications   and   beneficial  
consequences   of   new   pharmaceutical   agents.   And   finally,   because   he   has  
treated Murdoch  for 20 years,  he could provide the individual  assessment,  
which   Carstens   had   considered   unnecessary   in   the   light   of   the   prevailing  
blanket ban.
34. In approaching Prof Bonnici’s evidence I am mindful of the need to do so  
with   some   measure   of   caution.   Prof   Bonnici   made   no   bones   about   his  
advocacy work in the field of diabetes and his fond regard for as well as his  
pride in Murdoch, who clearly has succeeded under his tutelage to become a  
model  patient and optimally  controlled diabetic.  A  mild  taint of partisanship  
seems unavoidable in the circumstances. But this must be measured against  
Prof Bonnici’s high calibre as a witness, a professional and a man of evident

Prof Bonnici’s high calibre as a witness, a professional and a man of evident  
unimpeachable   integrity.   While   ardent   in   his   testimony,   he   remained  
courteous under sustained cross­examination, open to criticism, respectful of  
the views of his colleagues with whom he disagreed, but firm in his conviction  
to   speak  his  mind   about  the  truth   of  diabetes  and  society’s  need   to  move  
beyond   prejudiced   stereotypes.   His   evidence   has   been   of   invaluable  
assistance for the purposes of this decision.
35.   Diabetes   mellitus   is   a   disorder   of   carbohydrate   metabolism   caused   by  
insulin deficiency, which results in glucose or blood sugar disorder. A person  
with diabetes cannot control his or her blood glucose and he or she becomes  
hyperglycaemic, meaning that they have abnormally high levels of glucose in  
the blood. Hyperglycaemia on a sustained basis produces long­term adverse  
medical side effects. It is a complex metabolic syndrome, which can cause  
damage to body organs, particularly the kidneys and eyes. The most effective  
means   of   treating   diabetes   is   through   the   subcutaneous   administration   of  
insulin by means of injections. Together with adjustments to diet and lifestyle,  
the   intake   of   insulin   enables   a   diabetic   to   control   blood   glucose   levels  
throughout   the   day   and   night.   Advances   in   pharmacology   and   technology  
10

have   made   insulin   injections   more   effective   and   easy   to   administer   and  
control.  The modern,   more  improved injections  came  in  ball­point­pen  type  
and size. They are easy and convenient to carry around and need not (unlike  
previous   products)   be   kept   in   a   refrigerator.   Murdoch   demonstrated   their  
handy application for the benefit of the court.
36. Most diabetics, including Murdoch, use two types of insulin, namely fast  
acting and long acting. Fast acting insulin becomes effective immediately after  
an   individual   has   injected   himself   and   is   used   normally   before   meals.   As  
eating will quickly elevate blood glucose levels, the fast acting insulin serves  
to   reduce   blood   glucose   to   normal.   The   long   acting   insulin   is   intended   to  
sustain the individual’s needs for a longer period of time and is usually taken  
every 24 hours. Murdoch administers long acting insulin before retiring to bed  
in the evening.
37. There are two types of diabetes. Type 1 diabetes is a disorder of sugar  
metabolism in which the pancreas is no longer able to produce insulin. This is  
due to the destruction of the insulin producing beta cells of the pancreas by  
auto­immune process. Type 1 commonly develops before the age of 40, with  
peak incidence around 14 years. The aetiology of the condition is not fully  
understood, but may follow from a genetic pre­disposition. Type 1 diabetes  
cannot be prevented but it can be detected and controlled, and is normally  
treated by means of insulin injections. Long term complications can be well  
controlled by using a tight control regime in which blood glucose is measured  
several times each day and the insulin dose adjusted accordingly. The risk of  
complications can be minimised by training Type 1 diabetics to take proper  
care of their condition by monitoring their blood glucose levels all the time. If  
they are physically active, they usually will check their glucose level before

they are physically active, they usually will check their glucose level before  
and during any physical activity and use insulin and carbohydrates to achieve  
the appropriate level. Murdoch is a Type 1 diabetic.
38.  Type   2  diabetes,   on  the  other  hand,   is   a  disease   in   which   the  person  
develops resistance to and relative deficiency of the hormone insulin resulting  
in high blood glucose levels. It is the most common variety of diabetes, and is  
most prevalent in people over 40 who are overweight. It is often associated  
with high blood pressure and elevated levels of fat in the blood. Unlike Type 1  
diabetes, Type 2 can often be prevented. Although it is not easy to control, the  
mainstay of treatment in Type 2 diabetes is lifestyle change, weight loss, a  
structured exercise programme and a low fat diet. However, in most people,  
even the correct diet along with exercise will eventually not be sufficient to  
control their blood glucose, accordingly drugs including insulin and agents to  
control   blood   pressure   and   high   blood   fats   have   to   be   used.   Prof   Bonnici  
speculated   that   in   the   not   too   distant   future   between   10­50%   of   the   adult  
population  over 50 will  be affected with the illness ­ a  telling indictment  of  
modern lifestyles.
11

39. As indicated, the long­term complications of   hyperglycaemia  in diabetics  
are   blindness,   heart   problems,   strokes,   amputations   and   kidney   failure.  
However,   the   risk   with   which   we   are   here   concerned   is   not   the   effects   of  
sustained hyperglycaemia ; rather it is the dangers of a  hypoglycaemic  attack 
(“a hypo”) within an occupational environment, especially a hazardous one.  
Hypoglycaemia occurs when blood glucose levels drop too low (between 2­4  
mmols/l). All people can experience a hypo. They are caused routinely by the  
intake   of   too   little   carbohydrates   (through   missed   meals),   hot   weather,   too  
much exercise, the use of alcohol and too much insulin. Because the very  
purpose of insulin is to reduce blood glucose levels it poses an added risk of  
causing   a   hypo.   Most   people,   diabetics   and   non­diabetics,   have   some  
awareness of the symptoms of a hypo. They include headaches, shaking or  
dizziness, anxiety, mood changes, sweating, palpitations, hunger, tingling lips  
or fingers, confusion, inability to concentrate and lack of co­ordination. Prof  
Bonnici likened the experience to that of pre­examination nerves. A mild hypo  
can be treated quickly by simply ingesting glucose tablets, soft sweets or slow  
acting carbohydrate such as a biscuit, sandwich, fruit or a glass of milk.
40. Beyond the normal discomfort brought on by a hypo, there are two areas  
of particular concern. When the blood glucose drops precipitously to a low  
level it may cause the sufferer to become unconscious. This is known as a  
“severe hypo”, and as already explained, is defined medically to mean a hypo  
requiring   the   assistance   or   intervention   of   a   third   party,   normally   a   friend,  
relative or colleague. The third party would have to administer glucose to the  
inert   patient.   The   second   problematic   situation   is   where   the   sufferer   has

inert   patient.   The   second   problematic   situation   is   where   the   sufferer   has  
become afflicted with hypoglycaemic unawareness as a result of either very  
severe and frequent hypoglycaemic reactions or autonomic nerve damage. A  
person   with   hypoglycaemic   unawareness   is   unable   to   recognise   the  
symptoms of a hypo and will not take preventive action, leading to a loss of  
consciousness. Prof Bonnici testified that Type 1 diabetics typically recognize  
the body warnings when the blood glucose levels are falling and are quick to  
take   preventive   action.   He   stated   also   that   hypoglycaemic   unawareness   is  
rare in Type 1 diabetics. Where it does occur, the patient needs special care  
and   re­training   by   experts.   He   accepted   that   hypoglycaemic   unaware 
diabetics should not drive or perform hazardous occupation.
41. On the basis of this general scheme, Prof Bonnici tendered the following  
opinion in relation to Murdoch’s fitness to perform the functions of a firefighter.  
Murdoch  is  a  Type 1 insulin  dependent diabetic who  uses  a  ballpoint  type  
insulin injection and a portable glucose meter. He has had tight control of his  
diabetes since the age of 9. Having only experienced one severe hypo at the  
age of about 10­11, he has hypoglycaemic awareness. He checks his blood  
glucose level and injects himself with insulin at least four times per day (at  
mealtimes and before going to sleep). Apart from that he takes blood glucose  
readings before physical and strenuous activity and has regular snacks. He  
suffers no complications from his diabetes, with no evidence of micro vascular  
12

disease or kidney problems. He regards Murdoch as highly motivated in his  
self­treatment and generous with his time and willingness to participate as a  
volunteer   in   several   studies   related   to   diabetes   and   the   development   of  
insulin. He described him as optimally controlled, and well educated in the  
management of his disease. He has known him since his childhood, saw him  
grow up into a strong and fit adolescent and believed he would encounter little  
difficulty in any physically demanding job, despite his diabetes. Hence, he was  
of the view that Murdoch was fit to perform firefighting functions.
42.   Prof   Bonnici’s   testimony   and   opinion   was   premised   obviously   on   his  
understanding  of  diabetes,   its  different  manifestations  and  the advances  in  
treatment.   This   body   of   knowledge   has   led   him   to   the   conclusion   that   a  
blanket ban on diabetics entering the firefighting profession is not justified on  
the basis of the medical evidence and modern treatment modalities. During  
his testimony he was referred to a number of studies, medical literature and  
policy   guidelines   upon   which   he   passed   comment   and   made   learned  
observations.   The   studies   provide   informative   insight   into   the   debate  
internationally.
43. As I understand his evidence and opinions, Prof Bonnici shares the views  
of HG Vaile and DA Pyke, the authors of the chapter “Diabetes mellitus and  
thyroid disorders” in RAF Cox   et al :  Fitness for work: The Medical Aspects  
(“Cox”). This text was canvassed with all the medical witnesses. Prof Bonnici  
confirmed his agreement with the following pertinent observation:
Despite   recent   advances   in   the   control   of   diabetes   the   condition   remains   poorly  
understood   and   it   is   sometimes   feared   by   employers   and   even   by   their   medical  
advisers. As a result, some diabetics still encounter unjustifiable difficulties in finding

advisers. As a result, some diabetics still encounter unjustifiable difficulties in finding  
and keeping work because of their condition. There is a paucity of published scientific  
data on the work experience of diabetics in general or in particular situations, e.g. in  
shift work, but there under representation in the work place suggests that there is  
continuing prejudice against their employment. The risk of hypoglycaemia and visual  
impairment may legitimately debar  poorly controlled  insulin dependent diabetics from  
jobs where safety or vigorous physical effort is an important factor, but diabetics are  
not invalids and most can work normally and should not be discriminated against in  
job selection.
44. Later in the text the authors look at special work problems and the work  
records of diabetics. Their remarks in this regard are equally instructive. Thus,  
they say:
There   has   been   a   great   change   in   recent   years   concerning   the   employment   of  
diabetics. Employers used to be frightened of diabetes and diabetics were frightened  
of employers. Both attitudes were due to ignorance. Now, the general public is better  
informed about diabetes, as about most other diseases. The result has been wholly  
beneficial; most employers realize that few occupations should be barred to diabetics.  
Diabetic employees are better  able to manage their  condition and less inclined to  
conceal their diabetes.
Occupations closed to insulin­taking diabetics are those in which a sudden loss of  
13

control or consciousness would be dangerous e.g. airline pilots or large goods vehicle  
drivers. The risk here comes not from the diabetes itself but from its treatment leading  
to hypoglycaemia. It may be unwise for insulin­taking diabetics to work in jobs where  
the danger would not be to others but to themselves e.g., with moving machinery, in  
foundries, on scaffolding, and firefighting.   But even here there is room for latitude.  
Much depends on the exact nature of the work, control of the diabetes, in particular  
the frequency and abruptness of hypoglycaemic attacks, and the good sense of the  
patient.
45. In their conclusions and recommendations the authors state:
Although much progress has been made in improving employers’ understanding of  
the   problems   of   diabetics   and   the   sickness   record   of   well   controlled   diabetics   is  
comparable   to   that   of   non   diabetics,   there   is   still   some   evidence   of   continuing  
employment prejudice against diabetics. Regrettably this seems to be due to lingering  
ignorance and fear of the condition among employers and their medical personnel. A  
continued  effort  is   necessary   to   educate   employers  and  persuade   them  to  take  a  
more objective view of diabetics…
It is essential that each individual case be assessed on its merits with full consultation  
between all medical advisers. Diabetes  per se  should not limit employment prospects,  
because the majority of diabetics have few, if any, problems arising from the condition  
and make perfectly satisfactory employees in a wide variety of occupations. 
46.   Taking   this   perspective,   Prof   Bonnici   was   critical   of   the   appraisal   of  
diabetes   contained   in   NFPA   1582   of   the   USA   National   Fire   Protection  
Association. NFPA 1582 classifies medical conditions into categories A and B.  
A   category   A   medical   condition   precludes   a   person   from   performing   as   a

