1
 CHAPTER 1 
 
1.0  INTRODUCTION 
1.1 BACKGROUND INFORMATION 
1.1.1 Alcohol 
Alcohol has played a major role in the socio-cultural and religious lives of people in most 
parts of the world.  In Africa, the consumption of alcohol predates the colonial era as beer 
was used to celebrate various religious ceremonies and feasts as recorded by oral and 
written tradition.1  
 
There were guiding norms that controlled the use of alcohol and these norms were largely 
effective.  These norms restricted the use of alcohol to adult males as drinking was 
largely communal and only took place on special occasions.1   
The advent of colonialism brought a change to these norms as alcohol was seen by the 
colonial masters as a means of mobilizing and controlling the cheap labour force in the 
agricultural sector.  This led to the introduction of the ?dop? or ?tot? system.2   
The advent of gold mining in South Africa led to the extension of this practice to the 
mining industry.2 
 
Alcohol consumption in the mines has been on the increase.3  Social problems such as 
violence and medical problems associated with excessive alcohol use such as 
hypertension, gastritis, liver cirrhosis and depression have also increased alarmingly.4  
 
 2
 1.1.2 Cannabis 
Cannabis is the most widely used illicit drug in the world with an estimated 144 million 
people using it annually.5  It is the main drug of abuse on the African continent.6  Its use 
has been linked to occupational accidents, an increase in health care services utilisation 
and costs as well as decreased productivity.7,8 
 
1.2 STATEMENT OF THE PROBLEM 
Alcohol misuse is a significant public health challenge in today?s South Africa.  It is 
estimated that 6-16% of the workforce is likely to be alcohol-dependent and that a further 
20% might likely experience drug or drug-related problems.9,10   
 
Cannabis (46%) and alcohol (32%) have been implicated as causative factors in patients 
with trauma injuries.11 They have been shown to cause road traffic accidents 12, 13 as well 
as increased mortality, morbidity, accidental injury, homicide, assault and suicide.14, 15,16  
 
Alcohol and to a lesser degree Cannabis are popular drugs in the mining industry and 
have been shown to contribute to employee illness, injury, increase in health service 
utilisation and costs.17  They have also been shown to be related to several workplace 
outcomes such as decreased productivity and poor labour market performance.18   
 
The impact of the high prevalence of alcohol and Cannabis on the medical services and 
the economy of the mines can be quite negative as it imposes a substantial financial 
burden on mine management in particular and the society at large.13, 18  
 3
 Alcohol and Cannabis use lead to an increase in sickness, absence and absenteeism, 
which for the mines increases costs and decreases productivity.  It has been shown that 
significant costs incurred for medical reasons in the mining industry are directly 
attributable to alcohol abuse.17  Thus, for effective cost containment being currently 
emphasized in the mining industry to be realizable, one of the factors amongst others that 
have to be reduced might be the high prevalence of alcohol and Cannabis.  
 
In South Africa, 77% of patients attending the Tygerberg Hospital Trauma Unit and 
65.9% of adults that were involved in fatal collisions in Cape Town Municipal area had 
blood alcohol levels higher than the legally accepted normal of 0.08g/100ml at the time.13  
 
1.3 JUSTIFICATION FOR THE STUDY 
Anecdotal evidence suggests the use of alcohol and cannabis to be widespread in the 
general population.  Though this is suspected to hold true for the working population, a 
paucity of published data on cannabis and alcohol use and abuse in the mining industry 
currently exists.  Alcohol has been recognised as a problem in the workplace since the 
1940?s 24 but few studies have documented the magnitude of this problem, particularly 
with regards to the mining industry.    
 
Although certain factors such as boredom, poverty and weak legislation 25 as well as poor 
social support 26 have been implicated as contributory factors to high alcohol 
consumption in the general populace, the underlying social and economic determinants 
that drive miners, particularly on diamond mines, to smoke cannabis and use alcohol 
 4
 before and/or during working hours have not been extensively investigated. Alcohol and 
cannabis use lead to psychomotor disturbance, thus workers under their influence are not 
only a danger to themselves but also to others.27 Also, the costs that result from lost shifts 
due to alcohol abuse as well as treatment costs for alcohol-related problems run into 
millions of Rand per annum.17  These high costs and the potential risks to the public 
health of workplace alcohol and cannabis abuse are now of major concern to the mining 
industry. 8 
 
This study aims to determine the knowledge, attitudes, practices and prevalence of 
alcohol and cannabis use on a South African diamond mine. By doing this, it is hoped 
that the extent of the problem will be estimated, and that appropriate intervention 
strategies which can be integrated into existing social programmes/structures to help 
combat the threat that alcohol and cannabis use poses to the health and safety of the 
employee, productivity and the overall cost-effectiveness of the industry will be 
determined. 
It is also hoped that information gathered from this study will help form the basis of a 
uniform policy framework and ultimately to laws that will make the mining industry free 
from the adverse consequences resulting from the misuse of alcohol and the use of 
cannabis and other illicit drugs.  In the absence of readily available national survey data, 
the United Nations has often relied on reports of treatment seeking, convenience samples 
and narcotics trafficking to estimate the intensity of international drug use .28 
 
 5
 This study will therefore not only contribute to the formation of a reliable data base, but 
will also help establish a baseline of reference for other studies as well as a reference 
point for monitoring intervention programmes that may be instituted in the South African 
mining industry. 
 
1.4 LITERATURE REVIEW 
For many years, the abuse of alcohol and mood altering drugs (amongst them cannabis) 
has been a major source of concern in the workplace.29 This is because the prevalence 
figures for alcohol and drug use have been alarming both locally and internationally.  The 
consequences of this for business and the workforce could be dire if not checked in time.  
Negative effects of such abuse such as absenteeism from work, increase in use of medical 
benefits and workers compensation claims, poor productivity, high job turnover, 
interpersonal conflicts, workplace injuries and damage to workplace property have all 
been documented.30 
 
 
1.4.1 Alcohol Use  
The rise in the prevalence of alcohol consumption in the general populace has been 
attributed to a lack in public knowledge towards the use of alcohol and drinking 
guidelines.31 In a New Zealand study, 74% of the population did not know about drinking 
guidelines and felt that it was not publicised enough.31     
 
 6
 In the workplace, adverse working conditions such as high job demands, high hazardous 
physical working conditions as well as other stressful circumstances have been associated 
with high levels of alcohol intake. 32 This study noted that although there was significant 
linear association between alcoholism and workplace accidents, the subject of an 
association between work stress and alcoholism remains inconclusive. Bennett and 
Lehman 21 in their study showed that employee exposure to co-worker alcohol use have 
been associated with lack of work group cohesion and increased accidents amongst other 
problem indicators. They also stated that social drinking sometimes serves as a primary 
means of support for employees who work under stressful conditions. It has been argued 
that adverse working conditions are associated with excessive or problematic alcohol use, 
and that the stressful condition extends outside the work-life of the individual. 19, 32 
Support for the ?spill over effect? of adverse working conditions being associated with 
problematic alcohol use has been weak 19 This weakness has been explained by numerous 
reasons which include under-reporting and denial by drinkers. 8 26,21 Greenberg 23   noted 
that despite the volume of studies conducted in this field, results are often contradictory. 
 
 
 
 
 
 
 
 
 7
 Alcohol use in the workplace  
The highest percentage of alcohol abuse and indeed alcohol-related deaths in the industry 
were recorded amongst industries associated with wholesale trade, transportation/utilities, 
mining and manufacturing. 33 
 
With regards to the contribution of alcohol to injuries in the workplace, the Centres for 
Disease Control in the USA reported in 1991 that 7% of workers are injured annually 
(because of alcohol) in an occupational setting in the USA. 34 
Between 1872-1973, alcohol was implicated in 19% of industrial injuries treated at the 
University Central Hospital in Helsinki, Norway. 35 
In 1978, a study on fatally injured workers in various jobs reported that about 11% of the 
deaths had a blood alcohol level greater than 0.08mg% in Maryland, USA 36 In a study of 
fatal workplace injuries in Pennsylvania (USA) between 1983 and 1984, the prevalence 
of workplace injuries associated with blood alcohol levels greater than 0.08mg% was 
estimated to be 17%. 37 
 
A retrospective study of reported fatal occupational injuries in North Carolina from 1978 
to 1984 showed a positive blood alcohol level in 11% of the injured workers.  This 
included both motor vehicle fatalities and homicides. 38 In Alberta, Canada, a study of 
373 fatal work-related injuries between 1979 and 1986 revealed a 10.7% 8 prevalence of 
blood alcohol level equal to or greater than 0.08mg%.  
 
 
 8
 In West Germany, a study carried out between 1963 and 1974 implicated alcohol in about 
10% of all industrial accidents.  It was also found out in the same country that all fatal 
injuries in industrial settings had a blood alcohol level greater than the accepted level for 
that country of 0.05mg% between 1962 and 1970. 39 
 
In Argentina, studies carried out amongst high altitude miners between 1976 and 1977 
showed that 34% of the respondents consumed alcohol on a weekly basis, 40 the same 
study reported a prevalence of 30% amongst accident cases and 64% for cases referred 
for disciplinary purposes. 
 
Alcohol use in South African mines 
In South Africa, a study on a Free State Gold mine reported the prevalence of alcohol 
abuse to be 32%. 17 This same survey also showed a prevalence that is higher amongst 
mine workers than what is obtained for the general populace, at between 10 to 35%. 41 In 
a 1994 survey of another Free State gold mine, it was reported that the main reasons for 
drinking alcohol amongst miners were to deal with worries, loneliness and boredom as 
well as for recreation and socialization. 3 In this study, 42% of the respondents who took 
alcohol reported drinking in order to deal with the stress of living. Rocha Silva et al. in 
1995 reported that a majority of the respondents in their study i.e. 49.6% drank for the 
purposes of enjoyment, fun and entertainment. 42 
Kew in his 1994 study of alcohol consumption patterns on a gold mine in South Africa, 
reported that high-risk drinkers preferred to drink alone.   
 
