Safety and Health at Work 10 (2019) 151e165

Contents lists available at ScienceDirect

Safety and Health at Work

journal homepage: www.e-shaw.org

Original Article

Analysis of Injuries in the Ghanaian Mining Industry and Priority Areas

for Research
Eric Stemn 1, 2
1
Mineral Industry Safety and Health Centre, Sustainable Minerals Institute, The University of Queensland, Brisbane 4072, QLD, Australia
2
Environmental and Safety Engineering Department, University of Mines and Technology, Tarkwa, Ghana

a r t i c l e i n f o a b s t r a c t

Article history: Background: Despite improvements in safety performance, the number and severity of mining-related
Received 10 April 2018 injuries remain high and unacceptable, indicating that further reduction can be achieved. This study
Received in revised form examines occupational accident statistics of the Ghanaian mining industry and identifies priority areas,
11 September 2018
warranting intervention measures and further investigations.
Accepted 13 September 2018
Available online 21 September 2018
Methods: A total of 202 fatal and nonfatal injury reports over a 10-year period were obtained from five
mines and the Inspectorate Division of the Minerals Commission of Ghana, and they were analyzed.
Results: Results of the analyses show that the involvement of mining equipment, the task being per-
Keywords:
Accident formed, the injury type, and the mechanism of injury remain as priorities. For instance, mining equip-
Injury analysis ment was associated with 85% of all injuries and 90% of all fatalities, with mobile equipment, component/
Mining part, and hand tools being the leading equipment types. In addition, mechanics/repairmen, truck op-
Mining equipment erators, and laborers were the most affected ones, and the most dangerous activities included mainte-
nance, operating mobile equipment, and clean up/clearing.
Conclusion: Results of this analysis will enable authorities of mines to develop targeted interventions to
improve their safety performance. To improve the safety of the mines, further research and prevention
efforts are recommended.
Ó 2018 Occupational Safety and Health Research Institute, Published by Elsevier Korea LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction fatalities [7,8]. In the United States, Marse and Layne’s analysis of
16-year fatality data indicated that the industry had the highest
The mining industry remains a vital contributor to the global fatality rate of 30.3 per 100,000 workers. Similarly, in Australia, Safe
economy. The products of mining have significantly improved hu- Work Australia [9] acknowledged that despite the reduction in fa-
man livelihood and are the bedrock of several other industries tality rate from 12.4 in 2003 to 4.4 in 2015, the number of death in
including aviation, power generation, electronics, cement produc- the industry still remains high at a yearly average of 9.
tion, steel production, agriculture, and even medicine [1]. In Ghana, As a result of the severity and frequency of mining-related in-
several sources indicate that the mining industry was the most cidents, they have been regarded as the costliest [10]. The cost is
important contributor to the nation’s economy in terms of usually categorized into two: direct cost and indirect cost [11]. The
employment, direct and indirect revenues, exports, and in- direct cost usually consists of cash payment under national laws
vestments [2e5] until the recent discovery and exploitation of oil in and regulation, in the form of compensation and other benefits. The
commercial quantities. However, despite these positive contribu- indirect cost that is usually greater consists of several things, such
tions, the industry is typically associated with hazardous working as the cost of time for treating an injured person, cost of lost time of
conditions, which affect the health and safety of workers. The in- an injured person, cost due to damage to property and equipment,
dustry has been regarded as one of the safety-critical domains with and even cost of investigating the accident. The direct cost has been
dangerous operations and an environment in which the operator is seen to significantly impact the economy of countries. For instance,
exposed to a plethora of risks and hazards [6]. The International the International Labour Organization estimates that mining deaths
Labour Organization estimates that mining employs around 1% of cost the global economy a staggering amount of $240 billion [12]. It
the global workforce; it accounts for 8% of the global work-related was estimated that mining-related accidents were costing the

E-mail addresses: e.stemn@uq.edu.au; estemn@umat.edu.gh (E. Stemn).

2093-7911/$ e see front matter Ó 2018 Occupational Safety and Health Research Institute, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
https://doi.org/10.1016/j.shaw.2018.09.001
152 Saf Health Work 2019;10:151e165

European Union 15-member states $527 million [13]. Leigh et al recorded in 2011 and 2012 for fatality and 2010e2012 for serious
[14] ranked the lignite and bituminous mining as the second in the injury. Although the cause of this rise has not been studied, thereby
US in terms of the average cost per employee for both fatal and making it difficult to offer a research-based explanation, it is
nonfatal injuries. possible that the increase in employment around those years could
Despite these disturbing statistics, significant safety improve- have contributed to this development because those high values
ments have been made in the mining industry over the last century. coincide with the boom period. It appears that the number and
Recent scholarly works indicate a continuous decline in the fre- frequency rate of incidents mimic the growth of the industry
quency and severity of mining accidents [1,10,15e22]. In the US, the because the least values were recorded during the downturn years
number of fatalities has declined from 164 in 1984 to 25 in 2017 with the highest numbers recorded during the prosperous periods.
[10]. In Australia, the fatality rate has reduced from 12.4 in 2004 to The figure shows that there has been a decrease in the frequency
4.4 in 2015 [23]. Similar reductions have been made in other rate of serious injury against that of fatality. The difference between
countries including India [24], South Africa [25], Spain [26], and the highest and the lowest serious injury rate is 1.54, representing
Poland [27]. Despite these records of improvements, the frequency 86% reduction, whereas that of fatality is 0.1, representing 75%
and severity of mining accident are still undesirable. As Kecojevic reduction. Reduction in serious injury rate is more than 10% better
et al [10] acknowledged, further progress can be made through a than that of fatality. A similar observation is made when the
synergy of traditional, fundamental, and innovative interventions. number of fatalities and the number of serious injuries are
However, such improvement strategies should target specific areas, compared. Thus, a careful examination of the figure indicates that
and those areas can be determined when past accidents are more people are fatally injured than involved in nonfatal injuries.
analyzed thoroughly. This study was therefore undertaken to In addition, comparing the fatality frequency rate of Ghana with
examine and more thoroughly characterize mining injuries in that of other major mining countries indicates that Ghana’s rate is
Ghana and perform descriptive analyses of injuries that had relatively higher than that of the other countries (Fig. 2). For
occurred over the past 10 years, from 2008 to 2017. Investigation instance, the 10-year fatality frequency rate of Ghana (0.0711) far
reports, for fatal and nonfatal injuries, obtained from underground exceeds that of Australia (0.0279) and the USA (0.0569). For half of
and surface mines were analyzed together. the 10-year period (2004e2013), the fatality rate of Ghana was
consistently higher than that of Australia and the USA. It can be
2. Brief overview of Ghana’s mine safety statistics observed from the figure that Ghana’s minimum frequency rate
(0.0353) exceeds that of Australia (0.0131) and that its maximum
In Ghana, by law, mining incident/accident reports are to be frequency rate (0.1471) exceeds that of both Australia (0.0556) and
submitted to the Inspectorate Division of the Minerals Commission the USA (0.085). Furthermore, correlation analysis indicates that
[28]. Thus, there should have been the existence of a database that there is a stronger positive correlation between the number of fa-
catalogs all reported incidents occurring within the industry, and talities and the hours worked in Ghana (r ¼ 0.607) and the USA
such a database should be readily accessible to the public. However, (r ¼ 0.609) than in Australia (r ¼ 0.056). This indicates that more
accessibility to such data still remains a challenge and has been people are fatally injured as the hours worked increase in Ghana
identified as the number one problem that hinders research in this and the USA than in Australia. The high work-related injury rate in
area [29]. It is therefore not surprising that accidents and injuries in the Ghanaian mining industry suggests that research is needed to
Ghana’s mining industry have been sparsely studied, although the offer an overall understanding and identify priority issues.
industry was long identified as a major safety-critical domain
[6,30]. Information on major topics such as causes, types, effects, 3. Materials and methods
and consequences of accidents is difficult to find. Most of the
research studies carried out so far consider only individual mines 3.1. Data source
[29,31,32]. Hence, studies that consider the entire industry across
different commodities and mine type using rich data will be a Currently, there are 12 active large-scale mines belonging to
significant contribution for improving health and safety. nine different companies that are members of the Ghana Chamber
Fig. 1 shows the number and frequency rate of fatalities and of Mines. Out of these 12 mines, five gold mines provided the data
serious injuries from 2004 to 2015. It depicts an average annual that were used for this research. The mines were selected based on
fatality of five and serious injury of 51, with the highest figures the following criteria:

Fig. 1. Injury statistics of the Ghanaian mining industry from 2004 to 2015.
 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 153

Fig. 2. Fatality frequency rate of the mining industry of Ghana, Australia, and the USA.

 being a member of the Ghana Chamber of mines and reporting analysis was specific to injuries, first was to select incidents which
incidents/accident data to the Inspectorate Division of the had resulted in an injury. Second, the analysis focused on injuries
Minerals Commission of Ghana that occurred within the mining lease/concession of the mines.
 operating for at least 10 continuous years Thus, reports related to accidents occurring outside the mining
 having instituted an internationally recognized safety man- lease/concession were rejected. The study also had a focus on
agement system such as OHSAS 18001 work-related injury. However, some reports had nothing to do
with work-related activities (for example, drowning of a local
After an initial invitation letter was sent to 10 mines with resident in a pond on the mining concession); such reports were
several follow-ups, five consented to participate in the research and also discarded from the analysis. Fig. 3 shows the screening and
gave the researcher access to their incidents/accidents data. A total selection process.
of 650 investigation reports from 2008 to 2017 were obtained from
the mines. The reports covered all categories of incidents investi- 3.3. Data classification
gated by the mine sites such as near miss, property damage, and
injuries of all types. There were differences in the details of the After selecting the relevant reports, the next stage was to clas-
reports because of the differences in investigation and reporting sify the reports by a coding process. Based on the research ques-
techniques used by different companies. However, all the reports tions, the content of the selected reports, and relevant literature
contained information about the victim(s), the task being per- [1,10,15], it was observed that information could be extracted into
formed, the incident itself, and the type of equipment if there was 17 user-centered classes grouped under the major topics of the
an involvement of mining equipment. injured, task/activity, equipment, and injury, as shown in Table 1.
For each class, several codes were identified based on the Mine
3.2. Data screening to select relevant reports Safety and Health Administration (MSHA) and the Department of
Natural Resources and Mines (DNRM), Queensland, Australia, ac-
A screening process was used to select samples of the inves- cident classification code. The codes were repeatedly refined by
tigation reports for further analysis. Because the focus of the constant comparison as the reports were read [33,34]. A flowchart

Fig. 3. Process of screening and selection of reports for onward classification.

154 Saf Health Work 2019;10:151e165

Table 1
Topics and classes identified in the reports for classification and coding

Topics

Injured Task/activity Equipment Injury

Age Worksite location Involvement of equipment Accident time
Experience Shift start time Equipment type Day of week
Employment type Hours into work Degree of injury
Job title Task being performed Accident type class
Body part affected
Injury type
Injury mechanism

detailing the classification and coding process including iterative engaged in those tasks have an increased risk of being injured [39,40].
changes to codes is shown in Fig. 4. Some have also found differences in the rates of injury among
different job titles and have identified specific occupations that merit
3.4. Data analysis attention [15,41,42]. Coding the task being performed may yield some
useful information, especially when combined with other classes. For
A single variable analysis was carried out on the individual example, it might allow a safety officer to determine the task mostly
classes, without any form of variable combination. This was carried associated with an injury affecting a particular body part.
out using basic descriptive statistics to identify broad patterns and
trends in accordance with the research questions. 3.4.3. Does the involvement of mining equipment in injuries remain
a priority? If so, which types of equipment should be prioritized?
3.4.1. Do the characteristics of the injured person show any trend The involvement of mining equipment in injuries has been
and deserve further examination? If so, which specific studied in detail [1,10,20,24]. Some authors have identified specific
characteristics should be considered? mobile mining equipment as high priorities [1,10]. Because the
Within the literature, there is divergent opinion whether certain injury investigation reports contained information on equipment, it
characteristics of workers are determinants of their injury experi- was coherent to determine if the situation in Ghana follows the
ence. Some authors have found associations between the severity global trend or had some deviations. This could provide informa-
of an injury and the age and experience of the injured [10,35,36], tion on the types of equipment that should be prioritized for
whereas others have said otherwise [18,37]. Analyzing the char- further studies and prevention efforts.
acteristics of the injured was necessary because the industry in
recent times has experienced significant growth with the expan- 3.4.4. Do some characteristics of the injury deserve priority
sion of existing operations and the commencement of new opera- investigation? If so, which specific issues remain significant?
tions. This expansion has led to high labor turnover, where old and Some scholars have found differences in the causal factors of
experienced workers leave old mines and move to new ones and fatal and nonfatal injuries [19,43e45]. Others have identified that
are replaced with young and less-experienced recruits [38]. By certain body parts, injury mechanism, and accident type remain a
examining injury experiences and workers’ characteristics, in- priority [15,16,22,42]. By analyzing characteristics of the injury,
terventions can be implemented to address vulnerable groups. broad trends could be identified for subsequent investigation. For
instance, priority body parts, injury mechanism, degree and nature
3.4.2. Does the task/activity being performed remain a priority of injury, and accident type associated with specific injuries can be
issue? If so, which specific areas ought to be considered? determined. Importantly, characteristics of the injury could yield
Some works indicate that within the mining industry, certain task more useful information and identify specific priority areas when
and occupation are more dangerous than others, and workers combined with other variables.

Fig. 4. A flowchart of classification and coding of selected injury reports.