A   category   A   medical   condition   precludes   a   person   from   performing   as   a  
member in a training or emergency operational environment by presenting a  
significant risk to the safety and health of the person or others. A category B  
medical condition is a condition that, based on its severity or degree, could  
preclude a person from performing as a member in a training or emergency  
operational   environment   by   presenting   a   significant   risk   to   the   safety   and  
health   of   the   person   or   others.   Unpacked,   category   A   conditions   invite   a  
blanket   ban   to   entry,   while   category   B   conditions   require   individual  
assessment   to   determine   whether   a   ban   should   apply   to   a   particular  
individual. Under paragraph 6.18 of NFPA 1582, dealing with Endocrine and  
Metabolic   Disorders,   diabetes   mellitus,   which   is   treated   with   insulin   and  
diabetes   not   treated   by   insulin,   which   is   not   controlled   as   evidenced   by  
Haemoglobin   AIC   (   Hbaic)   measurement,   are   both   category   A   conditions.  
Diabetes   mellitus   that   is   well   controlled   on   diet;   exercise   and/or   oral  
hypoglycaemic agents is classified as a category B medical condition.
47. Prof Bonnici rejected the premise in NFPA 1582 that a distinction should  
be   drawn   on   the   basis   of   treatment;   arguing   in   effect   that   insulin   taking  
diabetics are often better controlled than Type 2 diabetics who rely an oral  
agents and diet. With advances in analogue insulin and methods of injection  
he in actual fact favours the introduction of insulin to Type 2 diabetics earlier  
in their treatment, precisely to enable them to have better control. The idea  
that Type 2 diabetics on oral agents are better controlled and at lower risk is a  
14

myth. The distinction is not scientifically sound, principally because it takes no  
account   of   the   most   relevant   consideration   in   assessing   the   fitness   of   a  
diabetic,   namely   hypoglycaemic   awareness.   Indeed   he   went   so   far   as   to  
suggest that the opposite was true. Type 2 diabetics are probably at greater  
risk. They tend to be older, on less sophisticated medication and are not as  
well   informed.   Type   1   diabetics   are   better   controlled,   usually   seek   self­
management   training   and   have   greater   access   to   relevant   updated  
information. Accordingly, in his view, the distinction in NFPA 1582 is neither  
rational nor justifiable.
48. Prof Bonnici took a similar view of the classification contained in Guideline  
No 12 of the South African Society of Occupational Medicine (“SASOM”) of  
1996 where the following is recorded:
Diabetics controlled on diet or on diet oral agents are generally not prone to attacks of  
hypoglycaemia.   They   can   therefore   be   safely   employed   in   almost   every   situation  
exactly   like   a   non­diabetic.   They   are   not   however   allowed   to   become   commercial  
airline pilots.
Insulin dependent diabetics may be prone to hypoglycaemia but these attacks can be  
minimized with the help of a physician skilled in the management of diabetes. 
In   general,   insulin   dependent   diabetics   should   not   be   employed   where   they   may  
inadvertently   injure   themselves   or   others   in   the   event   of   a   hypoglycaemia   attack.  
They   should   not   work   on   scaffolding,   with   heavy   machinery   or   drive   heavy­duty  
vehicles.   They   should  not   normally   engage   in   underground   mining   or   where   easy  
access to food and medical help is denied. Insulin dependent diabetics should not  
normally work at heights i.e. in cranes or roofs. They should not work near open fires

normally work at heights i.e. in cranes or roofs. They should not work near open fires  
or furnaces or near high voltage installations. They should not work near unguarded  
moving machinery.
49.   Again   Prof   Bonnici’s   assessment   of   the   guideline   was   that   it   was  
unscientific   because   it   failed   to   attach   weight   to   considerations   of   greater  
relevance, namely the quality of control and the level of hypo­awareness. The  
proposition that diabetics controlled on diet, or on diet and oral agents, are  
generally not prone to hypoglycaemic attacks, he said, is simply scientifically  
incorrect. Once more the reverse is true. Those on oral agents may suffer  
more attacks, as they tend to be less controlled and more poorly managed.  
The distinction between insulin taking diabetics and those on oral agents is a  
false distinction or at least one of little predictive value. To the extent that the  
SASOM   guideline   advocates   the   individual   assessment   of   diabetics   for  
occupational   reasons   (although   seemingly   in   conflict   with   its   general  
proposition   about   diabetics),   Prof   Bonnici   agreed.   But   the   insulin/tablet  
dichotomy   is   merely   a   therapeutic   distinction   unrelated   to   the   causative  
factors, the type of disease or metabolic control.
50. The categorisations in NFPA 1582 and the SASOM guidelines were thus  
founded upon anachronistic and false assumptions. The more correct position  
is that reflected in Cox and by the International Diabetes Federation Europe in  
15

its   report   of   August   2003:   Discrimination   affecting   people   with   diabetes   in  
Europe – A survey of current status and initiatives , where it is stated:
Throughout Europe, political attitudes and legislation are reflecting a changing view,  
making discrimination unacceptable unless supported by convincing need (in which  
case it isn’t discrimination). The reality of life with diabetes is also changing rapidly,  
particularly in the past 5 years. The nature of the condition, new medical treatments,  
and the more appropriate focus of care on the individual with diabetes are all more  
clearly understood and more widely implemented in Europe, to the benefit of those  
with the condition, their colleagues, friends and families. In particular, new insulins, a  
wider   choice   of   oral   medications,   improved   delivery   systems   and   better   glucose  
monitoring have enabled tight glucose control while  reducing the risk of troublesome  
hypoglycaemia.
51.   Prof   Bonnici   was   furthermore   referred   to   a   report   emanating   from   the  
office of the Deputy Prime Minister in the UK dated September 2004 titled:  
Medical and Occupational Evidence for Recruitment and Retention in the Fire  
and Rescue Service,  which in Chapter 8 deals with endocrine disorders. The  
report   recommends   a   specific   policy   shift   and   approach   to   the   medical  
assessment   of   diabetics   seeking   employment   as   firefighters   in   the   UK.   It  
bemoans   the   fact   that   formulating   guidelines   for   the   safe   employment   of  
diabetics   in   the   firefighting   profession   is   difficult,   as   there   is   no   directly  
relevant   evidence­base.   It   draws   instead   however   upon   a   study   done   in  
Scotland   of   Type   1   diabetic   drivers   that   revealed   a   crash   rate   of   5.4   per  
100,000 miles driven, compared with 9.5 in a comparative non­diabetic group.

100,000 miles driven, compared with 9.5 in a comparative non­diabetic group.  
Other   studies   of   drivers   have   shown   no   major   discrepancy.   The   overall  
available   evidence   is   then   that   there   is   no   convincing   excess   of   accidents  
amongst   diabetic   drivers   on   insulin   treatment.   Accidents   due   to  
hypoglycaemia   do   occur,   but   the   report   concluded   that   they   are   relatively  
infrequent and are presumably offset by the majority of diabetic drivers acting  
in a particularly careful and responsible manner. Prof Bonnici concurred with  
this   conclusion,   arguing   that   diabetics   tend   to   be   overly   conscious   of   their  
condition   and   accordingly   act   more   responsibly   in   relation   to   potential   risk  
situations.
52. In the light of the evidence in relation to driving, the UK Government has  
accepted   the   position   of   Diabetes   UK   (a   diabetics’   interest   group)   and  
promotes a policy of individual evaluation for insulin treated diabetics being  
considered   for   potentially   hazardous   employment,   including   firefighting.   It  
accordingly   recommends   the   use   of   Diabetes   UK’s   Guidelines   for  
Employment   of   Insulin­Treated   Diabetic   Persons   in   Potentially   Hazardous  
Occupations   which   contain   strict   criteria,   emphasising   motivation   and   self­
care. The criteria are accepted in the UK and include:
• Physical   and   mental   fitness   in   accordance   with   non­diabetic  
standards.
• Diabetes   should   be   under   regular   (at   least   annual)   specialist  
16

review.
• Diabetes should be under stable control.
• Diabetic persons should monitor their blood glucose and be well  
educated and motivated in diabetes self care.
 
• There   should   be   no   disabling   hypoglycaemia   and   normal  
awareness of individual hypoglycaemic symptoms.
 
• There should be no advanced retinopathy or nephropathy, nor  
severe peripheral or autonomic neuropathy.
• There   should   be   no   significant   coronary   heart   disease,  
peripheral vascular disease or cerebrovascular disease.
 
• Suitability   for   employment   should   be   re­assessed   annually   by  
both an occupational and diabetes specialist physician.
53. The report then identifies specific coping strategies adopted by diabetic  
firefighters to minimize the risk of hypoglycaemic attacks and safely maintain  
a normal work pattern. The most common techniques and systems of self­
care are:
• In depth knowledge of diabetes and self care strategies.
• Commitment and motivation.
• Frequent and sensible self blood glucose monitoring
• The   ability   to   react   appropriately   to   particular   blood   glucose  
levels.
• Multiple insulin injection treatment (4 times per day).
• Use of analogue insulin which reduce hypoglycaemic risks.
• Available   supply   of   short   acting   and   long   acting   carbohydrate  
food on person.
• Running high (in terms of blood glucose) on duty e.g. perhaps 4­  
10  mmol/l off duty, but 6­12 mmo1/l on duty.
• Taking   carbohydrate   food   in   the   appliance   on   the   way   to   an  
incident.
17