 9
 Alcohol use in the South African populace 
With regards to the knowledge of the dangers of alcohol misuse, Mbombo et al. reported 
that most people in the general populace saw alcohol as potentially harmful to the 
individual, family and community. 43 Of particular concern to the participants was the 
danger alcohol misuse posed to pregnant women. The participants also believed that 
European alcohol was more dangerous than traditional beer. With regards to attitudes and 
practices towards alcohol consumption, the researchers in this study also reported that 
people drank in order to feel better, to socialize and to cope with problems.  The 
problems enumerated were general problems, health problems and oppression.  It was 
however worthy of note that several participants, although reporting that they drank 
alcohol as a coping mechanism towards problems, still believed that their problems did 
not go away and that even more problems were created by their drinking.   
Most studies have shown that people drink in the company of friends or relatives.  Rocha 
Silva et al. reported that 71.3% of drinkers drink in the company of friends. 42   
In the studies of Rocha Silva et al. in 1992 42 and Mbombo et al. in 1994, 43 shebeens 
were reported to be the most popular places to drink in both studies.  Other places such as 
bars, bottle stores, houses of friends and own homes were also reported as preferred 
places for drinking. 
These studies have shown that most people drink alcohol in the company of other people 
and usually do so outside their own homes. They also show that alcohol is used as a 
coping mechanism to combat stresses of living, loneliness and boredom. 
These different studies therefore advance the same reasons for drinking in both miners 
and the general populace.  
 10
 1.4.2 Cannabis Use 
A New Zealand survey revealed that 41.3% of male respondents and 43.3% of the female 
respondents began to smoke cannabis whilst at school. 31 Similarly, the high prevalence 
in the general populace with regards to cannabis use was blamed on peer pressure and 
lack of knowledge of its side effects. 44  
Cannabis use has been shown to produce impairment in a wide range of cognitive and 
behavioural performances.  It has also been shown to lead to a ?slowing of reaction time 
and information processing, impairment of motor coordination and performance, short-
 term memory attention, signal detection as well as tracking behaviour and time 
perception?. 45 These effects are dose-dependent and are larger and more persistent on 
tasks that require sustained attention.45 The dose-dependent relatedness of cannabis use 
and the production of these effects as well as the varying effects it produces on different 
individuals may well be responsible for the contradictory findings recorded in different 
studies.  This has led to the suggestion that cannabis use may play different roles in 
different occupational and cultural settings. 46 
 
Cannabis use in the workplace 
The use of cannabis and other mind-altering drugs in the workplace have been reported 
for quite a long time. 
At the end of the 19th Century, it was reported that the majority of Egyptian labourers 
used cannabis and ?relied on its stimulating properties to compensate for underfeeding, to 
combat fatigue and as a means of escape from their dreary lives?. 48  Relatively little is 
known about the role of cannabis in the workplace and its association with occupational 
 11
 injury has not been fully established. 46 However, the major concern about the effects of 
cannabis use in the workplace has to do with the possibility of accidents if users drive or 
operate machinery. 48 
 
Safety concerns in the workplace as well as the rising use of cannabis in the general 
populace led the United States government to institute the National Household Survey on 
drug abuse from 1971 to 1972.  The result from this survey showed that 6% of the 
American workforce used illicit drugs and that cannabis was the most abused drug..46 
In 1999, a survey carried out in the USA showed that 5.1% of full time and 6.8% of part-
 time employees aged 18 years or older used cannabis.  This result showed cannabis to be 
the most commonly used illicit drug in the workplace. 18 Also in the USA, 3.4 million 
workplace drug tests were conducted from January to June 2000 for various reasons 
which included random, post-accident, periodic, pre-employment and return to duty.  The 
prevalence for drug use was 4.7% and cannabis accounted for 65% of these positive 
tests.18 
In Europe, concerns over workplace safety also led to the institution of workplace drug 
testing even though it was not as efficient and as widespread as that of the USA. 49 
In Greece, pre-employment testing done from 1998 to 2000 showed 7% prevalence of 
drug use.  Of these, cannabis made up 87% 49 In the United Kingdom, a survey of British 
railway workers between 1994 and 1997 returned positive drug results for approximately 
0.36% of the workforce.  Cannabis was the number one drug detected and accounted for 
68% of the positive results. 49 
 12
 In France, a survey of the Petrochemical industry revealed that 57% of workers used 
cannabis.50 A 1998 Swedish survey amongst 24,000 workers from 25 companies 
recorded a cannabis abuse rate of 2.7%, which was the highest of all drugs of abuse. 51 
In 1999, Germany recorded a prevalence of approximately 9% drug use in the chemical 
industry.  Of all the drugs of abuse recorded in this study, cannabis was the most 
frequently encountered. 52 
 
With regards to cannabis use and occupational fatalities, Lew et al 53 reported a 
prevalence rate of 3.1% whereas Alleyne et al 8 reported a prevalence of 8.5%.   
Although many studies have reported that cannabis use leads to impaired mental 
function, this has not been shown to be the case with occupational fatalities as 
cannabinoids are cleared slowly from the body and clear relationships between 
metabolite levels and effects are difficult to establish. 
 
1.5 STUDY AIM AND OBJECTIVES 
1.5.1 Study Aim 
The aim of this study is to evaluate the knowledge, attitudes, practices and prevalence of 
alcohol and cannabis use by anonymous interview and unlinked, non-invasive 
biochemical tests among South African diamond miners. 
 
1.5.2 Study Objectives 
? To determine the factors that may predispose miners to alcohol and cannabis 
use on a South African diamond mine. 
 13
 ? To determine the prevalence of alcohol and cannabis use on a South African 
diamond mine. 
? To evaluate the knowledge, attitudes and practices regarding alcohol and 
cannabis use amongst miners on a South African diamond mine. 
? To make the results, findings and recommendations available to all stakeholders 
for relevant intervention strategies. 
 14
 CHAPTER 2 
 
2.0 METHODOLOGY 
2.1 DEFINITION OF TERMS 
With regard to this study the terms knowledge, attitudes, practices and prevalence were 
defined as follows by Webster?s Third New International Dictionary 54 and by the 
textbook of Epidemiology, Biostatistics and Preventive Medicine Review. 55 The terms 
use, misuse and abuse were defined by Oxford Dictionary Thesaurus. 56 
Knowledge:  The fact or condition of knowing something with a considerable degree of 
familiarity gained through experience of or contact or association with the individual or 
thing so known. 54 It is also the fact or condition of being cognisant, conscious, or aware 
of something. 54   
Attitudes:  This is behaviour representative of feeling or conviction.  It is also a 
disposition that is primarily grounded in affect and emotion and is expressible of opinions 
rather than belief. 50 
Practices:  To do something habitually.  It also means habitual conduct that is socially, 
ethically or otherwise unacceptable. 54 
Prevalence:  The number of persons in a defined population who have a specified 
condition at a point in time, usually the time a survey is done. 55 
Abuse:  The wrong use of something.  In this study, this means the wrong use of alcohol 
or the use of cannabis. 56 
Misuse:  Use wrongly 56 as with abuse in this study. 
 15
 Use:  Consume, drink, eat.  With regards to this study it means to drink alcohol or smoke 
cannabis 56. 
 
The ?CAGE screening questions? 57 have been validated as a screening tool for chronic 
alcohol use.  It is made up of four questions (see questions 23-26 of the questionnaire).  
Two or more positive responses to these questions suggest possible dependence on 
alcohol. 
2.2 STUDY DESIGN 
The study was a cross sectional analytical study. The knowledge, attitudes and practices 
of the miners to alcohol and cannabis use were done through face-to-face interviews of 
those selected for the study. 
 
The prevalence of alcohol and cannabis on the mine was done by biological testing of 
breath for alcohol and urine for cannabis. 
 
 
2.3 STUDY METHODS 
2.3.1 Study Site Description 
The study was carried out on a diamond mine in the Republic of South Africa. The mine 
was randomly selected from a list of mines. As a pre-condition for approval of this study, 
the Committee for research on human subjects (medical) of the University of the 
Witwatersrand stipulated that the mine should not be mentioned by name.  As a result of 
this, the mine will be referred to as mine X in this report. 
 16
 Mine X is a conventional open cast mine with two adjoining kimberlite pipes covering an 
area of about 18 hectares.  The mine was officially opened in the early 1990?s but its 
history dates back to 1903 when diamond-bearing gravels were discovered on high 
grounds close to a nearby river. The mine is located at about thirty-four kilometres from a 
small farming town and about eighty kilometres from the nearest big town. It has a 
workforce of seven hundred and fourty. Of this number, one hundred and fifty are 
administrative staff whilst the remaining five hundred and ninety are miners. There are no 
available hostels on site, but most workers stay in the small farming or the nearest big 
town. Transportation is provided by mine management. 
 
2.3.2 Alcohol and Drug Testing Policy 
Mine X has an existing alcohol and drug policy.  The objective of the policy is to 
facilitate the treatment of alcoholism and drug dependency and to provide for the testing 
of employees for the presence of alcohol and/or drugs. 
The mine?s policy states that ?no person in a state of intoxication nor any condition likely 
to render him incapable of taking care of himself or persons under his charge be allowed 
in the proximity of or allowed to enter a mine or to work on or near any machinery on the 
surface of a mine or at a works?. 
In this regard, and recognising that alcohol and drug dependency are complex illnesses 
and thereby encouraging its early identification and treatment, the company has 
undertaken to provide financial assistance in terms of rehabilitation and counselling to 
employees requesting treatment of medically documented alcohol or drug dependence 
problems or those identified by the company as displaying symptoms in keeping with 
 17
 problems of this nature. The company bears the financial costs of counselling, treatment 
and rehabilitation for first offenders. The costs for subsequent consults are borne by the 
employee. The policy also provides for routine checks as part of a preventative and pro-
 active measure. This policy applies to all employees of mine X. 
 
2.3.3 Testing Procedures 
The alcohol and drug policy of the mine makes the supervisors responsible for detecting 
employees who are under the influence of alcohol and cannabis.  Co-workers although 
not explicitly mentioned also share in this responsibility.  A detection of alcohol or 
cannabis smell, bloodshot eyes/dilated pupils, unsteady gait, incoherent speech, moody or 
an aggressive behaviour coupled with an observed consumption of alcohol or cannabis or 
the possession of these substances on their person in the mine will lead to a request for 
the employee to submit himself for tests.  Employees may also be requested to submit 
themselves for tests following involvement in an accident or at random when entering or 
leaving the mine premises. 
 
In the presence of stipulated parties representing the employee and the mine 
management, a Lion Alcometer test with regards to alcohol and an ?On Track? urine test 
for cannabis are administered on the suspected employee. 
A positive test for alcohol is a value equal to or higher than the legal limit of 0.10g per 
1000ml of breath. This is the legal limit for driving in South Africa. 58 With regard to 
cannabis, the employee will be required to undergo a urine test and an additional ?Dover? 
test, which will indicate the employee?s reaction speed to various stimuli.  The urine 
 18
 sample is also sent to a laboratory in the provincial capital for a confirmatory test (gas 
chromatography).  Employees refusing to take the test stand the risk of having a negative 
inference being drawn at any future proceedings that may follow. 
 
2.3.4 Rehabilitation 
The company bears the full cost of rehabilitation through the employee?s department.  
The company also bears the transportation costs in terms of the Medical Travel policy to 
and from the rehabilitation centre. 
 
A leave of absence is also granted to the employee during the duration of his 
rehabilitation by a company recognised medical practitioner.   In addition to all of these, 
the company arranges for and bears the cost of a counselling session involving the 
employee, his/her supervisor, his/her representative (if requested), an employee relations 
official and the senior human resources officer.   
 
Rehabilitation is offered only to permanent employees and is for one treatment only.  It is 
only repeated at the discretion of the head of department in exceptional circumstances.  
Subsequent costs will have to be borne by the concerned employe. 
 
2.4 STUDY POPULATION AND SAMPLING 
2.4.1 Sample Size Estimation 
No previous study has been done in this area on diamond mines in South Africa.  
However, literature that was reviewed showed a prevalence of alcohol use of between 25 
 19
 to 30% in studies carried out on the general populace in South Africa 25 Also, in a study 
carried out on a Welkom goldmine a prevalence of alcohol use of 32% was reported. 3 
 
Assuming a 30% prevalence with a 5% margin of error, the sample size required for the 
study was 282 at the 95% confidence level.  This was calculated with Epi Info version 6.1 
using Statcalc.  However, to make allowance for non-response, a 20% increase in sample 
size was deemed adequate.  Thus, the sample size calculated was 338. 
 
2.4.2 Sampling Method 
A simple random sample of miners was drawn from the mine register. 338 miners were 
selected to participate from a sampling frame of 740 miners. The sampling frame 
comprised of all the workers in the general staff list.  
Samples were collected pre-shift with the cooperation of the mine management, the 
access cards of selected participants were inactivated through the computer, a system 
normally used in this mine for similar purposes and referred to as ?parading?. 
 
Paraded miners were instructed about the study and written informed consent was sought 
and obtained for each person that took part in this survey.  On obtaining them, they were 
required to submit to the tests and a urine sample was also required from them at the 
same time. 
 