 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 155

4. Results years). Fig. 8 shows that more contractors (29.8%) and underground
employees (18.3%) were involved in fatal injuries than operators
4.1. Severity of selected injury reports (10.3%) and surface employees (13.4%).
The breakdown of the job title of the injured at the time of the
After screening all 650 incident investigation reports, 202 re- incident is shown in Fig. 9. Mechanics (22.5%), welders (10.6%), and
ports of different degrees of injuries were selected for subsequent truck operators (18.3%) make up over half of the surface mining
classification and coding. The selected reports covered a period of injuries, whereas supervisors (15%), drillers (26.7), and blasters
2008e2017. Two-thirds of the selected reports were from 2012 to (11.7%) make the majority of the injuries logged from underground
2014, with 2008 and 2017 having the lowest number of two reports locations. The figure shows several differences, such as injury to
each. The reports included 30 fatalities and 172 nonfatal injuries as underground drillers (26.7%) and surface drillers (1.4%). Similarly, a
shown in Fig. 5. mechanic/repairman shows a significant difference, with 22.5% of
all surface injuries against 10% of all underground injuries. Overall,
4.2. Do the characteristics of the injured person show any trend the top five most affected job titles were mechanics/repairmen
that deserves further examination? If so, which specific (18.8%); truck operators (13.9%); and drillers, laborers, and super-
characteristics should be considered? visors (8.9%). This figure generally identifies specific job titles that
merit consideration for prioritizing research and prevention
There were more operators (76.7%) and surface operations efforts.
workers (70.3%) than contractors (23.3%) and underground workers
(29.7%). The age and experience (Table 2) of the injured miners show 4.3. Does the task/activity being performed remain a priority issue?
differences with respect to surface/underground locations and If so, which specific issues ought to be considered?
operator/contractor workers, although the differences are marginal.
Most injured operators aged 38e47 years (34.8%), whereas most Fig. 10 shows details of the task/activity being performed and
injured contractors aged 28e37 years (36.2%). The most affected age the location of the task at the time of the incident. For surface
group for both surface and underground operations was 38e47 operations, the top five tasks were machine maintenance/repair
years (35.9% and 23.3%, respectively). Similarly, for the whole cohort, (19.7%), operating mobile equipment (16.9%), cleaning up/clearing
the most affected age group was 38e47 years (32.2%). Generally, (10.6%), lifting/lowering by hand (7.7), and directing moving
it can be observed that the top affected age group is 38e47 years, equipment (7%), altogether making up 61.9% of all surface injuries.
followed by 28e37 years, with 58 years being the least. Hence, the The top five tasks for underground locations were drilling (20%),
age distribution of the injured miners (see Fig. 6) appears symmet- charging up (10%), walking (8.3%), barring/scaling (5%), and
rical. This is such that the percentage of injured workers increases changing/adjusting (5%), totaling 48.3% of all underground injuries.
with increase in age from 18e27 years until a peak is reached at 38e This is logical as occupations associated with such tasks were
47 years, and then there is a decrease in the percentage of injured related to most of the injuries. The top five tasks in descending
workers with an increase in age from 38e47 years to >58 years. order for the whole cohort were machine maintenance/repair,
More than 30% of the reports had no information on age; this was operating mobile equipment, cleaning up/clearing, drilling, and
more pronounced in underground mines than surface mines and lifting/lowering by hand, which are responsible for 51.5% of all in-
among operators than contractors. juries. Sixty percent of the underground injuries occurred at the
Similar to the age class, most of the reports had no information stope mining area (45%) and the shaft area (15%). For surface op-
on the work experiences of the injured miners. This was mostly erations, the majority of the accidents (54.2%) occurred at work-
associated with underground mines than surface mines and with shops (23.2%), in processing plants (17.6%), and on haul roads
operators than contractors. Information on this would have (13.4%). This gives a broad indication for prioritizing prevention
contributed to understanding the data better. Despite this lack of measures as specific tasks and locations related to most of the in-
information, the data show that workers with less experience (9 juries have been determined.
years) were involved in injuries more than their experienced Details of hours of work before the accident occurred and the
counterparts (>10 years). Fig. 7 indicates that the modal age group shift are shown in Fig. 11. The trend is somehow similar for both
for both fatal and nonfatal injuries was 38e47 years. Similarly, the surface and underground mines. Incidents predominated in the
modal experienced group for both fatal and nonfatal injuries was initial 8 hours of a workday for both surface (73%) and underground
>9 years. In addition, old workers (38 years) were often involved (85%) mines. However, a large portion of surface mines compared
in both fatal and nonfatal injuries than young workers (<38 years). with underground mines experienced injury after 8 hours of work.
Similarly, less-experienced miners (<5 years) were often involved A further breakdown of the first 8 hours of a workday shows that
in both fatal and nonfatal injuries than experienced workers (20 most injuries occurred after 4 hours but less than 8 hours into a

Fig. 5. Injury severity of selected reports.

156 Saf Health Work 2019;10:151e165

Table 2
Age and experience of injured miners by employment and mine type

Class Operators (n,%; 155, 76.7) Contractors (n,%; 47, 23.3) Surface (n,%; 142, 70.3) Underground (n,%; 60, 29.7) Overall (n,%; 202, 100)
Age (y)
18e27 8 (5.2) 3 (6.4) 10 (7) 1 (1.7) 11 (5.4)
28e37 28 (18.1) 17 (36.2) 41 (28.9) 4 (6.7) 45 (22.3)
38e47 54 (34.8) 11 (23.4) 51 (35.9) 14 (23.3) 65 (32.2)
48e57 10 (6.5) 2 (4.3) 7 (4.9) 5 (8.3) 12 (5.9)
58 3 (1.9) 3 (6.4) 5 (3.5) 1 (1.7) 6 (3)
Unknown 52 (33.5) 11 (23.4) 28 (19.7) 35 (58.3) 63 (31.2)
Total mining experience (y)
<1 4 (2.6) 3 (6.4) 7 (4.9) 0 (0) 7 (3.5)
1e4 21 (13.5) 13 (27.7) 34 (23.9) 0 (0) 34 (16.8)
5e9 38 (24.5) 10 (21.3) 45 (31.7) 3 (5) 48 (23.8)
10e14 17 (11) 1 (2.1) 16 (11.3) 2 (3.3) 18 (8.9)
15e19 1 (0.6) 4 (8.5) 2 (1.4) 3 (5) 5 (2.5)
20e24 3 (1.9) 2 (4.3) 1 (0.7) 4 (6.7) 5 (2.5)
25 6 (3.9) (0) 4 (2.8) 2 (3.3) 6 (3)
Unknown 65 (41.9) 14 (29.8) 33 (23.2) 46 (76.7) 79 (39.1)
Current mine experience (y)
<1 10 (6.5) 19 (40.4) 24 (16.9) 5 (8.3) 29 (14.4)
1e4 38 (24.5) 14 (29.8) 46 (32.4) 6 (10) 52 (25.7)
5e9 39 (25.2) 1 (2.1) 36 (25.4) 4 (6.7) 40 (19.8)
10e14 6 (3.9) 2 (4.3) 5 (3.5) 3 (5) 8 (4)
15e19 2 (1.3) 3 (6.4) 2 (1.4) 3 (5) 5 (2.5)
20e24 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
25 3 (1.9) 0 (0) 2 (1.4) 1 (1.7) 3 (1.5)
Unknown 57 (36.8) 8 (17) 27 (19) 38 (63.3) 65 (32.2)
Total experience in job title (y)
<1 7 (4.5) 5 (10.6) 12 (8.5) 0 (0) 12 (5.9)
1e4 25 (16.1) 15 (31.9) 40 (28.2) 0 (0) 40 (19.8)
5e9 27 (17.4) 6 (12.8) 29 (20.4) 4 (6.7) 33 (16.3)
10e14 21 (13.5) 1 (2.1) 21 (14.8) 1 (1.7) 22 (10.9)
15e19 1 (0.6) 2 (4.3) (0) 3 (5) 3 (1.5)
20e24 7 (4.5) 2 (4.3) 3 (2.1) 6 (10) 9 (4.5)
25 3 (1.9) 3 (6.4) 5 (3.5) 1 (1.7) 6 (3)
Unknown 64 (41.3) 13 (27.7) 32 (22.5) 45 (75) 77 (38.1)

Fig. 6. Age distribution of injured miners based on employment and mine type.

workday. Most injuries were recorded during the morning shift for 4.4. Does the involvement of mining equipment in injuries remain a
both surface (82%) and underground (63%) mines; however, more priority? If so, what types of equipment should be prioritized?
injuries occurred during evening shift in underground mines (35%)
than in surface mines (18%) (Fig. 11). There was no incident during Mining equipment was involved in 96% and 62% of the surface
afternoon shift in surface mines because almost all surface mines in and underground mining injuries, respectively. Of all mine types,
Ghana operate a two-shift system of morning and evening. 86% of the injury reports involved mining equipment of some sort,
 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 157

Fig. 7. Age and mining experience of injured miners by injury category.