54. Both the criteria and the self­care techniques furnish occupational medical  
practitioners   with   the   tools   of   assessment   for   application   in   evaluations   of  
individual diabetics seeking employment in potentially hazardous occupations.
55. As already explained, what is strikingly noticeable about the assessments  
conducted   by   Drs   Woolley   and   Carstens   is   that   none   of   these   tools   of  
assessment   were   brought   into   play   and   hence   considerations   which   one  
might have thought to be highly relevant were for all intents and purposes  
ignored. When Prof Bonnici was asked to comment on Woolley’s report he  
observed   that   nothing   in   it   referred   to   Murdoch’s   self­care   techniques,   his  
motivation and treatment regimen. Instead, he said, the occupational medical  
practitioners   had   simply   labelled   the   condition   and   automatically   excluded  
Murdoch   from   employment   as   a   firefighter.   He   noted   also   that   the   report  
reflected   no   analysis   or   comment   on   how   Murdoch   had   coped   with   his  
condition   during   his   pervious   experience   while   fighting   fires.   He   similarly  
criticized   Carstens’   report,   which   more   overtly   recorded   not   doing   an  
examination because of  his application  of  the blanket ban.  Accordingly, no  
assessment was made of Murdoch’s blood levels or consideration given to his  
hypo­awareness, his access to regular specialist review, the absence of any  
organ damage and the lack of a history of disabling hypoglycaemia.
56.   Carstens and Woolley offered some justification for the approach they  
took. Carstens in particular performs health risk assessments at workplaces  
and   is   often   required   to   give   recommendations   to   employers   regarding   an  
employee’s   ability   to   work   from   a   medical   perspective.   This   involves   an  
assessment of the specific requirements relating to medical fitness and the

assessment of the specific requirements relating to medical fitness and the  
general   inherent   requirements   of   the   job   in   order   to   qualify   the   risks   and  
hazards and the probability of an adverse factor having an adverse impact.  
For   this   purpose   he   applies   risk­rating   tools   usually   employed   by   risk  
management advisers (such as Mr. Rowen, whose evidence I discuss later),  
though   he   readily   acknowledged   that   epidemiological   evidence­based   tools  
are invariably more reliable. Normally when assessing the risks imposed by  
employment, in line with the approach in  Cox, an  individual  assessment is  
required. An exception to this, he felt, is when the nature and extent of the  
impairment   is   known   and   where   the   specific   requirements   of   the   job  
automatically   exclude   employment   of   a   person   with   an   impairment   of   that  
nature and extent. The occupational medicine practitioner then simply verifies  
the condition and recommends non­employment of that person. An obvious  
example would be where a blind person is considered not fit to drive a car.  
The function of the occupational medicine practitioner would be to ratify the  
blindness   and   to   recommend   non­employment.   There   is   no   need   to  
investigate   the   specific   history   and   circumstances,   or   to   do   an   individual  
assessment,   because   a   blanket   ban   applies.   In   Carstens’   opinion   the  
employment   of   an   insulin   dependent   diabetic   as   a   firefighter   falls   into   this  
category. Consequently, he saw his function as limited to verifying Murdoch’s  
insulin dependent diabetes and then recommending non­employment, which  
18

he did. 
Although he somewhat contradictorily accepted Cox’s individual assessment  
approach, Carstens believed the blanket ban was appropriate because of the  
unpredictability of the hypoglycaemic effects of decreased blood glucose level  
possibly occasioned by insulin intake. While he accepted the risk was less in  
well controlled diabetics, and that it could be further minimized by “running  
high”   (i.e.   eating   sufficient   glucose   before   attending   an   incident)   he   still  
considered   there   to   be   a   real   risk,   principally   because   a   firefighter   is   not  
always in a position to take supplements while fighting a fire. This, he felt,  
yielded an unacceptable risk of employee failure in a critical situation which  
could endanger the health and safety of the employee and others.
57. In support of his conclusions, Carstens referred the Diabetes Control and  
Complications Trial (DCCT), a study conducted in the USA during the 1980’s  
describing the epidemiology of severe hypoglycaemia and identifying patient  
characteristics or behaviour associated with severe hypoglycaemia in patients  
with insulin dependent diabetes. The study claims to have demonstrated that  
intensive  therapy,  with  the  aim of  achieving  glucose  levels as  close  to  the  
non­diabetic range as possible, was accompanied by a threefold increase in  
severe   hypoglycaemia   compared   with   conventional   therapy   and   that   there  
was   a   high   incidence   of   severe   hypoglycaemia   occurring   without   apparent  
warning   symptoms,   as   some   diabetics   lose   their   hypo   awareness.   Prof  
Bonnici, in his evidence, expressed reservations about this study, which he  
regarded   as   dated,   saying   also   that   many   in   the   international   diabetes  
community have raised ethical questions about it. The message that greater  
control   causes   more   hypoglycaemia   has   not   been   translated   into   the

control   causes   more   hypoglycaemia   has   not   been   translated   into   the  
international   guidelines,   particularly   in   Europe   and   Scandinavia,   where   the  
view remains that the intensification of control prevents hypoglycaemia. The  
study’s   reliability   is   open   to   debate   also   because   it   relied   on   old   types   of  
insulin   rather   than   the   recent   analogue   insulin,   which   has   proven   more  
effective and less prone to producing hypoglycaemia.
58. While Carstens readily admitted to not being an expert in diabetes, he  
remained convinced, despite Prof Bonnici’s evidence, that Type 2 diabetics  
dependent on diet control and oral agents were at less risk of hypoglycaemia,  
saying   that   there   was   no   overwhelming   evidence   to   the   contrary.   Still,   he  
acknowledged that he had not evaluated Murdoch’s level of hypo awareness,  
nor did he take account of the fact that in 13 years of firefighting Murdoch had  
never had to deal with the onset of hypoglycaemia during an incident which  
required   him   to   ingest   glucose   as   a   preventative.   Furthermore,   he   was  
unaware that Murdoch had not had a severe hypo in 20 years, or that his only  
severe hypo was as a 10 years old child at the early stages of the disease,  
prior to his being introduced to a tight control regimen. As I have explained, he  
considered   these   matters   irrelevant   because   of   the   blanket   ban.   He   held  
likewise in relation to other aspects of Murdoch’s individual experience and  
19

attached no weight to the fact that Murdoch was subject to regular reviews by  
the Diabetes Unit at UCT, that he had stable control, was highly motivated,  
educated in self­care and had no organ damage or vascular diseases. Nor  
was he particularly influenced in his opinion by Prof Bonnici’s evidence that  
modern   developments   in   insulin   therapy   and   self­management   justified   a  
revision  of   prevailing   attitudes   and  that   this  had  been   accomplished  in   the  
United Kingdom. He was adamant that his judgement still held and that even  
today   he   would   not   advise   the   employer   to   employ   a   diabetic   firefighter  
despite   the   revised   UK   guidelines.   In   his   view   there   was   not   enough  
epidemiological evidence to justify a change in the blanket ban. Yet ultimately  
he was compelled to concede that firefighters in employment developing Type  
2 diabetes, obesity, asthma or heart problems could pose a risk at least equal  
to that of a Type 1 diabetic and that the respondent had no policy to detect the  
onset of such medical conditions, to assess them or manage them by means  
of a blanket ban or otherwise.
59. Woolley’s support of the blanket ban was much in line with Carstens’ view.  
He   too   accepted   that   individual   assessments   should   be   the   norm   but   that  
some conditions justify a blanket ban. The fact that Murdoch was optimally  
controlled   and   that   modern   developments   indicated   a   shift   in   policy   and  
prevailing attitudes did not sway his opinion that there was an undisputed risk  
of hypoglycaemia with potentially disastrous consequences for the employer. 
60. Woolley was influenced in his decision to recommend non­employment by  
a study conducted at the Joslin Clinic by Bhatia and Wolfsdorf, referred to in  
NFPA 1582, involving the assessment of 196 insulin dependent adolescents  
and in which all of them experienced hypoglycaemia at least once during the

and in which all of them experienced hypoglycaemia at least once during the  
2­year   observation   period.   Of   these   15   percent   were   classified   as   severe,  
based on loss of consciousness, seizure, or the clinical need for therapeutic  
glucagons or intravenous glucose. It was of particular concern to him that 24  
percent of the hypoglycaemic episodes detected by blood glucose monitoring  
were   not   apparent   to   the   patients.   Thus   Woolley   concluded   that   about   a  
quarter of Type 1 diabetics were in danger of being hypo­unaware and this  
alone   justified   the   blanket   ban.   This   approach   stands   in   contrast   to   Prof  
Bonnici’s   view   that   hypo­awareness   should   be   assessed   individually   with  
reference to the personal history of the patient. Like Carstens though, Woolley  
accepted   that   there   was   merit   in   Cox’s   view   that   well   controlled   diabetics  
should   not   be   automatically   debarred.   Nevertheless,   puzzlingly,   in   his  
assessment of Murdoch he too attached no weight to his personal history, his  
hypo­awareness, access to regular review etc.
61.   Woolley   further   conceded   that   the   Joslin   Clinic   study   was   dated   and  
besides   indicated   that   76   percent   of   the   participants   were   indeed   able   to  
detect   the  onset   of   the  symptoms   of   mild   hypoglycaemia  and  as   such  the  
study seemed to support an individualized approach aimed at detecting a job  
applicant’s level of hypo­awareness.
20

62.   Carstens   and   Woolley   both   agreed   that   had   an   individual   assessment  
been   done   in   accordance   with   the   UK   Guidelines,   Murdoch   would   have  
qualified   for   employment   as   a   firefighter.   Prof   Bonnici’s   testimony   in   this  
regard   stands   unchallenged.   To   sum   it   up:   Murdoch   comes   from   a   stable  
family unit with parents highly motivated to support him in his illness, he is  
well educated about his illness, he has access to the best diabetes research  
unit   on   the   continent,   he   participates   regularly   in   clinical   trials   and  
experiments,   he   regularly  monitors   his   blood   glucose   levels   using  the  best  
innovative techniques, he uses insulin intelligently adapting his dosage to his  
immediate circumstances, thus avoiding the danger of fixed doses, he has no  
history of disabling hypoglycaemia, there is no organ or vascular disease and  
he uses the best analogue insulin.
63.   The   medical   evidence   reveals   that   modern   advances   point   towards   a  
different approach to the employment of diabetics in hazardous occupations.  
Those on the cutting edge, like Prof Bonnici, regard the fears of employers as  
over   cautious   and   unnecessarily   restrictive.   Occupational   medical  
practitioners,   on   the   other   hand,   are   less   influenced   by   the   grievances   of  
diabetics. They argue, albeit regretfully, that diabetics’ rights should yield to  
the employer’s duty to provide safe working environments in the interests of  
the public and its employees. They believe that individualized assessments  
can   only   ever   mitigate   the   risks   associated   with   employing   diabetics   in  
potentially hazardous occupations. The advantage of a blanket ban, properly  
conceived and consistently applied, is that it removes the risk entirely. This  
leads to the question of what precisely is the risk against which the employer

leads to the question of what precisely is the risk against which the employer  
needs to guard: how probable is its materialization quantitatively and what is  
the legitimate and proportional evasive response to it? 
The risk assessment
 
64. Looked at precisely, the risk facing an employer who employs a diabetic  
firefighter is in essence that the firefighter under the strenuous conditions of  
fighting   a   fire   may   suffer   a   severe   hypo   (requiring   third   party   intervention)  
resulting in injury to himself, his colleagues or members of the public. How  
real, probable or fanciful is such a risk in the light of advances in medicine and  
techniques of self­care and coping strategies, looked at not only in general but  
also taking into account the personal history of an applicant for employment?
65. All the medical experts accepted there was indeed some risk. Even Prof  
Bonnici conceded that there is always the possibility of a hypo. But, as he saw  
it,   health   problems   always   pose   risks   in   any   employment   scenario.   For  
instance,   he   reminded   us,   President   Kennedy   suffered   from   Addison’s  
disease. Even young, apparently healthy persons can experience sudden and  
unanticipated strokes, embolisms, or seizures. Asthmatics have been known  
to have such severe attacks that they lose consciousness or even die. In his  
21

view,   there   is   simply   no   feasible   or   fair   way   to   restrict   employment   to  
individuals who will  never have a calamitous health incident while on duty.  
Whether health risks materialize is determined primarily by the parameters of  
control.   A   neglectful,   poorly   controlled   diabetic   is   at   greater   risk   than   an  
optimally controlled one. Likewise a hypo­aware, optimally controlled Type 1  
diabetic   is   at   less   risk   than   an   undiagnosed   or   poorly   managed   Type   2  
diabetic.
66.  Sparks, the former Chief Fire Officer, described the possible risks on the  
fire ground most graphically. As stated earlier, he saw the dangers as most  
acute in fighting ship fires, fires in high rise buildings and hazardous material  
incidents,   especially   when   the   firefighter   would   need   to   wear   breathing  
apparatus   or   a   hazardous   materials   suit   (“hazmat   suit”)   for   a   prolonged  
period, or when operating a charged hose line. All these would lead to high  
levels of perspiration, excessive physical exertion and less  ability to detect  
hypoglycaemia and to treat it by popping a glucose tablet into the mouth. At  
this point, I pause to recall that the antidote to a hypo attack is merely the  
ingestion of a sweet or some glucose. Still, Sparks contended that there was  
no   guarantee   that   a   firefighter   would   remember   to   keep   glucose   on   his  
person. But even assuming he or she did have glucose, the exigencies of the  
situation might prevent its ingestion, for instance where it would be dangerous  
to remove the breathing apparatus or a glove, where it is impossible while  
wearing a hazmat suit, where the firefighter was trapped, while he was on a  
turntable ladder or where taking glucose out of his pocket would cause him to  
lose control of a charged hose. There are, in other words, many situations in  
which   a   firefighter   would   have   no   hands   available   or   would   lack   the