In order not to disrupt production at the mine (a condition agreed upon by both parties 
during prior negotiations) and also in order to keep sampling strictly pre-shift, study 
 20
 participants were required to submit themselves for the structured interviews during 
shifts. 
Refusals were replaced by the next name on the register.  The same individuals 
undergoing the breath and urine tests also completed the structured interviews. 
 
2.5  DATA COLLECTION METHODS 
2.5.1 Structured Interviews 
Face-to-face interviews were carried out by means of a structured interview schedule, 
which was administered by trained interviewers.  The team comprised of three 
undergraduate medical students at the University of the Witwatersrand, one post-graduate 
engineering student and three matric holders experienced in questionnaire administration 
from similar studies done in the immediate past.  In addition, all the interviewers were 
trained to administer the questionnaires in relevant local languages to ensure correct 
interpretation of miners? knowledge, attitudes and practices as regards alcohol and 
cannabis use and their perception to its health and safety risks. 
 
The questionnaires were anonymous and unlinked to both the breath and urine results.  
The questionnaire was divided into three parts (see Appendix A) and contained questions 
relating to the following: 
 
 
 
 
 21
 2.5.1.1 Section 1 of the questionnaire 
This part contains questions relating to demographic information.  Participants were 
asked to state their age, sex, main language, religion, educational level, type of 
accommodation they live in and people living with them.  They were also asked if they 
worked above ground or underground.  Questions relating to their perception on the 
dangers associated with their jobs were also asked. 
 
2.5.1.2 Section 2 of the questionnaire 
This is the alcohol use interview section.  Questions asked here related to current and past 
use of alcohol.  Also questions relating to types of alcohol consumed and quantity of 
alcohol consumed within two weeks prior to data collection were asked. 
Respondents were also required to state who they usually drink with, the times they 
usually drink more and the reasons behind their taking alcohol.  CAGE questions were 
included to determine the prevalence and attitude of chronic alcohol users. 
 
2.5.1.3  Section 3 of the questionnaire 
This was the substance use interview section.  Participants were asked about their 
cannabis usage status.  Questions relating to current use, if usage predates work on the 
mine or started after beginning work on the mine were also asked.  Also questions to 
determine how often respondents used cannabis, their perception of the relationship 
between cannabis usage and accidents as well as their perception of health risks 
associated with cannabis usage were asked. 
 22
 Options to the questions ranged from ?yes?, ?no? to ?I don?t know?.  Open-ended 
questions were also included to further clarify the responses. 
 
2.5.2  Assessments 
2.5.2.1    Alcohol assessment 
Breath alcohol was assessed using the Alca test 7410 RS breathalysers?, the calibration 
of which was verified against another breathalyser the Alca test 7110?, certified by the 
Council for Scientific and Industrial Research (CSIR) and the South African Bureau of 
Standards (SABS). For each test, one litre of air is blown over a period of two seconds or 
1.7 litres over four seconds is blown into a mouthpiece.  Each mouthpiece has a 
unidirectional air valve so that air blown into the instrument can only flow into the 
screening device and cannot be inhaled by the individual whose test is being done or that 
of other individuals whose tests will be done later.  This is aimed at limiting the 
transmission of infections.  The test uses an electrochemical sensor that is sensitive to the 
breakdown of ethanol in the liver.  Precision of the test is high because whatever the 
amount of air that is blown into the mouthpiece, only 1cc is analysed every time. Since 
there is no legislated alcohol limit in the mining industry, the current legal driving limit 
of 0.10mg/1000ml of breath for professional drivers as contained in section 65(6) of the 
South African National Road Traffic act of 1996 was used. 58 
 
Because a very high breath alcohol concentration, thick tobacco smoke and alcoholic 
mouth sprays, medicines and drops as well as recent vomiting may lead to false 
 23
 estimation, a standard pre-test questionnaire asking various questions to detect any of 
these was included (see Appendix D). 
 
For those who smoked immediately prior to the test, an interval of two minutes was 
allowed before tests were administered.  For alcohol mouth sprays, 15 minutes interval 
was given.  These requirements were specified by the manufacturers (Dr?ger 
Sicherheitstechnik GmhH, L?beck, Germany). 
 
2.5.2.2 Cannabis assessment 
Urine specimens were tested for cannabis using the ?On Trak Testik THC??.  This is an 
in vitro diagnostic test for the detection of cannabinoids at a level equal to or above 
50ng/ml. This test relies on competition between cannabis or its metabolites present in 
the urine and conjugate immobilized on the membrane for binding to antibody coated 
coloured micro particles. A negative result is obtained when a blue band forms at the 
result window.  If the test is positive, the result window remains white.  There is a control 
band that forms at the ?TEST VALID AREA?.  A blue band indicates that the test has 
been completed, the reagents viable and the results valid and ready to be interpreted. As 
part of a larger study, the sensitivity of the ?On Trak Testik? was calculated to be 80% 
and specificity calculated to be 97%. 
 
2.6    PILOT STUDY 
The pilot for this study was successfully carried out in early February 2002 on eleven 
mineworkers. The mine where the pilot study was carried out was randomly selected 
 24
 from a list of mines and it is located in the same province where this study was carried 
out.  The aim of the pilot was to test the instruments used for this study and to refine them 
where necessary. 
 
2.7    DATA ENTRY, PROCESSING AND ANALYSIS 
The questionnaires were coded and data were entered into a computer using Microsoft 
Excel spreadsheet.  The data files comprised 250 variables on which the responses were 
obtained.  These files were converted into SAS format and responses were coded and 
descriptive statistics carried out. The analysis which included frequency distributions, 
chi-square analysis and cross tabulations yielding p-values were performed. Tables were 
also used to present and interpret findings. 
 
2.8  ETHICAL CONSIDERATIONS 
Written informed consent was sought and obtained prior to all interviews (see Appendix 
C).  All potential participants were explained the reason for and the methods employed 
for collecting the data (see Appendix B). Participation was voluntary and any participants 
who did not wish to continue at any stage of the study were allowed to withdraw without 
penalty. Results were anonymous and confidentiality was strictly maintained in all cases 
and at all times. 
 
 
 
 
 25
 2.9    POSSIBLE LIMITATIONS AND STEPS TAKEN TO REDUCE THEM 
2.9.1    Interviews 
Studies involving interviewer-administered questionnaires are fraught with difficulties 
particularly when it involves translating responses from one language to the other.  
Certain nuances peculiar to certain languages might be missed out during translation.  
Also, accurate translation of responses might vary from individual to individual. 
 
This limitation was controlled by rigorously training the interviewers and subjecting them 
to simulated interview sessions.  The trainees were then selected according to pre-set 
selection criteria.  The criteria were strictly adhered to with those not meeting up to set 
standard left out of the team.  In addition, the questionnaires were structured to be very 
simple and user friendly. The questionnaire was piloted and information gathered through 
the process utilized in enhancing the tool. Interviewers were people from a similar socio-
 cultural environment as the miners and spoke the same language as them in order to 
increase effective communication of questions and answers. 
 
2.9.2 Underestimation of Reported Prevalence 
The use of alcohol and especially cannabis in the workplace is a very sensitive and 
contentious issue.  Carrying out this study in a mining environment makes it even more 
contentious as alcohol, cannabis and other illegal substances are prohibited on the mine 
premises. 
The fear of social stigma associated with cannabis use as well as the fact that cannabis 
use is a criminal offence in South Africa might have resulted in underreporting of use in 
 26
 this mine.  Also, the fear of legal prosecution and threat to job security might have been a 
factor as well. The effect of underreporting in this study was reduced by having 
numerous meetings with workers representatives, unions and management to ensure buy 
in. Also guaranteeing anonymity and maintaining strict confidentiality throughout the 
study played a major role in helping to reduce underreporting.    
 
2.9.3 Underestimation of Tested Prevalence 
The healthy worker effect might have contributed to this.  Workers who probably were 
absent on the study days might have been suffering from the effects of alcohol and/or 
cannabis consumption and might have thus been left out of the tested population thereby 
giving a reduced rate of tested prevalence. 
 
Also, false negative results from rapid metabolism of alcohol or people with impaired 
liver function might have underestimated the true prevalence.18 Selection bias could have 
an effect here as well. Participants who knew that they were positive might have decided 
to opt out of the study. As mentioned before, this was reduced by having numerous 
meetings with the workers and their unions to ensure worker buy in. Also maintaining 
confidentiality and anonymity throughout the duration of the study helped in ameliorating 
this.     
 
2.9.4 Overestimation of Tested Prevalence 
Cannabis test reliability becomes a problem if the urine is allowed to stand for a 
considerable time period.  Tests for cannabis were therefore carried out immediately after 
 27
 urine samples were obtained.  When this was not possible, they were stored in ice-packed 
containers and were transported to the laboratory for testing as per the instructions of the 
manufacturers.  Cannabis metabolites can remain in the urine long after the effects of the 
drug have worn off. 53 Thus a positive urine test result does not necessarily translate to 
current use and possible impairment but might represent past use.  Also, unknown use 
whereby people consume cannabis through substances like herb tea or intermittent use in 
other ways will return a positive result and hence overestimate the true prevalence. 54 
 
 28
 CHAPTER 3 
 
3.0 STUDY RESULTS 
3.1 PARTICIPATION RATES 
Mine X has a workforce of 740.  Of this number, 590 are miners whilst 150 are 
administrative staff personnel.  The sample size calculated for the study was 338. Of the 
338 miners randomly selected into the study sample, the participation rates in the 
different aspects of the study are as shown in Table 3.1 below. 
 
Table 3.1: Participation Rates 
Type of test Sample Size Frequency Percentage 
Breathalyser 
Urine Cannabis Test 
Questionnaire (Interviews) 
338 
338 
338 
318 
316 
322 
94.1 
93.5 
95.2 
 
 
There were 318 breath samples collected over the five-day period.  The total number of 
questionnaires administered was 322. More questionnaires were administered because 
four people who did not earlier subject themselves for breathalyser tests turned up for 
questionnaire interviews. This was not detected during data collection and since it was 
impossible to link individual questionnaires to breath samples due to the study design, 
these questionnaires could not be identified in order to be removed so they were included. 
The total number of urine samples collected was 316.  
 
 29
 3.2 FACTORS THAT MAY PREDISPOSE MINERS TO ALCOHOL AND 
CANNABIS USE 
A large majority of the participants (59.35%, 95% CI: 53.7 - 64.9) had secondary school 
education of between standards 6 to10.  This group made up more than half of the 
participants.28.3% (95% CI: 23.4 - 33.8) of the participants had post-matric education 
whilst 8.39% (95% CI: 5.6 -12.0) had primary school education between standards 1 to 5. 
Miners with no formal education made up 3.87% (95% CI: 2.0 - 6.6) of the entire study 
population (see Table 3.2). 
 
The majority of the participants were male (86.98%, 95% CI: 82.8 - 90.5), as opposed to 
female participants who made up only 13.02% (95% CI: 9.5 - 17.2). Most of the 
participants (72.33%, 95% CI: 67.1 - 77.2) were married, whilst 20.75% (95% CI: 16.4 - 
25.6) were single. The remainder of the participants (6.92%, 95% CI: 4.4 - 10.3) were 
either unmarried, divorced/separated or were cohabiting with live-in partners (see Table 
3.2). 
 
A majority of the respondents (57.1%, 95% CI: 51.4 - 62.6), owned their 
accommodation; 42.59% (95% CI: 37.1 - 48.2) either shared off site accommodation or 
lived in single or more than one hired rooms; A small percentage of the participants 
(0.32%, 95% CI: 0.01 - 1.7), lived in hostel accommodation. 82.14% (95% CI: 77.4 - 
86.3) of the participants had family members living with them whilst 17.86% (95% CI: 
13.7 - 22.6) did not have family members living with them (see Table 3.2). 
 