Fig. 8. Injury category based on employment and mine type.

that is, 90% and 86% of fatal and nonfatal injuries, respectively, underground mines, the injuries lead to a minimum of days away
involved equipment. This identification of a large proportion of from work, whereas for surface mines, the minimum outcome of
accidents involving mine machinery is consistent with other works the injuries was a restricted work activity. The percentage of per-
[1,10,20]. A breakdown of the specific equipment types as per the manent disability is identical for both underground and surface
DNRM classification of equipment/tools is shown in Fig. 12. Among mines (8%), whereas there were more temporary disabilities in
the equipment subclasses, haul trucks (16.8%), components/parts surface (14%) than in underground mines (8%). Injuries resulting in
(13.9%), nonpower hand tools (7.3%), and light vehicles (5.8%) made days away from work showed a large difference, with 84% of all
up a majority (43.8%) of the surface mining injuries, whereas underground mining injuries versus 40% of surface mining injuries.
component/parts (17.9%), drill rigs (12.8%), rock drills/borers Fig. 14 shows broad peaks in the percentage of injury for the
(12.8%), and other earth-moving equipment (10.3%) made up a following times.
majority (53.8%) of the underground injuries. Mobile equipment
such as haul trucks, excavators, graders, and drill rigs were the most  11 AMe12 PM, that is 5e6 hours into the morning shift
identified equipment. The involvement of these equipment sub- (assuming the shift start at 6 AM) for both surface and under-
classes is not surprising because they are prevalent in the mining ground mining.
environment. Although the nature of component/part subclass can  1e2 PM for both surface and underground mining, this is,
make it difficult to focus prevention efforts, specific intervention however, more pronounced in surface mines than in under-
strategies can target the more specific drill rigs, nonpower hand ground mines.
tools, light vehicles, and haul trucks provided. The more specific  9e10 PM and 2e3 AM for underground mines.
equipment subclasses such as haul trucks have been extensively
studied, and several recommendations have been made to improve The day of the week during which the incident occurred is
the safety of truck operators [46e48]. shown in Fig. 15. Thursday (31.7%) remains the peak in under-
ground mines, whereas for surface mines, it is Friday (26.1%). The
4.5. Do the characteristics of the injury deserve priority percentage of incidents for these two days is quite significant, and
investigation? If so, which specific issues remain significant? further investigation is required to identify explanations for these
peaks. Saturday and Tuesday appear to be the safest days for both
Concerning the degree/severity of injury (Fig. 13), there were underground and surface mines. The top three days for surface
more underground mining fatalities (18%) than surface mining fa- mines were Friday (26.1%), Wednesday (15.5%), and
talities (13%). The breakdown of the nonfatal injuries shows that for Monday (14.1%) and accounted for 55.7% of the total surface
158 Saf Health Work 2019;10:151e165

Fig. 9. Job title of the injured at the time of the accident.

Fig. 10. Victim’s activity and location of activity at the time of the accident.

injuries. For underground mines, 65.1% of the incidents occurred on hand tools, slip/fall of the person, and handling of materials were
Thursday (31.7%), Wednesday (16.7%), and Friday (16.7%). the top five accident types. Concerning specific regions of the body
Fig. 16 shows the classification of the injuries based on the that injuries affected, the majority of the injuries for both surface
MSHA classification for the accident type and nature of an and underground locations affected the hand/finger (31% and 15%,
injury and the DNRM classification for the affected body part and respectively) and multiple body parts (16.2% and 25%, respectively).
mechanism of injury. For underground operations, machinery Regarding the injury type, the majority of the surface injuries were
(25%), falling/rolling rock/material (21.7%), and slip/fall of the per- laceration (29.6%), fracture (18.3%), and multiple injuries (17.6%).
son (18.3) were associated with more than half of the injuries. For Similarly, 70% of the underground injuries were laceration (43.3%),
surface operations, power haulage (21.1%), hand tools (19%), and contusion (11.7%), and multiple injuries (15%). Burns (9.9%),
machinery (17.6%) were equally associated with more than half of contusion (9.2%), and traumatic amputation (7%) were also signif-
the injuries. This trend agrees with previous results. For instance, icant in surface mines. The frequent injury mechanisms in surface
the identification of power haulage as a major issue is not sur- mines were being struck by a metallic object (26.1%), being hit by a
prising as earth-moving equipment such as haul trucks was iden- moving object (9.9%), motion of a moving vehicle (9.2%), and
tified as one of the major mining equipment commonly associated trapped between stationary and moving objects (8.5%). Fall/slip/
with injuries. For the whole cohort, machinery, power haulage, trip from height (18.6%), being hit by a moving object (15.3), struck
 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 159

Fig. 11. Shift start time and hours of work just before the accident occurred.

Fig. 12. Equipment-related injuries.

by a rock (13.6), and being hit by a fallen object (10.2%) contributed injuries than middle-aged (38e47 years) and young (<38 years)
to most of the underground injuries. For the whole cohort, struck miners. The middle-aged miners were the most affected group in
by a metallic object (20.9%), being hit by a moving object (11.4%), both fatal and nonfatal injuries. In addition, old miners were mostly
fall/slip/trip on the same level (8.5%), and motion of a moving involved in fatal injuries than in nonfatal injuries. A number of
vehicle (7%) were the top injury mechanisms. studies support these observations. For instance, Phiri [49] found
that young miners were 2% more probable of sustaining an injury
5. Discussions than old miners. In addition, Salminen [50] observed that young
workers had a higher nonfatal injury rate and a lower fatal injury
The findings of this study agree with as well as deviate from rate. On the contrary, Bennet [40] found no relationship between
those of previous studies. These agreements and deviations are injury severity and the age of workers. The results further indicate
discussed in this section. The section also focuses on areas that that less-experienced workers were involved in fatal accidents than
deserve attention for prioritizing research and prevention efforts. in nonfatal accidents. In addition, less-experienced workers were
To identify such areas, some form of variable combination was more involved in fatal injuries than their experienced counterparts.
carried out, particularly focusing on the various classes/categories Similarly, most of the nonfatal injury victims were less experi-
and their contribution to fatal injuries. Details of the identified enced; however, a significant portion (20%) of the fatal accident
priority areas have been listed in Table 3. victims had a total mining experience of more than 20 years. In an
analysis of equipment-related fatalities, Kecojevic et al [10] found
5.1. The injured person that the most affected group was of less-experienced miners, that
is, those with less than 5 years of mining experience. Similarly,
An examination of the age and work experience of the victims Butani [51] reported that the severity of injuries sustained by coal
indicated that old miners (>47 years) were less involved in fatal workers related more to their mining experiences than their age.
160 Saf Health Work 2019;10:151e165

Fig. 13. Degree/severity of injuries by mine type.

Fig. 14. Percentage of injuries by time of the day during which the accidents occurred.

Fig. 15. The day of the week during which the accidents occurred.