which   a   firefighter   would   have   no   hands   available   or   would   lack   the  
opportunity to administer the intake of glucose. Moreover, toxic substances  
might   pollute   the   glucose   itself   or   it   might   be   dropped   in   the   heat   of   the  
moment.   In   short,   the   peculiar   difficulties   of   firefighting   might   make   the  
ingestion of glucose impossible leading to a real risk of a severe hypo attack  
with the resultant danger enhanced by the hazardous environment.
67. The evidence of Murdoch and Hoy, referred to earlier, reveals some of  
these   legitimate   concerns   to   be   a   little   fanciful.   In   13   years   of   firefighting  
Murdoch has always carried liquid glucose (“glucogel”) in his tunic with the  
purpose of being able to counter the onset of hypoglycaemia by squirting it  
into his mouth if necessary. In all those years he has never needed to use it.  
He  has  worked  long  shifts   of  up  to  34  hours  without  incident.   Meals  have  
always   been   available   to   him   during   his   shifts,   but   he   takes   the   added  
precaution of storing his own food on the fire engine. He has from time to time  
been   required   to   use   breathing   apparatus,   but   pointed   out   that   standard  
procedure limited the use of breathing apparatus to 45 minutes and that he  
was able to ensure that his blood glucose levels run high for that period, if not  
longer.   He   admitted   to   not   having   used   a   hazmat   suit   and   conceded   that  
administering glucose while wearing one would be difficult, if not impossible.  
However,   he   explained   that   if   he   ever   had   to   use   one   he   would   before  
22

donning   it   simply   elevate   his   blood   glucose   (by   ingesting   glucose   and  
immediately testing) to a level sufficient to avert any risk of hypoglycaemia. In  
other words, to use the language of the UK Guidelines, he would “run high” for  
the   appropriate   period.   Here   we   see   the   significance   of   self­care   by   well  
educated, optimally controlled diabetics. In the past, diabetics were at greater  
risk   because   they   applied   fixed   doses   of   insulin.   Today,   because   of   short  
acting, analogue insulin and portable testing equipment, diabetics are able to  
adjust blood glucose levels to deal with the exigencies of a particular situation.  
That is why in 13 years Murdoch has never experienced a hypo on the job or  
had to leave the scene of an accident. Running high is thus a feasible option  
in all situations where the ingestion of glucose may prove problematic. The  
concern about handling charged hoses is diminished by the fact that many  
have shut­off valves at the nozzle, which can be done momentarily for the  
purpose of ingesting glucose.
68.   Hoy,   as   mentioned,   is   a   UK   based   firefighter   with   diabetes   and   the  
founder of  The International Register of Firefighters with Type 1 Diabetes . The  
purpose of the register is to show how firefighters with Type 1 diabetes cope  
with their illness in their work environment. He has been a firefighter since  
1985 and was diagnosed as a Type 1 diabetic in early 1987. Since then he  
has   performed   all   the   tasks   of   a   firefighter   and   currently   holds   rank   as   a  
Station Commander in London. Like Murdoch, he is an optimally controlled,  
educated diabetic. There are 200 firefighters on the register, 72 in the UK,  
who have clocked up 700 person years without a severe hypo – a fact born  
out to some degree by no witness being able to testify to having encountered  
any report or record of an incident or injury arising from a firefighter suffering a

any report or record of an incident or injury arising from a firefighter suffering a  
severe hypo while fighting a fire.  Hoy too has never experienced a hypo while  
on duty and has never had a severe hypo on duty or otherwise. He too will  
occasionally run high while on duty.
69. During his testimony, Sparks provided statistics on firefighter injuries while  
attending   emergency   incidents   in   Cape   Town.   In   the   three   years   between  
2002   and   2005   there   were   102994   emergency   calls,   169   injuries   at  
emergency calls and 63 injuries whilst performing station duties and training.  
There were no fatalities on duty. During that time there were 850 firefighters  
employed in Cape Town. This injury rate is commendably low. Expressed as  
an injury percentage of emergency incidents it translates as an injury rate of  
0,16%. Expressed as injuries per fireman it indicates one injury per fireman  
every 5 years. These statistics may be interpreted to mean that firefighting is  
less hazardous than one might otherwise assume. More likely, it reflects a  
laudable professional standard, a high level of safety awareness in the Cape  
Town   fire   service   and   that   the   hazardous   environment   is   in   the   main   well  
managed, and therefore generally less risky. 
70. This brings me to Rowen’s evidence. He is a consultant with 44 years  
experience   in   risk   management.   He   too   has   impressive   credentials.   He  
23

defined risk as the presence of uncertainty, measured as the variation from  
the   expected   outcome   of   a   given   situation.   Risk   assessment   involves  
identifying the elements of risk and establishing whether or not the elements  
are significant. Significance can be quantified on a scale of 1 to 5, 1 to 10 or 1  
to 100. One looks at the probability of occurrence and measures this against  
the likely severity of the outcome. He testified that it is impossible to use an  
exact   numerical   yardstick   of   measurement   of   probability   in   the   case   of   an  
insulin dependent diabetic firefighter being injured at a firefighting incident as  
there   can   be   no   assessment   prior   to   the   event   of   the   permutation   of  
circumstances   which   could   lead   to   a   hypoglycaemia   attack   coupled   with  
circumstances in which this could lead to severe injury or death. However, he  
relied on a risk­rating tool, which applied numerical values to three factors: the  
probability of an incident where the event occurs (“likelihood”), the frequency  
of occurrence of the event (“exposure”) and the consequence. In assessing  
the risk of a severe hypo on the fire ground causing injury or death, Rowen  
pegged its likelihood as “conceivable, but very unlikely (hasn’t happened yet)”  
with a corresponding value of 0,5 out of 10; its exposure as “continuous“ with  
a value of 10 out of 10; and its consequences as “very serious” with a value of  
15 out of 100. This generated a risk score of 75 (0,5 x 10 x 15 ) leading to a  
risk classification of “substantial risk: correction needed”. This then required  
the removal of the risk and the best way to achieve that was a blanket ban on  
the employment of diabetics as firefighters. 
71. Speaking frankly, I am not overly impressed with the scientific quality of  
this   risk   assessment   method.   Undoubtedly   it   can   be   applied   helpfully   in

this   risk   assessment   method.   Undoubtedly   it   can   be   applied   helpfully   in  
business   planning.   But   it   is   of   dubious   validity   in   the   justification   of  
discrimination impacting upon the dignity and identify of a class of individuals  
in   society.   It   relies   too   easily   on   generalized   assumptions,   and   in   this  
particular case attaches little significance to the critical factors of Murdoch’s  
optimal control and hypo­awareness or the possibility that his personal history  
might be the best predictor of future experience. Were one justifiably to adjust  
the exposure factor from “continuous” to “frequent”, on the basis that Murdoch  
always has glucose with him, runs high at incidents and is optimally hypo­
aware,   the   risk   classification   would   be   significantly   affected.   The   value   of  
frequent exposure being pegged at 6 out of 10, the risk score would be 45  
(0,5 x 6 x 15) resulting in a classification of “possible risk”. Likewise the risk of  
an overweight firefighter having a heart attack could be assessed as ”quite  
possible” (6), “unusual” (2) and  “very serious” (15) leading to a risk score of  
180 (6 x  2 x 15)  bringing it close to the high  risk classification justifying a  
blanket ban on hefty firefighters.
72. Accordingly, while I am certain the risk­rating tool has its valuable uses, I  
am loath to place much value on it for present purposes. In any event, Rowen  
acknowledged that the statistics drawn from the International Register pointed  
to the risk being minimal. 
24

73. The  proposition  that  the risk  of  a severe hypo on  the fire ground  is at  
acceptable   levels   is   supported   by   the   shifts   in   policy   witnessed   in   other  
countries, particularly in Europe. The epidemiological study relating to diabetic  
drivers in Scotland, referred to in the report emanating from the office of the  
UK Deputy Prime Minister, is probably the best (and perhaps only reliable)  
indicator   that   there   is   no   convincing   excess   of   accidents   among   insulin  
dependent diabetics. The International Diabetes Federation Survey of August  
2003   records   that   of   12   European   countries   surveyed   5   did   not   impose   a  
blanket ban on fire officers. At that time, the UK was one of the 7 imposing a  
ban. As Hoy explained in his testimony, with the acceptance of the Diabetes  
UK Guidelines, that ban was lifted in 2004, meaning that of the 12 surveyed  
European countries, 6 do not impose a ban. 
                                                                        
74.   Courts   in   other   jurisdictions,   by   upholding   a   requirement   of   individual  
assessments,   have   gauged   the   risks   posed   by   diabetes   in   potentially  
hazardous   occupations   as   not   warranting   blanket   bans.   For   instance,   in  
Bombrys   v   City   of   Toledo   849   F.Supp.1210 ,   the   City   of   Toledo   was  
permanently enjoined by the US Federal Court from implementing a blanket  
exclusion for persons with insulin dependent diabetes from employment as  
police officers.  In weighing up the risks, the court @ para 14 noted:
This court does not intend to belittle the very real concerns of the City of Toledo.  
Were Mr Bombrys in an emergency situation, he may not have the time to monitor his  
blood sugar. If he were to experience a drop in his blood sugar level, he may not  
have   the   opportunity   to   ingest   food   or   glucose.   This   court   recognises   that,   if   Mr

Bombrys were to become incapacitated while involved in an emergency situation, the  
consequences to him and to those around him could be tragic. However, this court  
also   recognises   that   officers   become   incapacitated   for   reasons   other   than   insulin  
dependent   diabetes.   There   are   police   officers   currently   on   the   force   who   are  
overweight   and   run   the   risk   of   heart   attacks.   Even   young,   presumably   healthy  
persons have been known to have sudden and unexpected strokes, embolisms, or  
seizures. Asthmatics have been known to have such severe attacks that they lose  
consciousness or even die. There is simply no viable or fair way to restrict a police  
force to individuals who will never have a catastrophic health incident while on duty.  
The best that the City of Toledo can do is to evaluate each police officer candidate on  
a case­by­case basis and determine what risks that individual presents to him/herself  
and   the   public.   An   individual’s   medical   history   and   record   of   compliance   with  
physician’s recommendations seem to be an ideal place to begin such an evaluation.  
In short, before the City may determine that an individual poses a threat to the health  
of safety of others, it must develop and apply an evaluation process that will comply  
with the three mandates of the [Americans with Disabilities Act] … An individualised  
assessment is absolutely necessary if persons with disabilities are to be protected  
from unfair and inaccurate stereotypes and prejudices.
75. Similarly, in  Kapche v City of  San  Antonio   304 F  3d  493  (2002),  the Court  
of Appeals held that an individualised investigation of the plaintiff’s ability to  
perform the job was required. In that case, the City had deemed the plaintiff  
ineligible for a police officer position due to his insulin dependent diabetes.  
The   Court   of   Appeals   reiterated   its   earlier   position   in   Chandler   v   City   of  
Dallas  511 US 1011  that:

Dallas  511 US 1011  that:
25

“We nonetheless share the hope…..that medical science will soon progress to the  
point that ‘exclusions on a case­by­case basis will be the only permissible procedure;  
or,   hopefully,   methods   of   control   may   become   so   exact   that   insulin   dependent  
diabetics will present no risk of ever having a severe hypoglycaemic episode.”   
In the interim, however, the court held that an individualised assessment of  
the applicant’s present ability to safely perform the essential functions of a  
police officer was required.  
76.   The   Canadian   courts   have   shown   similar   resolve   and   preference   for  
individualised   assessments.   In   Nowell   v   Canadian   National   Railway   Ltd  
[1987]   DLQ   8   the   complainant   lodged   a   complaint   against   the   respondent  
under   the   provisions   of   the   Canadian   Human   Rights   Act,   alleging   that   the  
respondent’s policy of excluding insulin dependent diabetics from the position  
of   trainman   was   a   discriminatory   practice.   The   Canadian   Human   Rights  
Tribunal   pointed   out   that   there   is   a   difference   between   an   occupational  
requirement  which sets  down a particular working condition  (i.e.  wearing  a  
certain piece of equipment on the job) that applies to all employees and an  
occupational requirement which excludes a whole group of individuals from a  
particular job because of a certain physical disability. It held @ para 33:
Where   the   occupational   requirement   excludes   a   whole   class   of   individuals   with  
varying   degrees   of   disability   within   the   class   (according   to   the   medical   evidence)  
there should be, in the interests of fairness and justice, individual assessment within  
the  group  that  is excluded  to determine if  there is sufficiency  of risk  to justify  the  
exclusion of that particular employee from the job. 
Nowell, an insulin dependent diabetic was held not to pose a sufficient risk of

Nowell, an insulin dependent diabetic was held not to pose a sufficient risk of  
employee failure to justify his exclusion from the position of trainman and he  
was  awarded  damages.  The tribunal’s comments  in assessing  the putative  
risk of Nowell having a hypoglycaemic episode are especially relevant:
The sufficiency of risk was not proved in this case. There was no evidence of the  
likelihood of Mr Nowell’s suffering an incapacitating reaction because of his diabetes.  
In fact, the evidence was to the contrary. In 15 years as a diabetic, he has never  
suffered an incapacitating reaction. He is a well­controlled diabetic who is physically  
fit to do the job of trainman.
77.   Similarly,   in   McKenzie   v   Quintette   Coal   Ltd   (1986)   8   CHRR   D/3762  
(BCCHR) ,   the   British   Columbia   Human   Rights   Council   found   that   Quintette  
Coal   Ltd  discriminated  against   McKenzie   when   it  refused   to  hire  him   as  a  
miner because he was an insulin dependent diabetic. McKenzie’s evidence  
and expert testimony indicated that McKenzie’s diabetes was stable and well  
controlled and that the harsh environmental conditions and long hours under  
which miners worked posed no greater risk to McKenzie than to non­diabetic  
workers. The following remarks of the tribunal are especially apposite:
Human rights legislation seeks to create equal opportunity in competition for jobs.  
26

Some risk of injury exists in all jobs and to deny a disabled person an opportunity  
solely   because   of   a   perceived   risk   to   the   individual   defeats   the   purpose   of   the  
legislation.
The tribunal found that McKenzie had learned to live with his diabetes without  
experiencing significant problems. His jobs had involved long hours and often  
extended   and   changing   shifts.   He   had   experienced   no   difficulty   with   his  
diabetes and had never lost consciousness, required assistance, had to leave  
work   or   taken   time   off   for   any   illness   or   injury.   The   tribunal   found   the  
respondent’s blanket policy of refusing to employ insulin dependent diabetics  
as   miners   was   not   a   bona   fide   occupational   requirement.   Additionally,   the  
respondent’s decision not to employ McKenzie was not based on an individual  
assessment   that   was   objectively   related   to   the   performance   of   the  
employment concerned.
 
The discrimination analysis
78. In South Africa the matter has to be determined within the framework of  
section 6 of the EEA. Section 6(1) reads: 
No person may unfairly discriminate, directly or indirectly, against an employee in any  
employment policy and practice, on one or more grounds, including race, gender,  
sex,  pregnancy,  marital  status, family  responsibility,  ethnic  or  social  origin, colour,  
sexual   orientation,   age,   disability,   religion,   HIV   status,   conscience,   belief,   political  
opinion, culture, language and birth.
79. Section 6(2)(b) provides as a defence that it is not unfair discrimination to  
distinguish, exclude or prefer any person on the basis of an inherent  
requirement of a job. Moreover, section 11 of the EEA provides that whenever  
unfair discrimination is alleged, the employer against whom the allegation is  
made must establish that it is fair. This in effect creates a rebuttable  
presumption that once discrimination is shown to exist by the applicant it is

presumption that once discrimination is shown to exist by the applicant it is  
assumed to be unfair and the employer must justify it­  Jooste v Score  
Supermarket Trading (Pty) Ltd (Minister of Labour Intervening)  1999 (2) SA 1  
(CC);  and  Hoffmann v South African Airways  2000 (2) SA 628 (W). Once  
discrimination has been established, the employer will have to prove that the  
discrimination was fair or have to justify the discrimination as justifiable under  
section 6(2)(b). It is common cause that it is the respondent’s employment  
policy or practice not to employ insulin dependent diabetics as firefighters.  
The question, therefore, is whether that constitutes unfair discrimination.
80. The approach to unfair discrimination to be followed by our courts has  
been spelt out in  Harksen v Lane NO and Others  1998 (1) SA 300 (CC).  
Although the  Harksen decision concerned a claim under section 9 of the  
Constitution (the equality clause), there is no reason why the same or a  
similar approach should not be followed under the EEA.  
81. The   Harksen   approach contains a specific methodology for determining  
27

discrimination cases. The first enquiry is whether the provision differentiates  
between people or categories of people. If so, does the differentiation bear a  
rational connection to a legitimate governmental purpose? If it does not, then  
there is a violation of the guarantee of equality. Even if it does bear a rational  
connection, it might nevertheless amount to discrimination. The second leg of  
the enquiry asks whether the differentiation amounts to unfair discrimination.  
This requires a two­staged analysis. Firstly, does the differentiation amount to  
“discrimination”? If it is on a specified ground, then discrimination will have  
been established. If it is not on a specified ground, then whether or not there  
was discrimination would depend upon whether, objectively, the ground was  
based on attributes and characteristics which had the potential to impair the  
fundamental   human   dignity   of   persons   as   human   beings   or   to   affect   them  
adversely   in   a   comparably   serious   manner.   Secondly,   if   the   differentiation  
amounted to “discrimination”, did it amount to “unfair discrimination”? If it is  
found to have been on a specified ground, unfairness will be presumed under  
the Bill of Rights by virtue of the provisions of section 9(5) of the Constitution,  
which transfers the onus to prove unfairness to the complainant who alleges  
discrimination on analogous grounds.  As I  read section 11 of the EEA,  no  
similar transfer of onus arises under the EEA. In other words, whether the  
ground is specified or not the onus remains on the respondent throughout to  
prove fairness once discrimination is shown. In the context of the EEA section  
6(2)(b) also permits justification on the basis of an inherent requirement of a  
job, in which event the discrimination is deemed not to be unfair. The onus in  
this respect is also on the employer.
82.   The   impact   of   the   discrimination   complained   of   on   the   complainant   is

generally   the   determining   factor   regarding   the   unfairness   of   alleged  
discrimination. Factors which must be taken into account include: the position  
of the complainants in society and whether they have suffered in the past from  
patterns   of   disadvantage;   the   nature   of   the   provision   or   power   and   the  
purpose sought to be achieved by it; the extent to which the discrimination  
has affected the rights or interests of complainants and whether it has led to  
an   impairment   of   their   fundamental   human   dignity   or   constitutes   an  
impairment of a comparably serious nature.
83.   In   the   pre­trial   minute   the   respondent   admits   differentiating   between  
Murdoch and other persons in an employment policy or practice on the basis  
that   he   is   an   insulin   dependent   diabetic.   However,   it   avers   that   its  
differentiation bears a rational connection to a legitimate government purpose.  
As   a  local   authority   the   respondent   bears   a   duty   to   provide   fire   protection  
services in terms of section 155(6)(a) and (7) read with Part B of Schedule 4  
of the Constitution, read also with the provisions of the Fire Brigade Services  
Act 99 of 1987. The respondent also has a duty in terms of section 8 of the  
Occupational   Health   and   Safety   Act   85   of   1993   (“OHSA”)   to   provide   and  
maintain, as far as is reasonably practicable, a working environment that is  
safe   and   without   risk   to   the   health   of   its   employees.   Section   9   of   OHSA  
28

creates the same duty on the employer in respect of third parties. In deciding  
what is “reasonably practicable” the employer must have regard to amongst  
other things the severity and scope of the hazard or risk concerned; the state  
of knowledge reasonably available concerning that hazard or risk and of any  
means of removing or mitigating that hazard or risk and the availability and  
suitability of means to remove or mitigate that hazard or risk. 
84.   As   under   the   common   law,   therefore,   the   respondent   must   take  
reasonable steps  to  prevent  reasonably  foreseeable harm.  If it does not,  it  
could be sued in delict for damages sustained by third parties under section 9  
as   well   as   the   common   law,   and   could   face   a   claim   for   increased  
compensation from the employee under the Compensation for Occupational  
Injuries and Diseases Act 130 of 1993. 
85.   Government’s   imposition   of   these   duties   on   employers   and   local  
authorities   aim   at   the   protection   of   the   community   and   advance   the   public  
interest. I accordingly accept that the purpose of the respondent in imposing  
the ban on the employment  of  Type  1  diabetics  as firefighters  was one of  
ensuring public safety and limiting public liability. 
86.   Much   of   the   documentary   evidence   reveals   that   blanket   bans   against  
insulin   dependent   diabetics   becoming   firefighters   have   been   customary  
elsewhere, and the respondent’s medical and risk advisers considered these  
before   making   their   recommendations.   Bans   were   recommended   by   the  
United Kingdom’s Home Office (prior to 1 October 2004); the United States  
National Fire Prevention Agency (“NFPA”), the Queensland Fire and Rescue  
Authority in Australia, and, as we have seen, were imposed by at least 6 of 12  
European countries surveyed by the IDF. It is a matter on which reasonable