 30
 Full-time employees made up 84.54% (95% CI: 80.1 - 88.3) of the participants in the 
study whilst the rest (15.46%, 95% CI: 11.7 - 19.9), were contract staff.  About half of 
the number of participants (50.51%, 95% CI: 44.7 - 56.4), was group 3-4 employees 
(these are the employees in the lowest category who carry out labour intensive jobs such 
as truck drivers, drillers, operators and cleaners). Officials made up 33.56% (95% CI: 
28.2 - 39.3) of the participants, whilst 13.56% (95% CI: 9.9 - 18.0) and 2.37% (95% CI: 
0.9 - 4.8) were union men and group 5-8 employees (these include staff in supervisory 
positions such as team leaders and team supervisors) respectively.  The age of the 
participants ranged from 16-65 years.  The average age of the participants was 35 years.  
Most of the participants, 44.65%, were between 31-40 years; 32.7% of the participants 
were above 40 years whilst 22.6% were below 30 years. 
 
Most of the participants in this study (95.9%, 95% CI: 93.1 - 97.8) were of South African 
origin as against non-South Africans, which accounted for 4.1% (95% CI: 2.2 -6.9) of the 
study population.  This largely corresponds to the demographic pattern of the entire mine 
(see Table 3.2). 
 
 
 
 
 
 
 
 31
 Table 3.2: Sociodemographic Profile of Participants 
Variable Response Frequency Percent 
Accommodation 
Hostel 
Other 
Own 
1 
135 
181 
0.32 
42.59 
57.1 
Total  317 100 
Country 
Non SA 
SA 
13 
304 
4.1 
95.9 
Total  317 100 
Education 
No schooling 
Primary (Std 1-5) 
Secondary (Std 6-10) 
Post Matric 
12 
26 
184 
88 
3.87 
8.39 
59.35 
28.39 
Total  310 100 
Gender 
Male 
Female 
274 
41 
86.98 
13.02 
Total  315 100 
Marital Status 
Single 
Married 
Other 
66 
230 
22 
20.75 
72.33 
6.92 
Total  318 100 
Member lives with you? 
Yes 
No 
252 
55 
82.14 
17.86 
Total  308 100 
Job Category 
Officials 
Union Men 
Group 5-8 
Group 3-4 
99 
40 
7 
149 
33.56 
13.56 
2.37 
50.51 
Total  295 100 
Type of Work 
Contract 
Full time 
49 
268 
15.46 
84.54 
Total  317 100 
 
. 
 
 32
 3.3 PREVALENCE OF ALCOHOL USE 
Of the 318 participants that were subjected to breathalyser tests, 1.57% (95% CI: 0.58 - 
3.84) of the participants tested positive for breath alcohol i.e. they had blood alcohol 
levels of equal to or greater than 0.10mg/1000ml. The majority (98.42%, 95% CI: 96.37 
? 99.49) tested negative. 
Of the 318 participants, only 0.94% (95% CI: 0.24 - 2.96) had blood alcohol levels equal 
to or above the legal limit of 0.10mg/100ml (see table 3.3). 
 
Table 3.3: Tested Prevalence of Alcohol Use (Results of Breathalyser Testing) 
Results Frequency Percent 
No alcohol 
Positive 
Above Limit 
313 
2 
3 
98.42 
0.63 
0.94 
Total 318 100 
 
 
3.3.1 Reported Prevalence of Alcohol Use 
When asked if they had ever used alcohol 48.47% (95%CI: 4 3.2 - 53.8) of the 
respondents answered in the affirmative. Of this number 90.8% (95% CI: 85.50 - 94.65) 
admitted to be currently using alcohol. This gives an overall reported prevalence of 
alcohol use as determined with the administered questionnaires to be 44% (95% CI: 38.8 
- 49.3) 
 
 
 
 33
 3.3.2  Chronic Alcohol Use  
The frequency of chronic alcohol use as determined by the use of the CAGE screening 
questions were as follows: 
Of the 322 participants who were assessed with the questionnaire tool, 315 of them 
responded to the ?CAGE? questions. Of these 3.49% (95% CI: 1.76 ? 6.16) screened 
positive for chronic alcohol use. This is in comparison to the breathalyser-screening test, 
which returned a 1.4% positive rate.  This could mean that about 3.49% of the 
participants are chronic users of alcohol as detected by the ?CAGE? questions.    The 
disparity   between the breathalyser and the ?CAGE? questions could be explained by the 
period in time that the tests were conducted. An analysis of the demographic variables of 
those participants who admitted to ever using alcohol was done with respect to their 
demographic characteristics. 
 
3.3.3 Lifetime Alcohol Use by Demographic Variables 
56.71% of South Africans (95% CI: 50.9 - 62.4) as opposed to 53.58% (95% CI: 25.1 - 
80.8) of non-South Africans were found to have responded in the affirmative when asked 
if they had every used alcohol.  This finding was however not statistically significant (p 
value = 0.8383) (Table 3.4). 
With regards to educational qualification, it was found that the likelihood to have ever 
used alcohol increased with the level of education obtained: 25% (95% CI: 5.5 - 57.2) of 
the people with no formal education were found to have ever used alcohol as opposed to 
71.59%, (95% CI: 61 - 81) of people with post matric qualifications (p value = 0.0013) 
(Table 3.4). 
 34
 Overall, full-time employees were found to have a higher likelihood of ever having used 
alcohol when compared with contract staff, 60.08% (95% CI: 54 - 66) and 37.5% (95% 
CI: 24% - 53%) respectively (p value = 0.0037) (Table 3.4). 
 
Table 3.4: Lifetime Alcohol Use by Demographic Variables 
Yes No  
Demographic Variable Number Percent Number Percent 
Country 
South African 
Non-South African 
169 
7 
56.71 
53.95 
129 
6 
43.29 
46.15 
Total 176  135  
P value = 0.8383 
Educational Group     
No schooling 
Primary (Std 1-5) 
Secondary Std. 6-10) 
Post matric 
3 
11 
94 
63 
25 
44 
52.51 
71.59 
9 
14 
85 
25 
75 
56 
47.49 
28.41 
Total 171  133  
P value = 0.0013 
Type of Work     
Contract Worker 
Full-time Employee 
18 
158 
10.22 
89.77 
30 
105 
22.22 
77.77 
Total 176  135  
P value = 0.0037 
 
 
3.3.4    Current Use of Alcohol by Demographic Variables 
Table 3.5 shows an analysis of the responses of those who reported to be currently using 
alcohol against their demographic profiles.  The only statistically significant finding from 
this analysis is from the type of work done i.e. whether full time or contract work. Full 
 35
 time employees who had ever used alcohol were still currently using alcohol were 
92.76% (95% CI: 87 ? 96) as opposed to 76.19% (95% CI: 53 - 92) of contract staff (p 
value = 0.014). 
 
The country of origin, age, marital status, type of accommodation, educational 
qualification, the influence of living with a family member, where the staff member 
works as well as the type of job done when cross tabulated with current use of alcohol 
was found not to be statistically significant. 
 
Table 3.5: Current Use of Alcohol by Demographic Variables 
Current Use 
Yes No 
 
Demographic  
Variables Number Percent Number Percent 
Total 
Country: 
South African 
Non-South African 
 
150 
7 
 
90.91 
87.5 
 
15 
1 
 
9.09 
12.5 
 
165 
8 
Total 157  16  173 
P value = 0.7451 
Age 
30 and less 
31 to 40 
40 and above 
 
43 
59 
56 
 
95.56 
86.76 
91.8 
 
2 
9 
5 
 
4.44 
13.24 
8.2 
 
45 
68 
61 
Total 158  16  174 
P value = 0.27 
Type of work: Contract 
Worker 
Full-time employee 
 
16 
141 
 
76.19 
92.76 
 
5 
11 
 
23.81 
7.24 
 
21 
152 
Total 157  16  173 
P value = 0.014 
 
 
 36
 Marital status 
Single 
Married 
Others 
 
41 
106 
11 
 
91.11 
90.6 
91.67 
 
4 
11 
1 
 
8.89 
9.4 
8.33 
 
45 
117 
12 
Total 158  16  174 
P value = 0.989 
Current use 
Yes No 
 
Demographic  
Variables Number Percent Number Percent 
Total 
Accommodation 
Own 
Hostel 
Others 
 
75 
1 
82 
88.24 
100 
93.18 
 
10 
0 
6 
 
11.76 
0 
6.82 
 
85 
1 
88 
Total 158  16  174 
P value = 0.5044 
Educational 
group 
No schooling 
Primary (Std 1-5) 
Secondary Std. 6-10) 
Post Matric 
 
 
3 
9 
86 
55 
 
 
100 
75 
93.47 
91.67 
 
 
0 
3 
7 
5 
 
 
0 
25 
7.53 
8.33 
 
 
3 
12 
93 
60 
Total 153  15  168 
P value = 0.225 
Does any member of your 
family live with you? 
Yes 
No 
 
 
122 
29 
 
 
90.37 
90.63 
 
 
13 
3 
 
 
9.63 
9.83 
 
 
135 
32 
Total 151  16  167 
P value = 0.9649 
Where work 
Underground 
Above Ground 
 
11 
145 
 
100 
90.06 
 
0 
16 
 
0 
9.94 
 
11 
16 
Total 156  16  27 
P value = 0.272 
 
 
 
 37
 Current use 
 
Yes 
No 
 
 
Demographic 
Variables Number Percent Number Percent 
Total 
Type of job group 
Officials 
Union Men 
Group 5-8 
Group 3-4 
 
58 
17 
3 
68 
 
93.55 
77.27 
100 
90.67 
 
4 
5 
0 
7 
 
6.45 
22.73 
0 
9.33 
 
62 
22 
3 
75 
Total 146  16  162 
P value = 0.1542 
 
 
3.3.5    The Distribution of Drinking Companions 
The majority of the participants (46.98%, 95% CI: 41.02 - 52.99) who admitted to 
drinking said that they did so in the company of their friends. Some of the respondents 
(8.9%, 95% CI: 5.8 ? 12.85) said that they drank alone (Table 3.6).  
 
A small percentage of the respondents (1.78%, 95% CI: 0.58%-4.1%) drank with their 
family members whilst a few (1.42%, 95% CI: 0.38 - 3.6) drank both alone and with 
friends. One person (0.3%, 95% CI: 0.009 - 0.02) claimed that he drank with anybody 
that was available (Table 3.6). 
It is however worthy of note that 40.57% (95% CI: 34.8 - 46.6) of the respondents either 
did not answer this particular question (question number 18) or had incorrectly filled 
responses. 
 
 
 38
 Table 3.6: Distribution of Drinking Companions 
Who with Frequency Percent 
Not answered 
Alone 
With Friends 
Family Members 
Alone and With Friends 
Anybody 
114 
25 
132 
5 
4 
1 
40.57 
8.9 
46.98 
1.78 
1.42 
0.36 
Total 281 100 
 
3.3.6    Places of Alcohol Usage 
The majority of the respondents admitted to drinking at home (63.54%, 95% CI: 56.1 - 
70.6).  The rest either drank at shebeens or taverns (11.05%, 95% CI: 6.9 -16.5), clubs 
(9.94%, 95% CI: 6.0 -15.3) or at parties with friends (5.52%, 95% CI: 2.7 - 9.9). A 
similar number of participants claimed to drink in bars or hostels and everywhere i.e.    
1.1% (95% CI: 0.13 - 3.9) for both groups (Table 3.7). 
 