On the contrary, Bennett and Passmore [37] observed that the efforts. Concerning the job titles of the injured, mechanics/re-
severity of injury does not relate to the mining experience of the pairmen, truck operators, drillers, supervisors, and laborers were
victims. In this study, it is hard to conclude whether any significant identified as the most affected ones, which is supported by other
relationship exists between the severities of injury a miner sustains studies [15,18,39,49,52]. These occupations remain a priority war-
and his age and experience because no such analysis was carried ranting intervention efforts; however, such efforts should be spe-
out as a large portion (more than one-third) of the reports had no cific to the mine type as there were major differences between
information on the age and mining experience of the accident surface and underground location. The study further showed that
victims. However, broad patterns and trends were observed, and it contractors have an increased risk of being fatally injured than
may be important to focus on such patterns and trends. Consis- operators because contractors were involved in fatal accidents
tently, among different injury severity, mine types, and employ- more than operator workers. Randolph and Boldt [42] observed
ment types, 38e47 years remained the most affected age group and that contractors consistently had a higher rate of being involved in
may merit consideration for prioritizing research and prevention haul truck fatalities than operators. Similarly, analysis of injuries
 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 161

Fig. 16. Surface and underground injury classes: accident type, body part injured, and nature and mechanism of injury.

statistics by Muzaffar et al [52] indicated that contractors had a workers’ interaction with equipment/machinery, especially mobile
higher proportion of fatal injuries than operators. These results equipment, continue to remain as a priority. Therefore, further
support the need to focus on improving contractor safety, especially research is required in this area to better understand the situation,
with the recent increased employment of independent contractors enabling the development of specific and targeted intervention
in the industry [53,54]. that will limit such interaction. It was further observed that most of
the injuries were specific to particular locations, such as the stope
5.2. The task being performed mining area and shaft area for underground operations and pro-
cessing/treatment, haul road, crushing station, and open-cut pit for
Consistent with other works, this study found that machine surface operations. More than 70% of the fatal accidents occurred in
maintenance, operating mobile equipment, drilling, cleaning up/ these locations, indicating that some work locations are more
clearing, and lifting/lowering by hand were the most dangerous dangerous than others. This is strongly supported by the works of
activities. In their study of machine-related injuries in US mines, Muzaffar et al [52] and Coleman et al [15]. To safeguard the safety of
Ruff et al [20] observed that operating machine and maintenance/ workers who work in such hazardous locations, there is an urgent
repairs were the most dangerous activities and accounted for 46% need to implement and improve methods of detecting worker
and 26%, respectively, of all the accidents that they analyzed. proximity to those locations as and to ensure accountability for
Similarly, Muzaffar et al [52] and Coleman et al [15] identified following prerequisite safe practices. In addition, additional state-
machine maintenance and operating mobile equipment as part of of-the-art controls including devices that sense human presence
the top activities resulting in both fatal and nonfatal injuries. causing workers to stay in a safe buffer zone while working in
However, unlike other works, this study observed that directing dangerous locations should be explored [20]. The injuries domi-
mobile equipment (13%), moving equipment (10%), inspection nated during the morning shift and mostly after 4 hours but less
(10%), and connecting equipment/machinery/hoses (10%) resulted than 8 hours into the day’s work. The relationship between risk of
in fatal injuries more than any other activities. These indicate that injury and shift schedule and overtime has been reported in other
162 Saf Health Work 2019;10:151e165

Table 3
Summary of suggested areas deserving attention and focus for prioritizing research and prevention efforts

Topic Class/Category Suggested areas meriting consideration for prioritizing research and prevention efforts
Characteristics Age and mining experience 37e38 y was the most affected age group; it was the modal group for operators in surface and
of the victim underground locations as well as for fatal and nonfatal injuries. Old miners (>57 y) were involved
in fatal injuries than in nonfatal injuries. Less-experienced workers were equally involved in fatal
injuries than in nonfatal injuries
Employment type More contractors (29.8%) were involved in fatal accidents than operators (10.3%). There were more
injured young contractors (<38 y ¼ 42.6%) than young operators (<38 y ¼ 23.3%). Similarly, there
were more old injured contractors (>47 y ¼ 11%) than old injured operators (>47 y ¼ 8.4%).
Therefore, there should be a focus on the safety performance of contractors.
Job title Mechanics/repairmen, truck operators, drillers, supervisors, and laborers were the most affected ones
at both surface and underground locations and remain a priority for research and intervention
efforts. However, efforts should be specific to the mine type as there were major differences
between surface and underground mines. For instance, surface mechanics were affected more than
those at underground; and underground drillers were also affected more than those at surface
mines. Dump controllers also remain a priority occupation as 13% of the fatalities affect them, same
as supervisors and drillers.
Characteristics of the Activity being performed Machine maintenance, operating mobile equipment, clean up/clearing, drilling, and lifting/lowering
task being performed by hand accounted for 51% of all injuries. Directing moving equipment (13%), moving equipment
(10%), inspection (10%), and connecting equipment/machinery/hoses (18%) resulted in more
fatalities than any other activity. These activities remain a priority and warrant further
investigation and intervention strategies. Job titles related to these activities were also identified as
a priority area, which further supports the need to focus on these activities.
Location of activity 54.1% of all the injuries occurred at the stope mining area, processing/treatment plant, haul road, and
workshops. 70% of all fatal accident occurred at five location: preparation/treatment plant (20%),
stope mining area (16.7%), shaft area (13.3%), breaking/crushing station (10%), and open-cut pit
(10%). It may be important to focus on these areas as they are the dangerous locations within a
mining environment.
Shift start time and hours Injuries occurred more in the morning shift (75.7%) and mostly occurred after 4 hours but less than 8
into work hours of work. Most (43.6%) of the injuries occurred within the second 4 hours of work. This trend is
similar for both fatal (50%) and nonfatal (45.3%) injuries. Over 75% of both fatal and nonfatal injuries
occurred during the morning shift. Thus, the morning shift and second 4 hours into work merit
consideration for prioritizing further investigation and improvement efforts.
Equipment Involvement of equipment Mining equipment was associated with over 85% of the injuries with respect to both mine type and
and equipment types injury severity. Thus, the involvement of equipment/machinery should receive special attention.
Specific equipment that deserves focus is mobile equipment, component/part, and nonpowered
hand tools. Specific mobile equipment is haul truck, drill rigs, cranes, dozers, and excavators. Haul
trucks and portable rock drill/borers were involved in fatalities more than any other equipment
type.
Characteristics Severity of injury There were more underground fatalities (18%) than surface fatalities (13%). In addition, there were
of the injury more severe (disability) nonfatal injuries in surface (20%) than in underground (8%) locations.
Therefore, underground fatalities and surface severe injuries were identified as priority areas.
Time of accident 10e11 AM, 11 AMe12 PM, and 1e2 PM, which are 4e5 hours, 5e6 hours, and 7e8 hours into the
morning shift (assuming morning shift begins at 6 AM), respectively, were the peak times for both
underground and surface locations. In addition, 9e10 PM remained a peak time at underground
mines. 26.7% and 30% of all fatalities occurred at 10 AMe12 PM and 1e2 PM respectively, further
indicating that those periods deserve further attention.
Day of the week of accident Fri (21.3%) and Thu (18.3%) were identified as the peak days for injuries. Similarly, most fatal
accidents occurred on those days, 26.6% for Fri and 23.3% for Thu. A further investigation of those
days may yield important results to ensure improvements.
Accident type Machinery (19.8%), power haulage (15.8%), hand tools (15.3%), slip/fall of the person (14.4%), and
handling materials (11.4%) accounted for 76.7% of all injuries. The top accident types for the fatal
injuries were machinery (36.7%), power haulage (16.7%), and slip/fall of person (16.7%), indicating
that these accident types remain priorities.
Affected body part The hand/finger/thumb (26.2%), multiple parts (18.8%), and the lower leg (9.9%) were the most
affected body parts. 60% of all fatalities affected multiple body parts, whereas 23.3% affected the
neck and head. Hand injuries affected mechanics/repairmen (32.1%) and truck operators (22.6%)
more than any other occupation.
Injury type 75% of the injuries were laceration (33.7%), multiple injuries (16.8%), fractures (14.9%), and contusion
(9.9%). All permanent disability injuries (14) were traumatic amputation, affecting either the hand/
finger/thumb (13) or the foot/toe (1).
Mechanism of injury Being struck by a metallic object (20.9%), being hit by moving object (11.4%), fall/slip/trip on the same
level (8.5%), and motion of a moving vehicle (7%) accounted for 47.8% of all injuries. Fall/slip/trip
from height (20%), being struck by a metallic object (16.7%), being struck by rock (13.3%), vehicle
rollover (13.3%), and being hit by a moving vehicle (3.3%) caused 76.5% of the fatalities. This shows
that falls/slips/trips and workerevehicle interaction merit consideration for prioritizing research
and prevention efforts