European countries surveyed by the IDF. It is a matter on which reasonable  
people   may   differ,   meaning   in   the   final   analysis   that   a   ban   is   not   entirely  
beyond   rationality.   Prof.   Bonnici   also   confirmed   that   in   the   past   insulin  
dependence  carried  a  high  risk of  hypoglycaemia  due  to  the nature  of  the  
treatment, particularly before the invention of fast­acting insulins. The history  
of the blanket ban against Type 1 diabetics therefore comes not so much from  
unfair prejudice against people with diabetes, but from the factual history of  
treatment of diabetics dependent on insulin. In the past there was a real risk  
of hypoglycaemia. Although the respondent concedes that insulin treatments  
have improved and have reduced the risk, there are, in its view, insufficient  
statistics   or   clinical   trials   to   show   the   extent   to   which   the   risks   have   been  
reduced.   In   the   absence   of   evidence   to   the   contrary,   Mr   Kantor,   who  
appeared   for   the   respondent,   submitted   that   government   agencies   act  
reasonably in adopting a risk­averse approach.
87.   The   differentiation,   as   such,   can   be   seen   as   a   legitimate   method   of  
guaranteeing public safety and certainly bears a rational relationship to that  
objective.   At   this   stage   of   the   enquiry   the   question   is   not   whether   the  
government may have achieved its purposes more effectively in a different  
29

manner, or whether its regulation or conduct could have been more closely  
connected to its purposes. The test is simply whether there is a reason for the  
differentiation that is rationally connected to a legitimate government purpose  
­   East   Zulu   Motors   (Pty)   Ltd   v   Empangeni/Ngwelezane   Transitional   Local  
Council and Others   1998 (2) SA 61 (CC) @ 73 C­D. The respondent has a  
legitimate government purpose in taking steps to reduce the risk of harm to  
employees in its fire service. Nor has it been suggested by the applicants that  
the   respondent’s   purpose,   to   ensure   public   safety,   is   not   legitimate.   In   the  
premises,   I   accept   that   the   respondent’s   differentiation   bears   a   rational  
connection   to   a   legitimate   government   purpose   and   is   therefore   mere  
differentiation   and   not   per   se   a   violation   of   the   right   not   to   be   unfairly  
discriminated   against.   Had   it   been   otherwise   the   matter   would   have   been  
resolved in favour of Murdoch on that basis alone.
89.   The   next   leg   of   the   inquiry   invites   consideration   of   whether   the  
differentiation is on a specified or listed ground. If it is, it will be discrimination  
and   can   be   presumed   unfair.   If   not,   the   respondent   contends   that   the  
applicants   will   bear   the   burden   not   only   of   proving   it   is   on   an   analogous  
ground but also that it is unfair. As I have indicated, I doubt whether the shift  
of the burden applies in the context of the EEA. The shift of the burden in  
constitutional cases is the result of the unambiguous language of section 9(5)  
of   the   Constitution   which   provides   expressly   that   discrimination   on   one   or  
more of the grounds listed in section 9(3) of the Constitution is unfair unless it  
is established that the discrimination is fair. No similar provision exists in the

is established that the discrimination is fair. No similar provision exists in the  
EEA. Nevertheless, it is still necessary to determine whether there has been  
differentiation   on   a   ground   specified   in   section   6(1)   of   the   EEA,   namely  
“disability”, or whether the applicant has established his medical condition to  
be  an   analogous  ground.   It   must   be  kept   in  mind   that   the   list   of   specified  
grounds in section 6(1) is not exhaustive, and in so far as the onus was on the  
applicant to prove discrimination, he needed to persuade the court that his  
medical condition constitutes a ground contemplated within the scope of the  
prohibition.
90. The word ”disability” is not defined in the EEA, but item 5 of the  Code of  
Good Practice: Key Aspects on the Employment of People with Disabilities , 
enacted in terms of the EEA, defines “people with disabilities” as “people who  
have   a   long   term   or   recurring   physical   or   mental   impairment   which  
substantially   limits   their   prospects   of   entry   into,   or   advancement   in,  
employment”. Item 5 commences with the following statement in item 5.1: 
The scope of  protection for people  with  disabilities in employment focuses on the  
effect of a disability on the person in relation to the working environment, and not on  
the diagnosis or the impairment.
The definition is therefore not based on the medical model of disability but  
rather on the effect the impairment has in limiting the complainant’s entry into,  
or advancement in, employment. (Dupper  et al:   Essential Discrimination Law  
30

Juta 2004, 60 @163; Thompson and Benjamin CC 1­47.)
91.   Prof   Bonnici   testified   that   Type   1   diabetes   is   a   long­term   physical  
impairment.   There is no cure for it and it is a lifetime disease.   People who  
suffer   from   Type   1   diabetes   are   dependent   on   insulin   that   has   to   be   self­
administered or administered by others, for the rest of their lives. They cannot  
function without it; in fact, if they are not given insulin, they will die. There is,  
therefore,   no   doubt   that   it   is   a   long­term   physical   impairment.   The  
respondent’s   expert   witness,   Dr   Carstens,   confirmed   this.   However,   in   my  
opinion, the matter does not end there. Item 5.1 requires that before being  
classified as a person with disabilities, an applicant must satisfy all the criteria  
in   the   definition.   Hence,   in   addition   to   showing   a   long­term   physical  
impairment,   the   applicants   need   to   show   that   such   substantially   limits  
Murdoch’s prospects of entry into or advancement in employment. In terms of  
item 5.1.3(i) an impairment is substantially limiting if, in its nature, duration or  
effects,   it   substantially   limits   the   person’s   ability   to   perform   the   essential  
functions of the job for which they are being considered. Additionally, items  
5.1.3(ii)   and   (iii)   give   important   content   to   the   term   “substantially   limiting”.  
They provide:
(ii) Some impairments are so easily controlled, corrected or lessened, that they have  
no limiting effects. For example, a person who wears spectacles or contact lenses  
does not have a disability unless even with spectacles or contact lenses the person’s  
vision is substantially impaired.
(iii)   An   assessment   to   determine   whether   the   effects   of   an   impairment   are  
substantially   limiting,   must   consider   if   medical   treatment   or   other   devices   would

control   or   correct   the   impairment   so   that   its   adverse   effects   are   prevented   or  
removed.
89. I am of the view, especially in the light of Prof Bonnici’s evidence, that fast  
acting,   analogue   insulin   controls   or   corrects   the   long   term   physical  
impairment,   diabetes   mellitus,   so   that   its   adverse   effects   in   relation   to   the  
working environment  are largely  prevented or removed.  Indeed, that  is the  
applicants’   case.   It   must   follow   that   although   diabetes   mellitus   can   be  
accurately described as a long­term impairment, in our law, a sufferer of it is  
not regarded as a person with a disability under the EEA. Murdoch lives a  
normal   life   apart   from   his   medication   regime,   and   there   is   no   substantial  
limitation of his abilities to carry out tasks. He does therefore not fall within the  
definition   of   “people   with   disabilities”   in   the   Code   of   Good   Practice.   The  
respondent   for   that   reason   did   not   differentiate   on   the   listed   ground   of  
disability within the meaning of that term in section 6(1) of the EEA. My finding  
in this regard, I would venture, accords with the view taken by diabetics of  
themselves.   Many   surely   would   prefer   not   to   be   stigmatised   by   the   brand  
“disabled”.   A similar  conclusion  was  reached  by  the  US Supreme  Court   in  
Sutton   v   United   Airlines   Inc   527   US   471   (1999)   which   held   that   the  
determination of whether an individual is disabled under the ADA Disability  
Standard requires consideration of the individual’s impairment in its mitigated,  
31

or medicated state.
90.   Absent   proof   of   differentiation   on   a   listed   ground,   the   applicants   were  
burdened to prove that the ground of differentiation is based on attributes and  
characteristics having the potential to impair Murdoch’s dignity or to affect him  
adversely in a comparably serious manner as to amount to discrimination. I  
am   satisfied   that   Type   1   diabetes   is   an   analogous   ground   to   the   listed  
grounds of disability, HIV status and, given its genetic origins, perhaps even  
birth. Controlled diabetics seek dignity with the demand that their capacity to  
function as normal members of society now be recognised to the extent that  
modern   pharmacological   and   technical   advances   make   that   possible.  
Arbitrary,   irrational   and   unfair   exclusions   predicated   upon   anachronistic  
generalised   assumptions   impair   their   dignity   and   seriously   affect   them  
adversely by limiting the full enjoyment of the right, guaranteed by section 22  
of   the   Constitution,   to   pursue   a   chosen   trade,   occupation   or   profession.  
Accordingly,   the   respondent’s   differentiation   does   indeed   amount   to  
discrimination, and in terms of section 11 of the EEA, the respondent must  
establish that it is fair.
91.   Various   factors   should   be   considered   in   making   the   determination   of  
unfairness   or   otherwise.   They   include   the   position   of   the   complainant   in  
society;   the  nature  of  the  practice  or   policy  and  the  purpose   sought   to   be  
achieved by it; and the extent to which discrimination has affected the rights of  
the complainant and to which it has led to an impairment of his fundamental  
dignity ­  Harksen @  para 51.  
92. The  respondent has countered the  applicants’ arguments of unfairness  
with a number of submissions aimed at legitimating its policy and conduct. In  
the first place it contends that the impact on Murdoch was not severe: he was

the first place it contends that the impact on Murdoch was not severe: he was  
not   seconded   to   a   lower   post;   his   current   employment   prospects   are  
unaffected;   he   is   left   in   a   job   where   the   risk   of   danger   arising   from   a  
hypoglycaemic episode at work is lower; the long terms health prospects are  
better in his present job, given that he will not be doing 24­hour shift work  
where he keeps his blood sugar level elevated for one third of his time, which  
is likely to raise his Hbaic levels. In addition he may not be able to progress  
up the ranks if he declined hazardous materials course training or is barred  
from driving the fire apparatus on account of his diabetes. The respondent  
recognizes that the impact on Murdoch may have been acutely disappointing,  
but claims no more so than any aspirant applicant whose application is turned  
down. 
93. The respondent further asserts that the position of diabetics in society is  
not notoriously disadvantaged. As a group they may have experienced some  
pattern   of   disadvantage   in   the   past,   but   not   nearly   to   the   same   extent   as  
persons who are HIV positive, for example. The history of the blanket ban  
against Type  1 diabetics comes not so much  from  unfair  prejudice  against  
32

people   with   diabetes,   but   from   the   factual   history   of   treatment   of   diabetics  
dependent on  insulin. The disadvantage  arises  primarily from their medical  
condition itself, and the extent to which it rendered them incapacitated. The  
disadvantage as a group is a factor to be considered, but not on its own a  
weighty one.
94.  Harksen   also   refers   to   the   nature   of   the   provision   or   policy   and   the  
purpose sought to be achieved by it. As discussed earlier, the respondent’s  
policy was not aimed at impairing Murdoch’s dignity. The worthy societal goal  
sought   to   be   achieved   was   the   safety   of   Murdoch   as   well   as   his   fellow  
employees and members of the public.
95. A further relevant circumstance, in the opinion of the respondent, is that  
as government  it has  discretion  to determine its  practices in regard to  risk  
management   and   to   err   on   the   side   of   caution.   The   legal   duties   on   the  
respondent, particularly in terms of OHSA, favour the blanket ban. There was,  
according to the respondent, no other reasonably practicable way of removing  
or reducing the risk in this matter given the operational constraints.
96. These submissions, while prudent, in some respects suffer a degree of  
inappropriate  paternalism.  All   else   being   equal,   Murdoch  should  be  free  to  
choose his career path relying on his own understanding of his health and  
earning prospects. And insofar as his disappointment has relevance, it should  
be kept in mind that but for his diabetes he would not have been disappointed.  
It is common cause that he would have qualified and moved on to pursue his  
calling.   He   has   not   been   disappointed   because   of   a   deficiency   in   skill,  
qualifications or performance. Even though the arguments display rationality,  
on balance they are diminished by considerations of legitimacy and fairness.