Table 3.7: Places of Alcohol Usage 
Where drink Frequency Percent 
Clubs 
Everywhere 
Party/Friends 
Home 
Hostel/Bar 
Party 
Shebeen/Tavern 
Total 
18 
2 
10 
115 
2 
14 
20 
181 
9.95 
1.1 
5.53 
63.54 
1.1 
7.73 
11.05 
100 
 
 
 39
 The vast majority of the respondents (89.96%, 95% CI: 83.3 - 94.2) reported using 
alcohol during the evenings, at weekends or after work; 5.41% (95% CI: 2.4 - 10.4) 
indicated that they used alcohol only at occasions, such as parties whilst 4.05% (95% CI: 
1.5 -8.6) of the people who admitted to currently using alcohol admitted to doing so 
during the daytime. When the respondents were asked to give reasons why they use 
alcohol, 43.7% (95% CI: 35.2 - 52.5) of the respondents to this particular question 
claimed to do so for relaxation and enjoyment. Other reasons included: fun/entertainment 
and party, a response given by 25.19% (95% CI: 18.11 - 33.4) of the respondents, 5.19% 
(95% CI: 2.1 -10.4) said they did so for experimentation whilst 0.74% (95% CI: 0.02 - 
4.1) said that they drank alcohol only at braais. 
 
The results show that a majority of the respondents i.e. 75% (95% CI: 68 - 81.2) started 
taking alcohol before commencing work on the mines as opposed to 25% (95% CI: 18.9 - 
31.9) who started taking alcohol after commencing work on the mines. 
 
3.4 PREVALENCE OF CANNABIS USE 
A total of 316 urine samples were available for cannabis testing. Of these 2.85% (95% 
CI: 1.31-5.3) were positive for cannabis whilst 97.15% (95% CI: 94.66 - 98.69) were 
negative. Of the participants, 6% (95% CI: 3.7 - 9.2) admitted to ever using cannabis.  
However, only 2% (95% CI: 0.7 - 4.1) of the participants admitted to be currently using 
cannabis - the overall reported prevalence for current cannabis use (Table 3.8).  
 
 
 40
 Table 3.8:  Tested Prevalence of Cannabis Use 
Urine results Frequency Percent 
Negative 
Positive 
307 
9 
97.15 
2.8 
Total 316 100 
 
3.4.1 Practice of Lifetime Cannabis Use  
Table 3.9 shows the overall reported prevalence for cannabis by demographic variable. 
From the results, a higher proportion of non-South Africans use cannabis when compared 
to South Africans: 15.38% (95% CI: 1.92 - 45.3) versus 5.61% (95% CI: 3.3 - 8.8).  The 
difference in proportion was however not statistically significant (p value = 0.1466). 
 
With regards to the influence of education on cannabis use, participants with an 
educational level of standard 6-10 had the highest rate of cannabis use of 4.2% (13 of 
309; 95% CI: 3.8 ? 11.8).  This was followed by participants with post matric 
qualifications with a rate of 1.62% (5 0f 309; 95% CI: 1.9 - 12.9).  Participants with 
primary qualification of Standard 1-5 had only a rate of 0.32% (1 of 309; 95% CI: 0.10 - 
19.6) (Table 3.9). 
 
People with no formal educational qualification did not admit to ever using cannabis.  .  
Contract workers were more likely to have ever used cannabis. 6.12% (95% CI: 1.35 - 
16.8) of them admitted to this when compared to 5.99% (95% CI: 3.5 - 9.5) of full-time 
employees.  This result was also not statistically significant (p value = 0.9719) (Table 
3.9). 
 
 41
 Table 3.9:  Lifetime Cannabis Use by Demographic Variables 
 Ever use of cannabis 
Demographic Variable Yes No 
 Number Percent Number Percent 
Country: 
South African 
Non-South African 
 
17 
2 
 
15.38 
5.61 
 
11 
286 
 
84.62 
94.39 
Total 19  297  
P value = 0.1466 
Educational group 
No Schooling 
Primary (Std 1-5) 
Secondary (Std 6-10) 
Post Matric 
 
0 
1 
13 
5 
 
0 
3.85 
7.07 
5.75 
 
12 
25 
171 
82 
 
100 
96.15 
92.93 
94.25 
Total 19  290  
P value = 0.7251 
Type of work: 
Contract Worker 
Full-time Employee 
 
3 
16 
 
6.12 
5.99 
 
46 
251 
 
93.88 
94.01 
Total 19  297  
P value = 0.9719 
 
3.4.2    Current Cannabis Use 
An analysis of the demographic profiles of the participants who admitted to current 
cannabis use shows that 75% (95% CI: 19.4 - 99.4) of the people living without family 
members currently use cannabis when compared to 25% (95% CI: 0.63 - 80.6) living 
with families.  For the participants who currently live with family members; 14.29% 
(95% CI: 1.8 - 42.8) currently use cannabis when compared to 85.71% (95% CI: 57.2 - 
98.2) who do not.  These results are statistically significant (p = 0.0168). The rest of the 
analysis of demographic variables against current use did not show any statistically 
significant findings. 
 42
 A vast majority of the participants who conceded to using cannabis, claimed to do so with 
friends (82.35%, 95% CI: 56.5 - 96.2). About 12% of users said they did so in the 
company of friends, while 5.9% admitted to cannabis use with a family member. A great 
majority of the participants (72.22%, 95% CI: 46.5 - 90.3) started using cannabis before 
beginning work on the mines. 
 
3.5 PERCEPTIONS, KNOWLEDGE OF, AND ATTITUDES TOWARDS 
ALCOHOL 
Of 309 respondents 66.67% (95% CI: 61.11 ? 71.90) stated that they thought that their 
fellow workers used alcohol, 12.26% (95% CI: 9.1 - 16.8) of the respondents answered 
no to this question whilst 20.71% (95% CI: 16.3 - 25.7) claimed not to know if their 
fellow workers used alcohol. 
A majority (32.89%, 95% CI: 26.8 - 39.4) of the participants believe their fellow workers 
use alcohol when they are off duty or after work hours.  28.51% (95% CI: 22.7 - 34.8) of 
workers believe that those who use alcohol do so during special occasions or when very 
happy.  Some others i.e. 16.23% (95 CI: 11.7 - 21.7) believe the use of alcohol to be more 
during month end or on payday. 
About 7.02% (95% CI: 4.1 - 11.1) of the respondents believe that their fellow co-workers 
use alcohol before work or during work; 15.35% (95% CI: 10.9 - 20.7) gave answers that 
varied from not being sure, no specific times and when having personal problems. 
As for their perception of the dangers associated with their work, 50.5% (95% CI: 44.69 
? 56.3) reported that they thought their jobs were dangerous.  40.8% of them did not 
think that their work as being dangerous while 7.8% said they did not know whether their 
 43
 work was dangerous. With regards to knowledge of alcohol use on duty as a cause of 
work accident, the vast majority of the participants (98.06%, CI: 95.84 - 98.29) believe 
that drinking alcohol can lead to accidents on the mine. Only 1.94% (95% CI: 0.7 - 4.2) 
of the participants think otherwise (Table 3.10 below). When asked if they thought that 
anything could be done to influence the use of alcohol among miners, 74.5% (95% CI: 
69.33 ? 78.25) answered in the affirmative, 7.64% (95% CI: 7.61 - 14.8) of the 
participants thought that nothing can be done to influence the use of alcohol amongst 
miners whilst 17.83% (95% CI: 13.8 - 22.5) of the participants were uncertain. 
 
Table 3.10:  Knowledge of Danger of Alcohol 
Do you think that drinking alcohol can lead to 
accidents on the mine? 
Frequency Percent 
Yes 
No 
304 
6 
98.06 
1.94 
Total 310 100 
 
When asked to make suggestions on methods of reducing alcohol use amongst miners, 
education about the effects of alcohol was the commonest suggestion. This was made by 
18.85% (95% CI: 14.9 - 23.2) of the respondents, 13.11% (95% CI: 9.8 - 17.0) of the 
respondents believe that providing entertainment and recreational facilities was the 
solution, 16.67% (95% CI: 12.9 - 20.8) of the respondents answered that the introduction 
of stringent testing procedures will dramatically decrease the use of alcohol amongst 
miners (Table 3.11). 
 
 
 44
 Table 3.11:  Methods of Reducing Alcohol Use 
What can be done to influence the use of alcohol 
amongst miners? 
Frequency Percent 
Education 
Entertainment 
Testing 
Others 
69 
48 
61 
188 
18.85 
13.11 
16.67 
51.37 
Total 366 100 
 
However, 51.37% (95% CI: 46.1 - 56.6) of the miners gave reasons such as increasing 
religious activities on the mines, the shifting of the responsibility of monitoring alcohol 
consumption to co-workers, to introduce severe disciplinary actions and to totally ban the 
sale of alcohol beverages around the mine premises as means of reducing alcohol use. 
  
3.6 PERCEPTIONS, KNOWLEDGE OF, AND ATTITUDES TOWARDS 
CANNABIS 
About 15% (95% CI: 11.42 - 19.63) of the participants in this study answered in the 
affirmative when asked if they thought that their fellow workers used cannabis. Almost 
forty percent (37.97%, 95% CI: 32.60 - 43.58) of the participants believe that their fellow 
workers do not use cannabis whilst a majority of the participants (46.84%, 95% CI: 41.2 - 
52.5) claimed not to know if their fellow workers used cannabis. 
 
 
 
 
 
 45
 Table 3.12:  Perceptions Regarding Use of Cannabis by Fellow Workers 
Do your fellow workers use cannabis? 
Response Frequency Percent 
Yes 
No 
Don?t Know 
48 
120 
148 
15.19 
37.92 
46.84 
Totals 316 100 
 
The stated reasons why fellow workers used cannabis include relaxation/to forget their 
problems (25.53%, 95% CI: 13.9 - 40.4), to relieve stress (17.02%, 95% CI: 7.6 - 30.8), 
and as a habit (10.64%, 95% CI: 3.6 - 23.11).  Some (46.81%, 95% CI: 32.1 - 61.9) of the 
respondents gave other reasons such as to keep busy and reduce boredom, for fun, to help 
them avoid stressful situations at the workplace etc (Table 3.13). 
 
Table 3.13:  Perceptions of Why Fellow Workers Use Cannabis 
Reason Frequency Percent CI 
To relax and forget their problems 12 25.53 13.9-40.4 
Habit 5 10.64 2.6-23.11 
Relieve stress 8 17.02 7.6-30.8 
Others 22 46.81 32.1-61.9 
Total 47 100  
 
More than half (56.16%, 95% CI: 44 - 67.8) of the respondents indicated that they 
thought that fellow miners used cannabis when working hard. 21.92% (95% CI: 13.1 - 
33.1) of the respondents felt that miners used cannabis more after work, when off duty 
and during weekend. About a fifth (21.92%, 95% CI: 13.1 - 33.1) of the respondents gave 
various other reasons such as to satisfy themselves, usually after meals and at any time. 
 46
 About half (58.10%, 95% CI: 52.43 - 63.60) of the participants felt that working in the 
mines is dangerous to their health, 38.10% (95% CI: 32.7 - 43.7) of the respondents do 
not perceive any danger to their health as a result of working in the mines whilst 3.81% 
(95% CI: 1.9 - 6.6) claimed not to know if working in the mines posed a danger to their 
health. 
 
Most of the participants i.e. 90.49% (95% CI: 86.63 - 93.54) agreed that using cannabis 
can lead to accidents on the mine, 9.18% (95% CI: 6.2 - 13.0) of the respondents disagree 
whilst 0.33% (95% CI: 0.008 - 1.8) claimed not to know. 
When asked if anything could be done to influence cannabis use amongst miners, 77.92% 
(95% CI: 72.87 - 82.43) of the respondents answered in the affirmative, 6.5% (95% CI: 
4.0 - 9.8) of the respondents do not believe that anything could be done to influence 
cannabis use amongst miners whilst 15.58% (95% CI: 11.7 - 20.1) of the respondents 
claimed not to know. 
 