studies [52,55e58]. Possible intervention to address this increased 5.3. The involvement of mining equipment
risk of injury with long working hours includes work hour re-
striction; an introduction of short breaks; development and The involvement of equipment/machinery in mining fatalities
implementation of a comprehensive fatigue management plan; and and serious injuries have been studied in depth and have long been
increased awareness of the worker’s right to voice safety concern identified as a priority area deserving further research
related to shift schedule, working hours, and fatigue. [1,10,15,20,22,24,41,42,52]. Mining equipment was involved in 86%
 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 163

of all injuries and 90% of all fatalities. These figures exceed those most frequent accident types. Additional safety interventions
reported in other studies, such as in Australia (46% of all under- should be directed toward these areas, particularly machinery and
ground injuries from 2005 to 2008) [59] and the USA (69% of un- power haulage. Most of the nonfatal injuries affected the hand/
derground fatality from 1995 to 2007 and 37%e88% of total mining finger/thumb (26.2%) and the lower leg (9.9%); however, the fatal
fatalities from 1995 to 2005) [10,60]. Despite these difference be- injuries frequently affected multiple body parts (60%) and the head/
tween the figures reported here and those of other works, they give neck (20%). The hand injuries mostly affected mechanics/repairmen
an indication that mining equipment remains a priority causal (32.2%) and truck operators (22.6%). By combining classes/cate-
factor of accidents, and the situation is even more conspicuous and gories in the injury analysis, it was possible to reveal more specific
severe in Ghana. This emphasizes the need to address equipment issues to address them. Many human factors models emphasize
safety issues, as highlighted in other countries [20,22,61e63]. that it is only the combination of a number of factors that
Specific priority equipment subcategories identified include mobile impact real-world performance. For instance, by combining the
equipment (36.3% of injuries), component/part (14.8% of injuries), affected body part with the occupation of the victims, it was
and nonpowered/powered hand tools (10.7% of injuries). Among possible to identify that hand injuries frequently affect mechanics
the mobile equipment subcategory, haul trucks (14.2% of all injuries and truck operators. Similarly, by combining the mine type with the
and 20% of all fatalities), drill rigs (6.3% of all injuries), cranes (2.8% injury severity, it was observed that there were more fatalities in
of all injuries), dozers (2.3% of all injuries), and excavators (2.3% of underground locations (18.3%) than at surface (13.4%) locations. In
all injuries) had the greatest proportion of injuries. Other studies addition, underground injuries had a minimum of days away from
agree with these findings. Analysis of equipment-related fatalities work, whereas injuries from surface mines resulted in a minimum
in the US by Kecojevic et al [10] identified haul trucks to be involved of restricted work activity only, indicating that underground in-
in 22.3% of the fatalities. Ruff et al [20] identified haul trucks and juries tend to be more severe than those that happen on surface
loaders as the most frequently involved in injuries involving mobile locations. There should be a focus on the safety of workers in un-
equipment. The need to focus attention on these equipment sub- derground mines, particularly with the recent transition of several
categories has been emphasized in the literature. By focusing on surface mines to underground operations [67e69] and the
them, in-depth knowledge can be obtained so that intervention can commencement of new underground operations [70] in Ghana.
be specific. For instance, Md-Nor et al [64] assessed the risk of
loader- and dozer-related fatalities and found that the two ma- 6. Conclusion
chines had different hazards. Failure in following maintenance
procedure and failure of machine components was the most Although significant improvement has been achieved in safety
frequent hazard for loaders, whereas failure to identify adverse in the mining industry, the frequency and severity of mining acci-
conditions was the most frequent hazard for dozers. By focusing on dents are still unacceptable. In Ghana, injury statistics of the in-
machine-related injuries, Ruff et al [20] identified that most of the dustry far exceeds that of major mining countries such as Australia
injuries occurred during either the operation of the machine or its and the USA. Unfortunately, little is known about the safety per-
maintenance/repairs. They further identified issues more specific to formance of Ghana’s mining industry although the country has long
particular equipment, such as loss of control/visibility issues of been a major producer of gold, ranking 10 globally. This research
haulage equipment in motion and safeguard of moving part of was, therefore, to provide a broad overview of the safety statistics of
stationary equipment. Unlike in other countries, there has been no the industry by analyzing 202 injury reports using descriptive
focus on machine-related injuries in Ghanaian mines, and this is statistics. Results of the analysis indicate that the involvement of
the first attempt that gives a broad overview. To ensure improve- mining equipment, certain characteristics of the task being per-
ment in machine safety, more research is therefore required. There formed, the accident victim, and the injury itself deserve attention
is the need to understand the specific hazards associated with these for prioritizing research and prevention efforts. It was identified
machines, the position of the injured person on the equipment, that at both surface and underground mines as well as for both fatal
specific tasks that were being performed, and the injury mecha- and nonfatal injuries, the most frequently affected age group was of
nism among others to support the development of intervention those between 37 and 48 years. Less-experienced workers were
strategies. Although the works of Zhang and Kecojevic [48], Zhang involved in fatal accidents more than nonfatal accidents. In addi-
et al [65], Kecojevic and Md Nor [66], and Md-Nor et al [64] offer tion, contractors had an increased risk of being fatally injured than
some insight into improving the safety of mining equipment, operators, supporting the need to focus on contractor safety
studies specific to the Ghanaian mining industry will be invaluable particularly with the recent rise in contract mining. Furthermore,
to improving the safety performance of the industry. certain occupations were most often involved in injuries than
others. The significant occupations included mechanics/repairmen,
5.4. The injury truck operators, drillers, and laborers.
With regard to the task being performed at the time of the
Consistent with other research [20,63], being struck by a injury, machine maintenance, operating and directing mobile
metallic object (20.9% of all injuries), being hit by a moving object equipment, drilling, and lifting/lowering by hand resulted in more
(11.4% of all injuries), fall/slip/trip on the same level (8.5% of all fatal and nonfatal accidents than other activities. In addition, in-
injuries), and motion of a moving vehicle (7% of all injuries) were juries dominated at the shaft and stope mining areas for under-
the major mechanisms through which the injuries occurred. The ground mines and at the treatment plant, crushing station, haul
major mechanisms through which the fatal injuries occurred were road, and workshops at surface locations. About 75% of the injuries
fall/slip/trip from height (20%), being struck by a metallic object occurred during the morning shift. There were more underground
(16.7%), being struck a by rock (13.3%), vehicle rollover (13.3%), and injuries during night shifts than surface injuries. The study showed
being hit by a moving object (13.3%). This suggests that falls/slips/ that 86% of the injuries and 90% of fatalities had the involvement of
trips and workerevehicle interaction merit consideration for mining equipment/machinery, the equipment that dominated was
prioritizing research and prevention efforts. Similarly, machinery mobile mining equipment (haul trucks, drill rigs, excavators, and
(19.8% of all injuries and 36.7% of fatalities), power haulage (15.8% dozers), hand tools (powered/nonpowered) and components/parts.
of all injuries and 16.7% of fatalities), hand tools (15.3%), and slip/fall For the mobile equipment, and most fatal injuries occurred during
of the person (14.4% of all injuries and 16.7% of fatalities) were the the motion of the machines. These emphasize the need to focus on
164 Saf Health Work 2019;10:151e165