on balance they are diminished by considerations of legitimacy and fairness.  
The position of insulin dependent diabetics in society is such that in practice  
they   are   denied   employment   and   inhibited   in   realising   their   ambitions   in  
various occupations simply because of their medical condition and a common  
misapprehension   of   its   nature.   The   blanket   ban   on   the   employment   of  
diabetics and the purpose sought to be achieved by it, the minimisation of risk  
to employees and others, is based on inaccurate, generalized assumptions  
about insulin dependent diabetics as a class or group and as such is open to  
the criticism of being disproportionate. The respondent has given insufficient  
regard to the compelling evidence and arguments of the kind presented by  
Prof Bonnici, and reflected in the Diabetes UK Guidelines, that well controlled  
insulin dependent diabetics are able to function effectively with minimal risk of  
a   severe   hypo,   shown   to   be   a   rare   occurrence   in   optimally   controlled  
individuals.   Accordingly,   the   blanket   ban   is   not   carefully   tailored   to   the  
legitimate objective of public safety and impairs Murdoch’s rights more than is  
reasonably necessary. A policy or system of individual assessments, along  
the lines envisioned in the Diabetes UK Guidelines, and as accepted by the  
courts in Canada and the USA, would be a less intrusive or drastic means,  
equally, if not more, conducive to the legitimate aim of the respondent, but  
33

reflecting   greater   proportionality   between   the   effects   of   the   policy   and   its  
objective.   Murdoch   has   been   unfairly   prevented   from   fulfilling   his   lifelong  
dream   and   calling,   namely   that   of   becoming   a   full­time   firefighter.   The  
discrimination affects the rights, not only of Murdoch in the present case, but  
also of diabetics generally.  
97. The respondent has made something of the fact that there was at the time  
of the decision no reliable data or statistics or reported clinical trials on which  
to   base   a   more   accurate   risk   assessment   than   that   performed   by   the  
respondent. There is admittedly little hard evidence, but I cannot agree that  
such   justifies   an   unmitigated   approach   to   managing   the   risk.   The   study  
regarding   drivers   in   Scotland   could   have   been   explored   more   thoroughly.  
There   is   also   the   International   Register   which   the   applicants   maintain  
presents reliable evidence of a lack of risk associated with employing insulin  
dependent diabetics as active firefighters. Even accepting Mr. Kantor’s valid  
criticism that the register is based on hearsay evidence, (in that it relies on the  
truth of representations made not in court but on forms filled in by unnamed  
firefighters at unspecified times and places, in unspecified conditions, collated  
by Tim Hoy), it still serves as some indication, one sufficient to influence the  
policy of the UK government. Nevertheless, I accept it should be treated with  
a measure of caution because most, if not all, the data in the register was  
recorded before 1 October 2004, the date on which the blanket ban was lifted  
in the UK, when there was more incentive for insulin dependent diabetics to  
report that  no  incident  was related  to their impairment,  for  fear  of possible  
exclusion. Furthermore, there is no indication of how many of the people on

exclusion. Furthermore, there is no indication of how many of the people on  
the register hold rank, and are more likely to do less strenuous work, and how  
this might affect the reliability of the data. These criticisms accepted, one fact  
is   certain:   as   appears   from   the   report   of   the   Office   of   the   Deputy   Prime  
Minister in the UK, the register had some influence in the uplifting of the ban  
in   the   UK.   What   is   more,   the   fact   remains   that   there   is   no   evidence   or  
documentary record of a single reported incident of any firefighter anywhere  
sustaining or causing injury as a result of a severe hypo while in a hazardous  
situation. Admittedly, this may follow from so few being in employment as a  
consequence of blanket bans, but then again 6 out of 12 countries surveyed  
in Europe did not have such a ban.
98.   Taking   account   of   the   preceding   analysis   and   in   the   absence   of   any  
convincing   evidence   pointing   to   its   efficacy,   the   blanket   ban   is   guilty   of  
overreach.   Accordingly,   the   ban   and   its   specific   application   to   the   first  
applicant constitute unfair discrimination.
99. Unfair discrimination can be justifiable in our law. The justificatory stage is  
where   the   respondent   seeks   to   justify   otherwise   unfair   discrimination.   In  
human rights or constitutional law the notion of “unfair” discrimination focuses  
on   the   holder   of   the   right,   whereas   justification   focuses   on   the   purposes,  
actions   and   reasons   of   the   government,   and   not   the   rights   of   the   holder.  
34

Factors that would or could justify interference with the right to equality are to  
be distinguished from those relevant to the enquiry about fairness. The one is  
concerned with justification, possibly notwithstanding unfairness; the other is  
concerned with fairness and with nothing else ­  President of the Republic of  
South Africa v Hugo  1997 (4) SA 1 (CC) @ 36 B­C. Justification or limitations  
analysis   happens   under   the   Bill   of   Rights   in   terms   of   section   36   of   the  
Constitution, which provides:
(1)  The  rights  in  the   Bill   of  Rights   may   be  limited  only  in  terms   of  law   of  general  
application to the extent that the limitation is reasonable and justifiable in an open and  
democratic   society   based   on   human   dignity,   equality   and   freedom,   taking   into  
account all relevant factors, including—
(a) the nature of the right;
(b) the importance of the purpose of the limitation;
(c) the nature and extent of the limitation;
(d) the relation between the limitation and its purpose; and
(e) less restrictive means to achieve the purpose.
(2)   Except as provided in subsection (1) or in any other provision of the Constitution,  
no law may limit any right entrenched in the Bill of Rights.
100.   Section   36   of   the   Constitution   has   no   application   in   this   matter.   The  
provision   explicitly   restricts   its   application   to   the   determination   of   the  
justifiability of limits upon the rights in the Bill of Rights. The applicants have  
not founded their claim on an alleged violation of the equality rights in section  
9 of the Constitution, but instead filed a statutory claim under the EEA. In the  
context of the EEA, some of the relevant factors considered in the justification  
enquiry under the Bill of Rights can legitimately be taken into account, as I  
have   done,   in   the   fairness   enquiry.   The   most   important   justification   in   the

employment   situation,   though,   falls   for   consideration   under   the   specific  
justification ground permitted by section 6(2)(b) of the EEA. Section 6(2)(b) of  
the   EEA,   it   will   be   recalled,   provides   that   it   is   not   unfair   discrimination   to  
distinguish,   exclude   or   prefer   any   person   on   the   basis   of   an   inherent  
requirement   of   a   job.   This   is   an   absolute   defence   against   unfairness   ­  
Leonard   Dingler   Employee   Representative   Council   &   others   v   Leonard  
Dingler (Pty) Ltd & others  (1997) 11 BLLR 1438 (LC) @ 148H.
101.   Dupper   &   Garbers   in   Employment   Discrimination:   A   Commentary   in  
Thompson   and   Benjamin,   South   African   Labour   Law   (Juta   2004)   submit that  
the   defence   should   be   applied   restrictively.   Any   legislatively   formulated  
justification   of   discrimination   constitutes,   in   effect,   a   limitation   on   the  
constitutionally   entrenched   right   to   equality   and   this   militates   against   an  
expansive  reading   of  the  phrase   “an   inherent   requirement  of   the  job”.   The  
term “inherent requirements” is not defined in the EEA but originates from the  
Discrimination (Employment and Occupation) Convention No 111 of 1958 of  
the International Labour Organisation, in respect of which the committee of  
experts has emphasised the need for a strict interpretation.  In  Association of  
Professional   Teachers   and   another   v   Minister   of   Education   and   others  
35

(1995)   16   ILJ   1048   (IC),  the Industrial Court held that this defence should be  
allowed only in very limited circumstances. 
102. The long title of the EEA includes as one of the purposes of the Act the  
giving   of   effect   to   the   obligations   of   the   Republic   as   a   member   of   the  
International Labour Organisation. Article 3 of Convention No 111 provides,  
inter alia:    
Any distinction, exclusion or preference in respect of a particular job based on the  
inherent requirements thereof shall not be deemed to be discrimination.
The   term   “inherent”   has   been   interpreted   as   “existing   in   something,   a  
permanent   attribute   or   quality;   forming   an   element,   especially   an   essential  
element, of something, essential” ­ Du Toit et al,  Labour Relations Law  (4 th Ed  
Butterworths)   at   569.The   ILO   has   identified   the   following   as   examples   of  
unacceptable requirements: 
•   the   evaluation   of   an   individual’s   competence   for   a   task   based   on  
stereotypes of the group to which the employee belongs;
• requirements based on the preferences of employees and clients;
• requirements that tasks should be performed in a particular way when  
there are other reasonable ways of doing so; and
• qualifications based on ‘light’ or ‘heavy’ work which amount to a veiled  
distinction   between   the   sexes   that   might   impede   the   promotion   of  
women.
103. The respondent avers that it is an inherent requirement of the job of a  
firefighter not to expose fellow employees, the general public or oneself to real  
risk of harm to their or one’s own safety. Further, in order to fulfil that inherent  
requirement,   due   to   the   dangerous   circumstances   in   which   firefighters   on  
active   duty   may   be   exposed,   they   must   not   be   at   risk   of   having   a  severe  
hypoglycaemic episode while on duty. This, as we have seen, according to

hypoglycaemic episode while on duty. This, as we have seen, according to  
Sparks,   Carstens   and   Woolley,   could   arise   where   the   insulin   dependent  
diabetic has a hypoglycaemic episode in circumstances where he cannot or  
does   not   take   immediate   steps   to   prevent   the   development   of   a   severe  
episode, because, for instance, he has no glucose or carbohydrates to take or  
he has glucose or carbohydrates to take but cannot or fails take them. Thus a  
Type 1 diabetic may forget to insert or replace glucose sachets in his tunic, he  
can become separated from his tunic, the glucose sachet could burst or be  
otherwise   destroyed   or   contaminated.   Likewise   the   firefighter   may   have  
glucose or carbohydrates to take but cannot take them because his mouth is  
covered with a breathing apparatus or he cannot access the sachet because  
his arms are trapped or injured, or he cannot get his glove off and there is  
nobody   on   hand   to   assist,   or   he   is   using   both   hands.   He   could   have   the  
36

glucose or carbohydrates to take but not take them because he is distracted  
by other events on the fireground or in the extreme heat and fatigue action he  
confuses the warning signs of the hypo with general fatigue.
 
104.   These   risks,   the   respondent   argues,   should   be   assessed   taking   into  
consideration   the   probability   of   the   occurrence   of   the   risk   event   and   the  
severity of the consequences should it occur, as well as the employer's legal  
obligations to co­employees and others, whether arising from a common law  
duty of care, occupational health and safety statutes, or other aspects of the  
employment regulatory regime.
105. As appears from the discussion of the medical evidence, it cannot be  
denied that hypoglycaemia is a fact of life for Type 1 diabetics, as it is for  
many non­diabetics, and it may be difficult to detect with any certainty whether  
a particular diabetic will suffer an incapacitating reaction. Whilst the risk is less  
for someone who is well controlled than otherwise, it is still there. Even the  
best­controlled and disciplined Type 1 diabetic could be prone to human error.  
Innumerable factors (including, according to Prof Bonnici, motivation, mood,  
family support, and complacency) could influence the person, and hence the  
risk. Both Carstens and Rowen testified that these risks are real, not fanciful,  
although  they  might   be  very  small.   In  their  assessment,   they  could  not   be  
rated   as   “acceptable”.   The   respondent   therefore   submits   that   any   Type   1  
diabetic   runs   the   risks   set   out   above,   and   in   the   light   of   the   inherent  
requirements of the firefighter job the blanket ban is justifiable.
106. These arguments, it would seem to me, support not the imposition of a  
blanket ban, but rather a policy of individual assessments. The respondent is  
in actual fact saying that the risk will vary from person to person. I agree. That

in actual fact saying that the risk will vary from person to person. I agree. That  
is   why   the   ILO   and   the   changing   policy   elsewhere   regard   individual  
assessments   as   necessary.   Under   the   Convention   the   evaluation   of   an  
individual’s competence for a task based on stereotypes of the group to which  
the employee belongs is an unacceptable requirement. The specific examples  
alluded   to   by  the  respondent  can   be  managed  on   an  individual   basis  with  
minimal   effort.   A   hypo­aware   firefighter   usually   will   be   able   to   leave   the  
fireground as he experiences the onset of hypoglycaemia; he can be assisted  
by his partner (firefighters work in pairs); he can run high; or he can limit his  
use of breathing apparatus to an acceptably limited period. These practices  
and safeguards, to be applied as a matter of individual policy, for a medically  
assessed individual, are proportionate means of reducing the risks and giving  
effect to the principle of non­discrimination. Although the applicants in the pre­
trial minute agreed that they seek no exceptional reasonable accommodation,  
the proportional practices suggested do not attain the level of special status or  
exemptions,   rather   they   are   coping   strategies   applicable   as   criteria   of  
assessment for determining suitability for employment. 
107.   In   Australia   the   test   is,   like   in   South   Africa,   whether   the   “inherent  
37