When asked to explain what could be done to influence the use of cannabis by miners, 
58.97% (95% CI: 53.06 - 64.68) of the respondents cited educating the miners as a means 
of doing this, it was followed by reasons such as regular testing, which was suggested by 
9.66% (95% CI: 6.5 - 13.6) of the respondents, entertainment and sports which was given 
by 1.38% (95% CI: 0.3 - 3.5) of the respondents. About a third (30%, 95% CI: 24.7 - 
35.6) of the respondents gave other reasons such as giving miners enough rest time, 
rehabilitating those already using or addicted to cannabis, the introduction of sports and 
recreation facilities etc as methods of influencing the use of cannabis (Table 3.14). 
 47
 Table 3.14:  Opinion on Influencing the Use of Cannabis Amongst Miners 
What can be done to influence the use of cannabis amongst miners? 
Response Frequency Percent 
Education 
Entertainment and Sport 
Testing 
Others 
171 
4 
28 
87 
 
58.97 
1.38 
9.66 
30 
 
Total 290 100 
 
 
3.7  KNOWLEDGE OF SPORT AND RECREATION 
Most of the participants (62.10%, 95% CI: 56.48 - 67.49) seemed to be aware of the 
availability of recreational facilities on the mine. A smaller number (28.66%, 95% CI: 
23.7 - 34.0) of the respondents seemed not to be aware of the availability of recreational 
facilities whilst a few (9.24%, 95% CI: 6.3 - 13.0) did not know if recreational facilities 
were existent or not on the mine. 
 
An overwhelming majority of the respondents i.e. 79.30% (95% CI: 74.4 - 83.6) 
answered that they believed that sports and recreational facilities will influence the use of 
cannabis and alcohol on the mine, 9.87% (95% CI: 6.8 - 13.7) of the respondents did not 
agree with this as they felt it will have no impact on the use of cannabis and alcohol. 
Some of the participants (10.83%, 95% CI: 7.0 - 14.8) of the participants were unsure if 
provision or awareness of sports and recreational facilities will influence the use of 
alcohol and cannabis on the mine (Table 3.15) 
 
 
 48
 Table 3.15:  Opinions Regarding Facilities 
Do you think that sports and recreational facilities influence the use of alcohol and cannabis? 
Response Frequency Percent 
Yes 
No 
Don?t Know 
249 
31 
34 
79.30 
9.87 
10.83 
Total 314 100 
 
 
 49
 CHAPTER 4 
 
4.0 DISCUSSION OF RESULTS 
This chapter explains the reasons behind the results and draws conclusions in line with 
the aims and objectives. It also compares the findings in this study with those from local 
and international studies. 
 
4.1  ALCOHOL 
As documented in the methods section, urine samples and breathalyser tests were carried 
out prior to the administration of questionnaires in this study. The calculated sample size 
for this study was 338 but a total of 318 people underwent breathalyser testing 
successfully while 316 urine samples were collected for cannabis testing. Majority 
(86.98%) of the participants were male with only 13.02% of participants being female. 
The predominance of male miners is generally representative of the gender distribution of 
the work force of most mines.  Socio-culturally, mining is seen as a very tough 
occupation meant for males.  Most women employed in the mining industry are mainly 
employed for administrative jobs.  
 
About half i.e. 50 .51% of the participants were employees who did actual mining job.  
Officials and artisans made up 37.9% whilst Union men made up the rest of the 
participants. The small proportion of the people who tested positive for alcohol in this 
study when compared with results from other studies17, 41 can be attributed to the 
stringent alcohol policy and the regular testing carried out on this mine. 
 50
 4.1.1 Prevalence of Alcohol Use 
The reported prevalence of ever use of alcohol on Mine X was 48.46% (95% CI: 43.20 - 
53.76). Of these, 90% (95% CI: 85.23 - 94.48) admitted to be currently using alcohol. 
This represents an overall reported prevalence of current alcohol use of 44% (95% CI: 
38.83 - 49.32).  This finding exceeds the prevalence for the general population which is 
put at between 10-35% 43 and the prevalence for risky drinking in a South African gold 
mine which was put at 32%. 41 
 
In comparison with studies done internationally, the reported prevalence of 44% recorded 
in this study exceeds that of 34% reported from Argentinean miners40 and 40% recorded 
for Zambian copper miners.59 However, a study on a Free State gold mine recorded a 
prevalence of reported alcohol use to be 60% in 1997.17 This far exceeds the value of 
44% in this study. Earlier studies done in 1994 gave a prevalence of 10-39%43 for the 
general populace and 32% 42 for a South African gold mine. In 1997, a similar study gave 
the prevalence to be 60%.  The differences in prevalence recorded in the different studies 
could be a reflection of the differences in prevalence of alcohol use in the general 
population. 25 
 
Full-time employees were found to have a higher likelihood to have ever used alcohol 
60.08% in comparison with contract staff 37.5%.  This result was statistically significant 
with a p-value of 0.004.  This might also be as a result of the company policy, which 
renders help to full-time employees who have alcohol problems.  The contract staffs, on 
 51
 the other hand, were employed by various organisations and policies on alcohol and 
alcohol-related problems differ for various recruiting company. 
 
71.59% of the participants with post-matric education were more likely to have ever used 
alcohol when compared to other participants i.e. secondary education 52.51%, primary 
education 44% and no formal education 25% (p value = 0.0013). This finding of a high 
prevalence of life time use of alcohol amongst people with higher educational 
qualifications in comparison with people of lower educational qualification is at variance 
with results from similar studies in the past 24 which found a higher prevalence amongst 
people with lower or no educational qualifications.  The reason that was put forward for 
such a finding in the above mentioned study was that people with lower educational 
qualifications were more likely to perform unskilled work hence the higher incidence 
among manual labourers in other sectors. 
 
In relation to this study, a more plausible explanation could be that the higher level of 
reported prevalence of ever use amongst people with higher educational qualifications i.e. 
skilled workers, could be as a result of the likelihood that those with higher educational 
qualifications apart from being more likely to be permanent staff members and hence 
have a higher notion of job security could also be aware that reporting alcohol use outside 
working hours would not affect their job. The converse is thought to be the case for the 
participants with a lower educational qualification. 
 
 
 52
 4.1.2  Pattern of Alcohol Use 
The majority of the participants who reported to have ever used alcohol, i.e. 75% (95% 
CI: 67.90 - 81.00) said that they started doing so before commencing work on this or any 
other mine. About a quarter of the respondents (25%, 95% CI: 19.00 - 32.10) reported 
that they started drinking after starting work on this mine.  
 
Most (40.57%) of the participants reported that they drank in the company of friends. 
This finding is in keeping with results from similar South African studies.3, 46  A small 
percentage of the participants reported that they drank alone i.e. 8.9% (95%CI; 5.8 - 
12.85%).  The low prevalence of people drinking alone might represent a socio-cultural 
norm amongst the participants.1 It might also represent a small proportion of high-risk 
drinkers. 52 Most of the participants i.e. 89.86% (95% CI: 83.8 - 94.2) reported drinking 
alcohol in the evenings, weekends or after meals.  A small percentage i.e. 5.4% (95% CI: 
2.4-10.4) drank in parties whilst a smaller percentage i.e. 4.05% (95% CI: 1.5 - 8.5) drank 
in the daytime.  A greater percentage of the people who drank alcohol did so in their 
homes i.e. 63.54% (95% CI: 56.1 - 70.6), when compared to 11.05% (95% CI: 6.9 - 16.5) 
of the participants who drank in shebeens; whilst 9.94% (95% CI: 6.0 - 15.3) drank in 
clubs, and 7.73% (95 CI: 4.3 - 12.6) drank at parties etc. 
 
The preferred place of drinking alcohol reported in this study seems to be in the homes.  
This is at variance with studies done locally which reported shebeens to be the most 
popular place to drink..3, 42, 46 This difference could have resulted because of a difference 
in the setting between the above-mentioned studies and this particular study.  One of the 
 53
 studies was carried out on the general population whilst the other was carried out on a 
gold mine with the majority of participants living in a hostel.42 In this study, the 
participants do not live in hostels.  Most of the participants live in their own homes 
situated far away from the mine premises. This high prevalence of reported alcohol use in 
own homes outside working hours might be as a result of lack of suitable leisure 
activities after work. 
 
4.1.3 Knowledge of and Attitudes Towards Alcohol Use 
The majority of participants (66.67%, CI: 61.11 - 71.90) responded that they thought that 
their fellow workers used alcohol. More of the respondents (32.89%, 95% CI: 26.8 - 
39.4) indicated that they thought that their fellow workers used alcohol after work and off 
duty; Some (28.51%, 95% CI: 22.7 - 34.8) thought that their fellow workers used alcohol 
on special occasions or when happy (7.02%, 95% CI: 4.1 - 11.1) responded that they 
thought their fellow workers used alcohol before or during work. About half the number 
of participants in this study (50.5%, 95% CI: 44.7 - 56.3) felt that their jobs were 
dangerous when compared to the rest (41.81%, 95% CI: 36.2 - 47.6) who felt that their 
job was not dangerous.  On the other hand, a vast majority of the participants felt that 
alcohol use could lead to accidents on the mine (98.06%, 95% CI: 95.84 - 99.29).  This 
represents a good insight amongst the participants of the dangers associated with alcohol 
use on the mine premises. 
 
On influencing the use of alcohol, it is encouraging to note that a majority (74.5%, 95% 
CI: 69.33 - 79.25) of the participants felt that something could be done to influence the 
 54
 use of alcohol on the mine.  The commonest suggestions on what could be done to 
influence the uses of alcohol amongst miners were: education (18.85%, 95% CI: 14.9 - 
23.2), regular testing of miners (16.67%, 95% CI: 12.9 - 20.9) and providing 
entertainment activities (13.11%, 95% CI: 9.8 - 17).   
 
About half the number of respondents gave various other reasons such as increasing 
religious activities, strict disciplinary measures against offenders and a complete ban on 
the sale of alcoholic beverages on the mine premises as a means to controlling alcohol 
use. 
 
4.1.4 Reasons for Alcohol Use 
Of the reasons cited by the participants when asked why they use alcohol, Most (43.70%, 
95% CI: 35.1 - 52.5) said that they used alcohol for relaxation and enjoyment.  Other 
reasons given were: fun and entertainment (25.19%, 95% CI: 18.1 - 33.4), when at parties 
(25.19%, 95% CI: 18.1 - 33.4), for experimentation (5.19%, 95% CI: 2.1 - 10.4) and 
when at braais (0.74%, 95% CI: 0.02 - 4.1). From the reasons advanced by the 
participants on why they use alcohol, it is clear that alcohol is used as a means of 
providing entertainment perhaps due to lack of or inappropriate recreational activities.  
The knowledge gained from this response will be vital in instituting pro active measures 
that will help in preventing alcohol use becoming a problem on this mine. 
 