equipment safety, and special attention must be paid to mobile [12] Brown M, Buehler M, Werna E. More than 2 million people die at work each
year. Here’s how to prevent it [Internet]. World Economic Forum. 2017 [cited
equipment and hand tools as they resulted in fatalities more than
2018 January 5]. Available from: https://www.weforum.org/agenda/2017/03/
any other equipment. workplace-death-health-safety-ilo-fluor/.
Finally, concerning the injury itself, underground fatalities far [13] Eurostat. Statistical analysis of socio-economic costs of accidents at work in
exceed surface fatalities, and 10 AMe12 PM and 1e2 PM (that is 4e8 the European Union. Luxembourg: Office for Official Publications of the Eu-
ropean Communities; 2004.
hours into work) were the peak injury periods at both underground [14] Leigh JP, Waehrer G, Miller TR, Keenan C. Costs of occupational injury and
and surface mines. Based on MSHA classification of accident types, illness across industries. Scand J Work Environ Health 2004;30(3):199e205.
machinery, power haulage, hand tools, slip/fall of the person, and [15] Coleman P, Brune J, Martini L. Characteristics of the top five most frequent
injuries in United States mining operations, 2003e2007. Trans Soc Min Metal
handling materials were the leading accident types. The leading Explor 2010;326:61e70.
injury mechanisms were being struck by a metallic object/rock, [16] Coleman PJ, Kerkering JC. Measuring mining safety with injury statistics: lost
falls/slips/trips, motion of a moving vehicle, and vehicle rollover, workdays as indicators of risk. J Saf Res 2007;38(5):523e33.
[17] Hull BP, Leigh J, Driscoll TR, Mandryk J. Factors associated with occupational
with the hand/finger/thumb and head/neck being the most affected injury severity in the New South Wales underground coal mining industry. Saf
body part. This study presents a broad overview using single vari- Sci 1996;21(3):191e204.
able analysis, with only a few variable combinations. However, to [18] Maiti J, Bhattacherjee A. Evaluation of risk of occupational injuries among
underground coal mine workers through multinomial logit analysis. J Saf Res
reveal specific human factor issues and specific injury-related in- 1999;30(2):93e101.
formation, further research effort that uses the combination of [19] Mine Safety and Health Administration. Equipment safety and health infor-
multiple variables would be a significant contribution. In addition, mation [Internet]. USA: MSHA. 2006 [cited 2017 November 10]. Available
from: www.msha.gov.
to ensure safety improvement of the mines, significant resources
[20] Ruff T, Coleman P, Martini L. Machine-related injuries in the US mining in-
must be allocated toward prevention efforts that address the dustry and priorities for safety research. Int J Inj Control Saf Promot
behavior, knowledge, and competencies of the workers; the design 2011;18(1):11e20.
of the task being performed; the work environment; and the [21] Sanmiquel L, Freijo M, Edo J, Rossell JM. Analysis of work related accidents in
the Spanish mining sector from 1982-2006. J Saf Res 2010;41(1):1e7.
equipment and machinery being used. [22] Ural S, Demirkol S. Evaluation of occupational safety and health in surface
mines. Saf Sci 2008;46(6):1016e24.
[23] Safe Work Australia. Notifiable fatalities monthly report November 2015.
Conflicts of interest
Australia: Safe Work Australia; 2015.
[24] Dash AK, Bhattcharjee RM, Paul PS, Tikader M. Study and analysis of accidents
All authors have no conflicts of interest to declare. due to wheeled trackless transportation machinery in indian coal mines e
identification of gap in current investigation system. Proc Earth Planet Sci
2015;11:539e47.
Acknowledgments [25] Leger JP. Trends and causes of fatalities in South African mines. Saf Sci
1991;14(3):169e85.
[26] Sanmiquel L, Rossell JM, Vintró C. Study of Spanish mining accidents using
The assistance of the Inspectorate Division of the Mineral
data mining techniques. Saf Sci 2015;75:49e55.
Commission of Ghana and the mine sites whose time, resources, [27] Kleczek Z, Malec M. Classification of hazards in underground mines-set
and information were vital to the research are duly acknowledged. against a background of experience in Polish mining industry. Coal Int
The author is a recipient of an Australian Government Research 1999;247(4):141e3.
[28] Minerals and mining (health, safety and technical) regulations, 2012 (L.I.
Training Program scholarship and Centennial Scholarship at The 2182); 2012.
University of Queensland, Australia. Acknowledgments are also due [29] Amegbey N, Ndur S, Adjei R. Analysis of underground mining accidents at
to the two anonymous reviewers who provided critical feedback. AngloGold Ashanti Limited, Obuasi Mine. Ghana Min J 2008;10(1).
[30] Gyekye SA, Salminen S. Causal attributions of Ghanaian industrial workers for
accident occurrence. J Appl Soc Psychol 2004;34(11):2324e40.
Appendix A. Supplementary data [31] Aidoo SJ, Eshun PA. Time series model of occupational injuries analysis in
Ghanaian mines-a case study. Res J Environ Earth Sci 2012;4(2):162e5.
[32] Sutherland D. Occupational injuries in a gold mining company in Ghana. Afr
Supplementary data to this article can be found online at Newslett Occup Health and Saf 2011;21:8e10.
https://doi.org/10.1016/j.shaw.2018.09.001. [33] Glaser BG. The constant comparative method of qualitative analysis. Soc Probl
1965;12(4):436e45.
[34] Lewis-Beck M, Bryman A, Futing Liao T. Constant comparison. The SAGE
References Encyclopedia of Social Science Research Methods; 2004.
[35] Council NR. Toward safer underground coal mines. National Academies; 1982.
[1] Groves WA, Kecojevic VJ, Komljenovic D. Analysis of fatalities and injuries [36] Root N. Injuries at work are fewer among older employees. Monthly Lab Rev
involving mining equipment. J Saf Res 2007;38(4):461e70. 1981;104:30.
[2] Aryee BNA. Contribution of the minerals and mining sector to national [37] Bennett JD, Passmore DL. Multinomial logit analysis of injury severity in U.S.
development: Ghana’s experiment. GREAT Insights; 2012. underground bituminous coal mines, 1975e1982. Accid Anal Prev
[3] Aubynn T. Mining and sustainable development: the case of Ghana [Internet]. 1985;17(5):399e408.
IM4DC. 2013 [cited 2016 18 April]. Available from: http://im4dc.org/wp- [38] Amponsah-Tawiah K, Ntow MAO, Mensah J. Occupational health and safety
content/uploads/2013/07/Mining-and-Sustainable-Development-Ghana.pdf. management and turnover intention in the Ghanaian mining sector. Saf
[4] Ghana Chamber of Mines. Performance of the mining industry in Ghana 2014. Health Work 2016;7(1):12e7.
Accra, Ghana: Ghana Chamber of Mines; 2014. [39] Barry T. Industrial engineering study of hazards associated with underground
[5] International Council on Mining and Metals. Mining in Ghana - what future coal mine production. Department of the Interior, Bureau of Mines; 1971.
can we expect?. London, UK: ICMM; 2015. [40] Bennet J. Relationship between workplace and worker characteristics and
[6] Gyekye SA. Workers’ perceptions of workplace safety: an African perspective. severity of injuries in US underground bituminous coal mines, 1975-1981.
Int J Occup Saf Ergon 2006;12(1):31e42. Pennsylvania State University; 1982.
[7] Duke PL. Mining safety [Internet]. Health and Safety Middle East. 2016 [cited [41] Aldinger J, Keran C, editors. A review of accidents during surface mine mobile
2017 November 10]. Available from: https://www.hsmemagazine.com/article/ equipment operations. 25th Annual Institute on Mining, Health, Safety and
mining-safety-1251. Research; 1994.
[8] Lang O. The dangers of mining around the world [Internet]. Latin America & [42] Randolph RF, Boldt CM. Safety analysis of surface haulage accidents. Blacks-
Caribbean: BBC. 2010 [cited 2017 November 6]. Available from: http://www. burg, VA (United States): Virginia Polytechnic Inst. and State Univ., Dept. of
bbc.com/news/world-latin-america-11533349. Mining and Minerals Engineering; 1996.
[9] Safe Work Australia. Mining [Internet]; 2017 [cited 2017 May 18]. Available from: [43] Lind S. Types and sources of fatal and severe non-fatal accidents in industrial
https://www.safeworkaustralia.gov.au/industry_business/mining#overview. maintenance. Int J Ind Ergon 2008;38(11):927e33.
[10] Kecojevic V, Komljenovic D, Groves W, Radomsky M. An analysis of [44] Maiti J, Chatterjee S, Bangdiwala SI. Determinants of work injuries in mines -
equipment-related fatal accidents in U.S. mining operations: 1995e2005. Saf an application of structural equation modelling. Inj Control Saf Promot
Sci 2007;45(8):864e74. 2004;11(1):29e37.
[11] Andreoni D. Accident cost. Encyclopaedia of occupational health and safety. [45] Yong Jeong B. Comparisons of variables between fatal and nonfatal accidents
3rd ed. Geneva: International Labour Office; 1983. in manufacturing industry. Int J Ind Ergon 1999;23(5):565e72.
 E. Stemn / Ghanaian Mining Industry and Priority Areas for Research 165