requirements of the position” are such as to justify the discrimination. In   X v  
The   Commonwealth   [1999]   HCA   63   the   High   Court   of   Australia   was  
concerned with the dismissal of a newly inducted recruit from the military due  
to   his   HIV   positive   status.   The   approach   of   the   court   to   the   inherent  
requirements of the job and risk assessment is instructive. The court held:
I   do   not   think   that   it   is   the   proper   approach   to   ask   whether   the   degree   of   risk  
emanating from the disease defines or can be prescribed as an inherent requirement  
of the employment. Rather the degree of risk is relevant in determining whether X  is 
able to carry out  an inherent requirement of the employment, namely, the requirement  
not to expose fellow soldiers and others to a real risk of harm to their health or safety.  
It is not a case of the employer seeking to impose a term or condition, but one where  
the inherent requirement arises as a matter of law. The real difficulty of the case lies  
in determining whether X can carry out that requirement with or without assistance.  
The issue of "inherent requirement" has become complicated only because, at all  
stages of the argument, the Commonwealth has insisted that the ability to "bleed  
safely" is the relevant inherent requirement.. …..
The circumstances for such transmission would need to be, as the Commissioner  
described them, "extreme". The risk of transmission in such extreme circumstances  
was "very low", although not "fanciful". By inference therefore, in the overwhelming  
majority of the circumstances in which a soldier such as X would be required to carry  
out the requirements of his particular employment, there would be no such risk,  
whether in training, or even in combat duties.
108. In the same way, while I readily accept that the consequences of the  
realisation   of   the   risk   of   a   severe   hypo   in   a   hazardous   situation   on   the

fireground could prove disastrous, I do not see the degree of risk as material.  
It is no more so than the risk of an overweight firefighter having a heart attack,  
or a surgeon suffering a cerebral episode during an operation. The relatively  
minimal risk of that happening should not disable the employee from carrying  
out   the   inherent   requirements   of   the   job   and   cannot   justify   a   total   ban   on  
employment. Both Murdoch and Hoy have shown that they pose little or no  
risk and have successfully carried out the inherent requirements of firefighting  
for a considerable and continuous period of time. Neither of them has had a  
severe hypo in their combined 27 years of firefighting. Their past experience  
is   the   best   available   forecaster   of   the   future.   The   72   UK   firefighters   with  
diabetes   on   the   International   Register   of   Firefighters   with   Diabetes   have,  
between   them,   700   person   years   of   active   firefighting   without   an   incident  
involving a severe hypo. Whatever evidentiary qualifications one may have  
about the hearsay nature of the evidence about the register, as I have said, it  
reflects an achievement to which the UK government was prepared to attach  
some weight in changing its policy.
109. In so far as the respondent has valid concerns about being sued in delict  
for   harm   arising   out   of   the   employment   of   Type   1   diabetics,   that   too   is  
misplaced. Firstly, if such employment is justified in order to observe the duty  
not   to   unfairly   discriminate   in   its   employment   practices,   the   respondent’s  
conduct   is   unlikely   to   be   considered   wrongful.   Should   it   conscientiously  
38

assess   firefighters   individually   and   proportionately,   it   further   will   have  
discharged   its   duty   to   take   reasonable   preventative   steps   to   minimize   any  
foreseeable   risk,   especially   if   it   makes   reasonable   adjustments   where  
required.   Its   legitimate   concerns   about   public   liability   must   yield   to   the  
constitutional principle of non­discrimination.
110. Therefore I agree with the applicants that the respondent has failed to  
justify its unfair discrimination (in the form of a blanket ban). Without in any  
way denying that firefighting is by its nature a hazardous occupation, to simply  
exclude all insulin dependent diabetics from the occupation on this ground is  
not justifiable. 
111.   The   respondent   is   guilty   of   assigning   characteristics   which   are  
generalised assumptions about groups of people to each individual who is a  
member   of   that   group,   irrespective   of   whether   that   particular   individual  
displays   the   characteristics   in   question.   It   is   treating   all   insulin   dependent  
diabetics the same and imposing a blanket ban on the employment of that  
group as firefighters, irrespective of whether the particular individual ­ such as  
Murdoch, who is physically fit and in optimal control of his diabetes – displays  
any susceptibility to uncontrolled hypoglycaemic episodes.
112.   Added   to   all   this,   the   Constitutional   Court   has   confirmed   that   an  
approach of an individualized assessment, rather than a blanket ban, should  
be   followed   in   cases   where   the   employer   seeks   to   differentiate   on   health  
grounds   in   an   employment   policy   or   practice.   In   Hoffmann   v   SA   Airways  
(2000) 21  ILJ   2357 (CC)  the Constitutional Court dealt with the situation where  
SAA adopted a policy of not employing HIV­positive cabin attendants. Holding  
that discriminatory practice to be unfair, Ngcobo J commented:

that discriminatory practice to be unfair, Ngcobo J commented:
The fact that some people who are HIV positive may, under certain circumstances, be  
unsuitable for employment as cabin attendants does not justify the exclusion from  
employment as cabin attendants of all people who are living with HIV. Were this to be  
the case, people who are HIV positive would never have the opportunity to have their  
medical   condition   evaluated   in   the   light   of   current   medical   knowledge   for   a  
determination to be made as to whether they are suitable for employment as cabin  
attendants. On the contrary, they would be vulnerable to discrimination on the basis  
of   prejudice   and   unfounded   assumptions   –   precisely   the   type   of   injury   our  
Constitution seeks to prevent.  This is manifestly unfair…
The need to promote the health and safety of passengers and crew is important.  So  
is the fact that if SAA is not perceived to be promoting the health and safety of its  
passengers   and   crew   this   may   undermine   the   public   perception   of   it.   Yet   the  
devastating effects of HIV infection and the widespread lack of knowledge about it  
have produced a deep anxiety and considerable hysteria. Fear and ignorance can  
never justify the denial to all people who are HIV positive of the fundamental right to  
be judged on their  merits.  Our treatment of people who are HIV positive must be  
based on reasoned and medically sound judgments. They must be protected against  
prejudice and stereotyping.  We must combat erroneous, but nevertheless prevalent,  
perceptions about HIV. The fact that some people who are HIV positive may, under  
certain circumstances, be unsuitable for employment as cabin attendants does not  
39

justify a blanket exclusion from the position of cabin attendant of all people who are  
HIV positive.
The constitutional right of the appellant not to be unfairly discriminated against cannot  
be   determined   by   ill­formed   public   perception   of   persons   with   HIV.   Nor   can   it   be  
dictated by the policies of other airlines not subject to our Constitution.
Prejudice can never justify unfair discrimination. This country has recently emerged  
from   institutionalized   prejudice.   Our   law   reports   are   replete   with   cases   in   which  
prejudice  was  taken  into  consideration  in  denying  the rights  that we  now  take  for  
granted.   Our   constitutional   democracy   has   ushered   in   a   new   era   –   it   is   an   era  
characterized by respect for human dignity for all human beings. In this era, prejudice  
and stereotyping have no place. Indeed, if as a nation we are to achieve the goal of  
equality that we have fashioned in our Constitution we must never tolerate prejudice,  
either directly or indirectly. SAA, as a state organ that has a constitutional duty to  
uphold the Constitution, may not avoid its constitutional duty by bowing to prejudice  
and stereotyping.
113.   Accepting   that   there   are   different   considerations   and   levels   of   stigma  
attached to HIV as compared to diabetes, justifying a stricter level of judicial  
scrutiny in relation to the former, the learned judge’s reasoning with regard to  
the unjustifiablity of relying on prejudiced and outdated medical knowledge is  
markedly apposite in this case. And the medically sound judgement of South  
Africa’s   leading   expert   on   diabetes,   Prof   Bonnici,   is   incontrovertibly   that   a  
blanket ban is irrational, unfair and unjustifiable in the light of current medical  
knowledge.
114. To repeat the general points made regarding fairness and justifiability:  
the   risk   justification   lacks   justifiability   for   not   being   rationally   connected   or

based upon the medical evidence. The range of factors informing the risk, as  
with   many   other   medical   conditions   such   as   obesity,   asthma   and   cardio­
vascular   disease,   requires   that   each   case   should   be   assessed   on   its   own  
merits and be based on an individualised assessment of the job applicant. If  
the job applicant meets all other requirements such as physical fitness and  
visual acuity, an individual assessment should be made as to that person’s  
control of is or her diabetes and the likelihood of that person having a severe  
hypoglycaemic   episode.   In   each   instance,   the   employer’s   medical   officer  
should assess the risk with regard to,   inter alia , the applicant’s past history;  
his control of his diabetes; hypo­awareness; and measures taken to minimise  
risks.
115. The purpose of the EEA is to achieve equity in the workplace by, inter  
alia, promoting equal opportunity and fair treatment in employment through  
the   elimination   of   unfair   discrimination.   The   Act   provides   that   it   must   be  
interpreted in compliance with the Constitution and so as to give effect to its  
purpose. Equality lies at the heart of the Constitution and aims to ensure that  
we achieve a diverse workforce with opportunities for all, including diabetics,  
to pursue their preferred calling.
 
40

116. In  the  premises, I  am persuaded  that the respondent,  contrary  to the  
provisions of the EEA, has unfairly discriminated against the second applicant  
in its employment policy and practice on the ground of his medical condition,  
namely Type 1 diabetes mellitus.
The order
117. The parties have made certain submissions with regard to appropriate  
relief, which are reflected in the order that follows. Both have requested an  
order   for   costs   on   the   basis   that   costs   should   follow   the   result.   The  
respondent, albeit perhaps too cautious, is a public authority accountable to  
the public and its ratepayers and has not acted unreasonably. It needed a  
judicial   decision   in   the   interests   of   certainty   in   un­chartered   territory.   The  
second applicant has been represented by his union, which continues in an  
industrial relationship with the respondent. In the circumstances a costs award  
is not justified. I accordingly make the following orders:
1. The respondent’s failure to transfer the second applicant from his  
position as law enforcement officer to that of firefighter within the  
Directorate: Protection Services is declared to be unfair discrimination.
2. The respondent’s employment policy of refusing to employ insulin  
dependent diabetics as firefighters is declared to be unfair  
discrimination.
3. The respondent is ordered to assess each applicant for the position of  
firefighter on the employee’s own merits and on objective criteria,  
including physical and medical fitness.
4. The respondent shall second the second applicant to Fire and  
Emergency Services (or the current equivalent) in the Cape Town  
Administration in the position of Learner Firefighter with effect from the  
date of the commencement of the first Firefighter 1 Training Course  
held by the Cape Town Administration following the date of judgment.  
The said secondment:
4.1 is conditional upon the second applicant successfully

The said secondment:
4.1 is conditional upon the second applicant successfully  
completing the aforementioned Firefighter 1 Training Course,  
failing which the said secondment shall be reversed; and
4.2 is not a permanent placement, irrespective of whether or not  
the second applicant successfully completes the said Firefighter  
1 Training Course.
5. The second applicant shall remain at his level of remuneration, with  
such notch increase as he would usually be entitled to from 1   July 
41

2005, until such time as he may successfully complete the  
aforementioned Firefighter 1 Training Course.
6. In the event that the second applicant does successfully complete the  
said Firefighter 1 Training Course, his remuneration shall be adjusted  
to the Cape Town Administration’s scale D 18 T, commencing on the  
notch R60 282, 00, as it is reflected as at 7   June  2005, or  
corresponding amount following annual increase. From the time of  
adjustment only, the firefighters’ standby allowance shall apply.
7. Save as is specified above, the current rules, policies, practices and  
collective agreements of the respondent, as they have been applied to  
other employees seconded as learner firefighters to Fire and  
Emergency Services in the Cape Town Administration in 2003, shall  
apply to the second applicant. 
8. There is no order as to costs.
MURPHY AJ
Date of trial:
Date of Judgement: 18 July 2005
Applicant’s   representatives:   A   Steenkamp   and   Z   Majamane   of   Sonnenberg  
Hoffmann Galombik 
Respondent’s representative : Adv Peter Kantor
42