 
 
 55
 4.2 CANNABIS 
4.2.1 Prevalence of Cannabis Use 
4.2.1.1 Tested prevalence 
The overall tested prevalence for cannabis use was 2.85% (95% CI: 1.31 - 5.3).  This was 
determined by the use of the Roche ?On Track? qualitative urine assay test.  This result 
closely matched a Swedish survey, which reported a cannabis use rate of 2.7% amongst 
24,000 workers drawn from 25 company employees. 52  Prior to this study, anecdotal 
evidence suggested that the prevalence might be higher than the 2.85% obtained in this 
study even though there was no recorded data on cannabis use amongst mine workers in 
South Africa.  There are occupational and socio-cultural differences between the 
population studied in Sweden and the study participants in South Africa.  Hence the 
estimate from this study might actually be an underestimation of the true prevalence.  
Some of the reasons for this assumption could be: 
 
? The social stigma attached to cannabis use could have prevented some people 
from participating in this study. This particular mining community is closely 
knit and despite assurances of confidentiality, participants might have feared 
that word might get around if they participate and are identified as possible 
users of cannabis. 
? Other forms of ingesting cannabis such as drinking tea or using it as a vegetable 
may lead to false negative results since absorption from the gastrointestinal tract 
is much slower. 
 56
 ? The ?healthy worker effect? where staff who are under the influence of cannabis 
may be absent from work resulting in a lower prevalence amongst the tested 
population. This could not be ascertained as the labour unions during 
negotiations for this study requested that no comparisons be drawn between 
those who participated in this study and those that were not selected or declined 
to participate. 
 
4.2.1.2 Reported prevalence 
The overall reported prevalence of cannabis use was 2.0%.  This was the proportion of 
people who answered in the affirmative when asked if they were still using cannabis. 
A lower reported prevalence rate of 2.0% when compared to the tested prevalence rate of 
2.85% could have resulted from: 
 
? Lack of trust in the anonymous nature of the tests and confidentiality of the 
interviews. 
? Fear of being ?found out? and either disciplined, sacked or socially stigmatised 
or a combination of all of the above. 
 
Under these circumstances, the tested prevalence remains a better reflection of the true 
prevalence of cannabis use in this mine. 
 
 
 
 57
 4.2.2 Practice of Cannabis Use 
4.2.2.1 Ever users of cannabis 
About 6.4% (95% CI: 3.9 - 9.8) of the total number of participants who answered this 
question answered in the affirmative when asked if they had ever used cannabis. The 
disparity between reported and tested prevalence is less than that seen for alcohol. This 
may suggest that there is a less clear understanding among workers that cannabis may not 
be used on duty or of the fact that many of them are not aware that cannabis use can be 
detected in urine by random testing. There may be a higher level of tolerating cannabis 
use in the mine environment as a genuine or even accepted practice necessary for coping 
with the demanding physical jobs miner do. On-the-spot alcohol testing of drivers and 
punishment of offenders is widely practised in South Africa. Together with on-going 
mass media campaign, the illegality of alcohol use when driving or working is widely 
understood. This may however not be the case for cannabis in South Africa.  Some 
religions even suggest that cannabis use is a necessary part of the religious practise of 
their adherents. 
 
The finding in this study that participants who lived with family members were less likely 
to be current users of cannabis (p value = 0.0168) is very important with regards to 
targeted health education. It may also suggest that putting miners in a proper family 
environment instead of hostels may be a necessary adjunct in the control of cannabis use 
among miners. In the same vein, the high prevalence of cannabis use with friends 
(82.33%, 95% CI: 56.5 - 96.2) reported in this study highlights its use as a social form of 
entertainment. This is felt to be a socio-cultural norm with this group of people. Hence 
 58
 peer group education may play an important role in spreading messages aimed at the 
control of cannabis use on the mines. 
 
4.2.2.2  Knowledge of and attitudes towards cannabis use 
A majority of the participants in this study (46.84%, 95% CI: 41.2 - 52.3) reported not to 
know if their fellow workers used cannabis, 37.97% (95% CI: 32.60 - 43.58) reported 
that they thought their fellow workers do not use cannabis and 15.19% (95% CI: 11.42 - 
19.63) of the participants reported that their fellow workers used cannabis.  The level of 
reported knowledge of cannabis use amongst mine workers in this mine is considered to 
be high   considering that 2.85% of the workforce tested for cannabis was positive.  The 
reason for this high level of knowledge of cannabis use amongst fellow workers is 
unknown. It might be as a result of a suspicion amongst miners in this particular mine 
that during competitions, miners take stimulants to enable them to work harder so that 
their groups come out tops. (Personal communication with a few miners during data 
collection). 
 
Some (46.81%, 95% CI: 32.1 - 61.9) of the participants who thought that their fellow 
workers used cannabis thought that they did so to keep busy and reduce boredom.  Other 
reasons advanced include helping avoid stressful situations at the workplace and for fun; 
25.53% (95% CI: 13.9 - 40.4) of the participants who felt that their fellow workers use 
cannabis think that they use it to relax and to forget their problems.  These responses 
indicate that stress in the workplace as well as lack of adequate recreational facilities 
might be an important determinant for cannabis use in this mine.  Therefore, reducing 
 59
 workplace stress as well as the provision of recreational facilities might be important 
targets if the situation is to be ameliorated. 
 
Of the participants who reported that they were taking cannabis, 72.22% (95% CI: 46.5 - 
90.2) started taking cannabis before starting work on the mines whilst 27.78% (95% CI: 
9.7 - 53.4) started taking cannabis after starting work on the mines.  Even though the 
proportion of people who started taking cannabis after commencing work on the mines is 
small when compared to those who started taking cannabis before joining the mines, it is 
still important to note that some people start taking cannabis after starting work on the 
mines.  It means that a lot has to be done with regards to health education especially with 
emphasis on the effects of substance abuse amongst miners. 
 
Most of the respondents (56.16%, 95% CI: 44.0 - 67.8) who believed that miners used 
cannabis reported that miners use it specifically when working hard, before and during 
working hours. 58.10% (95% CI: 52.4 - 63.5) of the participants in this study believe that 
working in the mines is dangerous to their health. 
 
An overwhelming majority of the participants (90.49%, 95% CI: 86.6 - 93.5) who took 
part in this study believe that using cannabis can lead to accidents on the mines. Most 
(77.92%, 95% CI: 72.8 - 82.4) of the participants believe that something can be done to 
influence the use of cannabis amongst miners.  Health education (58.9%, 95% CI: 53.1 - 
64.7) and regular testing of miners (9.66%, 95% CI: 6.5 - 13.6) represented the most cited 
means of influencing cannabis use. 
 60
  
The majority (62.10%, 95% CI: 56.5 - 67.5) of the participants were aware of the 
availability of sports and recreational facilities on their mine premises whilst a substantial 
number (28.66%, 95% CI: 23.7 - 34.0) of the participants seemed not to be aware. 
 
A majority (79.30%, % CI: 74.4 - 83.6) of the participants in this study felt that sports 
and recreational facilities could influence the use of alcohol and cannabis on the mines; A 
few (9.87%, 95% CI: 6.8 - 13.7) of the participants seemed not to agree with this whilst a 
similar number (10.83%, 95% CI: 7.6 - 14.8) were uncertain. 
 
 
 
 
 
 
 
 
 61
 CHAPTER 5 
 
5.0 CONCLUSIONS AND RECOMMENDATIONS  
5.1 CONCLUSIONS 
Alcohol misuse and cannabis use are a public health problem on South African diamond 
mines. The total prevalence of alcohol use on this particular mine was 1.4% and that of 
cannabis was 2.8%. This is low when compared to studies done internationally 31 59. The 
reason for this disparity might be because of the different times the tests were carried out 
and also because of the healthy worker effect, which varies from place to place. 
Alcohol was said to be used for relaxation and enjoyment by 43.70% of the respondents, 
fun and entertainment by 25.19%, when at parties by 25.19% and for experimentation by 
5.19%. 
Cannabis was reported to be used to help relieve stress and to cope with the physical 
demands of mining by 56.16%. 
Being a full time employee was found to be positively associated with alcohol use 
(p=0.004). Also, higher levels of education were positively associated with alcohol use 
(p=0.0013). 
Participants in this study felt that alcohol misuse and cannabis use can be controlled on 
the mine by: health education, (18.85% for alcohol and 58.9% for cannabis) and regular 
testing of miners, (16.67% for alcohol and 9.66% for cannabis). 
Recommendations made in this study were suggested by the participants and were aimed 
at managing the ongoing problems as well as preventing the problems in the future.     
 
 62
 5.2         RECOMMENDATIONS  
 
Alcohol and cannabis use among the workers in mine X has a potential to increase. 
The adverse effects of alcohol and cannabis use and the consequences of this on the 
workforce and workplace have been highlighted in the introductory part of this research.  
These consequences on the workforce have led employers to institute various measures to 
combat the problem.  Methods such as random testing, pre-employment tests, post-
 accident tests and tests on workers suspected to be under the influence of alcohol and/or 
cannabis have been employed. Mine X has an existing alcohol and drug policy but this 
policy lacks a preventive aspect. The findings of this study suggests that a more 
comprehensive policy that incorporates aspects of prevention, reducing alcohol and 
cannabis use among mineworkers as well as managing alcohol and cannabis related 
problems need to be in place. 
 
The following recommendations are made with the aim of managing the ongoing 
problem as well as reducing or preventing the problem in the future. 
 
5.2.1  MULTI STAKEHOLDER POLICY FORMULATION ON ALCOHOL 
AND CANNABIS USE 
There should be a uniform policy on the management of substance use on all the mines in 
South Africa.  All the stakeholders, such as mine management, labour unions, 
mineworkers, government representatives and other stakeholders including 
representatives from communities where mines are located should participate in this 
 63
 policy formulation to ensure better buy in. With regards to mine X, this policy should be 
comprehensive and should clearly state amongst other things the following: 
 
? What is considered to be alcohol and substance use on mine X. 
? The effects and consequences of alcohol and substance use on the worker in 
mine X and the workplace i.e. mine X   
? The mode of identifying offenders, rehabilitating them and if need be the 
disciplinary actions to be taken. 
 
The policy should be clearly displayed at each section of the mine and should encourage 
workers with alcohol and substance use problems to volunteer themselves for treatment.  
 
5.2.2 REHABILITATION/SUPPORT PROGRAMMES FOR USERS 
Effort should be made to provide a conducive working environment for the employees of 
this mine. It is important to target the root causes of alcohol and cannabis use and to 
address them. For example most participants who use alcohol started doing so before 
commencing work on this mine, social stressors such as financial and family problems 
might have contributed to this. In addressing this, employee assistance programmes with 
counselling services should be linked to such services that exist in the neighbouring 
towns since the workforce will largely be drawn from this population. 
 
Whilst having a policy that stipulates that the financial cost for first offenders be borne by 
the mine management is laudable, this policy is faulty in that it only caters for full time 
 64
 employees. Hence, it should be broadened to cater for all staff members and their 
dependants. It should also make allowance for second offenders. First offenders should 
be made to make a firm commitment to change. Co-workers can be used as mentors to 
help offenders make the desired change.   
  
5.2.3   HEALTH EDUCATION/PROMOTION ACTIVITIES 
Most miners on this mine spend a major part of their days on the mines, thus the mines 
are a perfect setting to institute health promotion activities, which may instil a positive 
attitude in the workers on the problems of substance use. These activities should be 
carried out in the local languages. 
 
These health promotion activities should aim at: 
? Publicising alcohol and substance use guidelines in mine X. 
? Involving all the workers and should be detailed on what is expected of the 
miners in combating this problem. 
? Health education activities that detail the effects of alcohol and cannabis use on 
the miners with special emphasis on the deleterious effects on the health and 
safety of the miners. 
? Correcting false information that leads to misuse of substances by miners e.g. 
most miners thinking that smoking cannabis gives strength and hence leads to 
increased production.  
 