[46] Kavuri A, Prakash B, Sabniveesu V, Nimbarte A, Kulathumani V, Kecojevic V. [58] Lockley SW, Barger LK, Ayas NT, Rothschild JM, Czeisler CA, Landrigan CP.
An adaptive, run-time navigation system for haul trucks in surface mines. Int J Effects of health care provider work hours and sleep deprivation on safety and
Min Reclam Environ 2017;31(5):364e74. performance. Joint Comm J Qual Patient Saf 2007;33(11):7e18.
[47] Sun E, Nieto A, Li Z, Kecojevic V. An integrated information technology [59] Burgess-Limerick R. Injuries associated with underground coal mining
assisted driving system to improve mine trucks-related safety. Saf Sci equipment in Australia. Ergon Open J 2011;4:62e73.
2010;48(10):1490e7. [60] Mine Safety and Health Administration. Equipment safety and health infor-
[48] Zhang M, Kecojevic V. Intervention strategies to eliminate truck-related fa- mation [Internet]. MSHA. 2008 [cited 2017 December 10]. Available from:
talities in surface coal mining in West Virginia. Int J Inj Control Saf Promot www.msha.gov.
2016;23(2):115e29. [61] Burgess-Limerick R, Steiner L. Injuries associated with continuous miners,
[49] Phiri JK. The development of statistical indices for the evaluation of hazards in shuttle cars, loadehauledump and personnel transport in New South Wales
longwall face operations. PA (USA): Pennsylvania State Univ., University Park; underground coal mines. Min Technol 2006;115(4):160e8.
1989. [62] Dhillon BS. Mining equipment safety: a review, analysis methods and
[50] Salminen S. Have young workers more injuries than older ones? An inter- improvement strategies. Int J Min Reclam Environ 2009;23(3):168e79.
national literature review. J Saf Res 2004;35(5):513e21. [63] Mitchell RJ, Driscoll TR, Harrison JE. Traumatic work-related fatalities
[51] Butani SJ. Relative risk analysis of injuries in coal mining by age and experi- involving mining in Australia. Saf Sci 1998;29(2):107e23.
ence at present company. J Occup Accid 1988;10(3):209e16. [64] Md-Nor Z, Kecojevic V, Komljenovic D, Groves W. Risk assessment for loader-
[52] Muzaffar S, Cummings K, Hobbs G, Allison P, Kreiss K. Factors associated with and dozer-related fatal incidents in U.S. mining. Int J Inj Control Saf Promot
fatal mining injuries among contractors and operators. J Occup Environ Med 2008;15(2):65e75.
2013;55(11):1337e44. [65] Zhang M, Kecojevic V, Komljenovic D. Investigation of haul truck-related fatal
[53] Dzawu MM. Gold fields Ghana to dismiss 1,500 staff as contractor hired accidents in surface mining using fault tree analysis. Saf Sci 2014;65:106e17.
[Internet]. Bloomberg. 2017 [cited 2018 January 20]. Available from: https:// [66] Kecojevic V, Md Nor Z. Hazard identification for equipment-related fatal in-
www.bloomberg.com/news/articles/2017-12-12/gold-fields-ghana-to- cidents in the U.S. underground coal mining. J Coal Sci Eng (China)
dismiss-1-500-staff-as-contractor-hired. 2009;15(1):1e6.
[54] Sabutey E. Goldfields justifies decision to operate contract mining in Tarkwa [67] Golden Star. Operations at a glance [Internet]; 2017 [cited 2018 January 10].
[Internet]. Accra, Ghana: JoyOnline. 2017 [cited 2018 January 18]. Available Available from: http://www.gsr.com/operations/wassa/wassa-main/default.
from: https://www.myjoyonline.com/business/2017/December-19th/ aspx.
goldfields-justifies-decision-to-operate-contract-mining-in-tarkwa.php. [68] International Mining. Great mine - Newmont Ghana [Internet]. International
[55] Barger LK, Cade BE, Ayas NT, Cronin JW, Rosner B, Speizer FE., Harvard Work Mining. 2011 [cited 2018 January 10]. Available from: http://www.infomine.
Hours, Health, and Safety Group. Extended work shifts and the risk of motor com/library/publications/docs/internationalmining/chadwick2011m.pdf.
vehicle crashes among interns. New Engl J Med 2005;352(2):125e34. [69] Ajarfor AS. Newmont invests US$200m in Subika underground mining
[56] Dembe AE, Erickson JB, Delbos RG, Banks SM. The impact of overtime and long [Internet]. Accra: Modern Ghana. 2017 [cited 2018 January 10]. Available
work hours on occupational injuries and illnesses: new evidence from the from: https://www.modernghana.com/news/816757/newmont-invests-
United States. Occup Environ Med 2005;62(9):588e97. us200m-in-subika-underground-mining.html.
[57] Dembe AE, Erickson JB, Delbos RG, Banks SM. Nonstandard shift schedules and [70] Resolute. Development - unlocking values [Internet]; 2016 [cited 2018
the risk of job-related injuries. Scand J Work Environ Health 2006:232e40. January 10]. Available from: https://www.rml.com.au/developments.html.