 
 65
 5.2.4 DISCIPLINARY ACTION 
Clearly spelt out disciplinary measures against people who fail to comply with policy and 
rehabilitation programmes should be agreed upon and disseminated to all the workers on 
mine X 
The following should be considered in this regard: 
 
? First offenders and volunteers should be offered all assistance by the 
management and staff. 
? Graded disciplinary action should be taken against subsequent offenders and 
habitual users. 
? Graded disciplinary measures could include for e.g. disciplinary hearings, 
suspension for a short period of time, suspension for a longer period of time and 
dismissal in this order. 
? Severe punitive measures like dismissal should only be considered as a last 
resort and it should only be resorted to when all else has failed. 
 
5.2.4.1   Sports and Recreation Activities 
Sports and recreation facilities exist on many mines.  The knowledge of the existence of 
these facilities is what is lacking, as was found out in this study.  Moreover, alcohol and 
cannabis were reported to be used mainly as a means of creating social activity (for fun 
and alleviation of boredom).  Thus: 
 66
 ? Adequate provision of sports and recreation facilities, which recognises the 
specific sports facilities and recreational activities favoured by the miners, 
should be set up. 
? Proper and regular utilisation of these facilities through innovative ways like 
organising intra and inter mine competitions with prizes 
? Family recreational facilities as well as family social events should be organised 
regularly by the mine management. 
? The promotion of the family as the nucleus of social stability should be 
undertaken.  Since this study showed that people living with family members 
were less likely to use cannabis, all effort should be made for miners to live 
with their family members.  Where this is not possible, regular visits of family 
members should be encouraged. 
 
 
 
 
 
 
 
 
 
 
 
 67
 6.0 REFERENCES 
 
1. Gumede V. Alcohol use and abuse in South Africa.  A socio medical problem.  
In Parry CDH, Bennetts AL.  Alcohol policy and public health in South Africa.  
Cape Town: Oxford University Press, 1998: 3-5. 
2. London L. The ?dop? system, alcohol abuse and social control amongst farm 
workers in South Africa: A public health challenge.  Social Science and 
Medicine 1999; 48: 1407-1414. 
3. Kew G.  A descriptive study of alcohol consumption patterns on a South 
African gold mine.  Urbanisation and Health Newsletter 1994; 21: 3-19. 
4. Cooper DB.  Assessment and monitoring of alcohol related problems.  British 
Journal of Nursing 1995; 4(14): 804-8. 
5. Bhana A, Parry CDH, Myers B, Pludderman A, Morojele NK, Flisher AJ.  The 
South African Community Epidemiology Network on Drug Use (SACENDU) 
project phases 1 to 8 (July 1996 ? December 2000):  Cannabis and Mandrax. 
6. United Nations Office for Drug Control and Crime Prevention.  World Drug 
Report 2000.  United Nations Office of Drug Control and Crime Prevention 
2000: 22-36. 
7. Jacobson M.  An alcohol agenda for the 90?s.  EAP Digest 1990, 10)6: 24-26. 
8. Allegne BC, Stuart P, Copes R.  Alcohol and other drug use in occupational 
fatality.  Journal of Occupational Medicine 1991 Apr; 33(4): 494-500. 
9. Wilcocks L. Exposure and extent of alcohol problem in industry (unpublished).  
SANCA, Johannesburg, South Africa  
 68
 10. Ronelle S.  Alcohol and drug use in the workplace ? implications of the 
amended Labour Relations Act.  Review.  Occupational Health SA 1996 
July/August. 
11. McDonald A, Duncan ND, Mitchell DIG.  Alcohol, Cannabis and cocaine usage 
in patients with trauma injuries.  West Indian Medical Journal 1999; 48(4): 200-
 202. 
12. Peden M, Knottenbelt JD, Van der Spuy JN, Oodit R, Scholtz HJ, Stokol JM.  
Injured pedestrians in Cape Town ? the role of alcohol.  SAMJ 1996; 86(9): 
1103-1105. 
13. Van der Spuy JW.  Alcohol related trauma.  VMO.CME 1991; 9(7): 859-868. 
14. Yach D, Joubert G.  Determinants and consequences of alcohol abuse and 
cigarette consumption in Mamre.  SAMJ 1988; 74: 348-351. 
15. Miller R, van Resnburg LCJ.  Alcohol levels in trauma victims.  SAMJ 1986; 
70: 592-593. 
16. Gillis LS.  The mortality rate and causes of death of treated chronic alcoholics.  
SAMJ 1969; 43: 230-232. 
17. Van der Linde, JA.  The cost of alcohol abuse at a Free State mine.  
Occupational Health South Africa 1997; 3: 9-14. 
18. Bush DM, Autry JH.  Substance abuse in the workplace: Epidemiology, effects 
and industry response.  Journal of Occupational Med 2002; 17: 13-25. 
19. Regland DR, Ames GM.   Current developments in the study of stress and 
alcohol consumption.  Alcohol: Clinical and Experimental Research 1996; 
20(8): 51A ? 53A. 
 69
 20. Martin JK.  Jobs, occupations and patterns of alcohol consumption; a review of 
literature.  In Roma PM 9ed).  Alcohol problems in the workplace 1990; New 
York: Quorum Books. 
21. Trice HM, Roman RM.  Spirits and demons at work (2nd ed).  Ithaca, New 
York: Cornell University Press. 
22. Moore S, Grunber L, Greenberg E.  The relationship between alcohol problems 
and well being, work attitudes and performance: Are they monotonic?  Journal 
of Substance Abuse 2000; 11(2): 183-204. 
23. Greenberg ES, Grunberg L.  Work alienation and problem alcohol behaviour.  
Journal of Health and Social Sciences Behaviour 1995; 36: 83-102. 
24. Pratt AE, Tucker MM.  Approaches to the alcohol problem in the workplace.  
Journal of Alcohol and Alcoholism 1989; 24: 453-463. 
25. Parry CDH, Bennets AL. Alcohol policy and public health in South Africa.  
Cape Town: Oxford University Press, 1998. 
26. Steptoe A.  A longitudinal study of workload and variations in psychological 
well being, cortisol, smoking and alcohol consumption.  Ann Behav Med 29(2): 
84-91. 
27. Bennett JB, Lehman WE.  Workplace drinking climate, stress and problem 
indicators: assessing the influence of teamwork (group cohesion).  Journal of 
Studies on Alcohol 59(5): 608-18. 
28. Adrian M. Substance use and multiculturalism.  Journal of Substance Use and 
Misuse 1996; 31: 1459-1501. 
 70
 29. Sator R.  Alcohol and drug abuse in the workplace ? implications of the 
amended Labour Relations Act.  Journal of Occupational Health SA 1996; 2(4): 
14-15. 
30. Rice DP, Kelman S, Miller LS.  Estimates of economic costs of alcohol and 
drug abuse and mental illness 1985-1988.  Public Health Reports 1991; 106: 
280-292. 
31. Sellman DJ, Ariel WG.  Public knowledge and attitudes towards the use of 
alcohol and drinking guidelines.  New Zealand Med J 1996; 109: 337-9. 
32. San Jose B, Van der Mheen H, Van Oers JAM, Mackenbach JP, Garrestem 
HFL.  Adverse working condition and alcohol use in men and women.  
Alcoholism, Clinical and Experimental Research 2000; 24: 8. 
33. Stallones L, Kraus JF.  The occurrence and epidemiologic features of alcohol 
related occupational injuries.  Journal of Addiction 1993; 88: 945-951. 
34. Centers for Disease Control.  National Center for Environmental Health and 
Injury Control. . Proceedings of the 3rd National Injury Control Conference.  
Occupational Injury Prevention 1991 April 22-25; Denver, Colorado, USA. 
35. Konkanen R, Visuri T.  Blood alcohol levels in a series of injured patients with 
special reference to accident and type of injury.  Annals of Surgery and 
Gynaecology 1976; 65: 692-697. 
36. Baker S, Samkoff J, Fisher R, Van Buren C.  Fatal occupational injuries.  
Journal of American Medical Association 1982; 64: 692-697. 
 71
 37. Parkinson D, Gaws W, Perpen J, Elliot S.  Traumatic workplace death in 
Allegheny Country, Pennsylvania 1983-1984.  Journal of Occupational 
Medicine 1986; 28: 100-102. 
38. Sniezek J, Horiagon T.  Medical examiner reported fatal occupational injuries in 
North Carolina 1978-1984.  American Journal of Industrial Medicine 1989; 15: 
669-678. 
39. Arner O. The role of alcohol in fatal accidents amongst seamen.  British Journal 
of Addiction 1973; 68: 185-189. 
40. Schinder EO, Ruder AM.  Epidemiology of coca and alcohol use among high 
altitude miners in Argentina.  American Journal of Industrial Medicine 1989; 
15(5): 579-87. 
41. Parry CDH.  Alcohol misuse: a significant public health threat facing the new 
South Africa.  Paper presented at the Mental Health in Africa Conference, 
Harare, April 1995. 
42. Rocha-Silva L, De Miranda S, Erasmus R. Alcohol, tobacco and other drug use 
among black youth. Human Sciences Research Council. South Africa. 1996. 
43. Mbombo L, Parry CDH, Dingain N.  Alcohol use in an informal settlement 
(Noordhoek).  Urbanisation and Health Newsletter 1994; 21: 3-19. 
44. Logie P, Morley JE, Bensusan AD.  The ?dagga? smoker: A survey.  SAM J 
1972; 46: 1400-1403. 
45. Chait LD, Pierri J.  Effects of smoked marijuana on human performance.  A 
critical review.  In Murphy A, Bartke J. Editors.  Marijuana/canabinoids: 
Neurobiology and Neurophysiology 1992; CRC: Boca Raton. 
 72
 46. Zwerling C.  Current practice and experience in drug and alcohol testing in the 
workplace.  Bulletin on Narcotics 1996; XLV (2): 154-196. 
47. Nahas G. Hashish in Egypt.  Bulletin of the New York Academy of Medicine       
1985; 61: 5. 
48. Hlal W.  The public health implications of cannabis use.  Australian and New 
Zealand Journal of Public Health 1995; 19: 235-242. 
49. European Workplace Drug testing in Forensic Science International 2001 15th 
Sept; 121: 1-2.<http://www.sciencedirect.com > 
50. Constant P.  Workplace drug testing: the Esso/Exxon 10 years experience. 
Presentation given at the Second European Symposium on Workplace Drug 
testing Rimini 2000. 
51. Hermansson U,Beck O.,Yngen M, Hjemdahl P.  Experiences of  Drug testing in 
Swedish Workplaces. Presentation given at the first European symposium on 
Drug testing at the Workplace, Huddingae, 1998. 
52. Posselt J. A drug screening programme in companies of the German chemical 
Industry. Presentation given at the second European Symposium on Workplace 
Drug testing, Rimini, 2000. 
53. Lew JR, Cooper SP. Alcohol, other drugs and fatal work related injuries. 
Journal of Occupational Medicine 1998; 31: 23-28.                                                                             
54. Grove PB, Arten E, Bethel JP, Cook D, Briscoll AM, Goepp PH.  Editors.  
Webster?s New International Dictionary.  Cologne: Koneman. 1993. 
55. Katz DL. Epidemiology Biostatistics and Preventive Medicine Review. 
Philadelphia: WB Saunders.1997: 227. 
 73
 56. Soanes C, Spooner A, Hawker S.  Editors.  Oxford Dictionary, Thesaurus and 
Work Power Guide.  New York: Oxford University Press: 2001. 
57. Ewing JA. Detecting alcoholism, the CAGE questionnaire. Journal of the 
American Medical Association 1984; 252: 1905-7. 
58. National Road Traffic Act 93 of 1996.  Republic of South Africa.  Government 
Printers: Pretoria, 1996. 
59. Buchanan DJ.  Studies on blood alcohol in the workers of a Zambian copper 
mine.  Alcohol Alcohol 1988; 23(3): 239-42. 
 
 74